Jump to content
IGNORED

Myarc cards for sale/repair tips


Recommended Posts

1 hour ago, InsaneMultitasker said:

2. JM2 and JM3 jumper hole 2 is confirmed as connecting to the bottom, front 5v regulator output.  

-----------------------------------

Confirmed that JM3 jumper hole 2 is not getting continuity with the front bottom 5 volt reg, next to the led position.

 

3. The 384k Jumper changes to JM2 and JM3 are correct; both jumpers are to be moved from pins 1-2 to 1-3, enabling AB1 and AB0.  

-----------------------------------

Confirmed JM2 & 3 pins 1-3 are getting continuity.

 

4. Remove one end of the mystery red wire and then confirm continuity:  

-----------------------------------

Removed leg of wire, confirmed no thing on-screen.

a. JM3 hole 2 (test top AND bottom) must connect to the +5 output of the bottom, front 5v regulator. Both sides of the through hole/ must be tested due to the core hole problem you referenced. 

-----------------------------------

Confirmed no continuity to front bottom 5v reg, on either side of board at JM3 pin 2.

 

b. Between the regulator output and the resistor lead soldered to hole 2. 

-----------------------------------

No continuity

 

c. Confirm the through hole is NOT connected to the ground plane.

-----------------------------------

No continuity.

 

5. This red wire is correct, we see jumper between 1-3 on JM3 and the red wire appears to connect to memory pin 31. 

a. My earlier comment in post 389 re: JM3 was incorrect.

b. The white wire should tie into JM1 pin 3 (next to the 9995), it appears to be heading that direction from your pin.

 

image.png.0b108d77f6f700dd47e4357ba176a2d6.png

-----------------------------------

Confirmed continuity.

 

6. Based on picture below

a. Did you confirm the other red wire here provides continuity between memory pins 30-32?  This is necessary to power the SRAM. 

b. Also confirm the jumper wire is not shorting to pin 31.

image.png.1b9028f662ef8a48e520e8a0334d7b8c.png

 

-----------------------------------

Confirmed continuity between pins 30 and 32, 31 is not shorted to them.

 

 

Link to comment
Share on other sites

19 minutes ago, RickyDean said:

2. JM2 and JM3 jumper hole 2 is confirmed as connecting to the bottom, front 5v regulator output.  

-----------------------------------

Confirmed that JM3 jumper hole 2 is not getting continuity with the front bottom 5 volt reg, next to the led position.

 

3. The 384k Jumper changes to JM2 and JM3 are correct; both jumpers are to be moved from pins 1-2 to 1-3, enabling AB1 and AB0.  

-----------------------------------

Confirmed JM2 & 3 pins 1-3 are getting continuity.  good

 

4. Remove one end of the mystery red wire and then confirm continuity:  

-----------------------------------

Removed leg of wire, confirmed no thing on-screen.

a. JM3 hole 2 (test top AND bottom) must connect to the +5 output of the bottom, front 5v regulator. Both sides of the through hole/ must be tested due to the core hole problem you referenced. 

-----------------------------------

Confirmed no continuity to front bottom 5v reg, on either side of board at JM3 pin 2.

 

b. Between the regulator output and the resistor lead soldered to hole 2. 

-----------------------------------

No continuity

So my guess here - and you might be thinking the same - is that the mystery red wire is intended to bring power to JM3 hole 2, either through another connection point to the lower regulator, or by connecting it to the 5v regulator nearby (that powers the V9938).  My thinking is that you should:

 

1. Lift the resistor leg out of JM3 hole ; connect the removed lead, using wire, to the 5v regulator output, and cover the exposed lead/wire.  You can use the nearby regulator output to which you soldered the red wire during the earlier repair.  We are removing the resistor from JM3-2 to avoid intermittent connections between the bottom and top regulator outputs.

 

2. Check continuity between SRAM pins 30,32 and the bottom, front regulator output.  (the same regulator that should have been tied into JM3 hole 2).   If there is continuity, that's good.  If there isn't continuity between SRAM and the regulator, this means the damaged core is no longer passing power to the SRAM socket, and may be affecting other components.  I doubt the core damage will cause this issue, we just need to rule it out.

