VOGONS


SUNTAC 80286 Mainboard help

Topic actions

Reply 20 of 40, by kenrouholo

User metadata
Rank Member
Rank
Member

The type of cap you circled is a ceramic cap and they aren't really good types of caps for many reasons, but they also don't tend to go bad.

Your main caps are the yellow "drop" shaped things. Tantalum caps. They're similar to electrolytic caps in performance and I don't believe the construction is very different, either, though I don't know as much about the materials science side of them.

You can't test a capacitor with a regular multimeter. You must have an ESR meter. I mean you can use a multimeter to sometimes determine a bad cap, but only very rarely will a cap be so bad that you'll find it that way. It's pretty much a waste of time. You can try it with the tantalum caps if you want but don't expect much in the way of results from this method.

Edit: Keep meaning to reply and say 2 things, and forgetting to say the second thing...

What did the ceramic cap you circled act like?

It may be in parallel with other devices that throw your results off, which is fairly common "in circuit." Or is it appearing open? Maybe it has some charge on it if so. When your meter says infinite/open, what it means is "extremely high resistance" which a charged cap could potentially present.

Yes, I always ramble this much.

Reply 21 of 40, by creepingnet

User metadata
Rank Member
Rank
Member
Predator99 wrote:
OK, I checked most of the capacitors with this method as I only have a simple Multimeter: […]
Show full quote

OK, I checked most of the capacitors with this method as I only have a simple Multimeter:

https://www.reference.com/science/test-capaci … 83b105dd3ea036e#

All capacitors act like desribed. With exception of this one

cap.jpg

Unfortunately it is the only capacitor of this kind on the whole board so that I am not able to compare this behaviour with another one. Additionally, its the - in terms of size - biggest one - so maybe there is another explanation that both pins are not connected to each other?

Hmm, could be the faulty capacitor is the issue. Looks to be a 0.047uF 473 Ceramic Disc Capacitor, same kind used in electric guitar tone control circuits, and they are pretty common at places like Radio Shack, Should be able to find one, swap it out, and find out if that's truly the issue. Heck, could pull a mylar (the green plastic ones) if within spec....might work.

~The Creeping Network~
My Website - https://sites.google.com/site/thecreepingnetwork/home
My Youtube Channel - https://www.youtube.com/channel/UCc6sYw9FvwuKahBHE_06diA

Reply 22 of 40, by kenrouholo

User metadata
Rank Member
Rank
Member

Yes a film cap like a mylar cap will generally work fine to replace a ceramic cap. It'll cost more but if you don't need a large quantity it'll be fine. It'll also be physically larger, but not overly so if you get the right one. And yes most capacitors including ceramic do have standardized codes so interpreting it that way is correct.

That said, ceramic caps aren't that useful in many circumstances. They're used because they are cheap and small. But their characteristics suck for signal paths and these are probably bypass caps, and if open of them opened, I'm not sure it's going to be a huge deal. I'm not saying it's 100% definitely a bypass cap, nor am I saying 100% that a bypass cap failure would be OK. The clock speeds in a system like this are not high so it's possible it could be a coupling cap or something even though ceramic caps aren't great for that. Just not totally convinced that cap's the issue.

Last edited by kenrouholo on 2017-02-17, 19:49. Edited 2 times in total.

Yes, I always ramble this much.

Reply 23 of 40, by BloodyCactus

User metadata
Rank Oldbie
Rank
Oldbie
Predator99 wrote:
OK, I checked most of the capacitors with this method as I only have a simple Multimeter: […]
Show full quote

OK, I checked most of the capacitors with this method as I only have a simple Multimeter:

https://www.reference.com/science/test-capaci … 83b105dd3ea036e#

All capacitors act like desribed. With exception of this one

cap.jpg

Unfortunately it is the only capacitor of this kind on the whole board so that I am not able to compare this behaviour with another one. Additionally, its the - in terms of size - biggest one - so maybe there is another explanation that both pins are not connected to each other?

you cant test capacitors while they are on the motherboard. its a false reading (your testing the capcitance of the entire circuit), you also need to be careful as testing like this, puts voltage into the circuit, you must know which side of polarity you are on if you are going to be pumping voltage out into pins that might not want a reverse voltage coming in

--/\-[ Stu : Bloody Cactus :: http://kråketær.com :: http://mega-tokyo.com ]-/\--

Reply 24 of 40, by h-a-l-9000

User metadata
Rank DOSBox Author
Rank
DOSBox Author

This capacitor has such a small value that you won't see anything with a normal multimeter. On most other capacitors you can only measure something because they are at the supply rails.

1+1=10

Reply 26 of 40, by Predator99

User metadata
Rank Oldbie
Rank
Oldbie

Sorry for the stupid questions 😉

Regarding the Tantalum caps: I identified 2 different kinds...

