Myarc cards for sale/repair tips

[fabrice montupet]

Yes, I have already found bad solders. I also straightened many pins that were bent to 90° and made short-circuits.

[fabrice montupet]

Just for the pleasure to ear if some Geneve 9640 return to life, those who have acquired one of these non-working could tell us in this topic the progression of their repair?

I had a few time to begin mine. At this day, here are the status of the restoration:

 

Missing parts:
VDP V9938
RTC MM58274
Audio SN76496
SRAM M5M5256P-15
Logic 74LS04
Logic 74LS244 X3
Logic 74LS245
Transistor 2N4126 X2
Crystal 32.768 KHz
Battery CR2032
LED

Repairs:

- PCB deeply cleaned
- Bottom side of the board: Many pins were bent/crushed to 90° and made short-circuits to the neighbours pins. I have straightened all of them.
- Damaged Solders on some pins crushed: Repaired
- Bottom side of the board: All excessive length of pins cut
- Capacitor C200 damaged : Replaced
- Crystal 12MHZ : One leg was broken: Replaced
- A crystal trace to C82 was broken: Repaired
- 74LS04, 74LS244 X3, 74LS245 installed
- CR2032 Battery holder soldered
- VDP 64 pins support badly soldered: Repaired
- 32.768KHz crystal installed
- LED installed
- Transistor 2N3906 X2 Installed (in replacement for 2N4126)
- Transistor Q200 (2N4401) dead: replaced
- 22uf 35V X7 capacitors replaced.

I continue tomorrow my investigations and repairs :-)

 

[+InsaneMultitasker]

Great update, Fabrice! Brings back memories of repairing other cards from Myarc. You are certainly going about this properly and your report(s) may help others to resurrect equipment in the future.

 

Edit: Swim may be able to chime in here. I vaguely recall some Geneve cards had a few traces that were only connected at one point so there was often a small piece of wire added to fix this omission. Any functioning card has the traces intact or fixed, so it is very rare to even come across one that's missing. I do not have any notes in my repair logs that detail this change, nor is it on my working schematic. I will look at a few Geneves for clues as I can picture at least one of the "traces to nowhere".

Edited July 1, 2018 by InsaneMultitasker

[+InsaneMultitasker]

Repairs:

- Transistor 2N3906 X2 Installed (in replacement for 2N4126)

The picture on the Geneve board is flipped 180 degrees on most (or all) cards! Very the orientation and ECB. Notice the flat edge of the transistor and drawing do not match in most cards.

 

 

I have found one of the modifications and problem traces noted in my previous post. Unfortunately,the two Geneves are different. The first Geneve connects R200 to edge connector pin 58 using two jumper wires and removes the incorrect trace. The second Geneve has the right trace connection but is missing the connection to pin 58. I need to do some more research this week.

[fabrice montupet]

Yes, I have noticed this error on the Silkscreen, I have soldered the transistors in the right direction. For helping the repair of the board, I use the Geneve schematics and I use my working Geneve9640 too :-)

[fabrice montupet]

When we look at the schematics of the Geneve9640, we see that all the two transistors Q11 & Q12 are inverted according to the MM58274 datasheet (see Typical connection). A strange situation that it's interesting to study.
In the same time, I will verify all the schematics of the Geneve9640 to see if they contain eventual errors. It remembers me the schematics of the 99/4A that contains errors that I had to find all before developing my TIny-99/4A computer.

For this moment, and to repair quickly the Geneve 9640, I prefer to use jointly the two documents that show the 4 layers of the PCB. It's very interesting to see where all the traces go.

[fabrice montupet]

Important information:

I was surfing on the web to see pictures of the Geneve 9640 and I came across a drawing made by Ron on mainbyte.com. An error quickly caught my eye: A LM7812 has been placed at M57 location (near the Yamaha V9938) instead of a LM7805. Don't place a LM7812 at M57 because you will damage the major part of your Geneve 9640 components like the VDP :-(

Edited July 6, 2018 by fabrice montupet

[+InsaneMultitasker]

I have a set of schematics from Ron but I don't see the error. Which document are you referring to?

 

I will check my original schematic for comparison.

[+Ksarul]

It may be that he's referring to the Geneve picture on the site. . .

[fabrice montupet]

I saw the error on the drawing here: http://www.mainbyte.com/ti99/graphics/geneve.jpg

I haven't used its redrawn schematics as I use to work with original documents to prevent any potential added errors.

Edited July 6, 2018 by fabrice montupet

[+InsaneMultitasker]

Yes, the picture is certainly in error.

 

The schematics are also incorrect -- both the John Strain schematic (which I scanned from my days @ Cecure) and Ron's redraw show the wrong part number. It looks like the original schematic was drawn to show "5v" and amended later to u55v in error. The parts and power pages appear to be correct from a voltage perspective. If someone quickly scanned for u55v, they would either not find the part OR assume it meant u55. Dangerous if you are not familiar with the card. I wonder if there are any other similar snafus.

Edited July 6, 2018 by InsaneMultitasker

[fabrice montupet]

I also saw that the Page2R_ComponentLocation.jpg contains at least one error. But I don't use it, This Geneve 9640 board is a such beauty that since all these past years, I and never got tired of looking at it. So, I could place each component eyes closed ^_^

Some errors on the orginal schematics, errors in the component location document, and maybe in some other documents. To work with these documents, it is necessary to work too with the datasheet of all IC to verify the coherence of the cabling. It's what I do.

Edited July 6, 2018 by fabrice montupet

[+InsaneMultitasker]

I had some time to check the traces.

 

In my schematic (from John Strain) power connection #6 is associated with -16v. This association is not depicted in the power distribution so it is easy to miss. -16v is routed from edge connector to a throughole, which is then routed toward the top center of the card to another throughole, to a trace, and finally to R250/R251.

 

Here is the schematic depiction. As Fabrice and I have noticed, the EBC orientation needs to be reviewed. The print on the board is reversed.

 

Some Geneves do not carry the power from the edge connector to the first throughole. A jumper wire has been added here.

 

This is the throughole that connects to R250/R251.

 

This Geneve shows the incorrect trace, routed to the wrong throughold. The trace has been cut and a jumper wire added.

 

This Geneve shows the trace in its correct state. It appears to be a hand-drawn correction.

 

My repair notes from the late 90s indicate clock and startup problems if these traces are not correctly in place. Since most Geneves function properly, it is probably safe to assume the correction was made to most cards. I have one Geneve where the connection between the edge connector and throughole is missing yet the card still works, so a bit more research may be prudent to understand what else might be different on that card.

 

Edit: one point of clarification regarding the schematics. The Mainbyte schematics were redrawn from my paper copies. So if there are errors in my copy, the same errors will probably show up in Ron's version. I noticed that I added a few things to my schematics since scanning them though most comments were to help in troubleshooting. I typically don't look at the schematics, having worked on so many of these cards in the past, that I only refer to them for a refresher now and again... or some nasty problem that requires point-to-point validation.

Edited July 7, 2018 by InsaneMultitasker

[fabrice montupet]

My two Geneve 9640 have the different routing (one board has the wire, the other has the corrected trace), I was surprised when I saw that Myarc had modified the PCB, I thought that all the Geneve have this wire runing on the top of the board.

I announce you that the defective Geneve 9640 is now.... working! :-)
Here are the last works made on the board:

- 10uf 16V C12 Capacitor replaced.
- A broken trace has been already repaired with a long running strap. It's not secure, so I made a better repair.
- All the MB81256 IC replaced with KM41256-12 (It's just a personal choice, I don't really like this Samsung chip because I had many problems with it on some other old computers)
- Broken trace found under the EPROM IC support: Repaired. A former owner used a metallic tool to extract the EPROM and made a scratch on PCB, cutting a trace.
- Bad PAL IC - Replaced.

It is too late here to take some pictures with a good lighting (21:34) So I will take them tomorow and I will show you the Geneve displaying the beautiful swan :-)

 

Some other members who bought the defective Geneve has also begun the repair? I would be happy to get news about these boards!

[Shift838]

Great work Fabrice! Another Geneve is ALIVE!

 

 

Edited July 8, 2018 by Shift838

[fabrice montupet]

Thank you :-)
For those who are reparing their non-working Geneve: You can remove the PAL chip to test it. This component is not necessary for the boot sequence.

[Opry99er]

Amazing. That's a ton of work you put in.

[ti99iuc]

Wow Fabrice !! congratulation for this repair session !!

[RickyDean]

Yes Fabrice, mine was completely empty of chips except for having the Gate Array. I have replaced all but the sound chip and the PAL(interesting to know that it will boot without it).

It has some bad traces and the three lower 7805 regs are missing. The sockets where the regs go, have no solder pads and I will have to come up with a way to retain them, to add wires to the locations they are supposed to feed, also have to place another battery holder as the original is missing(and it has bad socket too). Other wise it was in good shape

[fabrice montupet]

Thank you to all :-)

 

@RickyDean: I wish you all the best for your repair. You can also remove the audio chip SN76496 and the RTC chip MM58274 for the main tests. They are not essentials.

I had no time to take some pictures today so here are pictures that I took yesterday. I'm sorry for the poor quality, my retrocomputing room lacks of strong lighting and I used an iPad 2 (no camera flash)

The Geneve 9640 received:

 

First picture after the Geneve has been repaired:

To simplify the repair of the Geneve 9640, I quickly made an adapter to be able to use my multimeter and logic analyzer outside the PEB.

 

Overview of the Geneve 9640 after repair:

[+InsaneMultitasker]

To simplify the repair of the Geneve 9640, I quickly made an adapter to be able to use my multimeter and logic analyzer outside the PEB.

 

 

 

Excellent work!

 

Be sure to label your adapter's orientation. I had one similar to this until a card was mistakenly inserted backwards in the slot....

[fabrice montupet]

We don't see on the picture, the backplane has only three slots (but enough for testing), the Geneve is plugged into slot "2" and between slot "1" and slot "2", I used a large permanent marker to draw a big arrow for the orientation of cards :-)

[fabrice montupet]

A picture of the backplane that I quickly made:

As you can see, it only supplies the power lines and ground from the PEB, no data lines. The goal is only to use this external backplane, not the one of the PEB.

Edited July 10, 2018 by fabrice montupet

[RickyDean]

Excellent work sir, I have a extender card that brings a slot up to the level of the top of the PEB box top( with the lid off of course). I don't have a sound chip for it, but I probably have one stashed away somewhere, just have to search for it. BTW... do you have the ability to burn a PAL for the Geneve? I will be needing one for mine when I get it functional. If you do have the Jedec file to burn this would you part with it? If not what would you charge to burn one for me, so I can replace mine? I haven't contacted Richard to see if he has any available, yet.

 

I do have a question though, why did you change out the orange caps, the 35 volt ones? The one I am rebuilding, also has the same type, and they look very good, almost new. My other Geneve's have 16v ones'.

Edited July 10, 2018 by RickyDean

[+InsaneMultitasker]

 

I do have a question though, why did you change out the orange caps, the 35 volt ones? The one I am rebuilding, also has the same type, and they look very good, almost new. My other Geneve's have 16v ones'.

 

For reference, I am guessing those caps were replaced by Don as part of his attempts to resurrect cards or by me in the late 90s for the same reason. Either way, they are more than likely 20+ years old. I would recommend replacing all electrolytic caps if you have the ability to do so.