IBM PS/2 Repair & restoration (Type 1)
HTML-код
- Опубликовано: 8 фев 2025
- PCB Prototype the Easy Way. Full feature custom PCB prototype service. www.pcbway.com/
Support me on patreon.com/Epictronics
Buy me a coffee: ko-fi.com/epic...
Join me on Twitter: / epictronics1
-
In this "IBM PS/2 Repair & Restoration" video, I dive into a challenging vintage computer repair project. Watch as I restore and bring new life to an iconic IBM PS/2 Model 70. If you're passionate about retro computing, vintage tech, and hardware restoration, this is a must-watch for you!
-
Tools I regularly use
FNIRSI DWS-200 Solder station
PINECIL Soldering Iron
PinePowerPSU
AMTECH NC-559-ASM Flux
Kester 951 Flux pen
Hanstar 861DW Rework Station
Pro'sKit SS-331 Desoldering Station
Multicore 60/40. 0.38mm and 0.5mm solder
AOYUE int883 PCB Preheater
MaAnt Grinding Pen
RIGOL DHO800 70MHz four-channel digital scope
Tektronix 2246A 100 MHz four-channel analog scope
FNIRSI DPOX180H 180MHz Handheld Digital Oscilloscope
UNI-T UT61E Auto Ranging Multimeter
UNI-T UT890D Manual Ranging Multimeter
MESR-100 mk2 ESR meeter
TL866 II Plus Programmer
InfiRay P2 Pro Thermal Camera
DeoxIT D5 Contact Cleaner
PCBs from PCBWay.com
--------------------------------------------------------------------------------------
Support me on patreon.com/Epictronics
Buy me a coffee: ko-fi.com/epic...
Join me on Twitter: / epictronics1
Music by Karl Casey @ White Bat Audio
#powersupplyrepair #retrocomputerrepair #ibmps2restoration #ibmps2repair
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
You have to drill a hole right where the crack ends to stop it from spreading. This relieves the stress. The crack will continue to spread under the glue. It's not being caused by flexing.
Yeah, I've used that trick on cars. I'm sure it works on PCBs too
hi Aaren, what a lovely machine is that were you working on. Great job. I will look forward for part 2. Greetings from Steven from the Netherlands
The tiny capacitor (electrolytic) serves as boost (kickstart) for the main IC of the switch mode power supply. When it is dried, the smps would not start at all.
I think a full recap is probably the best way to move forward then
The orange caps are ROE for Roederstein. They were thought to be very high quality when new so they were not chosen to save money. Sometimes apparently high end companies will make disappointing products.
I think they are highly regarded in the hi-fi audio space. They were probably great back then
@@Epictronics1 A friend of mine used them in audio applications and they didn't live up to expectations. The technicians ended up replacing them in every unit that came back for any reason, although this is just what I remember from decades ago, so don't regard it as gospel.
@@jozsefizsak Just like any capacitor manufacturer, Roederstein made many different series of caps, with some of them aging badly and others aging quite well. Same even goes for high-end japanese brands like Nichicon; while Nichicon stuff is usually great, some series (e.g. PL series for PSU output caps or HM series for mainboard VRMs) are notoriously bad, causing leakage, high ESR, bulging etc.
The TI automotive rated power transistor you found in the power supply is a good thing since an automotive rating is given for parts of higher quality that are deemed to be more reliable (and usually able to take more severe punishment) than their non-automotive counterparts. They are also a fair percentage higher priced parts, a consequence of the automotive binning. (Edit: Should have mentioned that an automotive rating is considered better than industrial, and typically better than even mil-spec parts. Made to be punished by a vehicle for a couple hundred thousand miles!)
Dust bunny lived inside that ibm
Check the switch control IC output, if no signal check if the IC is getting power on VDD pin. There should be a 22-50 mF 50V cap between IC VDD pin and ground, it is good idea to replace it.
Thanks
7 minutes in - the LM339 is a quad comparator so a pretty good bet that this is the power good circuit
That was a pretty good guess :)
@Epictronics1 Not really - I had a power good problem on my Compaq Portable III - failed LM339.
BTB08 is an 8 amp triac (two SCRs in anti-parallel). Very common part to use for AC switching circuits.
And triacs are not testable by transistor tester, if I remember correctly.
The paper insulator may actually be Nomex sheet in which case it is good for the voltage
Ok, feels a bit cheap compared to the plastic sheet in the Astec branded PSU
40:43 - hFE 0 ??? I never saw anything like that.
Nice long video, surprising that you worked on IBM today haha
he ROE caps are a bit strange, some short and go bad, but others seems "better". I never saw any leak to be honest, but I dont like them.
This was a disgusting PC. Good cleaing effort.
Transformators often have a fuse (heat protection) in it - not sure if that is the case here but it is strange that it does "nothing". If it is a switch mode, it often needs a tiny cap to be started, so check maybe the smaller values those are more likely to go bad.
The thought accured to me after I had measured those caps, Can ESR be too low? I may actually just replace all the caps before we move on. Thanks
That thing looks like it came from a crematorium. 🔥
TYN116 - Category Discrete => Thyristors => Silicon Controlled Rectifiers (SCR) 16 A, 100 V, SCR, TO-220AB
Thanks!
DX2s are internally clock doubled. You don't change the crystal.
IBM were selling a huge number of these machines to corporate customers - they probably maxed out the production of the good power supply and had to fall back to the cheaper one.
That's quite possible. Maybe Astec couldn't keep up
40:57 why is it showing hFE 0? I think it's either wrong (lacking drive current) or the transistor is bad
BTB08-600BWRG - TRIAC ALTERNISTOR 600V 8A TO220
Thanks!
First thing, try to use yor old component tester. TYN116 is a tyristor, it should have been identifyed. BTB08 is a triac that some testers manage to identify and test. The transistor with hfe=0 and the rectifier that sholud have been identifyed as a double diode.. The tester seems scketchy.
The powergood circuit is made by the 339 comparator; it reads the 5V and maybe otrs voltages and outputs a 5V to the PowerGood line to let the motherboard know that it can boot. So yes, the PG fault was indeed that crack in the pcb.
For the power supply, measure the voltage across the big capacitors, if there is a DC voltage, the fuse, switch, diodes, and all the AC circuit is good. Beware, there is rectifyed mains so, high voltage. Check the datasheet for the main SMPS IC and see if it gets power, if not, check the circuit, may be a dried electolithic, may be a cracked resistor or a bad solder joint, a cracked trace on the pcb. Maybe use a 100W incandescent bulb in series with the PSU to limit the current.
Maybe a 40Mhz crystal would work since the P70 came with either 16 or 20mhz cpus...
I guess, that's what we'll try next
The large caps are ROE branded. They are old German made caps. They suck for the most part. So do the light blue plastic wrapped Axial Phillips caps~!
That last one was a double diode, so should not be a resistor. Are the leads correct ,ie, 123, because it looks 223.
Don't trust while resistors, they are wire wound or fusible resistors, ie, they blow open.
what kind of material are those clips. It's too rigid to be TPU. I'm thinking PETG maybe? PLA would melt too easily and probably crack too easily.
You mean the new replacement pegs? I ordered these to be made in ABS
What is that software on the very first frame of the video?
I don't know, but it looks freakin awesome
Great video
Thanks!
What frequency do the add in cards run. Can it be changed in settings?
I don't know actually, I need to check
"How disgusting!" You know, not everyone lives in a lab clean room...... 😉
I actually HAVE gotten a piece of gear FROM a lab clean room.... IT was astonishing that there was not one speck of dust in a 20 year old device, not even the fan!!!