Let's talk about RTC options once again
Вставка
- Опубліковано 1 гру 2024
- In the past I made some videos about RTC module repair and replacement and also open-sourced a drop-in replacement for Dallas DS1287, DS12887, Benchmarq BQ3287, Odin OEC12C887 and many more. And in this video I would like to share, which options do we else have.
My old videos about RTC module repairs:
• Ultimate RTC module r...
• Revised: Repairing Rea...
• Repairing Real-Time Cl...
Related GitHub Projects:
github.com/nec...
github.com/nec...
Music by Model Povedeniya
modelp.bandcam...
Patreon:
/ necroware
I now feel more educated than 24 minutes earlier.
Awesome!
@@necro_wareyou should be able to cook the resin casing in hot water too. Just be sure to put some glass shards on the bottom of the pan, so the chip won’t touch the bottom of the pan and burn itself.
The same could be done for the C64 power supply.
This man needs to be protected at any cost.
Now, question, would be possible to instruct an MCU to pretend to be an RTC chip?
Yes, of course. I already investigated that option, but it would need quite a lot of work to replicate all timings and compatibility hacks. Currently it is probably not worth the effort, especially where old modules can be reused in the shown way.
Clone him
@@emuboy85 In the Mac world, the custom 344-0042B RTC was replaced by a pin compatible ATTiny85.
@@ukmk3supra I think it'll get there eventually on the PC side, but for now this option is the very definition of "good enough" in the sense it's cheap, flexible, and relatively easy even for a novice hobbyist.
Awesome discovery about the temperature-specific removal of the resin body!! Heard it here first!
Same! Thank you!
Sorry to hear people gave you grief for something you didn't even sell them. You would think that someone with the requisite knowledge to seek out such a module would have something other than air between their ears.
This is a great tip though. Even without the mainboard spaces for battery and oscillator, that the modules can be disassembled and modified rather easily is good to know.
I was shocked by this. It even isn't so much money to lose, so anger and frustration should lead them to such rattish move.
They were probably full of anger of losing money and doing desperate things.
CUSTOMERS!!🤦 The other day they shouted at me because of the invoice amount from a company which sounds similar to the one I worked for. Even when I was the service engineer. Obvious violation of rule 142 "never mess with the service engineer"😁
Ugh these Dallas units make me shiver. I remember years ago I had to replace one of their SRAM units in an old digital HP oscilloscope. Maxim bought Dallas, but they discontinued that part. I had been emailing and calling them for a while, they didn't have the datasheet of the module, they got rid of a lot of the documentation of Dallas as a company in general. I took an old chisel to the module, managed to uncover what was in there. An SRAM chip, a battery management device from Dallas as well, and a button cell battery. Using the schematics of the scope I uncovered the pinout, within notime I had PCBs manufactured and now I'm slinging them to everyone having the same scope so they can keep them alive as well.
I understand it's fun to cast crap in epoxy into a neat little module, I love that way of thinking when it comes to analog hybrids, RF amplifiers, etc. But a battery??? These guys, I have no words for them.
It's just a product that you would presumably replace on roughly the same timeframe as a normal CMOS battery.
And the on-board SRAM allows you to use it to store CMOS settings or other parameters.
Best PC tech on YT. I never thought I'd watch 24 minutes of RTC content until this jewel. For the millionth time, well done sir!
You've surely rescued a lot of machines with this research, thanks!
Thank you very much!
Yes I just tried it and it worked. I used my hot air rework station; had to set it to 190F (I realized after a while that the 80 degrees you were talking about were Celsius; 190F = approx 88C) and used a screwdriver as a chisel. I found that the greatest danger is mangling the chip's pins if you use too much force and it slides from your grip. So after a little while I decided to insert it into a 24-pin sacrificial socket to protect the pins, and that worked great. I decided to leave the crystal oscillator soldered to the bent pins, and run a couple of wires to a remote battery holder because on the motherboard I want to install it, the height space is at a premium and I want to make it as slim as possible (It prevents inserting cards in a couple of PCI slots).
I had to heat up mine to a higher temperature (270°C) because otherwise the epoxy wouldn't break. I think my hot air station isn't powerful enough (I say this because I didn't even melt the plastic).