 

 

Link to comment
Share on other sites

44 minutes ago, InsaneMultitasker said:

So my guess here - and you might be thinking the same - is that the mystery red wire is intended to bring power to JM3 hole 2, either through another connection point to the lower regulator, or by connecting it to the 5v regulator nearby (that powers the V9938).  My thinking is that you should:

 

1. Lift the resistor leg out of JM3 hole ; connect the removed lead, using wire, to the 5v regulator output, and cover the exposed lead/wire.  You can use the nearby regulator output to which you soldered the red wire during the earlier repair.  We are removing the resistor from JM3-2 to avoid intermittent connections between the bottom and top regulator outputs.

 

2. Check continuity between SRAM pins 30,32 and the bottom, front regulator output.  (the same regulator that should have been tied into JM3 hole 2).   If there is continuity, that's good.  If there isn't continuity between SRAM and the regulator, this means the damaged core is no longer passing power to the SRAM socket, and may be affecting other components.  I doubt the core damage will cause this issue, we just need to rule it out.

 

 

Okay, will do that tomorrow, supposed to be raining here tomorrow, so no outdoor work. As a side note, I repaired the said Corcomp Ramdisk, been slowly removing old chips, socketing, and putting in new chips. Finally found the blown one, a 74ls00.

  • Like 5
Link to comment
Share on other sites

14 hours ago, InsaneMultitasker said:

So my guess here - and you might be thinking the same - is that the mystery red wire is intended to bring power to JM3 hole 2, either through another connection point to the lower regulator, or by connecting it to the 5v regulator nearby (that powers the V9938).  My thinking is that you should:

 

1. Lift the resistor leg out of JM3 hole ; connect the removed lead, using wire, to the 5v regulator output, and cover the exposed lead/wire.  You can use the nearby regulator output to which you soldered the red wire during the earlier repair.  We are removing the resistor from JM3-2 to avoid intermittent connections between the bottom and top regulator outputs.

 

2. Check continuity between SRAM pins 30,32 and the bottom, front regulator output.  (the same regulator that should have been tied into JM3 hole 2).   If there is continuity, that's good.  If there isn't continuity between SRAM and the regulator, this means the damaged core is no longer passing power to the SRAM socket, and may be affecting other components.  I doubt the core damage will cause this issue, we just need to rule it out.

 

 

Okay, I removed the resistor leg from pin 2 and left it hanging, then added a wire from that leg to the regulator- front - bottom. Then cheked continuity at 30 and 32 at the sram, and at between them and the reg. Have continuity. Then I plugged it into the peb, getting the initial boot screen as stated before again, but it goes no further, just hangs there.

Link to comment
Share on other sites

On 1/31/2021 at 1:44 PM, RickyDean said:

Okay, I removed the resistor leg from pin 2 and left it hanging, then added a wire from that leg to the regulator- front - bottom. Then cheked continuity at 30 and 32 at the sram, and at between them and the reg. Have continuity. Then I plugged it into the peb, getting the initial boot screen as stated before again, but it goes no further, just hangs there.

I've pretty much exhausted all remote hardware repair suggestions.  This presents as if the SRAM is bad or not accessible.  While there is a small probability the PFM is damaged, it is unlikely since the code is in one 8K bank and we can see the title screenn.

 

If you turn on the Geneve and hold the spacebar, does the PFM device show you the boot menu?  If so, select the various load options and see what happens. 

Link to comment
Share on other sites

28 minutes ago, InsaneMultitasker said:

I've pretty much exhausted all remote hardware repair suggestions.  This presents as if the SRAM is bad or not accessible.  While there is a small probability the PFM is damaged, it is unlikely since the code is in one 8K bank and we can see the title screenn.

 

If you turn on the Geneve and hold the spacebar, does the PFM device show you the boot menu?  If so, select the various load options and see what happens. 

Ok, I had forgotten about being able to hold the space bar, but do get the boot options. But not always, about 50 percent of the time. If I leave it on the boot option screen, then in about 5 seconds or so, I get a bunch of random pixels on the screen. If I select say Hfdc or Floppy, I get an f1-10 or default option, then when I press "D", I get a mosaic of elongated hexagons or a blank screen, and occasionally the random pixelization, seems to be the same pattern though. When I press "B" or "L" it goes to the random pixels, and then for some tries will not go to the boot options when holding the space bar, then will start allowing it again. Now the xt board I have is flaky and sometimes some of the keys work and sometimes not, so that may explain that. I took some pics, but can't load them yet due to data limits on straight talk till Sunday, or when I get near WiFi.

Link to comment
Share on other sites

On 2/5/2021 at 10:40 PM, RickyDean said:

If I leave it on the boot option screen, then in about 5 seconds or so, I get a bunch of random pixels on the screen

Menu response changes things slightly.  I will look through my old notes as I have the vague impression that I've encountered this before. Give me a few days to see what I can turn up. 

  • Like 1
Link to comment
Share on other sites

On 2/7/2021 at 9:08 PM, RickyDean said:

Finally was able to upload those pics, after holding the space bar.

At this point, pfm is just spinning in a loop waiting for you to press a key. 

 

The only notes I found suggest the following troubleshooting steps:

1. Hold for menu, press F for floppy, and D for default.  Does the disk controller activate and does it load the OS?  If so, 32K SRAM is likely ok. Possible vdp/vram / regulator issue.   Does the load stop part-way?  chip failure induced by heat or time, i.e., faulty chip.

2. Verify VRAM is getting 5v power.  If possible, swap chips with known working and/or re-seat them.

3. Swap 9995, 9901 and gate array to confirm they are not contributors to the problem.  

 

 

 

  • Like 1
Link to comment
Share on other sites

1 hour ago, InsaneMultitasker said:

At this point, pfm is just spinning in a loop waiting for you to press a key. 

 

The only notes I found suggest the following troubleshooting steps:

1. Hold for menu, press F for floppy, and D for default.  Does the disk controller activate and does it load the OS?  If so, 32K SRAM is likely ok. Possible vdp/vram / regulator issue.   Does the load stop part-way?  chip failure induced by heat or time, i.e., faulty chip.

2. Verify VRAM is getting 5v power.  If possible, swap chips with known working and/or re-seat them.

3. Swap 9995, 9901 and gate array to confirm they are not contributors to the problem.  

 

 

 

Will do that tonight, let you know the outcome.

Link to comment
Share on other sites

Tim,

 I might have asked this before, but... I have an original copy of SAMS Photo Facts for the 99/4a. Step 1 is always use your oscope to measure ...

 How can you fix the far more complicated Geneve's with out the aid of an oscope, in so many cases?

  Do you use an oscope / logic analyizer in your home shop?

Link to comment
Share on other sites

7 minutes ago, dhe said:

 How can you fix the far more complicated Geneve's with out the aid of an oscope, in so many cases?

  Do you use an oscope / logic analyizer in your home shop?

I do not own a scope;  I use a multi-meter and on the good days I try to use the built-in logic analyzer on my shoulders. ;)

 

I suppose that generally speaking, many folks who like troubleshooting - whether programming, hardware repair, appliances, cars, etc.- get a sense for where to start based on what they know and have learned before.  

 

I always physically inspect the card upon receipt to look for blown traces, missing components, incorrectly installed components, etc -- having seen many cards, when something is out of place it often stands out.  Then I check the power regulators for shorts or low resistance across ground/input/output.  I may replace caps and regulators before starting any work, depending on the age of the components and what the owner has told me.  If the card hasn't been smoked, I'll then place it into a PEB with a flex card (well, all cards except a Geneve).  If it seizes the bus and prevents a powerup, then I can usually focus on the bus connections; if I can access the ROM, the problem is somewhere beyond the buffer/eprom.  From there it depends on the card and what is wrong.  For example, if an HFDC can read/write a hard drive but fails to read a floppy, many of the components common to both drives can can be ruled out.

 

So much often boils down to considering the symptoms and looking for the cause based on direct feedback and deduction and even a bit of luck.   @Shift838 can probably attest to how I worked through the cards he brought over by stepping through what should work and tracing back through the schematic based on how the card reacted. 

 

 

 

  • Like 3
  • Thanks 1
Link to comment
Share on other sites

On 2/11/2021 at 11:27 AM, InsaneMultitasker said:

At this point, pfm is just spinning in a loop waiting for you to press a key. 

 

The only notes I found suggest the following troubleshooting steps:

1. Hold for menu, press F for floppy, and D for default.  Does the disk controller activate and does it load the OS?  If so, 32K SRAM is likely ok. Possible vdp/vram / regulator issue.   Does the load stop part-way?  chip failure induced by heat or time, i.e., faulty chip.

2. Verify VRAM is getting 5v power.  If possible, swap chips with known working and/or re-seat them.

3. Swap 9995, 9901 and gate array to confirm they are not contributors to the problem.  

 

 

 

1.Pressing F just goes to the next screen and just sets there generally, no matter how many times I press D. Pressing B or L or just lettling it set there for a few seconds, it generally goes to this screen with the corrupt looking header. All the rest just give me screens like the one pictured above, in post #407.

