🇦🇺 IBM PC/XT (Model 5160): Part 3 (Booting into Cassette BASIC) [TCE
HTML-код
- Опубликовано: 4 окт 2024
- It boots! The original symptoms were no video & no beep when booting - according to the list of POST steps this would indicate it's failing before step 21 when the video card is initialised.
minuszerodegre...
And from what I can see on the diagnostics ROMs it's looking quite likely that BANK 0 is failing - using a dedicated ROM to test this (TEST6099) indicates pretty much all of the 4164s have failed except for one. This is the step 15 test.
Supersoft diagnostic ROM (from the 80s):
minuszerodegre...
Ruud's diagnostics ROM (modern, based on Supersoft):
minuszerodegre...
TEST6099 ROM (tests just BANK 0 or the BASE 16/64KB):
minuszerodegre...
Replacing these suspect DRAMs and the machine finally boots to cassette BASIC with just a "601" error because no drive controllers are connected.
Progress!
I have upgraded the BIOS to the latest version (9-May-1986) and will try the drives next.
The Persyst "Time Spectrum 384" multifunction card requires some repair due to the Varta battery corrosion - replace some resistors and IC sockets.
Components of this machine:
[1] Twin half-height 360KB 5¼" floppy drives
[2] Single 20MB hard drive (Tandon TM-262)
[3] MDA video card
[4] Persyst "Time Spectrum 384" multifunction card: 384KB RAM, RTC, parallel & serial ports
[5] Floppy drive controller with external floppy connector
[6] Hard drive controller
[7] IBM Model 5151 monochrome (green) screen
More videos on troubleshooting IBM PCs & compatibles:
• IBM PC & Compatibles
Did not know about those diag ROMs. I will give those a try. There are also alternative BIOS ROMs. Makes the XT boot much faster.
I'm guessing if you remove a lot of the POST code then you could get it to boot much faster, i.e. don't check 640KB RAM every time you boot! Do you have any links? /Brett
Seven of nine😎
*groan* ... ;)
Just take them all out and test them individually (with other equipment)
I'll definitely be testing the ones I've removed. Probably worth reseating the other ones. I'm working on the Persyst multifunction card at the moment - have removed all 384KB (to give the card a good clean with white vinegar) and will add them back in banks to see what works
Are they including the parity chip as part of the fault beeps? @18:00
I'm still not entirely sure ... according to the TEST6099 documentation there are only eight beeps (one for each data bit), and a failure of the parity bit is indicated the same as no data bit failures (0x00) ... ?!
"If the BASE 16KB RAM test has executed and FAILED, then a few seconds after power-on, you will hear a series of 8 beeps from the speaker.
Some beeps are short, some are long. Short corresponds to a 0. Long corresponds to a 1.
The eight beeps (eight 0's and 1's) make up a byte. The order of beeps is bit 7 to bit 0 (MSB to LSB).
For example, if you hear short-long-short-short-short-short-short-short, that is 01000000 binary, or 40 hex.
00 hex (00000000) = Failure of parity bit
01 hex (00000001) = Failure of bit 0
02 hex (00000010) = Failure of bit 1
04 hex (00000100) = Failure of bit 2
08 hex (00001000) = Failure of bit 3
10 hex (00010000) = Failure of bit 4
20 hex (00100000) = Failure of bit 5
40 hex (01000000) = Failure of bit 6
80 hex (10000000) = Failure of bit 7
Anything else indicates multiple bits. For example, 28 hex (00101000) indicates bits 5 and 3."
@@Brfff That manual was incredibly not helpful lol. Happy you got it sorted. Just went into curious mode when I noticed 8 beeps mentioned but saw 9 chips :)
Now that is a terrible monitor....
What?! It's kinda growing on me ... nice, crisp, green text! I'm hoping to play some Pacgirl if I can get it to boot from one of the drives ;)
@@Brfff I can literally see letters fading over several seconds...
@@IkarusKommt Yeah, actually I noticed that too
nice job!.. strange the rest of the ram is fine and it was only that first bank...
There's a little bit of greenish corrosion on the BANK 0 socket pins from the Varta battery leak ... maybe due to that?
@@Brfff yeah maybe even the sockets?.. have u got a tester for the 4164s? ..it maybe just corrosion on the legs/sockets causing them to be bad?..new clean leg chips are making better contact?.. might be something to check out..
For sure I won't be discarding the chips until I'm certain ... there's some uncertainty about the parity chip (see discussion with Christian @Chriva)
@@Brfff yeah i was thinking that too..if the parity bits arnt working it could show up as the ram chips failing..
Put some deoxit or contactcleaner in the sockets. And clean the RAM chip legs. Maybe the sockets are just the problem.