Also yeah I had the same issue with the pin mangling. I think we should make some sort of tool to do the job properly and efficiently.
Some sort of big clamp that would hold the chip firmly in place while letting the hot air flow on it, and maybe also something hard to protect the pins
@@DxDeksor That is what I was wondering when I saw this. Might it be easier to use a socket anchored to a board as a jig while you heat it and lift the shell off.
Your desoldering technique is absolutely next level 🤯 looking forward to seeing the C64 psu depotting video (please do this!) 😁
Even if I never have a need for one of those modules myself, I value the contribution you've made available for free to so many people.
Thank you your upload! Just finished a conversion of a Dallas Rtc using the heatgun. Sacrificed a dip24 socket to save the legs of the crt ic from bending. Put the crt ic to the socket and push the legs of the socket invards. Used higher temperature - 200C degree - for shorter time to soften the outer plastic. Come off easily with the screwdriver. The resin needs a little bit more heat but it come off by chunks applying heat again after a chunk removal. Piggy backed a 2032 holder and done. The crystal and the IC survived the higher heat, it works fine. Thak you again, no mess at all.
awesome! thx for sharing this with us! 👍🏻
Oh, Peter! What a guest here! Glad to hear from you :)
WOW! am an engineer who never knew this heating trick! You are a great teacher. Thanks..
I find most "engineers" know a whole lot less than they think they do. Perhaps an element of humility is in order here?
What can enhance a Friday? A video from Necroware.
Greetings from Germany
Very interesting, I really didn't think it could be so easy and clean to de-pot those RTC bricks.
Speaking of companies doing bad things, a notable troll move to me is when a motherboard uses a CR2032 but the soldered version of it (like Compaq did for example). Yes it's trivial to replace it with a battery holder but it still requires tools and some skill
Soldering ensures no bad connection.
Soldering means a solid connection. Look at any pinball machine and you will quickly see that connectors are nothing but trouble. It's actually easy to remove a part that is soldered in. Smart people get some tools and learn. The rest (90%) throw it away and buy new.
@@g4z-kb7ctit makes sense for something such as a pinball, that is not meant to be inspected by the average joe and is going to be mistreated for its whole life.
It makes less sense for a PC where it's already filled with connectors.
You fear the battery is going to loose connection in a machine where the CPU is in a socket, expansion cards are in slots, memory is socketed or in slots as well and the hard drive is connected to a pin header ? 😁
I guess it could make sense in a POS-style computer made for an industrial environment though
@@DxDeksor I think you've misunderstood my comment. Pinball machines have dozens of connectors and fail very often due to bad connections between the boards or other parts (switches etc). Connectors in general are a bad idea for anything that is being used or abused by a human.
@@g4z-kb7ct ah yes maybe, though what I meant is that I agree that for something abused like a pinball machine does indeed suffer from having connectors.
However I think for a PC, it's just fine
When resoldering pins, I saw on another channel that using the leg of an LED works great. It's already tinned, and fits in a socket perfectly. And the LED is even unharmed just with shorter leads. The LED works like a handle while you're soldering, too ;-) I just thought that was a neat idea.
When I build some kit or do some other soldering with a bunch of new through hole components I usually save a handful of cut legs exactly to have replacement pins for such repairs
Yes, I have small can full of old legs from all kind of parts for such things. That is of course also a good option.
@@necro_ware That is exactly why I collect cut legs from all sort of components. Especially resistor legs (1/2 watt or more) and electrolytic capacitors I find especially useful. That doesn't come without dangers though. Just this morning while walking around in my work area in the basement I felt a pain in my leg, and when I looked I had one such capacitor leg stuck in one of my toes. :P Must have missed it when I collected the leftovers from the motherboard I did a recap on a couple days ago.
@@stamasd8500 Magnet is your friend, I always keep one big old speaker magnet just for that purpose on desk, these legs are useful when building any prototype circuit.
Awesome discovery! I hadn't realized that all these RTC ICs are not available anymore. A few months back you could still find the SSOP-24 ones in large quantities, but not anymore! This heating trick is a great alternative, thanks for sharing it.
