Искра 122 калькулятор ремонта

ЭВМ и ПЭВМ, производившиеся в СССР и странах СЭВ
system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 07.02.2021,22:52

With just a bench power supply I managed to repair the 7202618 PCB - I connected the supply directly to the edge connector, this is not ideal as Stefan pointed out above, but I don't have the parts for the transit adapter yet. Hopefully this explanation makes sense - I checked all the outputs of the K172TP1 chips, making sure that of each inverted pair one was 0V and one was "high" at around -12v or so (technically this is floating but it is a different state to 'stuck low'). Because there was no activity on the chips due to the board being isolated on the bench, I used fingers across the inputs of each chip or surrounding connected ones to cause random low level inputs - this is a trick I use on videogame repair because a floating or out of spec logic line can be disturbed by touch level capacitance. I was surprised that this worked on these too, it was enough to cause inverted pairs to flip. This allowed me to make sure that all 4 outputs on each chip were capable of 0v and -12v. It took some time... I found a K172TP1 where one output was nearly 0v all the time, whether the inverted version was high or low. I replaced this chip and it has solved the problem. As I was not sure my old stock replacement chips were working, I first tested them in the board using a turned pin socket, if pressed against the PCB these make good enough contact for basic checks without using solder.

In case this helps someone in future, here is a photo with the chip indicated in red (this is after repair, you can see the chip is a darker colour than the other TP1s), the failure of this led to a situation where all functions worked except negative calculations - for example:
2-4 = 2
-4 + 2 = 2
4 x -5 = 20
iskra_7202618.jpg
One fixed, one more to go (waiting for parts), I am also fully restoring the donor machine as it's cosmetically very nice.

system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 10.02.2021,17:51

Looking at the parts machine power supply today, not good. In fact all of them have burned looking wiring where the wires go from the filter box to the fuse and power switch, but the other two machines are clean aside from that. The parts machine had the power cable cut off, inside there are signs of electrical shorting, overheating or possibly fire. Large capacitor (I was told these should be removed as they are a common fault) already removed in these photos. There is black soot on the inside of the filter box, very bad.... Is the brown component a 270 ohm resistor?
iskra_122_77_psu1.jpg
iskra_122_77_psu2.jpg
iskra_122_77_psu3.jpg

Hunta
Advanced Member
Сообщения: 427
Зарегистрирован: 31.08.2020,21:16
Откуда: Самара

Искра 122 калькулятор ремонта

Сообщение Hunta » 10.02.2021,19:32

system11 писал(а): 10.02.2021,17:51 Is the brown component a 270 ohm resistor?
No, 270 kilo ohm with 10% precision

system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 10.02.2021,19:44

Also I don't know if this will be useful to anyone, I wanted to translate the manual for the Iskra 122. The manual I was sent was in dvju format, scanned images of pages making translation difficult. I have converted it to a PDF using some OCR software, it is still all in Russian but perhaps this will be useful to someone as text can now be copied / pasted. I will produce an English version later, but I am able to read this easily with machine translation now.

http://blog.system11.org/wp-content/upl ... al-WIP.pdf

system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 10.02.2021,20:28

Hunta писал(а): 10.02.2021,19:32
system11 писал(а): 10.02.2021,17:51 Is the brown component a 270 ohm resistor?
No, 270 kilo ohm with 10% precision
Thanks, this part measures OK. In fact even the large capacitor seems OK (I have a stock of new stock ones but will leave this part out). I can't find any shorts or faults that explain why this filter got so hot so far but the wire strands inside the filter output wires (going to fuse / switch) show that they were very hot and it was not just insulation discolouration. I think the sensible way forward is to reassemble with new wire, check for damage to the insulation on the filter coil wires (these seem to be single strand with insulation added manually). Then I can connect the power input to an isolated power supply with trip fuse (including connecting earth to the filter ground), connect the output wires to my multimeter and turn it on from a safe distance and see what happens... It's possible the problem happened in the past before the power cable was cut off and this was already repaired before (the solder was not original factory), but the filter casing was full of burned carbon.

system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 18.02.2021,02:30

Today I replaced the damaged wires and cleaned up the voltage filter. Unfortunately this power supply has another fault somewhere, I get stable -27V which proves the filter is ok now, but it seems like the other voltages are all dead. -100v, +100v, +6v, -12.7v all read just under 1v or -1v when tested against ground. Is there any common fault on the power supplies other than the filter? Power electronics are something I am very bad with, almost zero knowledge.