This one: "A 10µ 16"

A104Z-1.jpg
Filename
A104Z-1.jpg
File size
73.47 KiB
Views
860 views
File license
Fair use/fair dealing exception
A104Z-2.jpg
Filename
A104Z-2.jpg
File size
63.33 KiB
Views
860 views
File license
Fair use/fair dealing exception

And this one: "A 10 Z"

A10-16.jpg
Filename
A10-16.jpg
File size
528.14 KiB
Views
860 views
File license
Fair use/fair dealing exception

Can you tell me what the characteristics are and what replacement I have to look for? I did not find anything abou these values...

Reply 27 of 40, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++

I'm not sure what the letter code is for. Perhaps tolerance? Through hole tantalum caps aren't as standardized as SMD parts. The first cap is probably 10V, 47uF. The third is obviously 16V, 10uF.

All hail the Great Capacitor Brand Finder

Reply 28 of 40, by BloodyCactus

User metadata
Rank Oldbie
Rank
Oldbie

tantalums dont need to be replaced like electrolytics. tantalums dont loose thier capacitance.

--/\-[ Stu : Bloody Cactus :: http://kråketær.com :: http://mega-tokyo.com ]-/\--

Reply 29 of 40, by Predator99

User metadata
Rank Oldbie
Rank
Oldbie

So then...what is the advice? 😉 Should I replace the caps or not? I have learned the Tantalum will not fail and also not the Ceramics. Bute as far as I can see there are no others.

In the mean time I bought a replacement of the 473 with the differnt behaviour. I measured the resistance between the Pins and they dont seem to be connected on the new one, too. Therefore I think the one on the board is also OK.

I also fixed the RTC, but still no sign of life...

Reply 30 of 40, by h-a-l-9000

User metadata
Rank DOSBox Author
Rank
DOSBox Author

The small caps are not tantalum but film capacitors. Regarding the size the '10' could be nF. The voltage is coded into the letters, but 16V+ are safe.

But in your situation these caps would be one of the last thing to investigate. If you want to swap something, start with the crystals, they can be damaged by mechanical shock. Or the 286-10 may have burned out over time when ran at 12MHz.

Connecting a LED directly to 5V and GND will quickly burn it out. As to why the LED doesn't work on the power LED connector: It appears the board has the combined 5-pin power LED / keylock connector. The power led goes there with one free space between the pins.

Besides swapping parts, for a deeper error analysis you need better equipment (POST card, logic probe, oscilloscope).

1+1=10

Reply 31 of 40, by kenrouholo

User metadata
Rank Member
Rank
Member
Predator99 wrote:

So then...what is the advice? 😉 Should I replace the caps or not? I have learned the Tantalum will not fail and also not the Ceramics.

Tantalum can actually fail but I doubt it's happened here.

Try the diagnostic card Jo22 mentioned. You can find them as imports from various places and they are easy to find and quite cheap. Many have an ISA slot on one side and a PCI slot on the opposite side. They'll have a 2 or 4 digit display which will go through error codes and it will at some point stop and you will have a list (might need to find a list on Google if you get a cheap imported one that doesn't include a list of codes)... The problem is likely at the code it stopped on or the next code it would have gone to (which you may or may not be able to figure out).

BloodyCactus wrote:

tantalums dont need to be replaced like electrolytics. tantalums dont loose thier capacitance.

They can go bad with heat and time and excessive ripple like standard electrolytics, but they are considered better than most other types of earlier electrolytics. I'm not saying they're the problem, just that they also aren't timeproof. Some do have liquid electrolyte. They actually are electrolytic capacitors but they are somewhat unique so people keep them under a separate label much of the time... but they do suffer from most of the same problems that other types of electrolytic do.

Yes, I always ramble this much.

Reply 32 of 40, by Predator99

User metadata
Rank Oldbie
Rank
Oldbie

Yes, after reviewing the manual also noticed what i made wrong with the Power- and Turbo-LEDs, now they are working...
I replaced the CPU with another 80286-12, but nothing.

I ordered this POST card here
http://www.ebay.com/itm/142139186763

Maybe you are right, I will wait until it arrives to get a better idea...

Thanks again to you all for the moment!!! 😀

Reply 33 of 40, by Predator99

User metadata
Rank Oldbie
Rank
Oldbie

The diagnostics card arrived from China - I am very surprised how fast and cheap this was 😀 So the quest can go on....

The diagnostics card is working properly and shows the correct codes in a working PC.

However, in the 286, the tests do not even start and the display shows "----".

All Status-LEDs are lighting. Only the "Reset-" LED is off. It is flashing very short when presing the Reset-Key.
The "IRDY"-LED (LED that turns on and off when there is a message) is also lighting. In my working PC, it is off most of the time.

So for me this looks like everything is OK but the execution of the BIOS-code does not even start? Maybe the ROMs are damaged...but this seems to be impossible as I tested them as written above?

DSC_0257.JPG
Filename
DSC_0257.JPG
File size
233.91 KiB
Views
552 views
File license
Fair use/fair dealing exception

Reply 34 of 40, by creepingnet

User metadata
Rank Member
Rank
Member
Predator99 wrote:
The diagnostics card arrived from China - I am very surprised how fast and cheap this was :-) So the quest can go on.... […]
Show full quote

The diagnostics card arrived from China - I am very surprised how fast and cheap this was 😀 So the quest can go on....

