Elektronika MK-52: The MK-61 big brother?

[Sergei Frolov]

(02-10-2015 04:14 PM)Den Belillo (Martinez Ca.) Wrote:  Sergei; What functions do the 4 known ROMs preform? Were any of yours applicable to the Soviet space program?

I mean only mass produced or well-known ROMs.
BRP-Astro: Astronavigation module (not mass produced) http://www.leningrad.su/calc/docs/brp-2.rar
BRP-3: Mathematic software http://www.ru-hpc.narod.ru/calc/doc/mk-52_brp3.rar
BRP-4: Games and useful programs http://www.ru-hpc.narod.ru/calc/doc/mk-5...ns_doc.rar
BRP-Geo: Geodetic module (not mass produced)

[d b]

Cool stuff Sergei. Can you call them up to function as subroutines for your own programs in main memory?
If so, were you limited to something like a global label like in the 41 or could one break into it anywhere by line number or address?

[jhl]

(02-11-2015 12:47 AM)Den Belillo (Martinez Ca.) Wrote:  Can you call them up to function as subroutines for your own programs in main memory?

They're accessed the same as the EEPROM, so unfortunately not. I gather there was a switch on them to switch between expansion ROM and internal EEPROM.

As far as the other port goes: despite rumours about a printer, and external expansion controllers ("ВЗУ") with things like RAM, the definitive quote seems to be "Дополнительный разъем установлен с технологической целью и служит для диагностики изделия в процессе его изготовления." -- which says it's a diagnostic connector only. I suppose those EEPROMs might have been a bit finicky, given that they need more than 30 volts to take a write...

The port would be suitable for an externally controlled printer (ie. it could read all the registers and decide what to print), but it doesn't appear to have any way to disable the DD7 and so couldn't provide data to the calculator.

[Sergei Frolov]

(02-11-2015 12:47 AM)Den Belillo (Martinez Ca.) Wrote:  Cool stuff Sergei. Can you call them up to function as subroutines for your own programs in main memory?

You can download programs from ROM module to program area or memory registers depending of task.

[d b]

(02-11-2015 07:29 AM)Sergei Frolov Wrote:  

(02-11-2015 12:47 AM)Den Belillo (Martinez Ca.) Wrote:  Cool stuff Sergei. Can you call them up to function as subroutines for your own programs in main memory?

You can download programs from ROM module to program area or memory registers depending of task.

I never had a ROM for mine so I never knew. So are you saying that it can't run the actual ROM but needs to run a downloaded image of it in main? That sounds more Byzantine than Soviet.

[Sergei Frolov]

(02-11-2015 08:03 AM)Den Belillo (Martinez Ca.) Wrote:  I never had a ROM for mine so I never knew. So are you saying that it can't run the actual ROM but needs to run a downloaded image of it in main? That sounds more Byzantine than Soviet.

Yes, you need at first download program from module to user RAM (program or data area) and then execute or use it.

[jhl]

Thanks again jebem for your detailed repair writeup; it was very helpful in repairing my unit. The power supply needed recapping (all 4 large caps were the flag brand; all tested OK, but had terrible ripple under load). The switch in the power connector and the keyboard foam needed work too.

After I got everything else working, I found that some of the digits and some segments on the display were intermittent. All of the signals to the VFD were OK on the pins, and it changed if I twisted and pushed on the board - it turned out to be the lead-to-glass transition on the VFD itself.
The pins are connected to the traces on the glass with a black blob, then the whole thing is sandwiched with a brown glue and a piece of glass. I'm betting the blob is carbon-based glue and the brown is epoxy; it wouldn't budge after 12 hours in acetone, but some dichloromethane/methanol mixture dissolved it in a few hours.
After cleaning it up I mounted it back on the board with some 90 degree header pins and some carbon conductive adhesive of my own. By connecting the filament first, you can see the individual segments and digits come alive as you drop the glue on to the other pins! The filament pins need metal-to-metal contact, though, because the filament impedance is only 100R. You also have to be careful where the glue goes - a short between any two pins on the order of 50K will cause ghosting, and that's an invisibly small smear of glue.
I also found out the hard way that the front face of the VFD is also coated with a transparent conductive layer and connected to filament potential (-27V + AC). I assume this has to do with directing the electrons towards the rear of the device?

