Well, for the motherboard to make any beeps you'll need the processor. Have you checked the tantalum caps?

tsalat wrote on 2024-04-17, 07:16:

After this replacement I have added the CPU (i386DX 25), RAM, and my post card. After turning it on, the post card shows "----", and the IRDY LED is flashing. During the flashing the speaker beeps, it is a continuos short beep with a pause, and again, never stops, and the LED flashes during this.

That depends on BIOS but I would guess checksum error. Which does not mean the BIOS is bad (though it could be), merely that CPU can't fetch data from it without glitches. And since there are no codes on the POST card I would also suspect a problem with ISA bus address or data lines - which on many mobos is shared with BIOS ROM. You get beeps because that is decoded internally in the chipset so at least it works that far (CPU to chipset bus seems OK).

If you have EPROM programmer burn the "BIOS" that I've attached. It doesn't do anything except output CPU ID on the POST card, using as little code as possible. If you see that show up on the POST card then it's just the BIOS issue, if not it's also ISA as I've mentioned above.
Is there 74F245 (or ALS245) anywhere near the ISA slots? Could be under a socketed chip. If there is one - it is suspect of being bad. I would however remove all cache chips for testing before commiting to any desoldering.

dominusprog wrote on 2024-04-17, 10:13:

Well, for the motherboard to make any beeps you'll need the processor. Have you checked the tantalum caps?

thank you for your answer. The behaviour is different if the CPU is socketed or not, no CPU, no beep, CPU the one short beep. I have 4x i386 CPUs but I had never the opportunity to try them somewhere else - I guess they are fine, can not be sure 100%. The caps near the CPU seems to be fine although I am not sure how to :test: them while one the board. They are not shorted nor exploded, while being in capacity mode by the multimeter, the readings are either zero or around 400uF, the value is 10uF.

If you have any advice of how to measure them without extracting from the board, I am open.

br, tomas

Deunan wrote on 2024-04-17, 10:36:

That depends on BIOS but I would guess checksum error. Which does not mean the BIOS is bad (though it could be), merely that CPU […]
Show full quote

tsalat wrote on 2024-04-17, 07:16:

After this replacement I have added the CPU (i386DX 25), RAM, and my post card. After turning it on, the post card shows "----", and the IRDY LED is flashing. During the flashing the speaker beeps, it is a continuos short beep with a pause, and again, never stops, and the LED flashes during this.

That depends on BIOS but I would guess checksum error. Which does not mean the BIOS is bad (though it could be), merely that CPU can't fetch data from it without glitches. And since there are no codes on the POST card I would also suspect a problem with ISA bus address or data lines - which on many mobos is shared with BIOS ROM. You get beeps because that is decoded internally in the chipset so at least it works that far (CPU to chipset bus seems OK).

If you have EPROM programmer burn the "BIOS" that I've attached. It doesn't do anything except output CPU ID on the POST card, using as little code as possible. If you see that show up on the POST card then it's just the BIOS issue, if not it's also ISA as I've mentioned above.
Is there 74F245 (or ALS245) anywhere near the ISA slots? Could be under a socketed chip. If there is one - it is suspect of being bad. I would however remove all cache chips for testing before commiting to any desoldering.

thank you for your answer. I have tried your BIOS but no post on the card. I have tried another BIOS I have found, the same behaviour, short beep.
There is one 74F245PC between the ISA slots, I have measured the resistance using the red wire on gnd, readings below.

- the last IC 74f244n, 8Mohm exepct I4 - 28kohm (furthest from the DS1287)
- second IC 74f244n, 8Mohm
- first IC 74F245PC, A0-A6 4,5Mohm - A7 28kohm - B0-B7 28kohm (closest to the DS1287)

I want to try to remove the 74f244 that are between the DS1284 and memory, at least the one that looks shorted but I need to get one or few before.
Any other suggestion here what to measure or try?

thank you, tomas

The tri-state buffers (F244) are probably for the ISA address lines. The typical incidents, for example when someone pulls an ISA card while the system is still on affects mainly non-buffered signals like IRQ, DRQ, DACK - lines.
Desoldering any circuits without prooving them defective before, ist wasted time.
I´d suggest you take the ISA pinout
https://upload.wikimedia.org/wikipedia/common … SA_Bus_pins.svg
connect the red probe of your multimeter to ground and measure _ALL_ the ISA contacts´ resistances (except the supply voltages and ground).

tsalat wrote on 2024-04-17, 15:35:

thank you for your answer. I have tried your BIOS but no post on the card.