Thank you for open sourcing your RTC replacement. I've had great success with using the TI BQ3285S-SB2 with your design. It's still made and readily available from reputable sources. No hacking old parts needed. :)
I just tried a variation of this, without a heatgun, instead submerging it in almost-boiling water. It worked great!
Great video. Old module disassembly trick using heat is very unexpected! Thank you.
I've watched this video a number of times just to savour the elegance of simply removing the potting from the module. I doubt that I would ever have even thought about applying heat
I see Necroware, I click
Ditto 🦉
So do I
same here
Thank you so much for another brilliant solution NW!
Your videos are best of the best in retro computing... a hobby that more and more people come to love
I have several dallas modules around and I knew I shouldn't throw them away :)
Very cool video! It's one of the best channels about old computing. Hugs from Brazil.
I was going to try this method but I thought the epoxy would just melt and make a huge stinky mess. Thanks for diving in and giving it a try and man what a great result. 👍
I'll have to clean up the RTCs I have. I tried to remove the epoxy from a Dallas RTC by soaking it in acetone and it kinda works but not really worth all the time and effort.
Nearly all potting material is thermoset and has the properties described in the video. The only common exception is bitumen, it will melt and relatively cleanly leak out.
I have the exact same board. Guess what I'll do this weekend?! Great content. Thanks for sharing your adventures.
Wow! I like how you solve problems =) Very good user guide to safe RTC modules
Every video you post is a gift of knowledge and learning. Thank you, Necroware. :)
I really can't stand how effortless and easy you make soldering / de-soldering look - could you at least pretend to struggle a little to make the rest of us feel better 😝
Excellent video. Will be sure to try this with the pile of dead RTC chips I've amassed.
😂 thank you very much. You just not see me crying out of fear behind the camera. That's why I always have to do it with my eyes closed.
Amazing how cleanly the chips came out!
Excellent discovery! Thank you :)
I heard about this in the last episode of Adrian's Digital Basement with the Plexus!
I easily have 3+ dozen of these darn things to modify or replace, some soldered in, some not.
I've been ignoring the problem for years as doing the drop in replacements on everything would cost me a pretty penny in RTC chips alone and I was not comfortable coating an area in plastic dust from my dremel or leaving the original batteries in situ for the next 3 decades to short out or cause other problems if moisture should creep in.
Thank you for figuring this out, it sounds like I should build myself a little fume extractor, grab a pile of motherboards, extract the easy ones and decide if I am going the CR2032 route or do something closer to the original battery and buy a huge quantity of cheap coin cell battery holders. I will have to get creative for the network gear and analyzers that have all their stuff shoved in a card cage though, since height will matter.😅
Another fine video Sir, At: 8:52, Sorry to hear that people gave you grief for something you put your valuable time & effort into without making any money, Some people eh! 😔 Anthony - Birmingham/UK 🇬🇧
just tried this. it works great! not overheating the plastic is an important tip. thank you!
Thanks! Pretty cool tutorial!Just a few weeks ago I have salvaged all the RTC chips from crushed boards found on ine of the scrapp yards. I took dozens of bords , but many others were already in the FUBAR condition so I salavged all removable parts and now I see it had been a good idea! 👍
Again, I start thinking of a solution, never trying my theory, and Necroware just YOLOs it. Well done, sir!
Excellent! I have nothing more to say! :)
Definitely going to attempt this with the drained Dallas/Benchmarq modules I have laying around. Didn't think the potting compound would go brittle like that with heat applied.
I just tried it and it worked a treat. TBH it doesn't really become brittle, it is more like it becomes malleable and soft, you can pry it off like putty.
I love this sort of "finding new solutions to old problems" stuff! Thank you!
Thanks for the tip! I managed to replace the battery without desoldering the module from the board. I used a hot air soldering tool.
I recently had to deal with a Dallas DS5000. I had no idea that such a part existed! Fascinating little device!
A DS5000 is a 8051-compatible microcontroller with a battery-backed RAM inside. When the battery dies the RAM also clears and the code that was in there is gone forever. They are designed to self-destruct on purpose as a security device. We (as in MAMEDEV) had exactly the same issue when emulating/reversing the Gaelco arcade game World Rally Championship. Eventually the manufacturer provided the code and the game was emulated perfectly :-)
@@g4z-kb7ct ooo, interesting! I do know the System 16 boards, for example, with the dreaded "encryption modded" 68000. Seeing an arcade board with an 68k AND a DS5000 is quite cool :)
I was very lucky in two ways with the device that contained the DS5000. The Battery wasn't dead yet, AND protection was not enabled! The manufacturer even designed the thing in a way to enter programming mode via the DB9 serial port it had (Well, like the datasheet recommends...). So I managed to get the configuration and memory contents out, hacked on a new battery (luckyly there's an image of the unpotted chip, so I could confidently dremel away.)
I was quite stunned to find a device with such an advanced set of features from that time!
- programming directly over a serial connection by simply sending an Intel Hex file.
- Partionable Memory for different ROM/RAM configurations
- Program readout protection (seemingly implemented in a way that even physical access to the SRAM Chip would'nt help)
- integrated RTC
The Datasheet and the document about this line of secure microcontrollers is a quite interesting read.
FYI: The original programming software "KIT" runs flawlessly in DOSbox-X (tested on a Windows 10 host and FTDI USB to RS232 adapter)
@@g4z-kb7ct yes, I'm quite familiar with the DS5000 now :)
Interesting to see an arcade board with a normal 68k and the DS5000. I do know the System 16 boards, for example, with the dreaded "special sauce" 68k...
The DS5000 totally surprised me with its feature set, regarding the time when it was created. The Datasheet, and the "Secure MC Userguide" by Dallas are a quite fascinating read.
I was doubly lucky with the device I got, because the battery wasn't dead yet, AND the encryption was not enabled! So I dumped the content, installed a CR2032 holder onto the DS500 and re-uploaded the formware. (I even could merrily dremel away at the part, because there's a photo of the unpotted DS5000 without the battery and the 32kHz quartz).
I even found another guy with the same but dead device, wich he could revive with my firmware dump!
Initially,6 I did the backup/restore via a nortmal terminal program, but I can confirm, that the original config/programming/terminal software "KIT" runs flawlessly in DOSBox-X (Host machine was Windows 10 with a normal FTDI USB to RS232 adapter)
You sir, are amazing!
This is pretty incredible for those with weird butt motherboard that have non-standard RTC dallas chips. So clean, and I'm all for getting rid of the old battery hiding in the resin.
Amazing work 👏🏼 I really love how you find ways of reusing or recovering parts, I think this is game changing in restoring the reputation of the Dallas RTC chips, very salvageable and useful.
I think it is great the work you have done to make the replacement modules. I have seen others revive long dead systems with them but have yet to have a need to try it myself though I probably will at some point given the old systems I work on. Sorry you catch flack the times they don't work for people. Matching up the exact right part in something decades old in every situation is complicated. The work you are doing is appreciated and please keep it up.
Very useful information, I'm going to use the heat trick as I found out two of my older desktops/towers in the attic has one of those nefarious RTC modules.
Subscribed. You are explaining very well and distinctive !
Building some of your RTC modules has been on the todo list for awhile now, the biggest barrier being acquiring the BQ3285 IC's. After watching this I went into the drawer full of dead Dallas/Benchmark RTC's and revived a couple using this method. I used a sacrificial socket and my hot air rework station, a pair of old needle tipped tweezers worked wonders for chipping away at the epoxy. While I still want to build some of the RTC modules you designed, I do love being able to recondition existing RTC's with the parts I already have on hand. Cheers!
That's because the RTC (chip) isn't dead, just the on-board battery.
Excellent, very informative - and well put together. :)
Thank you for the video: the care and effort you put into finding ways to keep these old systems operating is heart-warming.
Now apologies in advance for the comment that's coming: please don't take it personally.
"Crystal oscillator" is the circuit which generates a periodic signal at a stable frequency. The component in the metal can is just a crystal (quartz crystal) and the inverting amplifier (or inverting logic gate) is inside the chip, across the two pins marked as OSC1/OSC2 (sometimes OSC-IN and OSC-OUT). It's a bit like having a clock with a pendulum and calling just the pendulum "pendulum-clock".
yeah most people think a crystal and an oscillator are the same thing lol. on the usual selling sites with smd 4 pin types they often say active / passive but most people don't know what that means either. The active ones are true 4 pin oscillators and passive ones are crystals with 2 pins not connected.
Yes, you are absolutely right. I even explained the difference between resonator and oscillator circuit in one of my own videos a long time ago, but I keep saying this. I have to stop doing it indeed. It took me also a while to stop calling the electrolyte in the batteries acid, despite that I know, that it is actually a base. People were always confused why I neutralize acid with acid, when I was using white vinegar. I'll try to improve next time ;)
@@necro_ware not to worry! 🙂
Great job finding the best way to remove case and potting from the RTC modules. The longer I think about them the more it seems they had a really daft idea at the time.
Great video as usual. Keep up the great work 👍
This is truly amazing, well done.
Awesome. This whole episode is full of gems. The heat trick was unkown to me. Thanks a lot!
Before you mentioned it, I had the same thought about the C64 power brick. That's a nice trick with the heat.
Many Thanks for this lesson. I always learn something new from your videos
Awesome stuff. thats is going to be a much neater option than using the dremal. Keep up all the good stuff
Lovely content, as always, friend.
I've lerned a lot, thank you Sir!
thank you very much for sharing this method, it's so much cleaner than the drilling method. i have just tried it and it worked great. i had to heat my module to around 100°C though. at 80°C i kept hammering at it for 20 minutes with my screwdriver, but only little bits came off. after turning up the temperature it came off very easily.
Great! Thank you too.
Amazing discovery. Great for a heavy-user of your solution to have the initial option (although I do love your circuit). So cool to be able to get rid of that horrendous resin. I felt sorry about you, your dremmel, and your workplace on the "basic attempt". It had to be quite hard to clean it. Thank you for another impressive video. Paraphrasing @emuboy85, "this man needs to be protected at any cost".
I've used this technique to replace the battery on ST M48T59 modules used by Sun SPARC systems, though it looks like you practiced more than I did 😀. Those chips have "internal" contacts for the battery and crystal on the ends of the package, rather than pins that were bent up on the side.
Yeah i've done exactly the same. It was really nice of them to make the contacts at the very ends on the bottom side. Really easy to attached an external battery and save $30... wow they are expensive hehe!
Remember seeing this when it was open-sourced, but failed miserably at soldering it. Designed a small carrier board to go on top of a DS12885 with pins bent up and soldered to it, but didn't make it public as I had no idea if it'd be compatible everywhere - never knew it was exactly the same chip inside 😂
Excellent video as always, seeing yours always make me want to tinker more with old PCs 😀
Thanks so much for this video. I recently picked up a PC with an ODIN Dallas RTC clone. Thankfully the motherboard has the spaces for the oscillator and CR2032 holder, so I'm going to be using this method to rebuild the timing circuit. I'll probably put out a video on my channel if successful (with full credit to yourself of course) and I'll let you know how I get on!
Great if this could help! Thank you for watching too. I already repeated this not only for Dallas, but also for ODIN and Benchmarq modules and it worked fine for all of those. On note, the crystal in the module is soldered directly onto the pins of the IC, but depending on the distance between the crystal and the RTC IC on the board the oscillation can get unstable. To work against that you can solder 10MOhm resistor between the pins of the crystal. Usually mainboards already have a place for that. Good luck.
That is great knowing how to salvage the ICs from the epoxy.
Great stuff as always!
Great video
Thank you for sharing!!!
Congratulations, that is a rewarding result.
Very well explained. thanks!
Awesome, the Dallas chip was a black box for me until today.
Great and valuable information.
This is amazing and a perfect solution. Thnx!
thanks for the information and demonstration of use
Very interesting instruction, thankyou 👍
Amazing trick about resin becoming brittle with heat! So those dallas can be recovered as you showed towards the end! It's kind of silly to see how they put them together, bending pins and flooding everything in resin! ahah! Thanks for showing us your findings.
PS: I'm thinking of buying one of your nwX287 but I think it won't work, can I please have a refund too? :D :D :D
LOL, of course! If you buy one from me you will. There is one problem though, I don't sell any :) Thanks for watching.
If it doesn't work, sell me.... I just need one.
Very cool. 🎉
This is very cool! Planned obsolescence has been a think since the early 90s I guess!
that happened way before 90's. Started at least 20-30 years earlier.
Excellent Stuff..
There are several "bugs" carried through the MC146818 and derivatives - Looking forward to a deeper dive.
That dive would be probably too deep. I'm certified only for sports and hobby diving up to 45m ;)
Beautiful.
Good video, thanks.
Awesome job! I regret throwing out my old rtc chips now 😅 could have reused them if it's this easy to get to the parts.
*THX 4 vid*
Chemical solution: Ethanol, paint thinner, adhesive remover. Heating a litte bit make the process faster.
*Use protective equipment*
Test this way.
I found an old post that says to use 1-methyl-2-pyrrolidinone. I bought some on eBay, soaked the chip, and a week later I came back and all the potting was dissolved.
@@owenmann1
THX 4 hint.
But in EU this in no more allowed in EU, in Germany look:
"The restriction stipulates that NMP may not be placed on the market as a substance or in mixtures in concentrations ≥ 0.3% from May 9, 2020"
Paint and paint thinner are available without restrictions in Germany and can be handed in at the recycling center (municipality).
Sry 4 my bad English.
15:30 I've extracted a few C64 transformers from the case by heating and chipping away at the brittle epoxy and also extracted security modules or microcontrollers on arcade boards the same way. It works well but you don't need to remove all the epoxy. For the C64 PSU just remove the pcb and replace with something better. I made my own SMPS board that fits inside. The transformer works fine as-is with epoxy around it. The wires are visible and just need to be connected to the new SMPS board with a direct pass-through for the 9VAC. I also drilled into a ST M48T58Y rtc module (used on a lot of Konami arcade games) and added an external battery. Inside is a special chip and a crystal and nothing else so they can't be built from parts. These M48T58Y chips are incredibly expensive now, over $30! From the factory the battery is disconnected and on first use or testing the battery is automatically connected. But if some seller tests them 10 years ago the battery is then working and when they put on the shelf they will be dead 10 years later even if New-Old-Stock. Best to drill into it and attached a battery holder on top.
Thanks for sharing your discoveries.
What about annealing or heating the pins, so that they don’t break when re-bent?
This reminded me of how I cooked transformer on kitchen stove to replace its winding 🙂
I recently cam eacross a couple of PcChips M560 boards, socket7, which have a sort of an intermediate solution for the RTC. They have the black boxes on them, but also a separate crystal oscillator soldered next to it on the board. There is no footprint for a battery though. The RTC modules have fewer pins bent up than a regular DS12887, in particular the pins for the oscillator are not bent. And in fact the modules are not marked at all except with a "burn-in: sticker. I think they were custom made by the same Chinese company that made the motherboards. In fact, on one of them they forgot to pour the epoxy in so the plastic cover came out freely. :) Inside they have only a Via VT82885 chip with a soldered battery, no crystal of course. Made it so much easier to fix. I will try your hot air method on the remaining one.
Thank you!
Brilliant, as always. AA++
The resine removal trick is something I have to try.
It looks like you could create a 'carrier' board that you put the dallas chip onto that just adds the little battery and crystal like your existing replacement. That'd look sleek and remove the need for board modifications.
Mr, very good!... I already had a module like this that I managed to disassemble and change the battery, it was on a board with an Intel vx chip, if I'm not mistaken, it was with an Intel 233mmx
Vor 15 Minuten online gegangen, das Video dauert 24 Minuten und schon 29 Likes. Das nennt man wohl ne treue Fanbase.
Absolut. Er ist halt auch einfach Klasse.
This is awesome, I didn't know the resin became brittle from heat without melting the components in the resin.
I have several dead modules as well and I think that might be a cleaner method than drilling/filing.
I am curious about the recency of manufacturing. I ordered a pair of the Dallas modules from Mouser (a reputable electronics supplier in the US) and received new modules with 2023 date codes. I wonder if there is a small operation still making them for industrial equipment.
I was also surprised to see how good the heat worked. In regards of the new modules, I heard, that they are still produced somewhere in Philippines, however I was not able to find any in Germany. I could order some of those from U.S. but the shipping costs and taxes were too high. For some rare mainboards it is probably worth it, but in such cases I prefer to use my nwX287 module anyway. For everything else, extracting the chip from the old module is a great option imho.
Yes new modules are still being produced. You have been able to buy them from Digi-Key for at least 10 years. I bought one back in 2015 and also recently. I did a little research on this and Analog Devices owns the tech now as they bought Maxim previously. These ICs are used in the Industrial sector so I am sure that is the consideration here. They also still manufacture the 12885 and if someone wanted they could easily bend the pins and add the XTAL and battery holder on the back. But yeah the video shows a great solution to be able to reuse the old modules.
It's confusing that the new manufacturers continue to make the modules in such a way, they own the IP to the circuit so they could easily make the same design as Necroware's and save themself money on the assembly line too!
@22:36 That's my preferred option. Best part of it: Depending on the mainboard layout, you can add the missing battery pin on a soldered 12C887 module without even desoldering it from the mainboard. Just drill it open, clean the one soldering pad, push a copper wire from the back side of the board, solder it to the pin in the module and solder it to the mainboard. Then add the battery holder and you are done.
For me at least, drilling precisely and cutting the internal connections with the spent battery is at least a 1h job. I just tried this method and it was only 10 minutes, and much less precision and stress were involved.
I have had success with other potted chips, heating them up as you described, the placing directly on ice in the freezer, usually the thermal coefficient of the chip material is different to the epoxy compound, giving a good, clean cleave line at the junction. Maybe this will give you good results without having to use force that can mangle the pins 🤞
God, you´re good!
❤❤❤
The resins are all formulated different and have different temperature endurance. The problem with C64 PSU is that the potting resin is potted into an ABS enclosure. ABS has an unpleasant trait that it starts to go a little floppy at around 100°C and simultaneously starts chemically degrading as well. ABS is injection moulded at temperatures around 200°C maybe a little higher not sure (they use lower temperature than we use for 3D printing it and a lot more pressure) and at the injection moulding temperature, it survives a total of vaguely around 20 minutes before it permanently degrades and becomes crumbly, and ours are already pre-degraded from decades since synthesis. Worth a try but don't expect it to go as smooth. So you might save the transformer, you might not save the enclosure.
I think a solution for the future might be a module with a microcontroller and an i2c RTC/NVRAM battery backed IC. Microcontroller would be mainboard powered and provide interface when the PC is running.
That's true, but it is not too complicated to get the resin block out of the case on the C64 PSU. If you go with a knife along the edge and wiggle the walls of the the enclosure a bit, the block of resin will just fall out. This is how people usually fix those PSUs. The internals get removed and a new transformer with a DC-DC voltage regulator goes in place of the old resin block. Once you have that resin block out you could try to heat it up and see if it reacts in the same way. (Un)fortunately I have currently no broken C64 PSU at hand to try.
@@necro_ware Cool didn't know the bond between potting and casing was weakened by now. I have one Commodore PSU and it's entirely healthy so i'm not very inclined to go poking in it :D
@@SianaGearz I think his point is that the resin and plastic aren't necessarily bonded, just a very tight fit. They probably used a mold.
0.3V sounds perfect for our clima neutral lemon based power supply.
Clima neutral does not exists. It is a politician's fairy tale.
Yeah, unfortunately the lemons didn't get into the video this time ;) May be another day.
If you cut the lemon in 4 and use 4 pair of electrodes, and connect them in series, you can even get to 1.2V. That ought to last for a while longer :)
@@necro_wareYes plz provide an open source lemon holder.
@@AncapDude He shouldn't do that! then people would get back to him complaining that their lemon is too SOUR!
This is amazing, I would bet you could buy buckets of "dead" rtc chips for pennies to fix up like this.
you cool 😎 that is what i need