| Posted By
unclouded
on 2020-04-30
08:40:21
|  C16 repair
Hi,
I recently recovered my original C16, which was not ever modified. When I turned it on the display was normal, showing 12277 BYTES FREE initially but after a few seconds it began to fade everything slowly to white. I turned it off but now it doesn't produce any display at all. Does this suggest a particular chip at fault?
I thought that I would recap it anyway and removed C16, C17 and C2 to begin with, but they all measured exactly what they should do (10u, 1u and 52u for the 47u). Does this mean they were OK? I replaced them with new anyway but should I bother with the rest?
Next I thought I'd check the regulator and the clock. The PSU showed 14.4 V open circuit and 12.1 V when switched on and the 7805 showed 5.8 V in its output, which didn't seem good. So I replaced the regulator with a new (also linear) one but it *still* showed 5.8 V. Is that normal or a result of the Vin being a bit high?
The clock looks like it's working though:
<img src="/dl/forum/20200430_083510_8505_C16-clock.png" width="800" height="480" alt="">
Is that what you'd expect or should it be cleaner?
Thanks!
|
|
Posted By
MCes
on 2020-04-30
15:23:38
 | Re: C16 repair
your PSU voltage is ok, but 5.8V on 7805 output can't be normal... (max 5.2V)
Have you pulled out chips from motherboard to read 5.8V? in this case the "current helping" of R10 can be too much and overfeed the 5Vcc.
Another component to check is diode D11, it is near heatsink and the ceramic resistor, after some seconds of power up put a finger on it: if it heat the diode is damaged (leakage reverse current).
|
|
Posted By
unclouded
on 2020-05-01
03:19:12
| Re: C16 repair
Yes! I had removed all socketed chips to prevent further damage. Thanks for heads up on D11. I couldn't get a finger in there but the thermometer indicated ambient temperature for D11 in this case. The 8708 chips (are these MOS 74257 equivalents?) U7 and U8 were 10°C above ambient but everything else was cool apart from R10 and bizarrely Q5 (to control the cassette motor) that were warmish.
It's good to know that the PSU open-circuit voltage is typical.
I put a DC-DC converter in-line with the original PSU anyway and set it to 7.5 V (the data sheet for the 7805 says drop-out is around 2.3 V with 1A at 25°C) and measured 4.9 V at the regulator Vout. I put the chips back in and it still showed 4.9 V, so I think I'll leave it in there because it ought to be easier on the regulator if it has to drop less.
Anyway, with the chips back in the machine came back to life! Briefly. This time after a few seconds the display showed only the left hand side of the characters so I switched off again. Might this implicate the KERNAL ROM or RAM rather than the TED or does it sound like a TED fault?
|
|
Posted By
MCes
on 2020-05-01
04:08:33
| Re: C16 repair
The overvoltage of 5.8V on 5Vcc line is a consequence of the pulling out the whole chipset.
If you want to permanently use a new stabilized power supply, it must not be less than 9 V, since it is necessary on the motor drive circuit that powers the cassette unit.
Issues that appear with heating are normally silicon (chips) problems.
If you have another C16 or PLUS4 you can try to swap chips to find who is the trouble.
If you are lucky it could be a ROMs or the PLA, otherwise it could be TED or CPU.
Can be useful to try the "diag264" kernal version (if your C16 work enough to perform same test)
You can also try to put a little ice cube over the center of a chip and after a pair of minute (it has to drop the internal temperature) you can switch on the C16, if the time that you have to wait for the symptoms will be significantly longer than the previous test: the chip is probably the damaged one.
(to be repeated for all socketed chips, and after also for the rams...)
|
|
Posted By
unclouded
on 2020-05-02
00:30:05
| Re: C16 repair
Good point about the cassette motor.
With the new DC-DC converter inline, the +5V line shows 4.9 V with all socketed chips pulled and 4.9 V with all inserted, which I guess is a good thing.
Thanks for the Diag264 suggestion. I put the PAL KERNAL version in the top half of a AT27C256R (32K OTP ROM). It's A14 pin is in the place of the CS3 pin of the 23128, which is tied to +5V hence the top half. This didn't help the sick machine. The screen twitches at power on but never shows the visible area of the video matrix. I put the Diag264 in to a working machine and saw Diag264 live for the first time. It rocks!
I didn't want to put known good chips in to a bad machine but then it occurred to me that I could put chips from the sick machine in to the working machine one at a time. Good news! The TED and CPU work (Diag264 pass) and it was the PLA that caused the machine to stop working!
So I need a PLA. What's the post like in Italy at the moment?
|
|
Posted By
MCes
on 2020-05-05
08:01:34
| Re: C16 repair
@unclouded
Now I can go to postal office for sending!
|
|
| |
Copyright © Plus/4 World Team, 2001-2024 |