I tested all the boards from this machine in the fully working one, every single board is fine, keyboard is fine, displays are fine (one nixie tube takes a couple of seconds to power up), so my remaining faults across all 3 calculators are the bad chips on one of the PCBs (still waiting for parts to arrive) and this bad power supply. It's a shame this one is faulty as it's very clean and looks like it has had the least use.

system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 18.02.2021,16:38

Parts arrived, PCB 7102622 is repaired now - similar to the other board I was able to identify failed chips by very low unstable outputs even isolated with a bench power supply. K172ЛИ1 indicated in red had failed, this caused the calculator to crash totally even with simple operations like pressing + twice, any calculations that did not cause a crash would simply return 0 or 10.
repaired_7102622.jpg
I also had to replace the chip in the bottom left, but this was my own fault and nothing to do with the original problem - oscilloscope probe slipped and shorted normal and inverted outputs together on that chip, destroying it.

I found some UR-231 equivalent varnish in the UK, so I can re-protect both repaired PCBs.

The only problem now is that bad power supply!

Аватара пользователя
Кай
Почётный пользователь
Сообщения: 19738
Зарегистрирован: 08.08.2010,21:44
Откуда: СПб, Ульянка-Лигово
Контактная информация:

Вклад в сообщество

Искра 122 калькулятор ремонта

Сообщение Кай » 18.02.2021,17:21

Не по теме
But I wrote before: "Be oversure to check K172ЛИ1. This is the least reliable IC's from that series. Produced "at an edge/kneeve".

:)
Great result. Congratulations.

system11
Newbie
Сообщения: 22
Зарегистрирован: 28.01.2021,20:57
Откуда: UK

Искра 122 калькулятор ремонта

Сообщение system11 » 22.02.2021,03:10

Power supply is proving to be a big problem.

I lost a lot of time looking at the Iskra 122 power supply diagram - unfortunately it seems that is only accurate for the 1975 model. The supply found in my 1977 & 1979 machines looks like the one also fitted to Iskra 123 in photos on leningrad.su. There is some information in the manual for the model 123, but it is incomplete - no full circuit diagram, no part numbers, and some of the descriptive text does not match what I actually find inside.
iskra_122_psuopen.jpg
I compared working and non-working supplies, and I see the same -27V (DC?) input to TP2 on both power supplies, but on the bad one no voltages are output. I'm not sure if the error is my measuring of the input as the circuit is unclear but it seems like both transistors T4 and T5 are connected to coils in TP2, but since there is connectivity between them via the coil I can't say for certain that both are working. I measured ohms across the various coil windings - unfortunately not ideal as it is still in circuit but readings matched across working and non working supply. There is no documentation on component TP2, it outputs -100v, +100v, -12.6v, +6v AC voltages but the input seems to be 2x -27V DC. Inside the markings are only on the coil casings, there is an input coil section and an output but the numbers printed on the side of both match: M2000HM1 160577 - but I think the second part of the number is manufacture date. Hopefully this part is not the fault as it seems like it would be impossible to replace and my only real solution would be to approach a custom power supply builder to build a complete modern replacement. Interesting item of note is that two of the connector poles in the component seem to have been soldered more recently than the others. I cannot see any part numbers on transistors T4 & T5, these are probably printed on the front so I will have to remove them.
iskra_122_tp2.jpg
While I was looking at the two supplies I noticed the working one seems to have a leaking capacitor, the giant 2000UF 50V one. There are no signs of moisture damage to explain corrosion like this. Since the 1977 power supply has a rebuilt filter wiring and a good capacitor, it's tempting to unsolder the wires from the power supply PCB and swap them over to make one good one.
iskra_122_capleak.jpg
The good power supply is actually putting out 29v, testing with the bad supply trimmer resistor R4 can be used to adjust this.

Sergei Frolov
Advanced Member
Сообщения: 2820
Зарегистрирован: 05.07.2003,15:30
Откуда: Питер
Контактная информация:

Конкурсы

Вклад в сообщество

Искра 122 калькулятор ремонта

Сообщение Sergei Frolov » 24.02.2021,22:32

Try soldering over the electrolytic capacitors with same ones.
I've had a few problems with faulty electrolytes. The problem was solved in this way.
Собираю в коллекцию советские калькуляторы и компьютеры.
Мой сайт: http://www.leningrad.su/museum/ ICQ=12370106 https://t.me/retrocomps

Sergei Frolov
Advanced Member
Сообщения: 2820
Зарегистрирован: 05.07.2003,15:30
Откуда: Питер
Контактная информация:

Конкурсы

Вклад в сообщество

Искра 122 калькулятор ремонта

Сообщение Sergei Frolov » 24.02.2021,22:42

https://www.radiodevices.ru/esr/esr5.htm

I don't know about your country, but we have a wonderful device that determines the capacitance of capacitors without unsoldering them from the board. It helps a lot with repairs.
Собираю в коллекцию советские калькуляторы и компьютеры.
Мой сайт: http://www.leningrad.su/museum/ ICQ=12370106 https://t.me/retrocomps

Ответить