VOGONS


First post, by keenerb

User metadata
Rank Oldbie
Rank
Oldbie

With the system on I don't get any frequency readings across the two legs of the oscillator, and my meter can read up to 60mhz. Is that more or less the proper way to read frequency on an oscillator, just directly across the legs?

Basically a short circuit due to battery corrosion burnt the 12v power connector completely through the motherboard. As far as I can tell I've successfully repaired the damage but previously it would boot to a "floppy drive failure" message; now I get nothing on screen.

I have spent hours and hours checking for shorts and bad traces and everything looks fine, I get 12v, 5v where I expect to get 12v, 5v.

I suspect one of the oscillators that was submerged in vinegar during my cleanup attempts might have gone bad.

Also, on the bottom oscillator in this photo, do you think that's 16.000 or 76.000 with part of the 7 gone?

oscillators.JPG
Filename
oscillators.JPG
File size
45.61 KiB
Views
890 views
File license
CC-BY-4.0

Reply 1 of 14, by retardware

User metadata
Rank Oldbie
Rank
Oldbie
keenerb wrote on 2021-10-11, 13:33:

With the system on I don't get any frequency readings across the two legs of the oscillator, and my meter can read up to 60mhz. Is that more or less the proper way to read frequency on an oscillator, just directly across the legs?

No, because you usually will stop the oscillator with that additional capacitive+resistive load.
Or you'll change its frequency to another harmonics.
So, tap its output.

keenerb wrote on 2021-10-11, 13:33:

I suspect one of the oscillators that was submerged in vinegar during my cleanup attempts might have gone bad.

Also, on the bottom oscillator in this photo, do you think that's 16.000 or 76.000 with part of the 7 gone?

Never have seen a 76MHz crystal. But 16MHz ones are common.

Reply 2 of 14, by keenerb

User metadata
Rank Oldbie
Rank
Oldbie

That's how I felt about the 16 vs. 76 as well, I've never seen a 76mhz on a board before. I'm going through and re-checking all my trace fixes and maybe replace that 16mhz oscillator. There's a much smaller round osc that I can't find any identifiers on, that might not be fun to troubleshoot...

Reply 3 of 14, by retardware

User metadata
Rank Oldbie
Rank
Oldbie
keenerb wrote on 2021-10-11, 15:10:

There's a much smaller round osc that I can't find any identifiers on, that might not be fun to troubleshoot...

These are usually the 32,768Hz crystals for the RTC. It is non-critical. If it doesn't work, the RTC will be frozen in time.

Reply 4 of 14, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie

With decent voltmeter it is usually possible to just measure voltage, it should read somewhere in the middle of the Vcc if the circuit is running. Yes, sometimes the parasitic capacitance of the probes can stop a low-power oscillator but usually even direct testing of the crystal pins works.

Stopped RTC can have some annoying side-effects like some programs freezing, and DOS kernel plain rejecting the time/date set and reverting to built-in defaults. And BTW I have had a case where I actually "broke" an oscillator circuit after washing the PCB. Turns out some wet dust got between crystal pins and dried there, and (when the air was not dry enough) had enough moisture and salts in it to conduct just barey enough to load the crystal down and stop it. It went away after I brushed the PCB clean again before soldering a new part in, and realized that could be the cause.

Reply 5 of 14, by dataino.it

User metadata
Rank Member
Rank
Member
keenerb wrote on 2021-10-11, 13:33:
With the system on I don't get any frequency readings across the two legs of the oscillator, and my meter can read up to 60mhz. […]
Show full quote

With the system on I don't get any frequency readings across the two legs of the oscillator, and my meter can read up to 60mhz. Is that more or less the proper way to read frequency on an oscillator, just directly across the legs?

Basically a short circuit due to battery corrosion burnt the 12v power connector completely through the motherboard. As far as I can tell I've successfully repaired the damage but previously it would boot to a "floppy drive failure" message; now I get nothing on screen.

I have spent hours and hours checking for shorts and bad traces and everything looks fine, I get 12v, 5v where I expect to get 12v, 5v.

I suspect one of the oscillators that was submerged in vinegar during my cleanup attempts might have gone bad.

Also, on the bottom oscillator in this photo, do you think that's 16.000 or 76.000 with part of the 7 gone?
oscillators.JPG

can u post a pic of the full board ?

and above all a detailed photo of the damaged area

Reply 6 of 14, by keenerb

User metadata
Rank Oldbie
Rank
Oldbie

Sure.

It booted but had a disk failure message with the burnt 12v connectors. After cleaning it, and before re-connecting the 12v it stopped booting.

I have identified one broken trace but it seems related to the missing battery charging circuit, and the batter has been gone for some time.

CPU gets warm and I get +5v and +12v, but now that I'm double-checking I dont seem to have -5v and -12v coming from power supply; checking those to ground just reads 0.

full board.jpg
Filename
full board.jpg
File size
185.45 KiB
Views
795 views
File license
Public domain
burnt side one.JPG
Filename
burnt side one.JPG
File size
368.09 KiB
Views
795 views
File license
Public domain
burnt side two.JPG
Filename
burnt side two.JPG
File size
105.85 KiB
Views
795 views
File license
Public domain

Reply 7 of 14, by dataino.it

User metadata
Rank Member
Rank
Member

the 37c65 chip (floppy controller) is missing or did you remove it to check the tracks?

Attachments

  • Filename
    datasheet.pdf
    File size
    1.54 MiB
    Downloads
    32 downloads
    File license
    Public domain
  • full board.jpg
    Filename
    full board.jpg
    File size
    145.54 KiB
    Views
    785 views
    File license
    Public domain

Reply 10 of 14, by dataino.it

User metadata
Rank Member
Rank
Member

the 37c65 chip needs a 16 Mhz clock on PIN 23 and it is certainly generated by the nearby components (74LS04 + the resistors and capacitors) check on the 74LS04 pins if there is a signal

Attachments

  • full board.jpg
    Filename
    full board.jpg
    File size
    39.8 KiB
    Views
    769 views
    File license
    Public domain

Reply 13 of 14, by keenerb

User metadata
Rank Oldbie
Rank
Oldbie

yeah I replaced all the capacitors in that area as well.

i'm going to replace the 28.63636mhz crystal that feeds the FE2010A; that's where the CPU clock and ISA OSC signal are generated, so maybe that's the problem.

Last edited by keenerb on 2021-10-12, 03:18. Edited 1 time in total.

Reply 14 of 14, by maxtherabbit

User metadata
Rank l33t
Rank
l33t
keenerb wrote on 2021-10-11, 21:12:

yeah I replaced all the capacitors in that area as well.

i'm going to replace the 28.63636mhz capacitor that feeds the FE2010A; that's where the CPU clock and ISA OSC signal are generated, so maybe that's the problem.

sounds like it, OSC should be exactly half of that