The diagnostics card is working properly and shows the correct codes in a working PC.

However, in the 286, the tests do not even start and the display shows "----".

All Status-LEDs are lighting. Only the "Reset-" LED is off. It is flashing very short when presing the Reset-Key.
The "IRDY"-LED (LED that turns on and off when there is a message) is also lighting. In my working PC, it is off most of the time.

So for me this looks like everything is OK but the execution of the BIOS-code does not even start? Maybe the ROMs are damaged...but this seems to be impossible as I tested them as written above?

DSC_0257.JPG

Hey, you bought the exact same card I have and was using on my 486 last week!

The Reset light is supposed to blink briefly when you press reset - it's sending a reset signal to the system for only a fraction of a second to cold-reboot the machine without turning it off.

Might pay to connect that card into the internal speaker port on the motherboard and see i that is making any noise if you've not done so already. IIRC, when testing the 486 last week, Might pay to plug the speaker cable that came with the card into the board and see if any beep codes are happening, then check the reference sheet that came with it.

The symptoms you describe though sound a lot like what I went through getting my 486 board towkring last week. Turns out the problem was an inaccuracy in jumper settings on TH99 regarding my CPU was causing the CPU not to work right. Ended up going to the OEM of my board whom still had the files for it on their FTP server (First Internation Computer (FIC)) and found out they had made revised jumper settings for my CPU. When it was not booting with the working CPU, it had IRDY on, it was off when the non-working CPU I had was in the machine. So that may be a sign your CPU is okay as you already tested it, and something else.

One thing that really helped me was looking up my BIOS and seeing what the POST sequence was from "Power Good" onward - that allowed me to track down possibilities of what was bothering the machine. in my case it was CPU, but it could be something else stopping it in yours that gets a light test/refresh before the BIOS code kicks in. Maybe a jumper set for a 287 co-processor that is not present or something.

~The Creeping Network~
My Website - https://sites.google.com/site/thecreepingnetwork/home
My Youtube Channel - https://www.youtube.com/channel/UCc6sYw9FvwuKahBHE_06diA

Reply 35 of 40, by Predator99

User metadata
Rank Oldbie
Rank
Oldbie

Hehe, yes the card is really cool! I do not know how to live without it before 😉 You will know its working quite simple to display the output of port 80, you just can write a small program..
mov al,EE
out 80,al
..and you see the EE on the LED..cool 😉

I also have this cable, but its not mentioned in the documentation that it can be used for the connection to the speaker-out? Therefore I have not tried it yet.

Back to my board...unfortunately there is not much to configure regarding the CPU. The relevant Jumpers are related to memory and keyboard configuration but these components do not seem to be the cause for the failure. I am sure the CPU works as I already tested 2 different CPUs and also the presence of the IRDY signal seems to confirm this as you already wrote.

My BIOS POST sequence will not help as the BIOS execution does not even start. The POST usually starts with a simple CPU test and outputs the result on port 80..therefore something should be displayed on the card and not a "----".

The problem seems to be related to the CPU or the BIOS...but the CPU and the BIOS itself seem to be working.

Currently I have no idea where I should look after next...

Reply 36 of 40, by Predator99

User metadata
Rank Oldbie
Rank
Oldbie

To give an update on this: Yesterday I received another very similar SUNTAC Board. It looks much worse than the other one. Dirty, very corroded and probably stored outside. But: it works! First I did was to remove the battery...

To check if the BIOS is the reason the first board does not Power up I exchanged the Chips: The new Board doesnt power-up with the old chips, but the Diagnostic card starts to count. Therefore the other board should do this, too. The old board does also not work with the chips from the new board, however...

Therefore still no idea what is wrong with the first one. But these boards seem to be very robust 🤣

IMG_2668-600.jpg
Filename
IMG_2668-600.jpg
File size
1.82 MiB
Views
376 views
File license
Fair use/fair dealing exception
IMG_2672-600.jpg
Filename
IMG_2672-600.jpg
File size
1.32 MiB
Views
376 views
File license
Fair use/fair dealing exception
IMG_2673-600.jpg
Filename
IMG_2673-600.jpg
File size
1.33 MiB
Views
376 views
File license
Fair use/fair dealing exception
IMG_2675-600.jpg
Filename
IMG_2675-600.jpg
File size
1.34 MiB
Views
376 views
File license
Fair use/fair dealing exception
DSC_0309.JPG
Filename
DSC_0309.JPG
File size
1.28 MiB
Views
376 views
File license
Fair use/fair dealing exception

Reply 37 of 40, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

That's a pretty unusual way to put a surface mount CPU on a board. I wonder what that little plastic stand is for.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 39 of 40, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

I wonder if it was a way to convert a board designed for a PGA 286 to one that uses surface mount...sort of like those surface mount am386-40 chips that came on the PGA adapters.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium