• Please review our updated Terms and Rules here

Restoring Micronics 09-00066-52 386/33 MoBo

gatewayrepairs1985

Experienced Member
Joined
Dec 10, 2021
Messages
104
Location
South Of Boston, MA
Hi all,

Hoping you are well. I really want to get this 33 mHz Micronics motherboard working again. When I got it, the VARTA battery had leaked, and there was corrosion surrounding it. I let the corroded part of the board sit in vinegar, then rinsed with warm water and let it dry. I still get no luck, (board powers on, no activity with POST card inserted). I have confirmed the BIOS chips are good. Here is some more info on the board:


I have photos attached of what the board looks like right now. I have a multimeter but no experience using it. I have some experience soldering and can desolder/solder things if necessary. How should I go about getting this board to POST again? Any advice from anyone would be greatly appreciated!

Thanks so much!

Shawn
 

Attachments

  • IMG_0159.jpeg
    IMG_0159.jpeg
    3 MB · Views: 23
  • IMG_0160.jpeg
    IMG_0160.jpeg
    1.8 MB · Views: 23

GearTechWolf

Experienced Member
Joined
Apr 24, 2015
Messages
264
Location
Washington
Looks like you have a bunch of traces that have been entirely removed by the battery leakage!
Fortunately, it seems the paths they used to take are still evident on the surface of the board.
Following those to their next viable solder-point (at a component or where the trace is large enough to solder to directly) and reconnecting them with thin wire (enamel-coated magnet-wire (for making transformers with, or harvested from an old one) may allow you to revive the board.
It does look like it may be a multi-layer board, which could complicate things if traces in the inner layers of the board have been damaged.
If their paths in the damaged area can be plotted out to accessible points (legs of components) you should be able to check them for continuity.
Checking continuity is easy, you just have to set your meter to the lowest resistance range or to the continuity/diode test.
If it's a newer/digital meter it should have a setting where it beeps when you touch the probes to each-other, this is the test you want.
If it's an older/analog meter it'll just peg the needle to full when the probes are crossed.
Aside from trying to repair all the obvious damage in the affected area and checking traces that aren't visually missing/damaged, you can also use a continuity-check to test if there's a good connection between the power-connector and various chips.
Looking up the part-numbers of the chips should net you a datasheet and that'll give you the pin-out, then you can match the positive voltage pin (VCC) to the appropriate pin on the main power-connector to ensure power is getting to the chips.
You may need to de-solder parts to get a better view of where the damaged/missing traces go, but as long as you document what parts go where and in what orientation (if applicable) you should be okay.
The corroded solder around the pins of components in the damaged area will likely need to be carefully scraped away before you can de-solder some parts.
It can be quite frustrating to deal with corroded solder, but patience, careful picking/scraping and plenty of flux to help wash away corrosion/transfer the heat to non-corroded solder should be enough to get you through it.
The plated through-holes in the board where the legs of the components go through may be quite fragile if badly corroded, so there is a risk of them pulling/stripping out when removing components.
With this in mind, it wouldn't be a bad idea to map out where the pins of a component connect to (at least one other point on the board for each trace seen leaving that leg, per layer) before de-soldering things.
That way, if a through-hole is damaged, there may be some hope of reconnecting things with bodge-wires.
It's no easy task repairing such damage, but I wish you great success if you choose to attempt it!
If you're lucky, and careful, just repairing the obviously missing traces will revive the board and any de-soldering you do won't cause more damage.
Good luck, and please share your progress here, as others are sure to have advice as well!
 

gatewayrepairs1985

Experienced Member
Joined
Dec 10, 2021
Messages
104
Location
South Of Boston, MA
Are all power rails working? Just asking because I see tantalum caps.
Hi,

Thank you for the reply. Could you advise me on how to check the power rails? My apologies, brand new to this. Still getting used to the multimeter. I have it set to test DC voltage, where should I go from there? Thanks in advance for any help you can give!
 

