VOGONS


Reply 20 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

I check the clock on the socket with and without cpu, and it's out of spec 😒 (~4.5ns and ~4ns)
The 74act08 get +5Vcc and 0.012Vss so power is OK
I have doubt about the reset : sometimes it goes straight 0V even if the clock is not OK. And i try to push the reset button, it does nothing to the state of the socket reset pin.
59476 59477

Reply 21 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

After un-plugin/plugin all thing one more time the SC425 came back to life bringing me a nice 33Mhz sine on socket.
Even meeting the 3ns rise time spec !
But still no beeps (my speaker is a true speaker so i can't plug it backward it will work anyway), no boot no VGA output.
At one moment without any chips (no cpu, no ram, no bios, no jetkey) i have seen "tops" on the bios addresses.
But when i plug back the cpu they were gone.
59530

Reply 22 of 46, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

Nice signal, are all 3 outputs looking ok now? Maybe this indicates some pin corrosion, so a bit more cleaning of sockets? Does the Reset pin still go low (I know you said it seemed to be going low anyway)? Can you get a probe on any of the CPU address pins on the back of the board and see if there's any activity there? I don't know exactly how the BIOS is connected to the CPU, I think it hangs off of the ISA bus somehow. If that's right I assume the signals would have to go via the SC9204A chipset, but I can't find a datasheet for that. You said the Reset signal doesn't change when you press the reset button, might be worth watching the NMI line as well. But without datasheets to look at I'm getting quickly out of my depth. It probably won't tell us anything more, but it might be worth getting hold of a working POST card, you can get some that have a few more status lights that the one you have that didn't work.

Oh, and what speaker are you using? You probably already know, but the PC speaker header has one side held at 5V and then it toggles the other side from HiZ to Ground, so it should work with pretty much anything. You can even stick something like a 1k resistor between pin 1 and 4 and then watch pin 1 with your scope.

Reply 23 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

I just check the other they're good, PD is 2.4V continuous
Yeah reset still goes low.
I have an old "buzzer" that look like a beeper (black cylindrical) but it's a real speaker that could play music if attached to a jack. (i actually test it that way)
I have measured 0 ohm between A0 on bios and A0 on the ISA bus, i check 2-3 other addresses pin and they were the same.
NMI is LOW and stay low when i push reset.
I have to say that what is cool with this MB is the double socket.
I can take all my measurements very easily by plugging probes into the 386 socket, every time i check, the pin is directly connected to the same pin on the 486 socket.

With what i see on this board, i doubt the a post card could help me, the addresses are stuck, don't the post card used it ?

Reply 24 of 46, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

So the CPU should be out of reset, but you're still not seeing any activity on its address lines? At that point I'm lost, I'm pretty certain that once Reset is released that the CPU should start fetching instructions, so low bit value address lines should be switching. Quick look at the 486 datasheet says there's also an Sreset pin which is similar to the Reset pin, so maybe something else is holding it in reset?

Snippet from the 386 datasheet:

2.10 RESET AND INITIALIZATION
When the processor is initialized or Reset the registers have the values shown in Table 2-7. The Intel386 DX will then start executing instructions near the top of physical memory, at location FFFFFFF0H. When the first InterSegment Jump or Call is executed, address lines A20-31 will drop low for CS-relative memory cycles, and the Intel386 DX will only execute instructions in the lower one megabyte of physical memory. This allows the system designer to use a ROM at the top of physical memory to initialize the system and take care of Resets.

So I think there should be something happening. Stumped. What does HOLDA look like? I think that can disconnect the CPU from the address bus.

Speaker sounds like one of those little piezo buttons often solder on to motherboard, so from what you've tested it should be fine. Just to check, it does have the 4 pin connector to go from pin 1-4 on the header?

[edit, partly because I got my HOLD and HLDA the wrong way around: I know some boards won't boot if there's no battery, so maybe try putting ~5v on the external battery connector and doing a CMOS clear with the jumper? Maybe something is somehow disabling the CPU if there's no CMOS battery. Seems unlikely. Pin 78 of the 82c206 is a Vcc pin, check the voltage there. That chip also has a HRQ pin, so it could affect the CPUs. Actually, since there is a datasheet for it, probably worth checking the Sysclk (pin 15), Osci (pin 75), Reset (should be same as CPU, pin 10)], HRQ (pin72). But I am just casing around at this point. Hey, at least I think I helped with the power LED.]

Reply 25 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

Yeah, thank for your valuable help !
Sreset seems to exist only on the embedded version
I have put 5V on the ext.battery
I have check HOLD (not HOLDA, i cannot find it on the 386 socket), RESET, NMI and INTR
I found that INTR is continuously HIGH on the scope., so an interrupt is present...
I don't check pin on the 82c206 it's not easy, i think i need to solder a wire each time i want to measure something on it 🙁
I'm a little desperate for this board.
I think i'm better waiting for the M919 i just bought on ebay (a full kit with ram and cpu certified working) and i will test my cpus on it.

Reply 26 of 46, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie
crazytiti wrote on 2022-01-18, 19:28:

I have check HOLD (not HOLDA, i cannot find it on the 386 socket), RESET, NMI and INTR
I found that INTR is continuously HIGH on the scope., so an interrupt is present...

Sorry, HLDA on the 386.

Maybe the INTR is something. OTOH, all the microcontrollers I've come across have interrupts disabled when they come out of reset, so they can deal with setting up registers and interrupt vectors before they start being interrupted. CPU must surely be the same, so it shouldn't stop the CPU trying to read the BIOS, so there should still be some activity on the address lines. I'd guess that when running the BIOS that it enables interrupts at some point.

With board off, can you check the resistance of INTR to +5 and to GND, and the resistance to pin 73 on the 82c206 (that should be safe to check with the board off, shouldn't matter if the probe accidentally touches two pins)? It's an active high signal, so might be pulled to ground via a resistor, but should be high resistance to +5. Maybe something has gone short, and it has then that part might have other problems.

I think i'm better waiting for the M919 i just bought on ebay (a full kit with ram and cpu certified working) and i will test my cpus on it.

That may be the way to go. It's always far easier being able to swap parts between two systems.

Reply 27 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

INTR to ground or +5 is open circuit even with the cpu installed
Short ? it seems more open ^^
Pin 73 is equally open to GND/5 but continuity to INTR.
maybe the 82c206 is dead 😒

I can't even remember where i've got this motherboard.

Reply 28 of 46, by Eep386

User metadata
Rank Member
Rank
Member

Symptoms of 82C206 death:
- Board is stone dead upon power up, but sometimes boots when you hit (sometimes repeatedly) the RESET button (bad reset logic - could also be external components related to the reset logic)
- Random BIOS crashes/lockups or errors in the lower 64K range regardless of the memory you use (bad internal 74LS612)
- IRQ or DMA channels non-working in any slot (bad internal IRQ/DMA logic)
- Bizarre uneven speed or unusually low speed regardless of turbo button or "green" function status (bad internal timer logic)
- RTC problems regardless of battery (bad internal RTC, could also be external components related to it - these external parts are often trashed by VARTA bombs)

Life isn't long enough to re-enable every hidden option in every BIOS on every board... 🙁

Reply 29 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

Ok, i'm stuck at the first one ^^
I even try to resolder the 82c206 (sometimes warming chips made them work) with some flux and the heat gun.
As usual nothing change.
Thank for all your help, i now know more deeply my electronic friends.

Reply 30 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

I have received my M919, it's working (at leat POST and counting ram) with its DX4-100
So i have tested all my CPU :
DX33 : OK (even boot at 3.3V but crash at counting ram)
DX2 66 : OK (even boot at 3.3V and counting ram)
AMD dx4 120 : OK
I think i will sell the motherboard on ebay "for parts", sadly it's the only 386 motherboard i have 😒

Reply 31 of 46, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

Just wish I knew more about CPU start up. Don't suppose you could make some comparison measurements, like on that interrupt line that seemed to be stuck high? If you wanted to do a speculative swap out, maybe try to find a replacement for the 206? Probably no chance of finding the main chipset chip, but the different versions of this motherboard have several different 206 chips, so they might be pin compatible and a bit more available (UM82C206F, 85C206, KS83C206Q).

Reply 32 of 46, by crazytiti

User metadata
Rank Newbie
Rank
Newbie

Now that i have one fully working 486 i will not bother swap IC.
Someone on the french forum hardware.fr wants to buy it for 20€ and try himself, maybe he will post here 😀
I find that i've done enough myself to try revive it.

Last edited by crazytiti on 2022-03-16, 00:24. Edited 1 time in total.

Reply 33 of 46, by Deksor

User metadata
Rank l33t
Rank
l33t

Hello, I'm the new owner of the board 😁

So I replaced the 82c206

IMG_20220315_195955.jpg
Filename
IMG_20220315_195955.jpg
File size
1.79 MiB
Views
989 views
File license
CC-BY-4.0

Unfortunately the symptoms remain exactly the same 🙁.

I have a post card with more LEDs and one reports the reset signal status. I noticed that the reset signal does never fire. Not at the startup, not when I use the reset button. Just never.

I also saw that the reset button goes straight into the chipset. This makes me wonder if the chipset itself isn't toast 🙁

Trying to identify old hardware ? Visit The retro web - Project's thread The Retro Web project - a stason.org/TH99 alternative

Reply 35 of 46, by Eep386

User metadata
Rank Member
Rank
Member

It would be nice if that USA chipset could get a positive ID.
I suspect it's either an OPTi or a Unichip, but without any further clues (a pin-out would be immensely helpful if one existed) it's hard to guess.

Life isn't long enough to re-enable every hidden option in every BIOS on every board... 🙁

Reply 36 of 46, by WJG6260

User metadata
Rank Member
Rank
Member
Eep386 wrote on 2022-03-17, 15:16:

It would be nice if that USA chipset could get a positive ID.
I suspect it's either an OPTi or a Unichip, but without any further clues (a pin-out would be immensely helpful if one existed) it's hard to guess.

I actually almost bought one of these boards a while ago, but before doing so, looked up and saved the manual. It happens to now be conveniently hosted at Ultimate Retro, and Page 6 details a pinout of the USA chipset. I'd be curious to see what your thoughts are; it might actually be proprietary? Anyway, here's the link!

-Live Long and Prosper-

Feel free to check out my YouTube and Twitter!

Reply 37 of 46, by Deksor

User metadata
Rank l33t
Rank
l33t

Oh my god, why didn't I see this !!

This needs some investigations/comparing with other chipsets.

Trying to identify old hardware ? Visit The retro web - Project's thread The Retro Web project - a stason.org/TH99 alternative

Reply 38 of 46, by WJG6260

User metadata
Rank Member
Rank
Member
Deksor wrote on 2022-03-18, 23:46:

Oh my god, why didn't I see this !!

This needs some investigations/comparing with other chipsets.

Any way I can help?

The most likely suspects are something from the OPTi 495 line or something from the Unichip line, as Eeep386 suggested. Are pinouts available for any of these?

-Live Long and Prosper-

Feel free to check out my YouTube and Twitter!

Reply 39 of 46, by Deksor

User metadata
Rank l33t
Rank
l33t

For now, most datasheets are stored here http://66.113.161.23/~mR_Slug/pub/datasheets/chipsets/. There's work being done in ultimateretro to move them to actual chipsets instead

Edit : the chip that reassembles the most to the USA chip is this one I think (look for boards using it) https://www.ultimateretro.net/en/chipsets/660

It has the same amount of pins, is the only chip of the chipset (except for the 82c206), is used in the same context (386/486 and VLB) and was released probably the same year (1992).

Last edited by Deksor on 2022-03-19, 00:37. Edited 2 times in total.

Trying to identify old hardware ? Visit The retro web - Project's thread The Retro Web project - a stason.org/TH99 alternative