[jebem]

(02-18-2015 06:22 AM)jhl Wrote:  I also found out the hard way that the front face of the VFD is also coated with a transparent conductive layer and connected to filament potential (-27V + AC). I assume this has to do with directing the electrons towards the rear of the device?

Excellent additional information on how to repair this interesting calculator!
We keep learning from others all the time.
Thank you for that.

Concerning the high voltage present on the front face, I didn't realize it was present.
Well, we know the VFD works much like a vacuum triode, where the cathode (filament) must be negative in respect to the anode.

So your explanation makes sense to me, but other expert people on physics here can give a more elaborate explanation.

[clarke2001]

Hi, I am new here on this great site, I am a happy user/collector of HP and other RPN calculators.

Now I have a problem with MK-52. I was just repairing one (thx for the useful infos above). It was pretty quite hard, keyboard, ribbon cable, elko replacements etc. Now it works fine but, the K pre-button functions don't work at all. They produce the "famous MK-like" Error message. Only the K-0 (NOP) button doesn't.

It is not the matter of K button or other keyboard issue, because writing it in a program, the proper instruction code appears in program listing. For example: K-5 (3H) -> 32 i.e. sgn(). Running this tiny program, I got the Error message again, as in manual mode.

All other functions, operations, EEPROM clear/write/read are OK.

Does it make a sense ? Maybe I have a corrupted ROM ? Or is there any way to put in and run some diagnostic, repairing code ?

The machine is made in year '92. And i think not in Ukraine but maybe in Belorussia. There is no 2nd slot at rear side (on jebem's pictures colored yellow). Only the blue slot is built in.

An other issue is the power voltages: My machine cannot work with -15V, only at approx -13,5 V. The unstabilized outputs (-27V, -40V) are very, very depending of input power voltage. Lower batterie voltage makes higher outputs.

Thanks for reading this long stuff, any info, advise are welcome.

Laszlo

Update: 2015.08.23.

Problem solved. The machine (DC-DC converter and the unit) is somehow defective. I digged out another MK-52 (keyboard broken), replaced the electrolic capacitors in DC unit, and it works perfectly, the input (battery) voltage doesn't influence the output, but:

when I adjust:

-V2 -15V (pin6) -> IC logic circuits

the voltages are as follow, approx:

-V3 -33V (pin5) (instead of - 27 V) -> to the VFD display
-V4 -17V (pin8) -> to the -12V regulator for Ucc1 pin 12 and Gnd pin 4 of EEPROM KP1601PP1
-V5 -56V !!! (pin4) (instead of - 40 V) -> to the -33V regulator for programming voltage Upr pin 15 of EEPROM KP1601PP1

and the machine doesn't work after switch on

on my machine V2 must be set to - 13,5 V ( V4 = -15,5 V) for working at all, and to have proper V3 ( - 28 V), V5 (-43 V) values.

Production year '92 maybe explanation for that ?? who knows ...

I plugged the good PS unit into the other machine, but the K prefix functions are still resulting Error. Quite strange case.

But after all, getting out the repaired keyboard, ribbon cable of this, and using them in the older machine, now I got a perfect working MK-52, that counts.

[Tugdual]

Just a quick note to say thanks for sharing this information. I had never heard about these vintage Russian calculators until a minute ago and find them fascinating.

[jebem]

(08-21-2015 03:21 PM)clarke2001 Wrote:  The machine (DC-DC converter and the unit) is somehow defective.

Hi, welcome to our forum.
The dc-dc converter is quite sensitive in its operation.
I would make sure all the components are ok, not only the caps.

Also, the output is sensitive to the load, so the dc-dc converter should not run without being properly loaded by the pcb components.
The risk here is that if the voltage is too high, even as a result of a voltage peak caused by a power-off followed by a power-on operation, the ic's may suffer damage.

You may try to contact Sergei Frolov, see his messages above. He might add some more useful information, depending if he is reading this pages (usually he is a very busy person).

[clarke2001]

Thx for reply, I tested the DC unit (converter) always plugged in.

Let's number the machines: 1. the working, good one. 2. the defective.

As mentioned above, for a few seconds I plugged the 1. DC-unit into 2. then turned on.The wrong operation (K-prefix issue) didn't change, so I quickly turned it off, to prevent possible DC-unit damage, and plugged it back into the 1. The 1. works good.

Without scope figures (provided on the original technical documentation), the only possibility is the comparison of 1. and 2. using oscilloscope, measuring voltages etc. But I'm afraid, something gone wrong in 2.