I think it's way more likely to be address problem than a data line, since if the lowest 8 bits of data were properly glitched you'd get a dead mobo. No beeps because any kind of code fetch from BIOS ROM would be corrupted, so no valid program would be there to command the speaker. So forget that F245 - for now at least. You can check what the ouputs are connected to on the ISA bus, but don't try to remove it yet.

It was already pointed out that you should measure the resitance on ISA slot pins. I'd focus mainly on the lower 8 bits part and mostly on address lines. It's not unusual to see different values here, some lines can be connected to other things on the mobo, but there should be no shorts. Anything lower than say 500 ohms is suspicious. You have a scope, right? You mentioned a square wave on the BIOS chip. If so try looking at the address lines, all lower 16 bits of address should be toggling. Some faster than others but if you see a line that is stuck, or indeed at weird voltage that is not in spec with TTL signals, that's a clue.

As for the '244 chips, see if you can figure out what is connected to those, esp. the pins that appear shorted. If it's ISA slot signals do make sure there's nothing in the slots that could be causing the short (perhaps something fell in there).
The beeping could also be a memory issue but the lack of POST card codes is unusual, I don't remember any 386 BIOS that would not output any.

BTW did you try a manual reset of the mobo by using the switch? In case there's an issue with power good signal and the mobo is not being properly reset. All kind of weird things can happen if the mobo thinks it's a warm reboot and doesn't init the hardware properly.

majestyk wrote on 2024-04-17, 15:48:

The tri-state buffers (F244) are probably for the ISA address lines. The typical incidents, for example when someone pulls an I […]
Show full quote

The tri-state buffers (F244) are probably for the ISA address lines. The typical incidents, for example when someone pulls an ISA card while the system is still on affects mainly non-buffered signals like IRQ, DRQ, DACK - lines.
Desoldering any circuits without prooving them defective before, ist wasted time.
I´d suggest you take the ISA pinout
https://upload.wikimedia.org/wikipedia/common … SA_Bus_pins.svg
connect the red probe of your multimeter to ground and measure _ALL_ the ISA contacts´ resistances (except the supply voltages and ground).

about the de-soldering - lesson learned for now not to jump in conclusions.
The resistance on all the ISA contacts ADR/DATA is about 10kohm while the source is connected to the AT (not powered on). I have used the layout from wiki to write down the rest below, some were around 350ohm, and the rest either 10kohm or 6-8Mohm.
Refresh, MEM CS16, IO CS 16, MASTER, 350ohm
IRQ, DRQ, ... - 6-8Mohm

tomas

Deunan wrote on 2024-04-17, 16:23:

I think it's way more likely to be address problem than a data line, since if the lowest 8 bits of data were properly glitched y […]
Show full quote

tsalat wrote on 2024-04-17, 15:35:

thank you for your answer. I have tried your BIOS but no post on the card.

I think it's way more likely to be address problem than a data line, since if the lowest 8 bits of data were properly glitched you'd get a dead mobo. No beeps because any kind of code fetch from BIOS ROM would be corrupted, so no valid program would be there to command the speaker. So forget that F245 - for now at least. You can check what the ouputs are connected to on the ISA bus, but don't try to remove it yet.

It was already pointed out that you should measure the resitance on ISA slot pins. I'd focus mainly on the lower 8 bits part and mostly on address lines. It's not unusual to see different values here, some lines can be connected to other things on the mobo, but there should be no shorts. Anything lower than say 500 ohms is suspicious. You have a scope, right? You mentioned a square wave on the BIOS chip. If so try looking at the address lines, all lower 16 bits of address should be toggling. Some faster than others but if you see a line that is stuck, or indeed at weird voltage that is not in spec with TTL signals, that's a clue.

As for the '244 chips, see if you can figure out what is connected to those, esp. the pins that appear shorted. If it's ISA slot signals do make sure there's nothing in the slots that could be causing the short (perhaps something fell in there).
The beeping could also be a memory issue but the lack of POST card codes is unusual, I don't remember any 386 BIOS that would not output any.

BTW did you try a manual reset of the mobo by using the switch? In case there's an issue with power good signal and the mobo is not being properly reset. All kind of weird things can happen if the mobo thinks it's a warm reboot and doesn't init the hardware properly.

I have checked the voltages and the signal from the ADR lines, again using the wiki layout. The scope is toggling with the beeps but no of the ADR lines seems to be stuck, they are toggling between noise and signal (I have made a screen of the noise as well).

adr-23-20, 0,16V
adr-19-17, 3V
adr-00-11, 4V
adr-12-19, 0,3V

The lines which seems to be shorted on the one 244 are going to the board, I have removed the IC with no difference in signal-still on the GND, not sure how many layers the board has (I would guess 4 at least) but the GND pin inside the board looks like a "cross" while looking through the light and these are looking the same while looking at the board from the ICs side, different from the bottom side - not sure if this make sense 😀. The GND looks the opposite from the bottom side like the cross, from the top side no. There are 0ohms between GND and those pins - no idea where they travelling or if they are on the GND. If I beep out the same pins from the other ICs next o it, they go to the chipset. I have tried the same memory in other boards, always working. I have tried different sticks, all 8x positions populated with the same result.

I have tried to start the board with holding reset, resetting the board while running, no difference.

br, tomas

Too lazy to decompile your bios, but usually bioses will write to POST card before turning on the siren. Are you sure your POST card is working correctly? have you tried it before in other boards?

tsalat wrote on 2024-04-17, 18:24:

The lines which seems to be shorted on the one 244 are going to the board, I have removed the IC with no difference in signal-still on the GND

which pins are those on 244, and where are corresponding 244 outputs going exactly? Its entirely possible those are unused pins grounded on purpose

tsalat wrote on 2024-04-17, 18:04:

...IRQ, DRQ, ... - 6-8Mohm...

Are you sure, with the red probe connected to ground??

rasz_pl wrote on 2024-04-17, 18:35:

Too lazy to decompile your bios, but usually bioses will write to POST card before turning on the siren. Are you sure your POST card is working correctly? have you tried it before in other boards?

tsalat wrote on 2024-04-17, 18:24:

The lines which seems to be shorted on the one 244 are going to the board, I have removed the IC with no difference in signal-still on the GND

which pins are those on 244, and where are corresponding 244 outputs going exactly? Its entirely possible those are unused pins grounded on purpose

The ones that are on GND are I6, I5 and I4.
The same pins I6, I5 from the IC next to it are going to the SL82C461. I4 is on GND.

br, tomas

majestyk wrote on 2024-04-17, 18:37:

tsalat wrote on 2024-04-17, 18:04:

...IRQ, DRQ, ... - 6-8Mohm...

Are you sure, with the red probe connected to ground??

yes, red probe on GND.

br, tomas

tsalat wrote on 2024-04-17, 18:46:

The ones that are on GND are I6, I5 and I4.

and corresponding outputs?

rasz_pl wrote on 2024-04-17, 19:05:

tsalat wrote on 2024-04-17, 18:46:

The ones that are on GND are I6, I5 and I4.

and corresponding outputs?

seems to be unconnected, i think you are right here, I am not able to trace them from the IC anywhere.

tomas

tsalat wrote on 2024-04-17, 18:24:

I have checked the voltages and the signal from the ADR lines, again using the wiki layout. The scope is toggling with the beeps but no of the ADR lines seems to be stuck, they are toggling between noise and signal (I have made a screen of the noise as well).

So all 24 address lines on ISA bus are toggling? Interesting. Some noise there is expected, this is still in spec so I wouldn't worry too much about it.
If it's not address then perhaps one of the control signals. Please also check (with the scope) RESET, IOR, IOW, MEMR, MEMW, AEN, ALE, IO CH (both), NOWS, CLK, OSC. Only the lower 8-bit part of the slot, and you can check or skip IRQ and DRQ/DACK lines for now, these would be ignored by the system this early into boot process so don't matter yet.

BTW did you try to discover what it is connected to the F245 input/output lines? One side should be tied to chipset, the other to ISA data bus, but the question is which byte - upper or lower. You don't much care for the upper byte at the moment, writes to POST card only use the 8-bit part of ISA slot. And speaking of POST card, did you try to put it into different slots? Try all of them, but above all the first and last ones. Sometimes a break in the copper trace can isolate some ISA slots from others but you can't be sure where the break is and what slots (if any) still work. So it's worth checking it.

Deunan wrote on 2024-04-18, 10:36:

So all 24 address lines on ISA bus are toggling? Interesting. Some noise there is expected, this is still in spec so I wouldn't […]
Show full quote

tsalat wrote on 2024-04-17, 18:24:

I have checked the voltages and the signal from the ADR lines, again using the wiki layout. The scope is toggling with the beeps but no of the ADR lines seems to be stuck, they are toggling between noise and signal (I have made a screen of the noise as well).

So all 24 address lines on ISA bus are toggling? Interesting. Some noise there is expected, this is still in spec so I wouldn't worry too much about it.
If it's not address then perhaps one of the control signals. Please also check (with the scope) RESET, IOR, IOW, MEMR, MEMW, AEN, ALE, IO CH (both), NOWS, CLK, OSC. Only the lower 8-bit part of the slot, and you can check or skip IRQ and DRQ/DACK lines for now, these would be ignored by the system this early into boot process so don't matter yet.

BTW did you try to discover what it is connected to the F245 input/output lines? One side should be tied to chipset, the other to ISA data bus, but the question is which byte - upper or lower. You don't much care for the upper byte at the moment, writes to POST card only use the 8-bit part of ISA slot. And speaking of POST card, did you try to put it into different slots? Try all of them, but above all the first and last ones. Sometimes a break in the copper trace can isolate some ISA slots from others but you can't be sure where the break is and what slots (if any) still work. So it's worth checking it.

hi, thank you all for your help, highly appreciate it!

The F245 lines are attached, CHP stands for chipset, the rest is mapped to the ISA slot as seen in here: https://upload.wikimedia.org/wikipedia/common … SA_Bus_pins.svg.
I have two new post cards, both are working in other systems but I have checked both and the behaviour is the same. I have tried all the slots, no difference.

The output from the scope attached.

br, tomas

*second batch of the scope

*third batch from the scope
IO CH RDY and IO CH CK are looking the same

tsalat wrote on 2024-04-19, 17:31:

IO CH RDY and IO CH CK are looking the same

That's expected in a system with no ISA cards, and in general these signals are rarely used. Both should be H and they are. RESET is noisy but held L and well below 0,8V so in spec. Everything else seems in spec. MEMR appears dead but it can't be, the other lines toggling mean there is a program executing. I guess you run this with cache chips populated? In which case there might not be memory fetches if the cache always hits. CLK looks odd but it could be either triggering issue or some clock streching to hide memory refresh maybe. Hard to tell from one photo.

The F245 is used here to drive the upper address bits, the 16-bit ISA slot extension works a bit differently. But as I've said you don't need to worry about that now, we want to get the 8-bit part of the slot working first.

I did notice you changing the trigger level on some signals, please don't do that. You are only capturing the ringing noise and it's hard to tell from the photo if there are any actual logic transitions. In fact I now that I look at your earlier report about the address signals I see the same thing so now I'm not so sure that all the bits flip or not. There is one photo that shows only noise and no state changes. So please re-test the ADR00 to ADR19 lines again, no need to photo everything. If you see something like on photos 1,2 and 4 then it's OK. But photo 3 is an example of inactivity that should not really happen during normal operation. The code might just be in such a loop that a particular address line will not change but's that's unusual so write all those down if you see it.