As far as I'm concerned, Jan repeatedly said that his power supply wasn't ideal and has been honest enough to show and address the potential problems. That is to be commended and is the essence of what makes his videos so interesting.
Ich finde es total Klasse, dass du nicht aggressiv versuchst den Aufbau des Netzteils zu verteidigen, sondern stattdessen den Input konstruktiv aufgreifst und den Aufbau verbesserst und die Grösse hast, zuzugehen dass das verbesserungswürdig war! Darum mag ich diesen Channel, denn so würde nicht jeder reagieren!
I watched / listened to this 4 times today while working on a project myself. It is always a joy to listen to Jan Beta. Make the videos as long as you like my man!
Great work, Jan! Man, I am having very similar issues with the power supply circuit section of my CoCo 2, as well. I have to say, the Service Manual for the CoCo 2 is quite impressive too. Tandy/Radio Shack did a great job with those documents. A great series thus far -- I'm glued to every moment.
Great video as always, love to watch the repair process mistakes and all. One video I'd like to see some day is if you could go into detail about using the oscilloscope to track down issues, like how do you know what you're looking at and whether or not it's what you expect to see and if it isn't what could that mean?
for some reason youtube didn't list your new videos in my subsriberlist... i went onto your channel to check to see...and i see you made lots of videos i haven't seen.. now i can chillout before bedtime with your great videos again...:)
(@41:12) - I’d just use an 8-position DIP switch there, and desolder the old socket; the pins look a little crusty. Then solder in the DIP switch, and throw s1, s3, and s5 to “on” to match the jumper config. 🤠
The chips are very voltage sensitive. Plugging in a vacuum cleaner or a slight power drop and it would lock up the computer. If you haven't saved your work, as you went along, it was gone.
It's not a Trash-80 video without having a power supply that MAYYYY seriously injure you and/or the computer lol great video, Jan. Need to order some new Jan Beta t-shirts soon.
Good show. Edit, you can swap the 4116 for 4164 chips by bending a few pins. Done it on a couple of Spectrums... That would negate the need for a -5v rail if you have the other rails present.
Great work! Went down a few rabbit holes there with that Tandy Trash-80. Inspiring me to get my Amiga 1000 & 2000 out of their boxes and up and running again. Got a bit of work to do in power supply re-cap, repair leaky battery corrosion on 2000, flaky floppy disk drive, dodgy mouse on 1000.
Patience of a saint! :) That being said it was really a dumb idea of Tandy to use a PSU port that is so similar to the video and cassette port. That's asking for trouble! lol
No worries about the PSU. No magic smoke was released. Your making progress. Damn that R70 I ran through that schematic a couple times. I did find a couple pictures of the board and I can see R70 but its hard to read the color bars.
One of my first computers. I have fond memories of it though I never had floppy drives until much later when someone gave me one as a retro computer. A great book for this is the The Custom TRS-80 & Other Mysteries by Dennis Báthory-Kitsz. I used it in the past to add lower case mode and to give the system a little speed bump.
I'm guessing that the cassette port resistors got fried because someone accidentally plugged power into the cassette port. IIRC the Model I uses 5-pin DIN connectors for all 3 connectors (power, video and cassette) making it easy for someone to accidentally plug power into somewhere where it doesn't belong. A rather poorly thought out design if you ask me.
I've recently built a new PSU for a 1977 signal generator and when I went to test it I was getting weird voltages. Same mistake: I had forgotten to connect the centre tap of the transformer! :)
I think the RAM chips may have been socketed as standard. Since there was also a 4K model. Though I'm unable to find a 4096x1 bit DRAM which is pin compatible with the 4116.
the Ram chips were not originally socketed ... when I upgraded mine I had to desolder.... it was one of the first times I ever did that... I made a mess and at that time not know what braid was....