Are there original schematics of MK-52 available ? I have one from net (quite hard to read), but without scope figures.

[jebem]

As I mentioned above, I couldn't find schematics with power supply waveforms; however, I have published my own scope reading from my working calculator.

Attached is the schematic diagram in good readable condition.

Also, as long as one can read/learn a little of Russian language (or just use Google translator for that matter), there is a excellent public domain book in djvu format, describing these Russian calculators in detail.
For example, the MK-52 power supply operation is described at pages 83 an 84.

ПРОГРАММИРУЕМЫЕ МИКРОКАЛЬКУЛЯТОРЫ Устройство и пользование, Под редакцией Я. К. ТРОХИМЕНКО, МОСКВА «РАДИО И СВЯЗЬ», 1990.
(Programmable micro-calculator - Design and operation, published by J. K. Trohimenko, MOSCOW RADIO AND COMMUNICATIONS 1990)

The above book url points to the well known site of Eugene Troitskiy at www.emulator3000.org.

[jebem]

(08-21-2015 09:30 PM)Tugdual Wrote:  Just a quick note to say thanks for sharing this information. I had never heard about these vintage Russian calculators until a minute ago and find them fascinating.

Il n'y a pas de quoi! You are very welcome!

[clarke2001]

Many thanks for the the attachment and references, great material. Whoever made the clear, readable schematics, is a good man

[clarke2001]

Hey guys, I have it:

the schematic and the oscilloscope waveforms of MK-52.

High resolution scanned, good readable very original material.

http://www.ru-hpc.narod.ru/calc/doc_1.html

on the page search this line for downloading:

MK-52 Cхема *.jpg Оттенки серого Загрузить! arbinada.com/pmk

generally, it is useful, to use keywords written in original russian form (cyrill letters)

[jebem]

Excellent repository of Elektronika information, thanks.

Concerning the schematics published here for download, I have it since long time and lost track of who post it in the Internet, but I have found it once again in one of the links from your previous post.

Talking about the published signal waveforms, it is good to have a reference to follow, however I would give it a wide tolerance margin concerning the amplitudes and timings, as these calculators can show a wide variance between production batches.

[Hlib]

I decided to rebuild the MK-52. Which one of them is big brother - I do not know. MK-52 were always of poorer quality than MK-61. This is just my experience. Sometimes to restore the old thing it is more interest, than to buy a new Prime :-)



In this case, the defective display.

[Adalin]

So

I have a mk-52 that seems to have a broken eeprom, at least I think its broken
I cant save anything to it in any case

I opened up the unit and measured the output of the voltage converter and all numbers where in the 5% range, I replaced the capacitors anyway and adjusted the voltage, but the eeprom still does not work

when I got the unit some time ago I wrote some programs and saved them to the internal eeprom, but I got some strange effects and thought that maybe I did something wrong, but I have tried many times now and it just doesn't save anything.

I start with clearing the memory and then I write the program and write it to the memory, then I load the program again, and its gone...

Has anyone replaced the eeprom chip in one of these calculators? I bought some replacement eeprom chips and some replacement VFD screens just in case, so I will try to replace the eeprom when they arrive.

or is it something I have missed?

its a later unit, from 1991. I got a PRB-4 external rom and it works fine when I load programs from it

[jebem]

(12-31-2015 01:38 PM)Adalin Wrote:  I have a mk-52 that seems to have a broken eeprom, at least I think its broken
I cant save anything to it in any case.

Hi, welcome to this forum!
According to Sergei Frolov referenced in post#5 above, he found a few machines having defective EEPROM chips.

If you managed to get these chips (I couldn't find any when I was looking for some samples some time ago), then replacing the suspect IC with a new one is the next obvious task to take imho.
EDIT: I have just found one seller here.


Just a word of caution: These printed circuit boards (PCB) are VERY fragile!
Removing components in a safe way is a destructive process - meaning that we either destroy the PCB or the component in the procedure.

Destroying the PCB is not an option, so we should destroy the EEPROM IC to protect the PCB.
Basically we cut the IC pin legs, one by one, and only then we remove the legs from the PCB by desoldering them, one by one.

The procedure is quite simple, but requires adequate soldering tools and skills.
A good 20/30Watt soldering iron with a fine tip, a desoldering station/pump, and a small cutter pliers are the minimum tools to be used here.
You may have a look here for an example of such procedure.

Good luck with your machine.