PDA

View Full Version : IBM 5160 does not start anymore...



Jan1980
November 22nd, 2017, 02:55 PM
Hi

I bought an IBM 5160 before years. It worked fine all the time. Now, i thought, dismantled the IBM completely and cleaned everything. I dismantled power supply and cleaned it properly. After assembling powersupply, i connected it to mainboard and speaker to mainboard. Then after switch on, i heard a long and 2 short beeps.. Ok, thatīs fine, thatīs normal without any cards..

After this, i decided me to change the capacitators. I removed the 10ĩ 16V tantalium capacitators and fitted 10ĩ 25V electrolyt capacitators. Now it does not start anymore.. :-(

I made the + from the new capacitators in the middle hole and - in one of the outer holes. I soldered properly. How can i now check what is happen ??

The fan on power supply is running. All voltages are fine. I removed the RAM chips and swapped the RAMs from bank 1 to bank 0. Nothing !! Do you have an idea to make trouble shooting ?

THX

Jan

Chuck(G)
November 22nd, 2017, 04:14 PM
Make certain that you have bridged the two outer holes of the capacitor pads--if in doubt use a meter to check continuity.

modem7
November 22nd, 2017, 09:06 PM
Do you have an idea to make trouble shooting ?
I suggest the use of a Supersoft/Landmark Diagnostic ROM.
See [here (http://minuszerodegrees.net/supersoft_landmark/Supersoft%20Landmark%20ROM.htm)].

Jan1980
November 23rd, 2017, 12:32 PM
So, i tried. I checked the resistance of the outer holes. There were no connection between 2 holes, so i soldered a small cable between them.. XT does not work.. :-(
Then, i wrote eprom 27c256 to make the supersoft diagnosis. XT does not start.. no beep...no screen...nothing..

I checked pins on processor with logic tester. Board seems to work ! There are signals on RAM, processor, eproms....

Do you have another idea ?

THX a lot !!!!!!!!!!!!!!!!!!!!

bobba84
November 23rd, 2017, 12:40 PM
Have you checked very carefully that you haven't bridged any tracks with solder?

Jan1980
November 23rd, 2017, 12:49 PM
Yes, i cleaned with isopropanol alcohol. Is this board multilayer ?

Stone
November 23rd, 2017, 01:40 PM
Have you checked very carefully that you haven't bridged any tracks with solder?


Yes, i cleaned with isopropanol alcohol. Alcohol has zero effect on errant (or any, for that matter) solder areas.

Jan1980
November 23rd, 2017, 02:00 PM
So, i checked again and cleaned with solder station. And then. i checked the lines from pcb. Everything fine.. but xt not start.. :-((

Chuck(G)
November 23rd, 2017, 03:11 PM
As a sanity check, have you checked to see if the correct voltages are present on the ISA bus?

http://pinouts.ru/Slots/ISA_pinout.shtml

T-R-A
November 24th, 2017, 12:08 PM
Is this board multilayer ?

No. The XT was way too early for that technology to be implemented.

ibmapc
November 24th, 2017, 12:54 PM
No. The XT was way too early for that technology to be implemented.


Ahem! I believe you are mistaken. The following is taken directly from the "IBM Personal Computer Hardware Reference Library Technical Reference";


System Board

The system board fits horizontally in the base of the system unit and is approximately 8-1/2 by 12 inches.
It is a multilayer, single-land-per-channel design with ground and internal planes provided.

So to answer the OP, Yes it is multi-layer. Not sure how many layers, but at least three.

T-R-A
November 24th, 2017, 01:47 PM
Ahem! I believe you are mistaken. The following is...


I stand corrected.

Chuck(G)
November 24th, 2017, 01:54 PM
Indeed--multilayer boards extend back into the 1970s. Initially, defect rate was pretty high, as registration was a bit rocky.

Jan1980
November 24th, 2017, 02:05 PM
I checked voltages on isa slots. Everything fine. BUT, i found the following:

Pin B11 MEMW should be high, but it is high and low on my logiktester
Pin B12 MEMR should be high, but it is high and low on my logictester

Do you have an idea ?

THX !!

ibmapc
November 24th, 2017, 02:09 PM
Indeed--multilayer boards extend back into the 1970s. Initially, defect rate was pretty high, as registration was a bit rocky.

Yup,
My original 5155 board has patch wires attached at the factory, presumably to fix bad traces, some of which are not visible on top or bottom of board. In fact I've seen several old 5155 and 5160 boards with patch wires for sale on ebay.

modem7
November 24th, 2017, 06:02 PM
My original 5155 board has patch wires attached at the factory, presumably to fix bad traces, some of which are not visible on top or bottom of board. In fact I've seen several old 5155 and 5160 boards with patch wires for sale on ebay.
The patch wiring that I have encountered so far in the IBM PC family is either an enhancement, or corrects a design error.

Some examples at [here (http://minuszerodegrees.net/patch_wires/patch_wires.htm)].

Jan1980
November 26th, 2017, 02:10 AM
Today, i was working with my oscilloscope. I have just 8kHZ on CPU clock ! The Reset pin on CPU should have 4 cycles in high, the jump to low. In my case these 4 cycles are 489ms, so nearly a half second !! What can i do ? do you have an idea ? THX !!

Kazblox
November 26th, 2017, 04:24 AM
The patch wiring that I have encountered so far in the IBM PC family is either an enhancement, or corrects a design error.
Some examples at [here (http://minuszerodegrees.net/patch_wires/patch_wires.htm)].
It's not just with IBM props either; My AT&T 6300 has a patch wire that I think deals with a certain trace that is attached to a RAM chip. an Evrex EV-391 MFM that I saw on ePay also had a patch wire.

Chuck(G)
November 26th, 2017, 07:50 AM
Today, i was working with my oscilloscope. I have just 8kHZ on CPU clock ! The Reset pin on CPU should have 4 cycles in high, the jump to low. In my case these 4 cycles are 489ms, so nearly a half second !! What can i do ? do you have an idea ? THX !!

The reset signal duration is a minimum of 4 clocks, not the maximum. In the case of the 5150, it's driven through the 8284 clock generator by the POWER GOOD signal from the power supply. What you're seeing sounds correct.

Have you reseated all of the socketed ICs and checked to see that there are no bent pins? Usually when something like this happens, it tends to be something rather simple.

Jan1980
November 26th, 2017, 12:48 PM
I made a mistake. CPU clock is at 4.766 MHZ, so perfectly !!! I checked all sockets. Everything seems to be fine, but XT does not work. Since i swapped tantalium capacitators to aluminum elektrolyt capacitators.. Is there a test board on the market to get trouble code on 7 segment for example ??