gatewayrepairs1985

Experienced Member
Joined
Dec 10, 2021
Messages
104
Location
South Of Boston, MA
Looks like you have a bunch of traces that have been entirely removed by the battery leakage!
Fortunately, it seems the paths they used to take are still evident on the surface of the board.
Following those to their next viable solder-point (at a component or where the trace is large enough to solder to directly) and reconnecting them with thin wire (enamel-coated magnet-wire (for making transformers with, or harvested from an old one) may allow you to revive the board.
It does look like it may be a multi-layer board, which could complicate things if traces in the inner layers of the board have been damaged.
If their paths in the damaged area can be plotted out to accessible points (legs of components) you should be able to check them for continuity.
Checking continuity is easy, you just have to set your meter to the lowest resistance range or to the continuity/diode test.
If it's a newer/digital meter it should have a setting where it beeps when you touch the probes to each-other, this is the test you want.
If it's an older/analog meter it'll just peg the needle to full when the probes are crossed.
Aside from trying to repair all the obvious damage in the affected area and checking traces that aren't visually missing/damaged, you can also use a continuity-check to test if there's a good connection between the power-connector and various chips.
Looking up the part-numbers of the chips should net you a datasheet and that'll give you the pin-out, then you can match the positive voltage pin (VCC) to the appropriate pin on the main power-connector to ensure power is getting to the chips.
You may need to de-solder parts to get a better view of where the damaged/missing traces go, but as long as you document what parts go where and in what orientation (if applicable) you should be okay.
The corroded solder around the pins of components in the damaged area will likely need to be carefully scraped away before you can de-solder some parts.
It can be quite frustrating to deal with corroded solder, but patience, careful picking/scraping and plenty of flux to help wash away corrosion/transfer the heat to non-corroded solder should be enough to get you through it.
The plated through-holes in the board where the legs of the components go through may be quite fragile if badly corroded, so there is a risk of them pulling/stripping out when removing components.
With this in mind, it wouldn't be a bad idea to map out where the pins of a component connect to (at least one other point on the board for each trace seen leaving that leg, per layer) before de-soldering things.
That way, if a through-hole is damaged, there may be some hope of reconnecting things with bodge-wires.
It's no easy task repairing such damage, but I wish you great success if you choose to attempt it!
If you're lucky, and careful, just repairing the obviously missing traces will revive the board and any de-soldering you do won't cause more damage.
Good luck, and please share your progress here, as others are sure to have advice as well!
Hi,

Thank you for the lengthy reply. I learned a lot from it. This does look like a multi-layered board, but I really hope I can revive it. Here I have what I think is the first trace I need to repair. I would appreciate if you agree with my synopsis. See the before photo, and then the marked-up photo - thanks!
 

Attachments

  • IMG_0203.jpeg
    IMG_0203.jpeg
    821.5 KB · Views: 7
  • IMG_0203.jpeg
    IMG_0203.jpeg
    2 MB · Views: 7

GearTechWolf

Experienced Member
Joined
Apr 24, 2015
Messages
264
Location
Washington
Looks like a good start, yeah!
Looking at the underside of the board, are there any other traces on the other side that connect at that through-hole where it changes layers?
There are a few ways you can route the wires that replace the traces.
You can follow the same path and use dabs of glue to hold the wire in place, you can run wires in straight/shortest paths between components, or you can parallel the original path so that you can see the wire and the original path at the same time.
The first method may be easier to diagnose later on, as the number of wires grows, the second tends to result in a difficult-to-follow tangle, and the third has better sanity-checking of correct pathing than the first, except for when you have a bunch of traces close together.
Hope you can get it going again, it's a great feeling when a board finally boots!
The 386 I fixed had very minor damage, all the traces managed to survive and I just had to replace a corroded chip-socket.
I also had to reattach the power-supply leads after I'd chopped them off, as I'd thought the power-supply was bad, but it was fine!
I'd just been testing it wrong. Anyhow, hope the soldering goes smoothly!
 

gatewayrepairs1985

Experienced Member
Joined
Dec 10, 2021
Messages
104
Location
South Of Boston, MA
Looks like a good start, yeah!
Looking at the underside of the board, are there any other traces on the other side that connect at that through-hole where it changes layers?
There are a few ways you can route the wires that replace the traces.
You can follow the same path and use dabs of glue to hold the wire in place, you can run wires in straight/shortest paths between components, or you can parallel the original path so that you can see the wire and the original path at the same time.
The first method may be easier to diagnose later on, as the number of wires grows, the second tends to result in a difficult-to-follow tangle, and the third has better sanity-checking of correct pathing than the first, except for when you have a bunch of traces close together.
Hope you can get it going again, it's a great feeling when a board finally boots!
The 386 I fixed had very minor damage, all the traces managed to survive and I just had to replace a corroded chip-socket.
I also had to reattach the power-supply leads after I'd chopped them off, as I'd thought the power-supply was bad, but it was fine!
I'd just been testing it wrong. Anyhow, hope the soldering goes smoothly!
Thank you for all the advice. This board will prove difficult to repair because it is multi-layered. Some traces seem to go nowhere - do you think you could offer an explanation for that? I will include the before photo and then the photo with the traces outlined that I think go nowhere. Thanks for any help you can give on this!
 

Attachments

  • IMG_0206.jpeg
    IMG_0206.jpeg
    1.6 MB · Views: 7
  • IMG_0206.jpeg
    IMG_0206.jpeg
    670.4 KB · Views: 7

GearTechWolf

Experienced Member
Joined
Apr 24, 2015
Messages
264
Location
Washington
Oh! Those don't go nowhere, they go to a plated through-hole called a via! It's a way of jumping the trace between different layers or sides of the board.
If you scrape away the solder-mask at those round points/dots you may be able to solder to them. They look to be in decent shape.
Happy to help!
 
Top