VOGONS

EIZO MD-B10 ISA VGA card repair attempt - Tseng Lab ET3000 chip

Topic actions

First post, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Hi,

I am a rookie on graphics card repair, practically this is the first bad one in my hand. It is a nice card, it would be great to be able to rescue it.

Picture of the card: see the pictures on the following link: https://www.ebay.de/itm/185311817171?itmmeta= … ABk9SR7DZorWkZA

Tried in a 486 config, got 8 beeps, indicating video card problem. No picture on the monitor.

The card is based on Tseng Labs ET3000 chip, have 512MB DRAM. I was not able to find any technical data, pinout, VGA BIOS, etc. for ET3000 and this card. No info about the card itself, just one thread in Vogons. Jump into the black hole.

Thermal camera showed no drama: after 15min of operation the max temperature is 50C at some ICs. Seems to be no shorted outputs in the card.

VGA BIOS seems to be functional: at the time of querying by the POST the OE-CE is pulsing, all the data lines are active, address lines are active as well, no stucked - floating pins.

It uses IMSG-176 palette DAC to generate the VGA signals. It seems this chip is kept in switched off state because BLANK is always low. Based on this no RGB is produced. Pixel clock is ok, data bus is active, no hard errors on these lines. Pixel address bus is low, no activity at all. RS0 RS1 are active all the time but RD-WR are high all the time. It does not add on: try to write - read the DAC's registers without WR, RD???

The DRAM 512MB is organised as 4 blocks, driven by 4 CAS lines. After power up, 1 second pause and RAS is active for each block. On the other hand, CAS signals are always high. I assume - maybe wrong - even in case of RAS only refresh cycle the CAS lines should be go down and up., ie.- pulsing. Write is alwasy high, no pulsing.

After MB reset the RAS stopped for 1 sec and start again, indicating ET3000 got the reset signal from the ISA bus. Unfortunately I was not able to trace down the reset signal, it dissapears under the PAL chip in a via.

Checked the 244, 245 bus drivers, seems to be OK, no signal conflicts. Checked the standard ttl ICs, seems to be ok: no signal conflicts.

At that point I ran out of options. Please advise what I missed and which directions worth to be investigated to diagnose this card. Does anybody know diagnostic eprom image for VGA card POST testing? It would be good to know in more detail what the POST is claiming regarding the VGA card.

Thank you!

Reply 1 of 31, by rasz_pl

Posted on 
User metadata
Rank l33t
Rank
l33t

ET3000 is unremarkable and not worth the effort.
8 beeps most likely means video bios didnt get executed at all

https://github.com/raszpl/FIC-486-GAC-2-Cache-Module for AT&T Globalyst
https://github.com/raszpl/386RC-16 memory board
https://github.com/raszpl/440BX Reference Design adapted to Kicad
https://github.com/raszpl/Zenith_ZBIOS MFM-300 Monitor

Reply 2 of 31, by zami555

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Check carefully the bottom side of PCB for any cuts in traces. If card was stored with other card without any protection there is a chance that some scratches happened on the surface of PCB and some could be deep enough to cut the trace.

Reply 3 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
rasz_pl wrote on 2024-08-05, 22:03:

ET3000 is unremarkable and not worth the effort.
8 beeps most likely means video bios didnt get executed at all

Thank you your reply! I checked the video bios activity by checking all the Eprom pins by a TTL probe. Data, address, RD, OE and CE pins are active pulsing when POST tries to communicate with the card, immediately followed by the 8 beeps. The next step would be to use a logic analyzer to check the timming between the conrol signals and the address-data bus to validate the MB reads valid bios content. Unfortunately I do not have a canche to check the validity of the BIOS, no downloadable files found on the web.

Reply 4 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
zami555 wrote on 2024-08-06, 18:27:

Check carefully the bottom side of PCB for any cuts in traces. If card was stored with other card without any protection there is a chance that some scratches happened on the surface of PCB and some could be deep enough to cut the trace.

Thank you your reply!
The card was quite dirty but after carefull cleaning it turned out the physical condition is quite good. No trace of pyhisical damage on both side of the panel. Payed attention to the possible broken traces due to some point like corrosion but none of found. Luckily the card was far from the battery vapor. Some solder joints have surface corrosion especially close to the ISA connector, but nothing serious. Checked the pins of ET3000 for movement and conductivity to the pads as well.

Reply 6 of 31, by Dorunkāku

Posted on 
User metadata
Rank Member
Rank
Member

Good luck!

Reply 7 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
Grzyb wrote on 2024-08-07, 12:26:
mita wrote on 2024-08-07, 11:26:

Unfortunately I do not have a canche to check the validity of the BIOS, no downloadable files found on the web.

You can find an ET3000 BIOS here - http://vgamuseum.info/index.php/companies/ite … -tseng-et3000ax

Thank you! As far as I know this Eizo card has different BIOSes than the standard Et3000 cards. This card has some additional feature the ET3000 cards missing. If the card will be reborn I will try it out to check this statement. Anyway got the BIOS from the community!

Reply 9 of 31, by Dorunkāku

Posted on 
User metadata
Rank Member
Rank
Member
mita wrote on 2024-08-07, 18:07:
Dorunkāku wrote on 2024-08-07, 12:31:

Good luck!

Man, you did the impossible! Thank you. Do you have access to this card?

Yes , I just tested it and it works.
I would like to find the 1.07 and 1.08 OS ROMs and the original EIZO drivers.

Reply 10 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
Dorunkāku wrote on 2024-08-08, 07:34:
mita wrote on 2024-08-07, 18:07:
Dorunkāku wrote on 2024-08-07, 12:31:

Good luck!

Man, you did the impossible! Thank you. Do you have access to this card?