2. Haven't verified voltage, but all regs are jumpered as this is in a peb that has an switching PS in it. And I am getting video. Have swapped VDP and VRAM too.

3. Have swapped the 9995, gate array, and 9901 also, did get a slightly different reaction, one time after pushing F, got a black blank screen, then another boot and hitting space or left shift, got a menu in black with white letters instead of blue, but the next time it was blue again.

 

Going back to the slightly corrupt header, when the system boots into the initial screen, then hitting space or left shift. When one hits B what bios is being loaded and in hitting L, is the PFM trying to load the internal stored system files I would have loaded 20+ years ago? If so where does it park it to display and show on the screen, could something there be corrupt, bad sram chip, flash eprom chip, perhaps?

20210212_222759.jpg

  • Sad 1
Link to comment
Share on other sites

Looking at Characters that are produced more than once, I'm looking at 9 and 4 - It appear the character definitions are corrupt, because the characters are corrupted in the same way over again.

 

So either VRAM or the source of the character definitions? Can you boot another version of MDOS?

Link to comment
Share on other sites

12 minutes ago, dhe said:

Looking at Characters that are produced more than once, I'm looking at 9 and 4 - It appear the character definitions are corrupt, because the characters are corrupted in the same way over again.

 

So either VRAM or the source of the character definitions? Can you boot another version of MDOS?

No, not allowing that option, as described in the post to Tim. It is a pfm+ and seems to be stuck, giving the published results. Is it possible to remove the srams and flash eeproms and place in a bios and sram to see if it goes further?

Link to comment
Share on other sites

18 minutes ago, RickyDean said:

No, not allowing that option, as described in the post to Tim. It is a pfm+ and seems to be stuck, giving the published results. Is it possible to remove the srams and flash eeproms and place in a bios and sram to see if it goes further?

Not without reversing some changes and jumpers.  Pop out that 384K chip and inspect the socket.  Is it a single-wipe?  Corroded or worn out wipes?  I'd suggest the same for the PFM (eprom) socket but check out the memory socket first, since PFM is gonna be a best to dismantle. 

 

I'm not in a position to do repairs right now but I would entertain taking a look at it later this year. It would be easy for me to swap both sets of chips with known working ones.   There is also some hardware-related work under way that might make updating the pfm and memory (with single layer 512k) viable.   We can continue to chip away in the meantime as long as you are willing and able.

  • Like 1
Link to comment
Share on other sites

1 minute ago, dhe said:

Tim,

  What do you think of spraying a socket with deox (contact cleaner?)  and letting it sit a couple of days, or do you just recommend replacing single wipe sockets?

The single wipes are problems waiting to happen, especially when the wipe can no longer exert enough force to provide proper contact.  The force  fatigue is not necessarily limited to single wipe sockets.  In either case, contact cleaner isn't going to help much - and my cleaner experience is limited so I can't offer much in the way of advice there ;)

Link to comment
Share on other sites

I had so many issues with the GBS8200 devices, that I decided to buy the OSSC device to see if that works better.  I made my own SCART to DB9 connector (and use the adapter of @Shift838 from DB9 to DIN8 pin plug).  The quality is very good now. (Still need to sort out interlacing mode etc. as it seems the screen is not stable in that mode, i tested Fractal-NTSC and -PAL.

 

However there is still an issue that I loose the signal for a split second when the Geneve card works longer, i think when it warms up  (but it is much better than with the GBS 8200 device as the screen comes back).

 

There is 1x regulator 12V I did not change yet. I only replaced the 5V regulators (which made the Geneve itself working stable).. Could the 12V  cause the issue loosing the signal?   In earlier posts I also had to replace some other components. 

 

PS. The OSSC device is arounf USD 137 on Lazada. 

More about http://junkerhq.net/xrgb/index.php?title=OSSC

 

20210220_184228.jpg

20210220_184235.jpg

20210220_184250.jpg

Edited by globeron
  • Like 5
Link to comment
Share on other sites

On 2/13/2021 at 6:39 PM, InsaneMultitasker said:

you are willing and able.

Okay, I have started pulling the sram and the flash eeproms from the board- have pulled if the old sram socket and will be redoing the bios socket, both with tooled sockets. Now I have a question on the attached photos, the top chip is the 20c010-15 chip, but I cannot find any information on the bottom chip. It says on top 20PC 9416 and on the bottom 4A0968-19108B, 1KOREA-H 4A9414.