At 39:05, I can't tell if your scope is set to the wrong time and voltage divisions or what, but that is _not_ a normal composite video signal. There is no pixel data. There is only what appears to be a horizontal sync signal. Look at the bottom of page 22 of this manual: www.1000bit.it/support/manuali/trs/TRS-80%20Technical%20Manual%20%281978%29%28Radio%20Shack%29%28pdf%29.pdf See the peaks following the trough of the horizontal sync that say "white level?" You don't have any (or at least your scope isn't showing any, which could be because you have the time division too short). Also, the signal should be about 2 volts peak to peak, and it should spend most of its time at around 0.5 volts (black). Yours _appears_ to be stuck at 2 volts most of the time. Compare your signal to this one: imgur.com/gallery/OJwQkS4 They do not look the same (unless your time and voltage divisions are all messed up and we're only seeing a tiny part of the signal... but even then I don't understand why your signal would be floating at 2 volts most of the time).
The ringing you saw on the CPU clock waveform is due to the long scope ground wire you are using. If you connect the short gnd wire on the probe itself to ground at the board, the waveform will look a lot better. Not saying you need to do that when just probing around. Just that it's nice to understand why the waveform looks the way it does.
Hi, can anyone help me find good quality assorted kits of components in Europe namely capacitors to start restoring vintage computers? Thanks in advance for the help.
Hello, I've somes questions : (I'm living in europe) where can I buy online good quality capacitors with a good price, I'm also asking somes advices about the "best bulk" to create some stocks.
Just a tip:the TRS-80's video hardware is completely independent of the Z80 CPU, so that even when you pull the Z80 out of its socket, the TRS80 will STILL generate a video signal, ( a random pattern of characters on screen) there are no programmable devices that need to be programmed in the video hardware. What I seemed to see in the video is that you did not actually get characters on the screen, just blocks of white or black. To me that means that either the character generator (ROM) isn't working OR that only the RAM bit that inverts a character position of not is working, Also, the TRS-980's canonical "graphics blocks" do not come out of a character ROM, but are directly generated out of the bit pattern coming from the video RAM. so if you can see that the video output consists of TRS-80 style graphics characters (in a 2 x 3 block pattern) instead of full character sized pattern's you can draw a conclusion on whether the character ROM (or 2x3 patter generator circuitry) is actually active or not . A closeup of the screen would be very helpful
As already noted in other comments, for a machine like a 16k TRS-80, A15 being permanently low is entirely expected - IIRC, the Z80 starts execution at 0x0000, so the ROM will be 0x0000-0x3fff, and if the RAM is 0x4000-0x7fff, there's no reason A15 would go high in normal operation.
Jan... as i remember there was a service bulletin for the cassette port that put that resister there... Further.... trs-80.org is up now anyway... please be aware that the memory switch assembly MUST be low height... a normal switch block will hit the case when you reinstall it...
Hmm, fried resistors in the cassette port circuitry, and the same connector used for power/video/cassette, I'm gonna go with somebody plugged the power adapter into the cassette port :)
You have a plastic drawer of transistors on top of the display venting holes of your oscilloscope, if you ever wonder why the display might get overheated.
@@Calphool222 Oh cool. Its always better to have confirmation from someone whose actually tried it and found it to work. Theres not a huge difference between CMOS and TTL logic levels and if its not critical....
Attach the keyboard and a tapedrive and blindly type the tapeloading command and see if the tape unpauses and starts loading.. That way you can find out if it is lack of a video signal of lack of booting the rom.
All things considered Jan there could have been much worse mistakes to make with the power supply connections, its the sort of thing we've all done at some time, I once made a mistake connecting up a switching power supply board I'd picked up from a electronics recycling company and it went bang. The power transistor on the primary side was just 3 legs sticking out of the board and the rest of it I found stuck in the ceiling tiles......
hello jan how are you any luck on my order for eprom bwack i tried and l0st 4 eproms i will pay you is u can program one for me it easy cheap to send to australia please help me im dying to install the eprom in my c 64 bread vin brown one .. thanks very very much .. bye
A 74Cxx device is a 3V to 15V device, while a 74HC(T)xx device is a 2V to 6V device, so you should first check by which power rail the 74C04 chip is powered on the board. BTW if you can’t source a genuine 74C04 chip you can find a replacement in the CD4000 (or equivalent) series: CD4049, HEF4049, MC14049, etc. should all be OK replacements in your case.
I very much doubt the 74Cxx or HC(T)xx devices even existed at the time, only 74LSxx logic (and 74xx logic) if the power rail is anything BUT 5V there is something wrong.
martin de jong That’s an interesting consideration. Although the 74Cxx series was already in production in 1973 (according to the date of a Fairchild AN) the chips on the board may not be the ones installed when the board was manufactured.
I have several functional Model I machines, so if you need someone to take a look at values on any burned out components, drop me a line. The garbage pattern you're getting on the display suggests bad RAM (or perhaps related to your missing A15 signal at the CPU... although as long as it's deselected, I don't think it's doing anything in a 16k machine, but that's just a guess :-). It's been a few decades, but sometimes the Model I would show similar (but not quite what you're seeing) garbage on the display, and you'd have to hit the 'reset' button (momentary switch next to the expansion edge connector on the opposite end of the MB from the power switch) a few times to get it to power up properly. Btw, if anyone happens to know why this occurred sometimes, I'd love to know.
eh Scheiße passiert, um eine amerikanische Phrase zu verwenden. So geht es, wenn Sie so etwas tun. Es wird wahrscheinlich nie beim ersten Mal korrigiert, was großartig wäre, aber das nimmt den Spaß daran, irgendetwas zu reparieren. Anyway before I screw up any more words in German, These old computers have to many oddities that can happen and parts that fail as well but at least you narrowed it down to somethin' and you got rainbows, never seen that but it's cool despite it bein' a bad signal. Danke Jan!
Hi Jan. You only show it briefly, but it *appears* you're using my TRS-80 Model 1 power supply design. (github.com/calphool/TRS80MXS) I'm interested, was there a flaw in the design itself, or just a mistake in how it was constructed?
I think you are way too hard on yourself. No need to be embarrassed that you missed a bad solder connection. It happens to everyone! The key is, you found it!
As far as I'm concerned, Jan repeatedly said that his power supply wasn't ideal and has been honest enough to show and address the potential problems. That is to be commended and is the essence of what makes his videos so interesting.
Ich finde es total Klasse, dass du nicht aggressiv versuchst den Aufbau des Netzteils zu verteidigen, sondern stattdessen den Input konstruktiv aufgreifst und den Aufbau verbesserst und die Grösse hast, zuzugehen dass das verbesserungswürdig war! Darum mag ich diesen Channel, denn so würde nicht jeder reagieren!
I watched / listened to this 4 times today while working on a project myself. It is always a joy to listen to Jan Beta. Make the videos as long as you like my man!
Great work, Jan! Man, I am having very similar issues with the power supply circuit section of my CoCo 2, as well. I have to say, the Service Manual for the CoCo 2 is quite impressive too. Tandy/Radio Shack did a great job with those documents. A great series thus far -- I'm glued to every moment.
I really appreciate you showing the error with the power supply. We can all learn together!
Thanks, I always try to show the mistakes because it's the best way to learn in my experience. :)
Great video as always, love to watch the repair process mistakes and all. One video I'd like to see some day is if you could go into detail about using the oscilloscope to track down issues, like how do you know what you're looking at and whether or not it's what you expect to see and if it isn't what could that mean?
3:11 NOOOOOOOooooooooooo!!!!!! :)
To make an omelette, you gotta break a few eggs (but it still hurts to watch it). ;-)
for some reason youtube didn't list your new videos in my subsriberlist... i went onto your channel to check to see...and i see you made lots of videos i haven't seen.. now i can chillout before bedtime with your great videos again...:)
Me too.
(@41:12) - I’d just use an 8-position DIP switch there, and desolder the old socket; the pins look a little crusty. Then solder in the DIP switch, and throw s1, s3, and s5 to “on” to match the jumper config. 🤠
The chips are very voltage sensitive. Plugging in a vacuum cleaner or a slight power drop and it would lock up the computer. If you haven't saved your work, as you went along, it was gone.
It's not a Trash-80 video without having a power supply that MAYYYY seriously injure you and/or the computer lol great video, Jan. Need to order some new Jan Beta t-shirts soon.
a good workman always comes clean with self criticism - way to go :)
Good show. Edit, you can swap the 4116 for 4164 chips by bending a few pins. Done it on a couple of Spectrums... That would negate the need for a -5v rail if you have the other rails present.
Electroboom would be happy you mentioned the full bridge rectifier
@@Retr0N1x Mother of all fullest bridge rectifiers :)
So easy to go down a rabbit hole looking for something that doesn't exist (gosh, that's philosophical 😊) It was educational as always. Cheers
Great work! Went down a few rabbit holes there with that Tandy Trash-80. Inspiring me to get my Amiga 1000 & 2000 out of their boxes and up and running again. Got a bit of work to do in power supply re-cap, repair leaky battery corrosion on 2000, flaky floppy disk drive, dodgy mouse on 1000.
Don't worry about "embarrassing mistakes." Anyone who does electronic troubleshooting has had forehead-slapping epiphanies.
and most UA-camrs edit theirs out, rather than taking the hit...
Learning is a wonderfull journey and making mistakes is a natural part of it. By the way, I love your videos. Good luck!!
Well done Jan Beta. Sticking to it. Love your method and tenacity and your work.
Patience of a saint! :) That being said it was really a dumb idea of Tandy to use a PSU port that is so similar to the video and cassette port. That's asking for trouble! lol
Retro Recollections That is definitely a lesson that seems to have been learned (eventually) in the industry going forward.
No worries about the PSU. No magic smoke was released. Your making progress. Damn that R70 I ran through that schematic a couple times. I did find a couple pictures of the board and I can see R70 but its hard to read the color bars.
Videos of Jan are always funny. Btw. I also did a #SepTandy video, but how can I submit it to come on the updated playlist?
we learn more from mistakes 'Jan.. honesty and openness is apricated. great video
One of my first computers. I have fond memories of it though I never had floppy drives until much later when someone gave me one as a retro computer. A great book for this is the The Custom TRS-80 & Other Mysteries by Dennis Báthory-Kitsz. I used it in the past to add lower case mode and to give the system a little speed bump.
At 7:20, you're saying 732-- don't you mean 723?
Yes! I misspoke there, sorry! :D
I'm guessing that the cassette port resistors got fried because someone accidentally plugged power into the cassette port. IIRC the Model I uses 5-pin DIN connectors for all 3 connectors (power, video and cassette) making it easy for someone to accidentally plug power into somewhere where it doesn't belong. A rather poorly thought out design if you ask me.
Yes, that might very well be. The 3 DIN connectors are all the same 5-pin type. So that could happen easily.
I've recently built a new PSU for a 1977 signal generator and when I went to test it I was getting weird voltages. Same mistake: I had forgotten to connect the centre tap of the transformer! :)
I think the RAM chips may have been socketed as standard. Since there was also a 4K model. Though I'm unable to find a 4096x1 bit DRAM which is pin compatible with the 4116.
the Ram chips were not originally socketed ... when I upgraded mine I had to desolder.... it was one of the first times I ever did that... I made a mess and at that time not know what braid was....
At 39:05, I can't tell if your scope is set to the wrong time and voltage divisions or what, but that is _not_ a normal composite video signal. There is no pixel data. There is only what appears to be a horizontal sync signal.
Look at the bottom of page 22 of this manual:
www.1000bit.it/support/manuali/trs/TRS-80%20Technical%20Manual%20%281978%29%28Radio%20Shack%29%28pdf%29.pdf
See the peaks following the trough of the horizontal sync that say "white level?" You don't have any (or at least your scope isn't showing any, which could be because you have the time division too short).