Yes , I just tested it and it works.
I would like to find the 1.07 and 1.08 OS ROMs and the original EIZO drivers.

Unfortunately I do not have other version of BIOSes than the 1.01 and 1.05. The original drivers is a hard case, I was not able to find online.

Well, I am asking if you can help me out with the diagnostic of this card. I am looking for an information to be able to determine wether the ET3000 chip is defective or not.
Regarding the on board memory handling I got valid RAS signals on the memory but not CAS signals. This memory supports hidden RAS refresh which could result no CAS signals but I am not sure. Maybe it is the intended operational mode.

It would be great if you can measure the CAS pin of any of the memories by a TTL probe or oscilloscope. If you do not have these tools maybe a multimeter is enough to check the voltage on this pin. The CAS line is the 16 pin of the 41464 chip.
After power up about one second the RAS became active on my card. It would be interesting to know what a CAS lines do after the power up , within the first 10 seconds time frame. If your card have a CAS line active after power up + 1 second than it is clear my card has a defective ET3000 chip. CAS line active means with a multimeter the pin's volatge below 5V but above 0V. Let see, maybe miracles happen.

Reply 11 of 31, by MikeSG

Posted on 
User metadata
Rank Member
Rank
Member

Have you tried lowering the ISA bus clock as low as it will go? The memory chips are 100ns, maybe there's a limit to ISA clock speed.

RAS-only refresh would mean CAS doesn't change, as you said.

Reply 12 of 31, by Dorunkāku

Posted on 
User metadata
Rank Member
Rank
Member

I meassured with a analog meter:
Pin 16 on IC1 goes to 5V on power on, after 5 seconds it drops to 2.3V and stays there when idle on the DOS prompt. While running "VIDSPEED +" it varies between 2.0 and 2.5V.

I hope this helps.

Regarding the driver, I found this: https://www.ricardo.ch/de/a/5x-eizo-md- ... 261581092/

Reply 13 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
MikeSG wrote on 2024-08-08, 08:36:

Have you tried lowering the ISA bus clock as low as it will go? The memory chips are 100ns, maybe there's a limit to ISA clock speed.

RAS-only refresh would mean CAS doesn't change, as you said.

OK, I will try to lower the speed of the ISA bus, let see what it do.

Reply 14 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
Dorunkāku wrote on 2024-08-08, 08:38:
I meassured with a analog meter: Pin 16 on IC1 goes to 5V on power on, after 5 seconds it drops to 2.3V and stays there when idl […]
Show full quote

I meassured with a analog meter:
Pin 16 on IC1 goes to 5V on power on, after 5 seconds it drops to 2.3V and stays there when idle on the DOS prompt. While running "VIDSPEED +" it varies between 2.0 and 2.5V.

I hope this helps.

Regarding the driver, I found this: https://www.ricardo.ch/de/a/5x-eizo-md- ... 261581092/

Thank you the measurement! Very important info: you measured active CAS after 5 seconds. It means the CAS was high until the POST checked the VGA card. When the check was completed the system started to write the VGA memory resulting active CAS.

For me it means the ET3000 may be not dead. The init of the VGA card was not sucessfull, pointing to the direction of the VGA BIOS. I will check the content of the BIOS on Sunday comparing it to your BIOS. If the BIOS is OK than I have to swallow the bullet and start to use my logic analyzer to figure out what is happening on the buses.

If nothing found on the hardare than I will try one more thing: try to disassemble the BIOS to check the init code segment of the VGA BIOS. I have never tried it before, uncharted territory for me except the assembly language.

Reply 15 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie
Dorunkāku wrote on 2024-08-07, 12:31:

Good luck!

Dumped my card's program BIOS and checked with yours. They are indentical, no BIOS corruption in my card.

Started to deal with the vga bios asm file. As far as I know ET3000 is register level compatible with IBM VGA adapter, therefore the card's POST polling code should be quite similiar. A lot to learn and refresh.

Reply 16 of 31, by srmeister

Posted on 
User metadata
Rank Newbie
Rank
Newbie

Hello mita,
have you considered simply dead RAM chips? Maybe it does some RAM testing in the VGA BIOS and determines bad RAM and stops working?
Btw. i do have 2 of the same cards, one is faulty but the other one works fine. So looking forward if you will be able to find the problem 😀

Reply 17 of 31, by mita

Posted on 
User metadata
Rank Newbie
Rank
Newbie

I can not close out bad ram chips but the thermo camera showed Uniform Thermal image of the ram chips. I suspect a bad chip shoud have a bit different thermo signature but it is not 100%.

The mb bios should be disassembled to.get the answer of the.cause of the 8 beep.

Reply 18 of 31, by mkarcher

Posted on 
User metadata
Rank l33t
Rank
l33t
mita wrote on 2024-08-09, 21:13:

I can not close out bad ram chips but the thermo camera showed Uniform Thermal image of the ram chips. I suspect a bad chip shoud have a bit different thermo signature but it is not 100%.

The mb bios should be disassembled to.get the answer of the.cause of the 8 beep.

8 beeps is obviously borrowed from AMI. The AMI BIOS expects a CGA compatible or an MDA compatible video card. This might either be an actual CGA or MDA card, or an EGA/VGA card intialized to test-mode 0-3 (CGA compatible) or mode 7 (MDA compatible). They typical check is to read/write the CRTC registers for the cursor position (on of the few 6845 registers that were not write-only) and read/write some bytes of video memory. If this fails for both the MDA compatible addresses (3B4 / B000) and CGA compatible addresses (3D4 / B800), the 8 beeps will be sounded.

Go to top of page Go to top of page
Back to General Old Hardware
Close window
Close window