 

20210225_235150.jpg

20210225_235129.jpg

Edited by RickyDean
trying to add pictures, but not able for some reason.
Link to comment
Share on other sites

Mmmmm, did you unsolder the two chips?   The legs look really uneven in the pictures (and I'm probably not telling you anything you don't know) so be sure to carefully straighten before inserting.

 

Both chips should be Atmel 29c010 flash.  The stacking order cannot be reversed as only one contains the actual boot code.  You can try reading the chips if your programmer supports the type.

 

For reference I did not program the PFM chips - Don did that with an old XT PC.  He typically removed the identification and markings from the top chip (and later both chips) and then soldered them together.  I do not recall if the OS and flashdisk  were loaded during the initial programming session or later to test the bootcode operation.  It's been too long for this particular model...

 

Link to comment
Share on other sites

8 hours ago, InsaneMultitasker said:

Mmmmm, did you unsolder the two chips?   The legs look really uneven in the pictures (and I'm probably not telling you anything you don't know) so be sure to carefully straighten before inserting.

 

Both chips should be Atmel 29c010 flash.  The stacking order cannot be reversed as only one contains the actual boot code.  You can try reading the chips if your programmer supports the type.

 

For reference I did not program the PFM chips - Don did that with an old XT PC.  He typically removed the identification and markings from the top chip (and later both chips) and then soldered them together.  I do not recall if the OS and flashdisk  were loaded during the initial programming session or later to test the bootcode operation.  It's been too long for this particular model...

 

Yes, I unsoldred them and it is easy to distinguish which is which. Just needed to know what the bottoms chip was. So if it's the same as the top one, only a different brand, I should be able to read them in my burner, we'll see. Tried googling the numbers and not finding any datasheets on that bottom chip. Thanks.

  • Like 2
Link to comment
Share on other sites

On 2/20/2021 at 12:55 PM, globeron said:

I had so many issues with the GBS8200 devices, that I decided to buy the OSSC device to see if that works better.  I made my own SCART to DB9 connector (and use the adapter of @Shift838 from DB9 to DIN8 pin plug).  The quality is very good now. (Still need to sort out interlacing mode etc. as it seems the screen is not stable in that mode, i tested Fractal-NTSC and -PAL.

 

However there is still an issue that I loose the signal for a split second when the Geneve card works longer, i think when it warms up  (but it is much better than with the GBS 8200 device as the screen comes back).

 

There is 1x regulator 12V I did not change yet. I only replaced the 5V regulators (which made the Geneve itself working stable).. Could the 12V  cause the issue loosing the signal?   In earlier posts I also had to replace some other components. 

 

PS. The OSSC device is arounf USD 137 on Lazada. 

More about http://junkerhq.net/xrgb/index.php?title=OSSC

 

20210220_184228.jpg

20210220_184235.jpg

20210220_184250.jpg

I‘m using an OSSC too for my EVPC2/V9938 TI-99/4a system. At the beginning I had the same issue (somtimes a black screen for a second or two). 
 

Then I played with the configuration of the OSSC with success and since about three months I never had a black screen. It is completely stable.


 

  • Like 1
Link to comment
Share on other sites

From this thread, in which I inquired about some mods I found on a Myarc HFDC and that made me wonder whether they're causing the controller to not work.  I've checked the TTL chips on drive side of the board and they appear to be OK.  After cleaning off the flux, I've reinstalled the mods that were present when I received the card, here's photos of its front and back:

 

20210227_191322.thumb.jpg.d3091f714bf39c5a08bc2e661f6a8633.jpg

 

20210227_191334.thumb.jpg.268a41bf6389256ebfa45501ead098c6.jpg

 

 

1 hour ago, InsaneMultitasker said:

2. U8 modification is suspicious; that connection is part of the floppy data separator circuitry.  Is it a blown trace or intentionally removed? The black mark (Sharpie marker?) obscure the board trace.

The trace between U9 (74LS373) pin 5 and U17 (7406) pin 5 was intentionally cut.  Here's a close shot of the trace:

 

 20120601_064459.thumb.JPG.cc224b2bc81ac284ed9cd356ffbb3bdb.JPG

1 hour ago, InsaneMultitasker said:

3. The disc capacitor may be correct; some HFDCs had them installed in this manner.  What does the wire connect?

It connects pin 5 of U17 to pin 12 of U9, so apparently the mod corrects an incorrect connection between the two?

 

Any further guidance will be appreciated.


Thanks,

Hans

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...