Early 5150 motherboard

[chris_nh]

I recently acquired this early 5150 motherboard into the collection. It was fitted with the original 5700051 BIOS, and the chip dates indicate production times in Sep. 81, so I'm placing production time for this board sometime in late Sep. / early Oct. 1981. It appears this board was produced just after the PC was announced in August, but prior to the 2nd BIOS release in Oct 81. Interesting point in the timeline.

Motherboard part number is stamped as 1802437 031 XM
BIOS 5700051 Date stamp 8140

There is an interesting chip in U34 (normally the 8253). Not the usual markings, so I'm wondering if perhaps anyone might recognize what this chip is? Obviously some form of an 8253...

I recently built up a light box for taking better quality photographs of circuit boards. Thanks to Dave EEVBlog guy for the inspriation on that one. Anyway, I tried to get some high res pics of this board, so check these out if you are interested:

Hi res photos:
https://drive.google.com/drive/folders/1DxzGDUMXgkKqGUz-U6tfgm3mOji2sm6-?usp=sharing



Some notes on the board ... it does not run, unfortunately. The CPU is good and the BIOS is good, as tested in another motherboard. This one will not run the landmark diagnostic ROM or anything. Scope shows CLK is present and ready/reset appear normal, and it appears the cpu starts to run, but fails immediately. There were short caps on the -12 and +12v rails. Clipped them off for now. Obviously there is another issue, either bad ram chip(s) or bad bus/logic related IC or both. However, given that this board is an early example, I'd rather not hack it apart such to shotgun replacement of the ram, etc. If it were a more common PC board perhaps, but not this one I think I will try a logic analyzer on it to see if I can further pinpoint the failure. Normally, I would expect the diagnostic ROM to at least start the display, but we get nothing from this board currently, and it looks nice hanging on the wall!

 

[Ruud]

Have a look at my PC debugger. It enables you to run the CPU step by step starting from a reset. The displays show the address and the data on the buses.
I already built one thirty years ago without the feature to stop the CPU at a set address and I am busy designing a PCB that has this feature. Think about it, then you don't have to drill holes in the wall.

 

[chris_nh]

Thanks Ruud, thanks for sharing the info about your debugger. I will probably give this a try ...

 

[Old_hitech]

Very nice early motherboard. It will be great to get it working. I've found these 16-64K motherboards a challenge to trouble-shoot and get working. I've used the Supersoft diagnostic ROM and it has helped some, but I was not able to get 2 different MBs working. I've replaced Bank 0 RAM, 8255 with no luck. I did find a working MB on eBay and that is what I am using now in my 5150 RevA. I'd like to know if there is a better way to get my 2 failed MBs working again. I'll be watching this thread closely.

 

[framer]

Please list any isa cards that you are using. The floppy controller is my main concern. My early MB will only boot with the FDC that has that big silver square chip on it. The more common PC floppy controller don't work. A shorted cap on any card could also bring it to a halt. I also had issues using newer BIOS in the MB until I had all 4 banks of RAM filled. All the above is from fixing mine 10 years ago so my memory might have holes... Once fixed, it has continued to work perfect and is turned on weekly.

Found my old thread:http://www.vcfed.org/forum/showthread.php?24223-I-bought-a-model-A-5150-cheap-on-FleaBay&highlight=framer Read post 17 and 18 it may give you some insight.

framer

 

[framer]

There is an interesting chip in U34 (normally the 8253). I opened my case and checked U34 and it is the same as yours. I was not aware of the chip difference until now. I have the same systemboard part number 1802437 as your board.

found:http://www.minuszerodegrees.net/5150/early/5150_early.htm scroll down and check out the early floppy controller card.

On the same page there is a photo of a later 16-64K systemboard that has the 8253 in U34.

 

[kdr]

Have a look at my PC debugger\. It enables you to run the CPU step by step starting from a reset. The displays show the address and the data on the buses.

That looks fantastic!

And thanks to your page, I have learned about the IOCHRDY trick. That should help as I continue to diagnose a couple of dead 5150 / 5160 motherboards where I suspect the CPU isn't getting the right instructions from the boot ROM. At least I'll be able to confirm whether or not the first JMP byte is being fetched from ROM.