• Please review our updated Terms and Rules here

DTK 286 AT systems don't boot.

J

joelgraff

Member
Joined
Jun 16, 2018
Messages
40
Location
Illinois, USA
I have two DTK 80286-12 systems that won't boot. One powers up, spins up the HDD and sounds like it's going through a BIOS POST, but no system beeps. The other does less than that.

Both systems had leaky barrel batters, which I removed and cleaned the PCB. The coating is eaten away in several spots, but the traces appear intact - they toned out, anyway.

Not sure what's wrong - do these need a battery to boot?

https://cloud.thinkopenllc.co/index.php/s/PPJjK67EBd7N39t
 
joelgraff said:
I have two DTK 80286-12 systems that won't boot.
Click to expand...
From the photo, it appears to be the motherboard that is detailed at [here]. (Or a very close version.)

joelgraff said:
... but no system beeps.
Click to expand...
The photo shows no speaker connected to JP3.

eswan said:
Shouldn't that have the ram chips in sets of 9? Probably 2 rows of 9 chips at minimum for 16-bit.
Click to expand...
Yes. It looks as though of the RAM chips, someone removed the data ones and left behind the parity ones.

Still, for a 'no RAM' situation on a 286 based motherboard such as the subject one, I would expect beeps from the JP3-connected speaker and a code on the POST card. Even the IBM AT displays POST codes in a 'no RAM' situation.

eswan said:
Both systems had leaky barrel batters, which I removed and cleaned the PCB.
Click to expand...
Like as shown at [here], the green 'gunk' may be in locations that you have not considered. Have you done a thorough examination?

eswan said:
The coating is eaten away in several spots, but the traces appear intact - they toned out, anyway.
Click to expand...
Six or so months ago, I investigated a remote control (for a Blu-Ray player) that had stopped working. No output from IR LED. The remote was normally stored button side down (batteries up) and one of the AAA batteries had leaked onto the PCB. On the PCB was a very long thin trace that the battery leakage had reached, discolouring part of the trace. Looking at the discolouration, I thought, "It's hit the coating on the trace, but that's as far as it's gone." However, the trace measured open-circuit, and use of a curved-blade scalpel revealed that under the 'discolouration', the track had been eaten away in parts.
 
eswan said:
Shouldn't that have the ram chips in sets of 9? Probably 2 rows of 9 chips at minimum for 16-bit.
Click to expand...

Two rows of 9 would be 18 bit. Parity chips aren't always required, it may be possible to use just 16 chips.

But there are also two SIPP sockets next to the DIP sockets. He could either use a SIPP adapter, or solder pins to two 30 pin SIMM modules and install them.
 
modem7, I don't think it's a memory issue. The chips are 44256 chips - which makes them 4-bit memory chips. So two chips = 256K. Since I have eight, that's a total of 1 MB on-board RAM. The two parity chips are for the upper banks that are filled, though they don't match the requirements. The only thing I noticed were the parity chips were also 4-bit and a higher capacity than is spec'd in the manual, perhaps. Seems like that shouldn't matter... I also haven't reseated the RAM, so there could be a problem there. It's just odd that both computers exhibit nearly identical behavior and both had leaky batters with roughly equivalent trace / PCB damage.

Thanks for the manual, BTW. I had most of it figured out from other references, but the actual manual was helpful.

As for system speaker - it's not attached in the photo, but there really is no beeping that takes place. The POST card never puts out a code or gives me an "F0" which isn't a code for any version of AMI BIOS so far as I can tell. The AMI BIOS chips in here appear to be of 1989 vintage.

Regarding leaking, I've been over the board pretty carefully and checked voltages in the ISA slots, so I know power is getting around. I'm just not really sure where to go probing next. I suppose I'll have to break out my oscilloscope at some point and start looking up logic chip data sheets to see what is and isn't getting signal, but it's uncharted territory for me.

I suppose I was sort of hoping for a general diagnostic rubric for clone AT machines like there are for the IBM PC ATs...
 
You must log in or register to reply here.
Back
Top