We can count ourselves lucky that the patent on "on" has expired. It's such a breath of fresh air being able to turn things on without having to pay royalties to some greedy corporation. We'll just have to leave them on until the "off" patent has run its course.
Board would have been laid out using black crepe tapes of various widths, which made it easy to make curved tracks by pressing down with a finger, guiding it where you want. Pads would have been stick-on symbols or possibly rub-down transfers ( Letraset) Done at either x1 to make direct artwork, or at 2x size and photographically reduced. Double sided boards were done using red and blue translucent tapes, and seperated photographically. For simple low-detail boards like this, freehand drawing with ink pens on drafting film was also sometimes used.
To bridge the gap in knowledge... The image of the circuit, once drawn or taped, was copied to acetate. The boards have 'photo resist', and black lines on the acetate translate after exposure to light to an acid-resistant film. The copper is then etched away in the rest of the area, leaving a "printed" circuit board. Then drill the holes for the components, populate the board, and solder. The heavy solder across all the traces shows that it was "wave soldered", where the board is moved across a machine which maintains a wave of molten solder. In higher-end PCBs at the time, and on most boards now, the standard green colour is "solder resist", a layer deposited after etching so that the solder only sticks to the exposed terminal pads.
@jrevillug I had to do hand-etched boards for a project in my highschool electronics class in the late 90's. This was still basically the process then, but we had print (from a computer) and transfer mask which made it easier. For the channel owner, it looks like the spaced the tube's as far apart as they did because putting down traces by hand with them any closer would be a PITA.
This was weirdly common in 70's hardware. I'm restoring a found electrostatic headstone amp and despite the original price it doesn't look much better than this inside.
Yeah, we're totally spoiled by the availability of complex ICs, dirt-cheap professional PCB fabrication, and ubiquitous CAD software. You can tell the PCB here was hand-drawn, hand-cut, and hand-assembled -- because nobody with a budget less than GE's could afford anything more sophisticated than that. This was only slightly after the majority of electronic devices were assembled with point-to-point wiring -- no PCB at all. I do not miss anything about this era.
@@orbatos Compared to a point-to-point wired TV or radio from just 10 years earlier, even this low-volume early 70s construction looks really high tech. It's definitely not as nice as HP or Tek in terms of build quality, but not much is.
Honestly they still look largely fine to me, even when displaying 88. The misplaced decimal point is a bit janky but nothing too unbearable by the usual standards of multi-segment displays. I’d guess that one thing which particularly helped to kill these was their power consumption. In Bob Johnstone’s _We Were Burning_ he mentions one reason why the Japanese carmakers were desperate for new illuminated-display technologies: the single incandescent bulb in their car radios was consuming more power than the entire rest of the radio system combined, and it had a much worse lifespan to boot. A display with multiple incandescent filaments for each single digit can’t have been too attractive to anyone building something which might need to run off a battery. See also the Apollo program messing around with promethium paint.
There weren't good alternatives at the time though. Nixies need high voltage, VFDs also need higher voltage than what's normally used in digital circuits and they also need AC for the filament (running the filament on DC especially with low anode voltage will make one side of the display darker than the other) .LEDs were tiny and expensive. As for the appearance - they look OK to me. I don't care if the numbers are a bit crooked or that the decimal point looks like an x. I'm not a perfectionist.
The gross inefficiency, for sure, over the utility. Because if you actually used it (used that instrument), you would know that's what 88 looks like. But I get the finer critique, too. It's funny to think of the original designers and fabricators and how they would've laughed heartily if you told them their products were going to get this level of scrutiny.
@@paulstubbs7678 I remember my first watch in 1077 or so had the res LED display which was unreadable in bright light. I lived in Phoenix, AZ at the time. The kind of place well known for being bright nearly every single day. So a wrist watch was a two-handed operation as the other hand had to shade the display.
@@Pentium100MHz I'm not particularly familiar with the VFDs of that time but I think it's interesting how car radio manufacturers largely adopted the technology at some point to where you couldn't find a radio that hadn't one.
RCA made another Numitron called the DTF104B which solves some of the problems these tubes exhibit. They are top viewing so reflections are much less of an issue. I don't think they made many of them, though. Probably never made it past pre-production. Edit: I put a short video of one on my channel for those who are interested in seeing it.
I have a bunch (about 20) of the top down Numitron tubes at home, new old stock that I found in my grandpas basement. I had planned to built a clock with them at some point, maybe I'll go ahead and do it
i have two clocks using DTF104B tubes they are quite difficult to find obtained 10 of them a few years ago made one 6 tube clock then was lucky enough to get 2 more recently and built another 6 tube clock
I have a few in flat glass lidded dual in line packages, should have no issues like these. Actually I like the round tube artefacts, ads a bit of character and depth to the display, unlike modern ones that have to be ultra flat.
Since this is a darkroom timer, and you like connections, when we stopped using track tapes and started using CAD for PCB design the resulting files (Gerber) would drive a photoplotter: essentially an enlarger/reducer where the head could translate in x and y. The files would select an aperture (like a trace width or component pad) from a wheel of shapes and either flash the bulb (for a pad/symbol) or draw with the bulb on (for traces) to expose the film. We had a darkroom these machines lived in and the usual monochrome chemicals. The resulting plots would then get couriered to the PCB manufacturing place where they'd be used as masks for the photoresist on the copper-clad boards or make silk screens for lettering and other layers. These days we still use Gerber files, but they get rasterised for plotting in something more like a laser printer. In defence of Numitrons, at the time these things were more "futuristic" than the alternatives, and despite the failings were "cool". You do quickly get used to their quirks and the decimal point when reading them.
Was looking to see if anyone else thought that. Putting a clouded filter on it might have literally covered up so many of the problems with this display technology.
@@j.donaldson2758 A filter like that would scatter the light - basically working like a projector screen. So either the filter would have be right in front of the filaments (which can't be done due to the round glass being in the way) or there would need to be lenses to make a rear-projection display (which would lose a lot of brightness).
Exactly, i was just about to say this, but i made sure i wasn't "stealing" the comment from someone else. But yeah, like nixel also says, it's a dark room AND it will be flooded with red light, so there probably would be no glare at all for this specific machine. Although of course, it is a problem for other devices, but just not this particular one.
I'm still not sure why I watch 26min videos on things I don't really care about... But your content is so relaxing and easy to watch that I now know more about dishwashers and turn signals then I ever needed to know :P
Thanks to these informative videos I now was my dishes with hotter water, some soap for the pre wash, dry by opening and letting the heat do it's magic. Cleaner dishes! Done faster!
I've always been fascinated by mechanisms, electronics etc. Even as a kid I liked taking apart things like old washing machines and to see how they worked - well for the part that I was able to (the screws were usually very tight and also quite often I had to e.g. ask my dad to explain stuff). Turns out I like also watch other people taking apart devices and figuring out how the work.
According to Wikipedia, the 7492 is a divide-by-12 counter. It actually has separate /2 and /6 sections, and the /6 makes perfect sense for turning the 60Hz line voltage into 10Hz. The 7490 is a /10 counter to give seconds.
74928 is a 4-decade counter and a multiplexed output 7-segment display driver. Fancy, right? Datasheet for the original part is hard to come by, but MM74C928 should be close enough to make sense of the circuit.
@SianaGearz I think video was saying 7492A rather that 74928. Some manufacturers slapped letters at the end to show a revision or variant. 1970 also seems too early for a 7400 series part with a high number of 74928. However, I could be wrong.
A 74LS92 is a divide by 12 counter. Usually the LS variant is the same as the standard TTL chip. Divide by 12 seems a bit odd in a device clocked by 60Hz - unless they used a full wave rectified waveform to give 120Hz.
@@jeffburrell7648 Of course they would. Ignoring one half-wave gives pretty choppy power, requiring fairly big capacitors to keep the DC from dropping out and bigger transformers to provide double the power during the non-ignored half-wave. We get away with this nowadays because we sprinkle the boards with capacitors and DC-DC converters everywhere.
Green can be used for certain types of monochromatic photo paper. One of the dark rooms I worked in used green safe light light since the eye is more sensitive to that wavelength over red.
Green safelights can also be used with most types of colour paper. When I did my degree, the college colour printing darkrooms were fitted with green safelights, although so dim as to be not especially useful. I preferred working in total darkness. The brightness of that display might be a problem if the device isn't carefully positioned. Also, the ease with which the filter comes off suggests to me that the device might have had optional red filters too.
Maybe that's why it was so easy to remove the display filter, maybe it also come with a red one when new. Left to the user to fit the right one for their photo lab.
Back in 1985 I put a new FM station on the air, complete with a studio automation system. The automation was an early microprocessor setup that used IIRC, ten Numitrons in the programming display. Each day, the commercials and other events were programmed in using this display terminal. Within weeks of taking delivery, the Numitron filaments began to burn out. That sucked, because the tubes were soldered in, and it was a bugger to get to the bottom side of the circuit board to unsolder them. What's more, soldering and unsoldering connections repeatedly on circuit boards is a destructive process. The operators learned to 'read past' the bad segments, but you can only do that for so long, before too many segments are blown. I looked into finding sockets for these tubes, but never found any. I also looked into replacing them with large 7-segment LEDs, which were beginning to become available, but that wasn't practical either, for various reasons, mostly economic. I moved on to another job, without ever resolving the issue. I often wonder what the next engineer did.
Now every time I'll get passed some work from a previous guy I will think of you. Sometimes people really tried to do it, but things kinda suck every once in a while
The weird devices are "snubber networks", which are an AC film capacitor with a series resistor. Used across switch contacts or thyristors to reduce the inductive spike when loads switch off.
@@Vaionko I had a strong "Replace those now!" reaction as soon as I saw them. LOL Messy stinkers! (Thankfully, they are easily replaced these days, so no biggie.)
Outputs will probably be switched by triacs. Big caps are snubbers for noise/spike suppression- internal 100R resistor is to reduce surge current when the switch closes
Bet you could find this project if you went through all the magazines we had back in the day! I mean a dark room timer is so 'Hobby' it's untrue. Practical electronics, ETI or everyday electronics....cheers.
I think you inadvertantly found the perfect application of numitron displays by accident. That green filter cover i will bet anything it was supplied with either 2 or 3 of them. Green for developing panchromatic materials, you can use a specific shade of green partway through developing if you need to inspect the development. They could have had red and amber covers red for orthochromatif film and amber will generally be fine for darkroom printing paper.
I agree, but they were clearly inferior to other displays that were contemporary or shortly to follow, like VFD and Panaplex. I like the idea of sticking them on a clock because they look unique, but if it were 1972 and I was choosing a display for my device? I'd probably use almost any other option.
those "Strange components" with 0.22uf cap + 100 ohm resistor is called Snubber Network. its a capacitor in series with resistor, they usually connect to relay contacts or switches to suppress arcing
I've never seen both parts in the same case. Interesting. With triacs a snubber is often needed with inductive loads or they'll fail to turn off. Modern triacs are much better.
Those capacitors with integrated resistors are snubbers for the (probably) Triacs. Without snubbers, certain loads can cause the Triacs to trigger (turn on) by themselves. They also look like they may be the type of capacitor that likes to soak up moisture over years of sitting on a shelf, ready to let the smoke out once they're put back into service.
The one Alec kept pointing to already has a crack. If it hasn't already blown in the past (perhaps explaining the smell he mentioned when opening the device), its demise is imminent.
personally I think the 8s look fine and the decimal makes sense in my head, works fine, maybe I'm just used to ugly fonts and weird displays. the glare on the filter bugs me a bit but I can get over it.
He is such a whiny soyboy about it "the glare from my studio lights is why this is actually a bad thing" The product in question being designed for use in a room with no lights at all...
I was trained in electronics in the late 1970s, and printed circuit boards were designed by sticking tape onto a clear plastic sheet. Then that was used photographically to effectively print a mask onto the plain copper board, which was then etched to replicate the pattern as a printed circuit, as only the unmasked areas were etched away by the acid solution. Even in those days, rudimentary digital design methods were beginning to be developed, so I got to use both the tape method and the C.A.D. (computer-aided design) method. (The burn on your filter cap there would have been caused by a soldering iron when the leads were soldered to the nearby tag.)
I used both methods. However the Racal Redac CAD system I spent many hours using at work cost more than my house. I don't know the price of the Daisy systems I used in the early 1990s but definitely remember the cup of coffee icon that would appear while it did anything complex.
Numitrons were commonly used in applications where the display might be exposed to direct sunlight, ensuring readability. Gas pumps and aircraft cockpits. Gilbarco gas pumps used them for about 20 years--they mostly started going away early 00s. Davtron made a clock/timer that was (and often still is) commonly found in just about every well-equipped smaller aircraft and biz jet. It might not actually be a RCA Numitron-brand display in these, though it is the same thing. Ebay always has dozens of them available, along with Numitron temp gauges. Tag Heuer made a sports Chronometer that is basically the exact same box as your darkroom timer, but has 3 or 4 Numitrons and a bright red painted case. I have 3 of them somewhere that I should dig out. Likely the same manufacturer as the darkroom timer. I also have a benchtop counter made by a company called "Holograf" called a "hi-gage display-counter" that has 6 DR2010 Numitrons and 6-position numeric thumbwheels to load a preset (just like your timer), so maybe plug these ideas into your ebay searches, for those on the hunt.
Yes, people forget (or weren't alive to remember) just how dim early LED displays were. There were several competing types, one had very tiny segments made of the LED dies themselves with a plastic "bubble" magnifier over each digit. Even with the magnifier they were tiny. Then there were LED displays with the segments made out of rows of individual LED dies which were looked at directly, those looked nice but due to the number of dies they were both very expensive and used a lot of power. Then there was the modern style of display with one or more dies illuminating a diffuser, those were dim and not usable in sunlight. It's only in the last 20-30 years that blindingly bright, high efficiency LEDs have been a thing.
From my memory Gilbarco replaced the Numitrons with those electromechanical seven segment displays similar to the station clock Alec showed. LCDs in gas pumps are a 21st century thing.
Another one of many interesting videos, Alex. I love it when you "go retro" and dissect old pinball and juke box guts. Electromechanical telephone line routers are REALLY amazing. I actually repaired photocopy machines in the early 70s that used these filament 7-segment displays. Problem was, when one filament burned out, the whole tube became useless, and they were NOT cheap. Large digit LED displays were still a few years off. Our chunky hand-held calculators had LED digits so tiny that they had magnifying lenses over them. I love seeing this "legacy" electronic stuff. I grew up learning vacuum tube circuitry, and discreet transistors were just coming into vogue. Amazing how yesterday's massive computing machines now fit into a pocket and do 100 times the computing.
There actually is a reason behind the green lights! Some ORWO Film stocks (ORWO DN21 and DP31) for example specify that you should use a dark green filter. I could imagine that the idea behind that is that green gives you better visual resolution than red, which can help with telling if something really is in focus or not
The fact that the filter was easily removable from the front suggests that it may have been an exchangeable part with red and green filters in the box to suit different film stocks.
@@mjdapp Which is probably precisely why numitrons were chosen as the display device. The white light of the filaments could be filtered to result in either red or green digits. Had they used LEDs this would not have been possible, red and green LEDs both emit their respective colors directly, and one can't be filtered into the other.
Sodium meta bisulfite. It’s the same stuff as IronOut. It’s also used as a bleaching agent, often in paper manufacturing (given that chlorine will make dioxins doing the same thing and peroxide probably costs more).
The weird capacitors with a series resistor is known as an RC snubbers for the AC power on the outlets to control the transient spikes. The 74192 is a BCD counter (up/down), the 74193's are for Hex(binary) 4-bit counters, the larger filter capacitor was singed with a soldering iron when soldering the wires on that terminal post. The thing you called a transistor on the back of the case is likely a power regulator in a TO3 case and they generate a lot of heat. I would guess it is a 7805 *(edited) or LM309K. The board was laid out on plastic (Acetate) and then was photographed as a negative and then the PCB would be dipped in KPR (Kodak Photo Resist) and exposed to UV light. Then the unexposed resist would be washed off and then etched. I could explain this in greater detail because I got my electronics degree in 1987 and they were still using this process at the college... and I had to make similar PCB's! Later on the CAD PCB layout software like PADS, Eagle, and KiCAD are available (affordable) for the home hobbyist but the big boy on the block is Altium... and prepare for sticker shock.
@@nezbrun872 Oh right... if you don't work on power supplies, especially linear ones, you forget. As soon as you said LM309K BINGO! I recall that now. So I think I have some LM309K's in my regulator junk box along with a lot of 78xx and 79xx regulators... Really? Those old parts are $60 bucks a pop? WOW! I could be sitting on a fortune! (Or not, I'm pretty sure I don't have a lot of them)
One main point is that the 74192 counters are loadable. So the value set by the thumbwheels are loaded into the counters and as long as there is no counting the value is fed directly to the numitron drivers. When counting is activated the load is inhibited and the numitrons display the count down value. The flip-flop is used to stop the count when zero is reached (and also to start the count) by using the flip-flop set/clear inputs .
@@ingehansen9069 Yes, that's correct. And just thinking about what I've seen in the video, I think I could design and build a similar unit with stuff that I already have in stock. I've got some 7447 7-segment display drivers for LEDs. Not that I need to, but just imagining I could. BTW: Did you ever get Forrest Mims, III Engineer's notebook at Radio Shack and see all those nifty projects? I build several of them. So there's one project in there that uses a CMOS chip that has 3 decade counters and it multiplexed 3 output (BCD) so then you needed a decoder chip, but it could replace a lot of that TTL logic. In his design you can cascade them so there was a 6 digit display, and with a timebase of 1 second and a gate you could make a frequency counter. I'm going back at least 40 years, or more... Fun Times!
Altium is for hobbyist as well with freeware CircuitMaker, and small business with CircuitStudio. But i don't see myself giving up on Kicad, it's gotten sort of... shockingly nice i would say? Though my frame of reference is Eagle 7, may it burn in hell.
The first PCB I have ever made was using an etch-resist pen to draw the traces before etching the rest. Another similar method which achieves more consistent trace width is taping the board, which allows you to put down longer perfectly straight and more tightly packed lines than what can be reliably drawn by hand.
Look in somewhere like Digikey and you can still buy them. About $10 each so aimed at professional use. We just put some on some bespoke test equipment.
Early-ish PCB design was done with 'letraset' transfer rub down component footprints. And a roll of track ribbon tape, to make a positive artwork. Then photographically reduced to give a positive print. That was used as a photo mask for UV sensitised resist on the blank PCB. It was common to solder flood the tracks, as there was no solder resist to protect the tracks from tarnishing. Sometimes a silver plating bath was used instead. The through holes were often copper studs that were soldered through.
Our eyes have their greatest sensitivity around 550 nm, which is green. For color photography dark rooms, you can use very, very dim green lights. They light up the room just enough so you can see outlines, paper boxes, the easel and so on, once your eyes were reasonably dark adapted. But not bright enough to haze color paper in the maybe few tens of seconds they were out. That's what we used back in my day (90s). The total light output of those Numitrons is likely very low, so at a bit of distance, they will appear as a very diffuse, very dim green, just like those green lamps we used in the darkroom.
The Items on the back panel next to the Triacs are called 'Snubber networks' they help manage switching noise and over voltage conditions. This helps protect the Triacs, and prevent them from latching on due to spurios noise.
The traces on the top of the board end in plate through holes, and big blobs of solder were use to make the connection from the top to the bottom. The "switches" used for the outlet are triacs and the cap/resistors are snubber devices to reduce any switching spikes from the triacs. The decade counters were in parallel load mode from the thumbwheel switches until you started the counter. Very simple logic compared to today's technology, but real interesting to see and brought back many memories for me! Thanks for posting!
@@GodmanchesterGoblin Indeed, those holes are NOT plated through. That's why there is a big solder blob. There may be an eyelet or a short post in the hole that is then soldered on both sides to complete the circuit.
@@Brian_Of_Melbourne Yes - and they can have a different coefficient of expansion with temperature compared to the pcb material, potentially resulting in hard to see cracks in the traces. Been there, repaired that, etc, back in the late 70s.
That board is a very interesting historical piece. It has a combination of point to point and PCB elements all on a hand laid board. That big cap is quite likely dead. I'd be careful about using that without checking the caps for leakage, it can quite easily destroy the most expensive part in there: the transformer. In fact that transformer already looks like it's been quite hot. The resistors are carbon comp and thus have probably also drifted (carbon comp usually fails with higher not lower values). The silver cans are likely germanium transistors, but could also be diodes. Edit: nevermind there is a full wave bridge rectifier in blue diodes in the back. Doh!
At 2:45, that display seemed crystal clear. Since it's meant for a dark room, perhaps the readability problem is just glare from operating outside of the design spec?
I made DIY PCBs in the 1980s at school; we would draw onto the unetched PCBs with marker pens, with rub-on transfers for component footprints, holes etc. Then you'd expose with UV, etch in a tank of some vile orange stuff, drill all the holes with a mini drill press, and you were done. It was actually very easy and effective. You could even do double-sided boards. Our teacher would always get us to minimise the amount of exposed copper which needed to be etched away as that used up the vile orange fluid, so we'd usually end up doing big ground fills by colouring them in with the marker pens or with masking tape or the like. I don't know if this board was home made. It could have been. There's no writing or silkscreen, there's no soldermask (whoever made it even forgot to tin some tracks), and some of the tracks are noticeably different in thickness from the others. OTOH if it was home made it was done very carefully. I'd expect hand-drawn curved tracks to be much wobblier than that. Also the corners are very nicely rounded.
In the late 1970's, we were using different width black tape to create the tracks, and this allowed you to make quite smooth curves etc. You would use the wider tape for the power/higher current routes, and/or you could over-solder the tracks to increase the CSA, hence load capacity. The rest was done as you say above.
There's a fairly professional-looking logo on the front, so I don't think it's a DIY project. It was probably made by a small company that didn't make very many of them and did a lot of the design and assembly using manual methods.
I did something similar, built a phone line noise filter out of discreet components, designed my board, markered it, UV then disolved the excess copper in a warm bromide? solution.. drill press for the component and wiring holes, soldered it all up and mounted it in a kit box from Tandy.
@@PurpleAlienPlanet I recall a kid in the electronics club at school (in the 1970s) saying "You know Stainless Steel is supposed to be stainless - it isn't). He had tipped used ferric chloride down his Mums sink.
Really puts the moon landings into perspective. I bet that power transistor on the back is a 2N3055, they were ubiquitous. The two pins poking through the holes reminds me of an amplifier I built in my teens in the early 70s; the "chassis" was sheet aluminium bent into an upside-down U shape, and the PSU was built around a zener and a 2N3055.
My first thought was a 7805, but given the date codes on the other parts this thing looks to be about a year too old for that, so yeah, more likely a 2N3055, those things were everywhere back in the day. Still have boxes full of them which are worth a small fortune today.
Bearing in mind that they had to integrate and test all the equipment in advance this is about a decade after the Apollo equipment was designed. having said that NASA had larger budgets than a small company making photography equipment.
4:36 You're complaining about reflections in a device that is literally designed to be used in a "dark room". There won't be any reflections there and the way that an enlarger is situated these won't be facing the light sensitive paper so they won't cause accidental exposure of the paper. This seems like a complaint about a problem which in practice is non-existent. This is a timer and those numbers are only there to parse the time as it ticks away, as the image is being exposed the numbers are counting away the time. If it works at a timer and with the rapid pace of development at the time, this product was still pretty remarkable and kinda great
So... someone doesn't get it. It's not abut the use in this particular device, but the issues with Numitrons in ANY device. this just happens to be the first thing that showed up that actually uses them. As for the color, the filter is swappable on purpose. You use green light on some types of papers instead of red. Given that a dark room is about the only place where numitron tubes many drawbacks don't actually matter, that is likely why this thing still exists some 50 years later where most other devices that used them found their way to the trash heap 45 years ago.
I agree, he's so whiny about things that aren't problems with the numitron. He complains about glare while he has it under studio lights, all tubes will have glare, even led displays and lcds will have glare, he complains about this (presumably kit built) product's quality as if it relates to the tube's quality, he complains that this particular product does not feature a divider between the tubes, which A. Not a problem with the tubes, a problem with the product, and B. You can literally add a piece of paper in the middle if it annoys you that much. Additionally, his eyesight must not be very good if he can't clearly read the numerals on the displays. He says the 8 looks like a blocky mess, perhaps he needs new glasses, it looks exactly how 7 segment displays always look.
I had an internship where I did printed circuit board (PCB) development using UV light. It's pretty fascinating. You start with a design on a transparency, and tape it down to this sandwich of UV-sensitive material / copper / plastic / copper / UV-sensitive material. After exposing it to the UV light, you'd put it in a solution that would remove the exposed bits. Then you used etching fluid to remove the copper in those spots. Then you used a solvent to remove the remaining UV-sensitive material leaving the copper traces underneath. Then you'd soak it in a tinning solution so the copper wouldn't oxidize. Then drill holes in the board as appropriate and solder your components in. I think you'd appreciate it given your fondness for photographic development.
If I remember rightly, the 9 without the bar is what you get from the commonly-used 7447 BCD-to-7-segment decoder chip. Some people used hacks to add the extra bar to it. This device seems to be using a different decoder-driver chip that I can't find on the internet.
Agreed. It's settled, the 6 with and 9 without are defacto the best way to do it. I wonder if some of it stems from handwriting. I'm not sure I've seen contemporary handwriting do the 9 with the tail. Only angled.
@@0Rookie0from my perspective as a European, basically everybody will do the tail on the numbers six AND nine in handwriting. Could the timer have been made for the European market?
Yeah the 7447 & 4511 didn’t have the bars but the neon DM8880 drivers did. Haha now that I program the microcontroller segment tables myself I could do the 6 and 9 differently. But I have preferred them without the bars.
The two silver components that are rectangular on the back panel are film capacitors probably for power safety, REFA branded ones tend to crack with age and will blow magic smoke when they go.
9:49 I actually think this out-of-focus blurry image looks pretty nice. The decimal point looks like a dot, too. Maybe that's why the back plate is gray. So that the reflection can make the segments look bolder. Apparently RCA forgot about the ambient light... Maybe try a frosty glass as cover plate? The sanded side can also fix the reflection issue too.
24:45 You essentially had thin black tape that you can put on transparent sheets to transfer it to a PCB with positive photo-resist. It was easier to lay curves than square connections (and you also wanted to avoid to steep angles anyway to prevent bubbles of air or etchant to collect there).
The 7492 is a 4-bit counter with two sections, a divide-by-2 and a divide-by-6. That's probably what they're using to get the 10Hz. You're probably right about the BCD encoders. Thumbwheels like that with BCD outputs are fairly common (or at least they used to be -- nobody wants to use physical controls any more!)
7492 seems to be divide by 12. sn74ls92 is, which is most likely a modern variant of the same logic. After rectifying ac 60hz with a bridge rectifier you get a 120hz which might be used here.
The SN means it's manufactured by Texas Instruments. LS stands for Low-power Schottky, which use less power and run faster than their original 1970's 74## counterparts. Functionally for most applications, LS, S, and standard TTL ICs are interchangeable.
Yep, that's about how I would have made it back then lol. Fun fact: those boards were designed using transparency and vinyl tape at 2 or 4 X then photographically reduced so you could do it with better resolution. I remember having layouts covering the entire kitchen table. When you brought them in to be reduced, you unrolled it then pressed the tape back down where it came up when rolling it up for transport lol! 😄
@@Hirnspatz basically, only it was about 8” tall. It plugged into the wall, you plugged the enlarger into it and then turned the dial to how long you wanted it to be on for. I’m pretty sure timer worked like an electric clock (not quartz but the constant motor speed type) and wasn’t spring powered the way a kitchen timer is, though. It survived decades of middle school students of varying ability and enthusiasm.
A dark room display might be the only use case those would work for. It's probably not bright enough to expose nearby film, and green filter I think was specifically chosen to reduce glare from amber lights in the dark room
The board looks like it has been "taped". You would use a mylar (or semi transparent paper) sheet and traces were done with sized rolls of black masking tape which you can stick onto that. Patterns for sockets could be done with dry transfer "rub on" symbols (similar to "lettraset"). Sometimes this was combined with black pens, similar to sharpies. This artwork then is used to photo expose the PCB.
I was wondering whether a polarising filter would help (in the correct orientation, obviously). Since the distracting reflections are largely from the vertical walls of the tubes, a pair of polarised sunglasses at 90 degrees to the normal orientation would be worth checking.
1:26 - The digits are simply TOO FAR APART resulting in *bad kerning* ! I believe they're that far apart due to the REALLY bad glare that would result if they were placed next to each other.
Hi! The burned spot on the main electrolytic capacitor isn't harmful, it was just not too carefully assembled. The pcb style is typical for the time (prior cad), but it was not made with the best of the designer skills. One point of concern is that pair of Rifa capacitors glued on the back plate. One of them is visibly cracked ( a typical mode of failure of this brand). This causes the moisture of the air enter into the capacitor and it leaks. They should be replaced by safety reasons. 🤗🇦🇷
@tookitogo Those caps (and other German brands of that time) were excellent , better than today's Chinese ones, but suffer the pass of time because the encapsulation technique. Too bad. 🤗🇦🇷
The big blob-trough contacts is a prototyping / low product count trick. You drill the trough contacts oversize (1.5 - 2.5 mm) with a copper ring around and put a big solder blob on both sides. This only works well with hand soldering, lead rich solder and wastes a lot of solder. I used to "collect" a bunch of through contacts at one place and used pin header to make the connection. But that is only useful for larger boards where you can route the connections easily.
Are you sure about this? Every darkroom timer i've used has had red numbers. Well, except a fully mechanical timer which was not illuminated at all. Not trying to be a smartass here btw you might be right. The ones i mean all used LEDs
That would only apply if they were passive segments under a red blocking filter, actively illuminated displays using the same colour are still going to be easily distinguished from black under monochromatic light
It’s more likely because it’s for colour paper, rather than monochrome. Colour usually has a sensitivity dip in the green which this should be close to.
It is possible I am assuming incorrectly. However, I think it would be worth examining this under the proper conditions of a dark room to truly see. And finding some red film shouldn't be difficult. Bonus if finding a variety to test with.
you draw with a special marker on a copper covered Fiberglas board (aka a blank circuit board) and then "develop" it in a acid bath that eats away all copper you did not draw on, if left in to long it will eat away the drawn lines to so timing is important. I made a lab power supply for my introduction to electronics course in college in 1995 using this method. Perfect for a one off diy build.
That's a home technique, this is commercial and is as others others have said hand laid out and then photographed (perhaps with reduction) to make the final transparency (maybe even a negative) to expose the PCBs with.
@@Brian_Of_Melbourne That board did not look professionally made. It is a teaching method as well as a "DIY" and small scale method, the photo/transparency method is more involved and not suitable for small scale DIY. As I said, it was used in a US college course as a teaching method/aid.
Circuit board traces were laid out with thin tape atop a transparent sheet, at scale. The component footprints would be pieces of transparency with the correct hole sizes and annular rings already on them, stuck onto the master transparency. Light would be beamed through it (or through copies) at light-sensitive etch resist, which would then be developed and etched. The package sticking out the back is a 5V regulator (they get hot), and the capacitor+resistor components are probably to act as noise filters for the sharp transients produced by the TRIACs. Full-wave rectified mains is 120Hz not 60Hz, so a divide-by-12 chip would make sense here.
The weird components on the back panel are X-class mains-rated film-type safety capacitors. If you intend to use the thing, it's prudent to replace those with modern X2 capacitors because some of them from that era are known to fail, and when they fail they catch fire. (X1 rated capacitors are for industrial 3-phase applications, X2 will suffice for stuff like this)
Correction: I missed the ohm rating. They’re snubbers. Same idea though, x class cap with a series resistor. Not terribly common in a single package anymore, so a replacement is tougher, but the resistor makes it less likely to spontaneously combust.
Years ago, you could buy circuit boards with the back completely covered with copper. You used a marker to draw all the tracks and connection pads over the copper, then you submerged the board in acid. The acid removed all the copper, except in the zones protected by the marker, living only the tracks. Then you used a drill to make the holes in the connection pads. That circuit board seemed to have been made following this procedure.
people used to make the masks by hand. not by hand for every board though, but using photo techniques. you can still buy raw boards to make your own pcb's tho of course. techniques differ, but back in university we'd use a laminating machine to transfer a laser print from paper onto the board.
That's how we used to solder and it was even a technique before my time but it's what you did if you didn't have much money but wanted to make something. I did some similar stuff when I was a kid in the 80s and then I decided that you could make cool jewelry out of solder too so I did that for a while. Then I studied other things and, the next time I needed a soldering gun, I found that my brother has lost all the tips I had for mine. Great video!
I don't understand the purpose of the rants, the display looks and works perfectly fine. in real life you'd get used to it in an hour and never think about it again. you'd be too busy working the device to do the actual job, not contemplate the shapes or decimal point :-)
The parts mounted next to the (presumably) triacs are surge suppressors (or snubber) to dampen transients from inductive loads like motors. Some are still made but can just be replaced with a discrete capacitor and resistor of the same values.
so my main clock is a segment clock and i honestly couldnt remember if the 6 and 9s had the tops and bottoms had to check and when you mentioned it and i realized i didnt know was just after the clock went from 16 to 17so thankfully i didnt have to wait long for 19
The signal can be from mains (or at least calibrated from mains) with the 7492. The 7492 is 3 counters in one - a /2, a /3/ and another /2 - letting you grab mains frequency and then divide that down into 30hz, then 10hz, with the last /2 ignored. The 7490 is a divide by 2 and divide by 5 in one package. 10hz output from the 7492 is fed into the 7490's divide by 10, getting you 1hz when on the "seconds" setting. The 7490 is bypassed in the ".1 second" setting, giving you the 10hz without change. The decimal activates when this is bypassed, denoting 10ths of a second. The selection of either 1hz or 10hz is fed into the the rightmost tube's 74192 counter. When the rightmost counter reaches 10, it instantly resets back to 0 and also decrements the left counter by 1, giving you a 2 digit BCD countdown of 99 to 0. The 74192 was selected because it has the additional feature of data input pins letting you "load in" any value (from the encoder). In the default state, anything on the encoder will push to the counters. When you press the operate button, one of the flip flops in the 7493 changes state, disabling the "load-in" from the encoder and enabling the 74192's to begin counting down. The flip flop continues holding the "countdown" state until zero is reached on both displays, causing it to immediately switch to pulling in the encoder values. The 74193's feed BCD output into the numitron tube drivers (however they happen to work) to display whatever the count is. The calibrate feature is pretty interesting to me, though. All of the exact required ICs are there to take 60hz and bring it to 10hz and 1hz but if its just grabbing mains, there's nothing to calibrate. The calibrate button might utilize the second flipflop to switch to an external signal using what looks to be a 1/4" jack on the side. Would love to hear anyone else's take on the calibrate feature.
I did put it back together
why
And there was much rejoicing?
that's a relief
Very glad to know that, actually.
Maybe you could send it to BigClive so that he could reverse engineer the circuit 😊
I like the implication that the concept of 'off' has a patent pending.
You know those people, they're always "on" it!
We can count ourselves lucky that the patent on "on" has expired. It's such a breath of fresh air being able to turn things on without having to pay royalties to some greedy corporation. We'll just have to leave them on until the "off" patent has run its course.
@@gcewing you know as well as i do that patent trolls are just as often moneygrubbing individuals.
Love to know what the patent application is
Ho ho, of course the patent pending is for the whole machine but a good joke anyway. 😊
18:51 Soldering iron melted the jacket on the large filter capacitor while the green wires was soldered onto the post.
Yep. A mark of quality construction....
Came to say the same - looks exactly like the side of a soldering iron tip.
Yep. I've left that same mark more than once.
Sign of the builder.
@@wobblysauce Used a real soldering iron, no wave or reflow junk here
Board would have been laid out using black crepe tapes of various widths, which made it easy to make curved tracks by pressing down with a finger, guiding it where you want. Pads would have been stick-on symbols or possibly rub-down transfers ( Letraset)
Done at either x1 to make direct artwork, or at 2x size and photographically reduced.
Double sided boards were done using red and blue translucent tapes, and seperated photographically. For simple low-detail boards like this, freehand drawing with ink pens on drafting film was also sometimes used.
I guess plated through hole wasn't a thing, or was prohibitively expensive! Presumably PCB rivets were used and soldered over.
@@nezbrun872 this exactly. Plated through cost more, through pins were cheaper
wouldn't surprise me if they etched the boards in house
To bridge the gap in knowledge...
The image of the circuit, once drawn or taped, was copied to acetate.
The boards have 'photo resist', and black lines on the acetate translate after exposure to light to an acid-resistant film.
The copper is then etched away in the rest of the area, leaving a "printed" circuit board.
Then drill the holes for the components, populate the board, and solder.
The heavy solder across all the traces shows that it was "wave soldered", where the board is moved across a machine which maintains a wave of molten solder.
In higher-end PCBs at the time, and on most boards now, the standard green colour is "solder resist", a layer deposited after etching so that the solder only sticks to the exposed terminal pads.
@jrevillug I had to do hand-etched boards for a project in my highschool electronics class in the late 90's. This was still basically the process then, but we had print (from a computer) and transfer mask which made it easier.
For the channel owner, it looks like the spaced the tube's as far apart as they did because putting down traces by hand with them any closer would be a PITA.
I like how the "actual product" is only marginally more professional looking (even on the outside) than your quick hackjob
This was weirdly common in 70's hardware. I'm restoring a found electrostatic headstone amp and despite the original price it doesn't look much better than this inside.
@@orbatos Agree - looks like a very typical '70s bit of electronics.
Yeah, we're totally spoiled by the availability of complex ICs, dirt-cheap professional PCB fabrication, and ubiquitous CAD software.
You can tell the PCB here was hand-drawn, hand-cut, and hand-assembled -- because nobody with a budget less than GE's could afford anything more sophisticated than that.
This was only slightly after the majority of electronic devices were assembled with point-to-point wiring -- no PCB at all.
I do not miss anything about this era.
@@orbatos Compared to a point-to-point wired TV or radio from just 10 years earlier, even this low-volume early 70s construction looks really high tech. It's definitely not as nice as HP or Tek in terms of build quality, but not much is.
that filter piece looks like an afterthought, even though it's pretty important for its job
Honestly they still look largely fine to me, even when displaying 88. The misplaced decimal point is a bit janky but nothing too unbearable by the usual standards of multi-segment displays.
I’d guess that one thing which particularly helped to kill these was their power consumption. In Bob Johnstone’s _We Were Burning_ he mentions one reason why the Japanese carmakers were desperate for new illuminated-display technologies: the single incandescent bulb in their car radios was consuming more power than the entire rest of the radio system combined, and it had a much worse lifespan to boot. A display with multiple incandescent filaments for each single digit can’t have been too attractive to anyone building something which might need to run off a battery. See also the Apollo program messing around with promethium paint.
There weren't good alternatives at the time though. Nixies need high voltage, VFDs also need higher voltage than what's normally used in digital circuits and they also need AC for the filament (running the filament on DC especially with low anode voltage will make one side of the display darker than the other) .LEDs were tiny and expensive.
As for the appearance - they look OK to me. I don't care if the numbers are a bit crooked or that the decimal point looks like an x. I'm not a perfectionist.
The gross inefficiency, for sure, over the utility.
Because if you actually used it (used that instrument), you would know that's what 88 looks like.
But I get the finer critique, too. It's funny to think of the original designers and fabricators and how they would've laughed heartily if you told them their products were going to get this level of scrutiny.
I heard the military liked these as they were kind of day light viewable, unlike LED's or the day that almost needed a darkened room.
@@paulstubbs7678 I remember my first watch in 1077 or so had the res LED display which was unreadable in bright light.
I lived in Phoenix, AZ at the time. The kind of place well known for being bright nearly every single day.
So a wrist watch was a two-handed operation as the other hand had to shade the display.
@@Pentium100MHz I'm not particularly familiar with the VFDs of that time but I think it's interesting how car radio manufacturers largely adopted the technology at some point to where you couldn't find a radio that hadn't one.
RCA made another Numitron called the DTF104B which solves some of the problems these tubes exhibit. They are top viewing so reflections are much less of an issue. I don't think they made many of them, though. Probably never made it past pre-production.
Edit: I put a short video of one on my channel for those who are interested in seeing it.
Im headed to check that out shortly, thanks! Seems you have some other stuff I need to check out. 😎
I have a bunch (about 20) of the top down Numitron tubes at home, new old stock that I found in my grandpas basement. I had planned to built a clock with them at some point, maybe I'll go ahead and do it
i have two clocks using DTF104B tubes they are quite difficult to find obtained 10 of them a few years ago made one 6 tube clock then was lucky enough to get 2 more recently and built another 6 tube clock
@@0106johnny there is a kit available if you have some of the DTF tubes from a fella called Richard scale here in the uk
I have a few in flat glass lidded dual in line packages, should have no issues like these.
Actually I like the round tube artefacts, ads a bit of character and depth to the display, unlike modern ones that have to be ultra flat.
Since this is a darkroom timer, and you like connections, when we stopped using track tapes and started using CAD for PCB design the resulting files (Gerber) would drive a photoplotter: essentially an enlarger/reducer where the head could translate in x and y. The files would select an aperture (like a trace width or component pad) from a wheel of shapes and either flash the bulb (for a pad/symbol) or draw with the bulb on (for traces) to expose the film. We had a darkroom these machines lived in and the usual monochrome chemicals. The resulting plots would then get couriered to the PCB manufacturing place where they'd be used as masks for the photoresist on the copper-clad boards or make silk screens for lettering and other layers.
These days we still use Gerber files, but they get rasterised for plotting in something more like a laser printer.
In defence of Numitrons, at the time these things were more "futuristic" than the alternatives, and despite the failings were "cool". You do quickly get used to their quirks and the decimal point when reading them.
I like how legible the digits are at 9:48 when the camera is out of focus. They almost look like regular 7-segment LEDs !
After reading your comment, I watched the video again without my glasses. Worked perfectly for me. 😉
Maybe the inventor needed, but never wore, glasses and thought it was brilliant.
Was looking to see if anyone else thought that. Putting a clouded filter on it might have literally covered up so many of the problems with this display technology.
@@j.donaldson2758 A filter like that would scatter the light - basically working like a projector screen. So either the filter would have be right in front of the filaments (which can't be done due to the round glass being in the way) or there would need to be lenses to make a rear-projection display (which would lose a lot of brightness).
The 7 segments were still legible regardless how much I unfocus my eyes ...
4:30 Fortunately, if it’s going to be in a dark room, room light glare is less of a problem.
Especially with red light through a green-pass filter.
Exactly, i was just about to say this, but i made sure i wasn't "stealing" the comment from someone else.
But yeah, like nixel also says, it's a dark room AND it will be flooded with red light, so there probably would be no glare at all for this specific machine.
Although of course, it is a problem for other devices, but just not this particular one.
Yes, also looked for this comment. The green filter was probably intentional, as it would cut the glare to near zero in a red-lit room.
I was about to make this comment, glad I looked for it first.
@@nixel1324 which is probably exactly why the filter wasn't red or amber!
I'm still not sure why I watch 26min videos on things I don't really care about... But your content is so relaxing and easy to watch that I now know more about dishwashers and turn signals then I ever needed to know :P
Same, it reminds me of when a friend says " let's take it apart to see how it works." But the friend was an electrician or something.
Thanks to these informative videos I now was my dishes with hotter water, some soap for the pre wash, dry by opening and letting the heat do it's magic.
Cleaner dishes!
Done faster!
@@jerodewert8334 Automagically! Through the magic of buying two of them!
I use some of his long videos as sleeping background noise😂
I've always been fascinated by mechanisms, electronics etc. Even as a kid I liked taking apart things like old washing machines and to see how they worked - well for the part that I was able to (the screws were usually very tight and also quite often I had to e.g. ask my dad to explain stuff). Turns out I like also watch other people taking apart devices and figuring out how the work.
According to Wikipedia, the 7492 is a divide-by-12 counter. It actually has separate /2 and /6 sections, and the /6 makes perfect sense for turning the 60Hz line voltage into 10Hz. The 7490 is a /10 counter to give seconds.
74928 is a 4-decade counter and a multiplexed output 7-segment display driver. Fancy, right? Datasheet for the original part is hard to come by, but MM74C928 should be close enough to make sense of the circuit.
@SianaGearz I think video was saying 7492A rather that 74928. Some manufacturers slapped letters at the end to show a revision or variant. 1970 also seems too early for a 7400 series part with a high number of 74928. However, I could be wrong.
@@ChrisDreher Ah right thank you, i brainfarted.
A 74LS92 is a divide by 12 counter. Usually the LS variant is the same as the standard TTL chip. Divide by 12 seems a bit odd in a device clocked by 60Hz - unless they used a full wave rectified waveform to give 120Hz.
@@jeffburrell7648 Of course they would. Ignoring one half-wave gives pretty choppy power, requiring fairly big capacitors to keep the DC from dropping out and bigger transformers to provide double the power during the non-ignored half-wave. We get away with this nowadays because we sprinkle the boards with capacitors and DC-DC converters everywhere.
Green can be used for certain types of monochromatic photo paper. One of the dark rooms I worked in used green safe light light since the eye is more sensitive to that wavelength over red.
Green safelights can also be used with most types of colour paper. When I did my degree, the college colour printing darkrooms were fitted with green safelights, although so dim as to be not especially useful. I preferred working in total darkness. The brightness of that display might be a problem if the device isn't carefully positioned. Also, the ease with which the filter comes off suggests to me that the device might have had optional red filters too.
Maybe that's why it was so easy to remove the display filter, maybe it also come with a red one when new. Left to the user to fit the right one for their photo lab.
I believe that infrared film used a green safelight.
Yep, we used green in one dark room I used dark for IR film and red-laser film (holography).
The green filter also means that there would be no glare from the red lights in a dark room
That glare seems like it would be really annoying in a well lit darkroom
😂
True.
at about 16:00 i'm hearing bigclive, "i will be back with the schematic, one moment please..."
👍😊
"Reverse engineering is complete..."
Back in 1985 I put a new FM station on the air, complete with a studio automation system. The automation was an early microprocessor setup that used IIRC, ten Numitrons in the programming display. Each day, the commercials and other events were programmed in using this display terminal. Within weeks of taking delivery, the Numitron filaments began to burn out. That sucked, because the tubes were soldered in, and it was a bugger to get to the bottom side of the circuit board to unsolder them. What's more, soldering and unsoldering connections repeatedly on circuit boards is a destructive process. The operators learned to 'read past' the bad segments, but you can only do that for so long, before too many segments are blown. I looked into finding sockets for these tubes, but never found any. I also looked into replacing them with large 7-segment LEDs, which were beginning to become available, but that wasn't practical either, for various reasons, mostly economic. I moved on to another job, without ever resolving the issue. I often wonder what the next engineer did.
Now every time I'll get passed some work from a previous guy I will think of you. Sometimes people really tried to do it, but things kinda suck every once in a while
That burn on the big capacitor looks for all the world like someone slipped with the soldering iron at some point and caught it.
Looks more like a dead and dried pillbug to me! :D
They didn't slip, they weren't paying enough attention to where the soldering iron tip was...until they smelled it burning. Been there!
Came here to say the same thing
The weird devices are "snubber networks", which are an AC film capacitor with a series resistor. Used across switch contacts or thyristors to reduce the inductive spike when loads switch off.
Also they are Rifa caps that have a very bad reputation for splitting open and releasing heaps of smoke and stink, so should be replaced NOW
I'm glad I checked the comments to see if someone else had already posted this, so I didn't repost it and look stupid. 😁
And look to be the same paper in epoxy construction as the infamous Rifa capacitors, bound to release their magic smoke after a while of use nowadays.
@@Vaionko I had a strong "Replace those now!" reaction as soon as I saw them. LOL Messy stinkers! (Thankfully, they are easily replaced these days, so no biggie.)
Someone must've thought they had a great idea to reduce BOM count. Guess they didn't really catch on though.
Outputs will probably be switched by triacs. Big caps are snubbers for noise/spike suppression- internal 100R resistor is to reduce surge current when the switch closes
Bet you could find this project if you went through all the magazines we had back in the day! I mean a dark room timer is so 'Hobby' it's untrue. Practical electronics, ETI or everyday electronics....cheers.
I think you inadvertantly found the perfect application of numitron displays by accident. That green filter cover i will bet anything it was supplied with either 2 or 3 of them. Green for developing panchromatic materials, you can use a specific shade of green partway through developing if you need to inspect the development. They could have had red and amber covers red for orthochromatif film and amber will generally be fine for darkroom printing paper.
I still think they're cute. And I like the little x for a decimal point, it seems like an interesting solution that you might get used to, with usage.
I agree, but they were clearly inferior to other displays that were contemporary or shortly to follow, like VFD and Panaplex. I like the idea of sticking them on a clock because they look unique, but if it were 1972 and I was choosing a display for my device? I'd probably use almost any other option.
those "Strange components" with 0.22uf cap + 100 ohm resistor is called Snubber Network. its a capacitor in series with resistor, they usually connect to relay contacts or switches to suppress arcing
I've never seen both parts in the same case. Interesting.