Also, the signal should be about 2 volts peak to peak, and it should spend most of its time at around 0.5 volts (black). Yours _appears_ to be stuck at 2 volts most of the time.
Compare your signal to this one:
imgur.com/gallery/OJwQkS4
They do not look the same (unless your time and voltage divisions are all messed up and we're only seeing a tiny part of the signal... but even then I don't understand why your signal would be floating at 2 volts most of the time).
The ringing you saw on the CPU clock waveform is due to the long scope ground wire you are using. If you connect the short gnd wire on the probe itself to ground at the board, the waveform will look a lot better. Not saying you need to do that when just probing around. Just that it's nice to understand why the waveform looks the way it does.
Ah, that might very well be. Thanks!
Hi, can anyone help me find good quality assorted kits of components in Europe namely capacitors to start restoring vintage computers? Thanks in advance for the help.
Hello, I've somes questions : (I'm living in europe) where can I buy online good quality capacitors with a good price, I'm also asking somes advices about the "best bulk" to create some stocks.
Just a tip:the TRS-80's video hardware is completely independent of the Z80 CPU, so that even when you pull the Z80 out of its socket, the TRS80 will STILL generate a video signal, ( a random pattern of characters on screen) there are no programmable devices that need to be programmed in the video hardware. What I seemed to see in the video is that you did not actually get characters on the screen, just blocks of white or black. To me that means that either the character generator (ROM) isn't working OR that only the RAM bit that inverts a character position of not is working, Also, the TRS-980's canonical "graphics blocks" do not come out of a character ROM, but are directly generated out of the bit pattern coming from the video RAM. so if you can see that the video output consists of TRS-80 style graphics characters (in a 2 x 3 block pattern) instead of full character sized pattern's you can draw a conclusion on whether the character ROM (or 2x3 patter generator circuitry) is actually active or not . A closeup of the screen would be very helpful
Here's hoping for a Commtober!
Orictober works better.
Don't be so hard on yourself. We've all been there haha. Great video as always.
Thanks, still felt a bit embarrassing, especially since I pointed the importance of the centre tap out previously... ;)
As already noted in other comments, for a machine like a 16k TRS-80, A15 being permanently low is entirely expected - IIRC, the Z80 starts execution at 0x0000, so the ROM will be 0x0000-0x3fff, and if the RAM is 0x4000-0x7fff, there's no reason A15 would go high in normal operation.
That video signal is really crazy... because it was color, but the model 1 is black and white!
Jan... as i remember there was a service bulletin for the cassette port that put that resister there... Further.... trs-80.org is up now anyway... please be aware that the memory switch assembly MUST be low height... a normal switch block will hit the case when you reinstall it...
Don't forget to add #Septandy to the video title, otherwhise it won't be found when fans search for #septandy
Dude nice buffy the vampire slayer shirt
🤘😝
This was my first PC. I sold it to buy a C64 when they first came out. Nice to see!
Hmm, fried resistors in the cassette port circuitry, and the same connector used for power/video/cassette, I'm gonna go with somebody plugged the power adapter into the cassette port :)
You have a plastic drawer of transistors on top of the display venting holes of your oscilloscope, if you ever wonder why the display might get overheated.
74HC04 would be a better choice of replacement for a 74C04 than the TTL compatible 74HCT04....Assuming theyre pin compatible.
I've used the HCT family for that specific spot. It works fine.
@@Calphool222 Oh cool. Its always better to have confirmation from someone whose actually tried it and found it to work. Theres not a huge difference between CMOS and TTL logic levels and if its not critical....
Should I ask what the roll of toilet paper was for? 😂
Great video. We all do mistakes.
Is A15 on the Z80 being used? You can address the first 32K without it, so depending on where ROM and devices reside, you may never use A15.
According to the documentation the level II ROM (12K) is 0000-2FFF
www.computerarcheology.com/TRS80/Hardware.html
Attach the keyboard and a tapedrive and blindly type the tapeloading command and see if the tape unpauses and starts loading.. That way you can find out if it is lack of a video signal of lack of booting the rom.
Cutting that keyboard cable was painful to watch 😳
Should be a lot more reliable with a detachable ribbon cable though.
Being Jan Beta, he should have somehow drilled it out. :P
Why?
All production Mod 1's had sockets for RAM.
All things considered Jan there could have been much worse mistakes to make with the power supply connections, its the sort of thing we've all done at some time, I once made a mistake connecting up a switching power supply board I'd picked up from a electronics recycling company and it went bang. The power transistor on the primary side was just 3 legs sticking out of the board and the rest of it I found stuck in the ceiling tiles......
Ian Rolfe Sounds like a serious BANG… 😮
55:17 We have a not working NOT-Gate... ;)
Hi Jan Beta!
hello jan how are you any luck on my order for eprom bwack i tried and l0st 4 eproms i will pay you is u can program one for me it easy cheap to send to australia please help me im dying to install the eprom in my c 64 bread vin brown one .. thanks very very much .. bye
Great video!
A 74Cxx device is a 3V to 15V device, while a 74HC(T)xx device is a 2V to 6V device, so you should first check by which power rail the 74C04 chip is powered on the board.
BTW if you can’t source a genuine 74C04 chip you can find a replacement in the CD4000 (or equivalent) series: CD4049, HEF4049, MC14049, etc. should all be OK replacements in your case.
I very much doubt the 74Cxx or HC(T)xx devices even existed at the time, only 74LSxx logic (and 74xx logic) if the power rail is anything BUT 5V there is something wrong.
martin de jong That’s an interesting consideration. Although the 74Cxx series was already in production in 1973 (according to the date of a Fairchild AN) the chips on the board may not be the ones installed when the board was manufactured.
74Cxx chips are in the parts list of the Technical Manual for the TRS-80 model 1, so they were installed on the board from the beginning
I have several functional Model I machines, so if you need someone to take a look at values on any burned out components, drop me a line. The garbage pattern you're getting on the display suggests bad RAM (or perhaps related to your missing A15 signal at the CPU... although as long as it's deselected, I don't think it's doing anything in a 16k machine, but that's just a guess :-). It's been a few decades, but sometimes the Model I would show similar (but not quite what you're seeing) garbage on the display, and you'd have to hit the 'reset' button (momentary switch next to the expansion edge connector on the opposite end of the MB from the power switch) a few times to get it to power up properly. Btw, if anyone happens to know why this occurred sometimes, I'd love to know.
Interesting watch
eh Scheiße passiert, um eine amerikanische Phrase zu verwenden. So geht es, wenn Sie so etwas tun. Es wird wahrscheinlich nie beim ersten Mal korrigiert, was großartig wäre, aber das nimmt den Spaß daran, irgendetwas zu reparieren. Anyway before I screw up any more words in German, These old computers have to many oddities that can happen and parts that fail as well but at least you narrowed it down to somethin' and you got rainbows, never seen that but it's cool despite it bein' a bad signal. Danke Jan!
Haha, danke. I was rather surprised to get a color signal from the Model 1. :)
"Let me take a couple of seconds to thank my... [skips 30 seconds)"
Next year, MaySX!!!
Hi Jan. You only show it briefly, but it *appears* you're using my TRS-80 Model 1 power supply design. (github.com/calphool/TRS80MXS)
I'm interested, was there a flaw in the design itself, or just a mistake in how it was constructed?
Das ist mal 'ne harte Nuss...
We won't get any sparks or dangerous behaviour from the power supply... probably.
Should have been Spectember
TRaSh-80
I think you are way too hard on yourself. No need to be embarrassed that you missed a bad solder connection. It happens to everyone! The key is, you found it!
not called the trash 80 for nothing lol
His neighbor is boring ;-)