With triacs a snubber is often needed with inductive loads or they'll fail to turn off. Modern triacs are much better.
Now I want to build some modern numitrons from those LED filaments you get in light bulbs. I like projects with such a low quality bar.
Yup, can buy some loose ones cheap too
That sounds like a great project. Good luck with that.
mikeselectricstuff already did this.
Those capacitors with integrated resistors are snubbers for the (probably) Triacs. Without snubbers, certain loads can cause the Triacs to trigger (turn on) by themselves.
They also look like they may be the type of capacitor that likes to soak up moisture over years of sitting on a shelf, ready to let the smoke out once they're put back into service.
They look like the forever accursed Rifa caps to me and could blow if a fly farts near them.
@@natejgee yep, look like them to me, they ought to be replaced with poly dielectric types, RIFA used and still uses paper dielectric in them!
When I first heard someone talk about Rifa caps, I thought he called them reefer caps. Made sense to me, they're rolled and go up in smoke.
The one Alec kept pointing to already has a crack. If it hasn't already blown in the past (perhaps explaining the smell he mentioned when opening the device), its demise is imminent.
@@pr0crastinatr 😜
It's important to always power your enlarger in the Darkroom
plugging it in first tends to help
@@smalltime0 But first ensure that the receptacle is clean and ready for completing the circuit.
personally I think the 8s look fine and the decimal makes sense in my head, works fine, maybe I'm just used to ugly fonts and weird displays. the glare on the filter bugs me a bit but I can get over it.
He is such a whiny soyboy about it
"the glare from my studio lights is why this is actually a bad thing"
The product in question being designed for use in a room with no lights at all...
As an old radio nut, I once saw an ad that read, "don't be vague, buy Sprague!"
Hope this helps.
I was trained in electronics in the late 1970s, and printed circuit boards were designed by sticking tape onto a clear plastic sheet. Then that was used photographically to effectively print a mask onto the plain copper board, which was then etched to replicate the pattern as a printed circuit, as only the unmasked areas were etched away by the acid solution. Even in those days, rudimentary digital design methods were beginning to be developed, so I got to use both the tape method and the C.A.D. (computer-aided design) method. (The burn on your filter cap there would have been caused by a soldering iron when the leads were soldered to the nearby tag.)
Hence, Printed Circuit Board, as in a photographic print.
I used both methods. However the Racal Redac CAD system I spent many hours using at work cost more than my house. I don't know the price of the Daisy systems I used in the early 1990s but definitely remember the cup of coffee icon that would appear while it did anything complex.
Numitrons were commonly used in applications where the display might be exposed to direct sunlight, ensuring readability. Gas pumps and aircraft cockpits. Gilbarco gas pumps used them for about 20 years--they mostly started going away early 00s. Davtron made a clock/timer that was (and often still is) commonly found in just about every well-equipped smaller aircraft and biz jet. It might not actually be a RCA Numitron-brand display in these, though it is the same thing. Ebay always has dozens of them available, along with Numitron temp gauges.
Tag Heuer made a sports Chronometer that is basically the exact same box as your darkroom timer, but has 3 or 4 Numitrons and a bright red painted case. I have 3 of them somewhere that I should dig out. Likely the same manufacturer as the darkroom timer.
I also have a benchtop counter made by a company called "Holograf" called a "hi-gage display-counter" that has 6 DR2010 Numitrons and 6-position numeric thumbwheels to load a preset (just like your timer), so maybe plug these ideas into your ebay searches, for those on the hunt.
Yes, people forget (or weren't alive to remember) just how dim early LED displays were. There were several competing types, one had very tiny segments made of the LED dies themselves with a plastic "bubble" magnifier over each digit. Even with the magnifier they were tiny. Then there were LED displays with the segments made out of rows of individual LED dies which were looked at directly, those looked nice but due to the number of dies they were both very expensive and used a lot of power. Then there was the modern style of display with one or more dies illuminating a diffuser, those were dim and not usable in sunlight. It's only in the last 20-30 years that blindingly bright, high efficiency LEDs have been a thing.
From my memory Gilbarco replaced the Numitrons with those electromechanical seven segment displays similar to the station clock Alec showed. LCDs in gas pumps are a 21st century thing.
Another one of many interesting videos, Alex. I love it when you "go retro" and dissect old pinball and juke box guts. Electromechanical telephone line routers are REALLY amazing.
I actually repaired photocopy machines in the early 70s that used these filament 7-segment displays. Problem was, when one filament burned out, the whole tube became useless, and they were NOT cheap. Large digit LED displays were still a few years off. Our chunky hand-held calculators had LED digits so tiny that they had magnifying lenses over them. I love seeing this "legacy" electronic stuff. I grew up learning vacuum tube circuitry, and discreet transistors were just coming into vogue. Amazing how yesterday's massive computing machines now fit into a pocket and do 100 times the computing.
There actually is a reason behind the green lights! Some ORWO Film stocks (ORWO DN21 and DP31) for example specify that you should use a dark green filter.
I could imagine that the idea behind that is that green gives you better visual resolution than red, which can help with telling if something really is in focus or not
The fact that the filter was easily removable from the front suggests that it may have been an exchangeable part with red and green filters in the box to suit different film stocks.
@@mjdapp Which is probably precisely why numitrons were chosen as the display device. The white light of the filaments could be filtered to result in either red or green digits. Had they used LEDs this would not have been possible, red and green LEDs both emit their respective colors directly, and one can't be filtered into the other.
"This is a darkroom timer"
Anybody else suddenly smell fixer? Just me?
well now that you brought it up....sour right? It's only been 40 years
I used to taste it after spending hours inthere.
Now that you mention it...yeah, 40 years for me too.
Sodium meta bisulfite. It’s the same stuff as IronOut. It’s also used as a bleaching agent, often in paper manufacturing (given that chlorine will make dioxins doing the same thing and peroxide probably costs more).
That "burn" mark on the silicone cover could be from the assemblers soldering iron....
19:38 pure gold. Delivery was on point. Perfectly describes the happiness of the average electronics enthusiast in the 70s.
The weird capacitors with a series resistor is known as an RC snubbers for the AC power on the outlets to control the transient spikes. The 74192 is a BCD counter (up/down), the 74193's are for Hex(binary) 4-bit counters, the larger filter capacitor was singed with a soldering iron when soldering the wires on that terminal post. The thing you called a transistor on the back of the case is likely a power regulator in a TO3 case and they generate a lot of heat. I would guess it is a 7805 *(edited) or LM309K. The board was laid out on plastic (Acetate) and then was photographed as a negative and then the PCB would be dipped in KPR (Kodak Photo Resist) and exposed to UV light. Then the unexposed resist would be washed off and then etched. I could explain this in greater detail because I got my electronics degree in 1987 and they were still using this process at the college... and I had to make similar PCB's! Later on the CAD PCB layout software like PADS, Eagle, and KiCAD are available (affordable) for the home hobbyist but the big boy on the block is Altium... and prepare for sticker shock.
FWIW, LM309K was the rage in TO3 in the 70s for 5V 1A regulators. LM323K for 5V 3A. You can still buy them, $60 a piece!!!
@@nezbrun872 Oh right... if you don't work on power supplies, especially linear ones, you forget. As soon as you said LM309K BINGO! I recall that now. So I think I have some LM309K's in my regulator junk box along with a lot of 78xx and 79xx regulators... Really? Those old parts are $60 bucks a pop? WOW! I could be sitting on a fortune! (Or not, I'm pretty sure I don't have a lot of them)
One main point is that the 74192 counters are loadable. So the value set by the thumbwheels are loaded into the counters and as long as there is no counting the value is fed directly to the numitron drivers. When counting is activated the load is inhibited and the numitrons display the count down value. The flip-flop is used to stop the count when zero is reached (and also to start the count) by using the flip-flop set/clear inputs .
@@ingehansen9069 Yes, that's correct. And just thinking about what I've seen in the video, I think I could design and build a similar unit with stuff that I already have in stock. I've got some 7447 7-segment display drivers for LEDs. Not that I need to, but just imagining I could. BTW: Did you ever get Forrest Mims, III Engineer's notebook at Radio Shack and see all those nifty projects? I build several of them. So there's one project in there that uses a CMOS chip that has 3 decade counters and it multiplexed 3 output (BCD) so then you needed a decoder chip, but it could replace a lot of that TTL logic. In his design you can cascade them so there was a 6 digit display, and with a timebase of 1 second and a gate you could make a frequency counter. I'm going back at least 40 years, or more... Fun Times!
Altium is for hobbyist as well with freeware CircuitMaker, and small business with CircuitStudio.
But i don't see myself giving up on Kicad, it's gotten sort of... shockingly nice i would say? Though my frame of reference is Eagle 7, may it burn in hell.
The first PCB I have ever made was using an etch-resist pen to draw the traces before etching the rest. Another similar method which achieves more consistent trace width is taping the board, which allows you to put down longer perfectly straight and more tightly packed lines than what can be reliably drawn by hand.
BCD thumbwheel switches were a commodity item, made sense to use them to save on interpreting circuitry
Look in somewhere like Digikey and you can still buy them. About $10 each so aimed at professional use. We just put some on some bespoke test equipment.
Does the fact that the filter can be easily removed imply you would have a choice of colour for different applications?
Early-ish PCB design was done with 'letraset' transfer rub down component footprints. And a roll of track ribbon tape, to make a positive artwork. Then photographically reduced to give a positive print. That was used as a photo mask for UV sensitised resist on the blank PCB.
It was common to solder flood the tracks, as there was no solder resist to protect the tracks from tarnishing. Sometimes a silver plating bath was used instead.
The through holes were often copper studs that were soldered through.
I reckon its early 'Tape' not even letrosett, some freaky curves on there and tape was easier to use, I didn't like the rub down stuff.
Our eyes have their greatest sensitivity around 550 nm, which is green. For color photography dark rooms, you can use very, very dim green lights. They light up the room just enough so you can see outlines, paper boxes, the easel and so on, once your eyes were reasonably dark adapted. But not bright enough to haze color paper in the maybe few tens of seconds they were out. That's what we used back in my day (90s). The total light output of those Numitrons is likely very low, so at a bit of distance, they will appear as a very diffuse, very dim green, just like those green lamps we used in the darkroom.
It's a series of tubes!
The Items on the back panel next to the Triacs are called 'Snubber networks' they help manage switching noise and over voltage conditions. This helps protect the Triacs, and prevent them from latching on due to spurios noise.
The traces on the top of the board end in plate through holes, and big blobs of solder were use to make the connection from the top to the bottom.
The "switches" used for the outlet are triacs and the cap/resistors are snubber devices to reduce any switching spikes from the triacs.
The decade counters were in parallel load mode from the thumbwheel switches until you started the counter.
Very simple logic compared to today's technology, but real interesting to see and brought back many memories for me!
Thanks for posting!
Yes, the holes are either pinned or have a short wire passed through before soldering - effectively a poor man's plated through hole.
@@GodmanchesterGoblin Indeed, those holes are NOT plated through. That's why there is a big solder blob. There may be an eyelet or a short post in the hole that is then soldered on both sides to complete the circuit.
@@Brian_Of_Melbourne Yes - and they can have a different coefficient of expansion with temperature compared to the pcb material, potentially resulting in hard to see cracks in the traces. Been there, repaired that, etc, back in the late 70s.
That board is a very interesting historical piece. It has a combination of point to point and PCB elements all on a hand laid board. That big cap is quite likely dead. I'd be careful about using that without checking the caps for leakage, it can quite easily destroy the most expensive part in there: the transformer. In fact that transformer already looks like it's been quite hot. The resistors are carbon comp and thus have probably also drifted (carbon comp usually fails with higher not lower values). The silver cans are likely germanium transistors, but could also be diodes.
Edit: nevermind there is a full wave bridge rectifier in blue diodes in the back. Doh!
Glare wouldn't be an issue in a darkroom though...
My thoughts exactly
@@christianseibold3369 nuance we needed
@@christianseibold3369 can we copy and paste your reply here to the dozens of other similar comments about "BuT tHeRe IsN't AnY gLaRe In A dArK rOoM!"
At 2:45, that display seemed crystal clear. Since it's meant for a dark room, perhaps the readability problem is just glare from operating outside of the design spec?
That timer looks like a nice thing to put in the Background
I made DIY PCBs in the 1980s at school; we would draw onto the unetched PCBs with marker pens, with rub-on transfers for component footprints, holes etc. Then you'd expose with UV, etch in a tank of some vile orange stuff, drill all the holes with a mini drill press, and you were done. It was actually very easy and effective. You could even do double-sided boards. Our teacher would always get us to minimise the amount of exposed copper which needed to be etched away as that used up the vile orange fluid, so we'd usually end up doing big ground fills by colouring them in with the marker pens or with masking tape or the like.
I don't know if this board was home made. It could have been. There's no writing or silkscreen, there's no soldermask (whoever made it even forgot to tin some tracks), and some of the tracks are noticeably different in thickness from the others. OTOH if it was home made it was done very carefully. I'd expect hand-drawn curved tracks to be much wobblier than that. Also the corners are very nicely rounded.
In the late 1970's, we were using different width black tape to create the tracks, and this allowed you to make quite smooth curves etc. You would use the wider tape for the power/higher current routes, and/or you could over-solder the tracks to increase the CSA, hence load capacity. The rest was done as you say above.
"some vile orange stuff" --> ferric chloride
There's a fairly professional-looking logo on the front, so I don't think it's a DIY project. It was probably made by a small company that didn't make very many of them and did a lot of the design and assembly using manual methods.
I did something similar, built a phone line noise filter out of discreet components, designed my board, markered it, UV then disolved the excess copper in a warm bromide? solution.. drill press for the component and wiring holes, soldered it all up and mounted it in a kit box from Tandy.
@@PurpleAlienPlanet I recall a kid in the electronics club at school (in the 1970s) saying "You know Stainless Steel is supposed to be stainless - it isn't). He had tipped used ferric chloride down his Mums sink.
did not deserve the hate :(
4:30 "That's coming from the room lights" ... wouldn't that be something less than a significant problem for a DARKROOM TIMER?
Really puts the moon landings into perspective. I bet that power transistor on the back is a 2N3055, they were ubiquitous. The two pins poking through the holes reminds me of an amplifier I built in my teens in the early 70s; the "chassis" was sheet aluminium bent into an upside-down U shape, and the PSU was built around a zener and a 2N3055.
My first thought was a 7805, but given the date codes on the other parts this thing looks to be about a year too old for that, so yeah, more likely a 2N3055, those things were everywhere back in the day. Still have boxes full of them which are worth a small fortune today.
Bearing in mind that they had to integrate and test all the equipment in advance this is about a decade after the Apollo equipment was designed. having said that NASA had larger budgets than a small company making photography equipment.
4:36 You're complaining about reflections in a device that is literally designed to be used in a "dark room". There won't be any reflections there and the way that an enlarger is situated these won't be facing the light sensitive paper so they won't cause accidental exposure of the paper. This seems like a complaint about a problem which in practice is non-existent. This is a timer and those numbers are only there to parse the time as it ticks away, as the image is being exposed the numbers are counting away the time. If it works at a timer and with the rapid pace of development at the time, this product was still pretty remarkable and kinda great
So... someone doesn't get it. It's not abut the use in this particular device, but the issues with Numitrons in ANY device. this just happens to be the first thing that showed up that actually uses them.
As for the color, the filter is swappable on purpose. You use green light on some types of papers instead of red.
Given that a dark room is about the only place where numitron tubes many drawbacks don't actually matter, that is likely why this thing still exists some 50 years later where most other devices that used them found their way to the trash heap 45 years ago.
I agree, he's so whiny about things that aren't problems with the numitron. He complains about glare while he has it under studio lights, all tubes will have glare, even led displays and lcds will have glare, he complains about this (presumably kit built) product's quality as if it relates to the tube's quality, he complains that this particular product does not feature a divider between the tubes, which A. Not a problem with the tubes, a problem with the product, and B. You can literally add a piece of paper in the middle if it annoys you that much. Additionally, his eyesight must not be very good if he can't clearly read the numerals on the displays. He says the 8 looks like a blocky mess, perhaps he needs new glasses, it looks exactly how 7 segment displays always look.
Ah, the smell of blown Rifa capacitors. Knew exactly what was probably causing that the moment you made mention of it.
I like that both these products looks like they were put together by electricians and not electrical engineers.
Looks pretty good to me honestly
I had an internship where I did printed circuit board (PCB) development using UV light. It's pretty fascinating. You start with a design on a transparency, and tape it down to this sandwich of UV-sensitive material / copper / plastic / copper / UV-sensitive material. After exposing it to the UV light, you'd put it in a solution that would remove the exposed bits. Then you used etching fluid to remove the copper in those spots. Then you used a solvent to remove the remaining UV-sensitive material leaving the copper traces underneath. Then you'd soak it in a tinning solution so the copper wouldn't oxidize. Then drill holes in the board as appropriate and solder your components in.
I think you'd appreciate it given your fondness for photographic development.
I like the 6 with the extra bar and the 9 without. And now I can never look at any seven segment displays ever again.
If I remember rightly, the 9 without the bar is what you get from the commonly-used 7447 BCD-to-7-segment decoder chip. Some people used hacks to add the extra bar to it. This device seems to be using a different decoder-driver chip that I can't find on the internet.
@@gcewing 74928. Datasheet for the original part is hard to come by, but MM74C928 should be close enough to make sense of the circuit.
Agreed. It's settled, the 6 with and 9 without are defacto the best way to do it.
I wonder if some of it stems from handwriting. I'm not sure I've seen contemporary handwriting do the 9 with the tail. Only angled.
@@0Rookie0from my perspective as a European, basically everybody will do the tail on the numbers six AND nine in handwriting. Could the timer have been made for the European market?
Yeah the 7447 & 4511 didn’t have the bars but the neon DM8880 drivers did. Haha now that I program the microcontroller segment tables myself I could do the 6 and 9 differently. But I have preferred them without the bars.
The two silver components that are rectangular on the back panel are film capacitors probably for power safety, REFA branded ones tend to crack with age and will blow magic smoke when they go.
7492 is a divide by 12 counter.
9:49 I actually think this out-of-focus blurry image looks pretty nice. The decimal point looks like a dot, too.
Maybe that's why the back plate is gray. So that the reflection can make the segments look bolder. Apparently RCA forgot about the ambient light...
Maybe try a frosty glass as cover plate? The sanded side can also fix the reflection issue too.
24:45 You essentially had thin black tape that you can put on transparent sheets to transfer it to a PCB with positive photo-resist. It was easier to lay curves than square connections (and you also wanted to avoid to steep angles anyway to prevent bubbles of air or etchant to collect there).
And for double sided boards you could use red and blue tapes on the one sheet, and expose it through filters to get the masks.
Ah, the Numitron: the very best of 1930s tech made 40 years late.
to be fair, everything looked janky back in the 70s
That burnt part is probably from them touching a solder pen to that solder point right next to the big cell :)
Yup. If you solder right-handed, that would be pretty much exactly where the soldering tip would be for that joint.
The 7492 is a 4-bit counter with two sections, a divide-by-2 and a divide-by-6. That's probably what they're using to get the 10Hz.
You're probably right about the BCD encoders. Thumbwheels like that with BCD outputs are fairly common (or at least they used to be -- nobody wants to use physical controls any more!)
my guess: the filter is green intentionally, so the red lights in an actual darkroom would not create the glare
hypothesis number 2: they put a grey background to lower the contrast, not of the display, but of the glare coming from the tube
7492 seems to be divide by 12. sn74ls92 is, which is most likely a modern variant of the same logic. After rectifying ac 60hz with a bridge rectifier you get a 120hz which might be used here.
The SN means it's manufactured by Texas Instruments. LS stands for Low-power Schottky, which use less power and run faster than their original 1970's 74## counterparts. Functionally for most applications, LS, S, and standard TTL ICs are interchangeable.
Yep, that's about how I would have made it back then lol. Fun fact: those boards were designed using transparency and vinyl tape at 2 or 4 X then photographically reduced so you could do it with better resolution. I remember having layouts covering the entire kitchen table. When you brought them in to be reduced, you unrolled it then pressed the tape back down where it came up when rolling it up for transport lol! 😄
I think that darkroom timer is still far more modern than the ones I had back in school.
And what was that? I suppose a wind up kitchen timer and a couple of wires?
@@Hirnspatz basically, only it was about 8” tall. It plugged into the wall, you plugged the enlarger into it and then turned the dial to how long you wanted it to be on for. I’m pretty sure timer worked like an electric clock (not quartz but the constant motor speed type) and wasn’t spring powered the way a kitchen timer is, though. It survived decades of middle school students of varying ability and enthusiasm.
I'm guessing the entire reason the filter is replacable -- when it sold, it came with optional filters, including paper-safe.
"Delightfully Crude..." That's one way to put it.😂
A dark room display might be the only use case those would work for. It's probably not bright enough to expose nearby film, and green filter I think was specifically chosen to reduce glare from amber lights in the dark room
The board looks like it has been "taped". You would use a mylar (or semi transparent paper) sheet and traces were done with sized rolls of black masking tape which you can stick onto that. Patterns for sockets could be done with dry transfer "rub on" symbols (similar to "lettraset"). Sometimes this was combined with black pens, similar to sharpies. This artwork then is used to photo expose the PCB.
Did I just watch 27 minutes about reflections in tubes? You're damn right I did.
AP clearly stands for Actual Product.
I was wondering whether a polarising filter would help (in the correct orientation, obviously). Since the distracting reflections are largely from the vertical walls of the tubes, a pair of polarised sunglasses at 90 degrees to the normal orientation would be worth checking.
i think they look cool af
I agree. Not a thing wrong with them.
1:26 - The digits are simply TOO FAR APART resulting in *bad kerning* !
I believe they're that far apart due to the REALLY bad glare that would result if they were placed next to each other.
“Green filter…?” was my first thought… But I’m 58, and did stills and super-8 as an expensive hobby before the world hit the 80’s. Silly me… 😂
4:39 it would be used in a dark room though. So they probably didn't bother with reflection control
The components with a capacitance and resistance marked are probably "snubbers" for the triacs.
Hi!
The burned spot on the main electrolytic capacitor isn't harmful, it was just not too carefully assembled.
The pcb style is typical for the time (prior cad), but it was not made with the best of the designer skills.
One point of concern is that pair of Rifa capacitors glued on the back plate. One of them is visibly cracked ( a typical mode of failure of this brand). This causes the moisture of the air enter into the capacitor and it leaks. They should be replaced by safety reasons.
🤗🇦🇷
I’ve seen tons of Rifa caps whose epoxy has cracked - on completely unused parts. They don’t even need to be ever connected to electricity to fail…
@tookitogo Those caps (and other German brands of that time) were excellent , better than today's Chinese ones, but suffer the pass of time because the encapsulation technique. Too bad.
🤗🇦🇷
@@DIY-valvular RIFA was Swedish, not German. Belongs to KEMET now, who still sells them.
@@tookitogo My bad! 🤗
Rifa capacitors have a habit of cracking and then "making poopie in their trousers".
I love this channel because sometimes it's just someone younger than me giving grumbling reviews about obscure technology from before I was born.
"So you thought I was too mean to the numitron?" pulls out gallagher sledgehammer
The big blob-trough contacts is a prototyping / low product count trick. You drill the trough contacts oversize (1.5 - 2.5 mm) with a copper ring around and put a big solder blob on both sides. This only works well with hand soldering, lead rich solder and wastes a lot of solder. I used to "collect" a bunch of through contacts at one place and used pin header to make the connection. But that is only useful for larger boards where you can route the connections easily.
The filter isn't red because the light in the dark room would be red, and you wouldn't be able to see the segments at all under that kind of filter.
Are you sure about this? Every darkroom timer i've used has had red numbers. Well, except a fully mechanical timer which was not illuminated at all.
Not trying to be a smartass here btw you might be right. The ones i mean all used LEDs
@Hamring Agreed. Red display is fine.
That would only apply if they were passive segments under a red blocking filter, actively illuminated displays using the same colour are still going to be easily distinguished from black under monochromatic light
It’s more likely because it’s for colour paper, rather than monochrome. Colour usually has a sensitivity dip in the green which this should be close to.
It is possible I am assuming incorrectly. However, I think it would be worth examining this under the proper conditions of a dark room to truly see. And finding some red film shouldn't be difficult. Bonus if finding a variety to test with.
I was today years old when I learned that you have a side channel. That'll be some sleepless nights to catch up.
you draw with a special marker on a copper covered Fiberglas board (aka a blank circuit board) and then "develop" it in a acid bath that eats away all copper you did not draw on, if left in to long it will eat away the drawn lines to so timing is important. I made a lab power supply for my introduction to electronics course in college in 1995 using this method. Perfect for a one off diy build.
So did they need a timer to build the timer?
@@TheRogueWolfyes
@@TheRogueWolf One mississippi, two mississippi... oh dang nabbit...
That's a home technique, this is commercial and is as others others have said hand laid out and then photographed (perhaps with reduction) to make the final transparency (maybe even a negative) to expose the PCBs with.
@@Brian_Of_Melbourne That board did not look professionally made. It is a teaching method as well as a "DIY" and small scale method, the photo/transparency method is more involved and not suitable for small scale DIY. As I said, it was used in a US college course as a teaching method/aid.
Circuit board traces were laid out with thin tape atop a transparent sheet, at scale. The component footprints would be pieces of transparency with the correct hole sizes and annular rings already on them, stuck onto the master transparency. Light would be beamed through it (or through copies) at light-sensitive etch resist, which would then be developed and etched.
The package sticking out the back is a 5V regulator (they get hot), and the capacitor+resistor components are probably to act as noise filters for the sharp transients produced by the TRIACs. Full-wave rectified mains is 120Hz not 60Hz, so a divide-by-12 chip would make sense here.
This feels like it's asking for a big clive collab.
Please send it to Big Clive!!! Heck ill pay the shipping!
The weird components on the back panel are X-class mains-rated film-type safety capacitors. If you intend to use the thing, it's prudent to replace those with modern X2 capacitors because some of them from that era are known to fail, and when they fail they catch fire. (X1 rated capacitors are for industrial 3-phase applications, X2 will suffice for stuff like this)
Correction: I missed the ohm rating. They’re snubbers. Same idea though, x class cap with a series resistor. Not terribly common in a single package anymore, so a replacement is tougher, but the resistor makes it less likely to spontaneously combust.
Years ago, you could buy circuit boards with the back completely covered with copper. You used a marker to draw all the tracks and connection pads over the copper, then you submerged the board in acid. The acid removed all the copper, except in the zones protected by the marker, living only the tracks. Then you used a drill to make the holes in the connection pads.
That circuit board seemed to have been made following this procedure.
people used to make the masks by hand. not by hand for every board though, but using photo techniques.
you can still buy raw boards to make your own pcb's tho of course. techniques differ, but back in university we'd use a laminating machine to transfer a laser print from paper onto the board.
My brain always parses 88 on a seven segment display as a diagnostic or error state unless proven otherwise
at this point this is just a pet peeve....this darkroom timer looks even better than the clock.
That's how we used to solder and it was even a technique before my time but it's what you did if you didn't have much money but wanted to make something. I did some similar stuff when I was a kid in the 80s and then I decided that you could make cool jewelry out of solder too so I did that for a while. Then I studied other things and, the next time I needed a soldering gun, I found that my brother has lost all the tips I had for mine. Great video!
I don't understand the purpose of the rants, the display looks and works perfectly fine. in real life you'd get used to it in an hour and never think about it again. you'd be too busy working the device to do the actual job, not contemplate the shapes or decimal point :-)
The parts mounted next to the (presumably) triacs are surge suppressors (or snubber) to dampen transients from inductive loads like motors. Some are still made but can just be replaced with a discrete capacitor and resistor of the same values.
so my main clock is a segment clock and i honestly couldnt remember if the 6 and 9s had the tops and bottoms had to check and when you mentioned it and i realized i didnt know was just after the clock went from 16 to 17so thankfully i didnt have to wait long for 19
Well, don't leave us in suspense -- which was it? :-)
@@gcewing it does have tops and bottoms
The signal can be from mains (or at least calibrated from mains) with the 7492. The 7492 is 3 counters in one - a /2, a /3/ and another /2 - letting you grab mains frequency and then divide that down into 30hz, then 10hz, with the last /2 ignored. The 7490 is a divide by 2 and divide by 5 in one package. 10hz output from the 7492 is fed into the 7490's divide by 10, getting you 1hz when on the "seconds" setting. The 7490 is bypassed in the ".1 second" setting, giving you the 10hz without change. The decimal activates when this is bypassed, denoting 10ths of a second.
The selection of either 1hz or 10hz is fed into the the rightmost tube's 74192 counter. When the rightmost counter reaches 10, it instantly resets back to 0 and also decrements the left counter by 1, giving you a 2 digit BCD countdown of 99 to 0. The 74192 was selected because it has the additional feature of data input pins letting you "load in" any value (from the encoder). In the default state, anything on the encoder will push to the counters.
When you press the operate button, one of the flip flops in the 7493 changes state, disabling the "load-in" from the encoder and enabling the 74192's to begin counting down. The flip flop continues holding the "countdown" state until zero is reached on both displays, causing it to immediately switch to pulling in the encoder values.
The 74193's feed BCD output into the numitron tube drivers (however they happen to work) to display whatever the count is.
The calibrate feature is pretty interesting to me, though. All of the exact required ICs are there to take 60hz and bring it to 10hz and 1hz but if its just grabbing mains, there's nothing to calibrate. The calibrate button might utilize the second flipflop to switch to an external signal using what looks to be a 1/4" jack on the side. Would love to hear anyone else's take on the calibrate feature.