Repairing Dreamcast (help)

Okay I picked up this dreamcast that had some damaged traces and now it's my pet project, and the bios chip had broken pins so I had to file down the chip to access the the pins again, I'll add more close-up pictures, I'd also love some critique on methods of achieving this fix in a more "professional" manner, since I'm just self taught and don't really have a good idea of how to achieve the trace repair in a better way at this scale.

Anyway here's the first image I have


You can see there was 3 damaged traces here and the filing I've done to access the pins, there's also one damaged on the inverse side

The dreamcast didn't boot before, so I repaired what damage I could see and now it boots, however it freezes after the logo, the set time and date appears but I can't interact with it, tried multiple controllers and even a different daughter board to ensure it wasn't an issue with the controller/daughter board. It seems to completely freeze. So I'm assuming the repair is either incomplete, or there's another underlying problem elsewhere that causes it to freeze. I intend to redo the joints on the bios chip with some of my solder

 

Looks like you have set yourself a challenge

Did you get this DC with the Bios pre modded? There's a lot of heat damage to it, did the previous owner take a blow torch to it? The flash chip IC502 isn't looking great either. What equipment are u using? U will need a temperature controlled iron and fine solder tips.

I would carefully remove the Bios chip, clean the board, improve the trace fixes for IC502 and then resolder a new preflashed Bios chip, @Bad_Ad84 sells them.

 

My god...

 

That's the idea! and yeah the bios was actually in a state where it had a wire hanging off of it and leg 1 + 44 were both completely missing, so I had to grind off the protective coating of the chip to access the legs underneath, there's a lot of mixed flux on this board too and what appears to be glue(?) so I'm not even sure how many hands have touched this, but I am aware of bad_ad84 selling bios chips I've bought a couple in the past and was very happy with them

I'll clean the flux with some ipa since the picture doesn't really show what I've done very well... But essentially the vias that the traces were damaged from scraped off the mask and poked through some kynar wire, soldered to both sides to ensure a good connection (though one of the vias was actually completely destroyed along with the trace) and soldered to the pins (as the pads below the pins were even damaged)

As mentioned in my first post, I intend on taking more pictures since that one was just one I took whilst I was still soldering and doesn't entirely represent current look (there was even a bridge on the bios chip pins in this image that has since been removed)

But I believe the DC to have been a botched mod (I buy in a lot of botched mod stuff and attempt to repair it for the experience)

I have a temperature controlled soldering iron though, as well as a fine tip, that's how I soldered the kynar onto the IC502, though I'm not too sure on their complete connectivity, maybe that's the issue I'm having with the freezing after the logo? I'm not sure, but I'll be doing some clean up tomorrow (today?, it's almost 5am and will get around to it some point in the PM).

Either way, I'm happy with the progress of having the machine now power on and display video - even if it does freeze on date/time.

 

well I had some stories with my dc projects as well when i was doing my hdd sata useless mod, I would suspect the bios misbehaving, also keep on mind that dc doesn't like long wires, try to make them as short as possible .

 

Ahh okay, I've not gotten around to doing this yet, or taking more pictures but I do intend on finishing the project. The middle wire that's fairly long is actually going to an alt point that I mapped out but it's further away from the original via, I'll see if I can't poke some wire through the original via to keep it shorter - maybe that's my problem!

I think I didn't really want to continue with it since I felt like I was out of options, but I'll tidy a little more and shorten wire lengths to the original trace length - then hopefully that resolves the problem

 

Redid the long wire back into the original via, keeping the trace length the same. Checked for continuity





Still no bueno, freezes on this screen, unable to do anything

 

Is the fuse in your controller board blown? It will make the controller stop responding.

If it is and you're in the USA shoot me your address I'll have a resettable fuse and a battery holder sent to ya.

 

Nah I've tried changing over the daughter board with a working one, same result. it just boots to this screen and the buttons never flash, like normally if you get this screen an option will alternate between orange and blue/greeny colour, but I get no flashing on the menu, it's really weird that it gets this far... But I don't know enough about the system

 

I've never seen it happen before, it's usually graphical issues in ones I've gotten that didn't work properly or that fuse being blown in the controller board. Sometimes the fan stopped working and it won't turn on past a black screen/ GD-ROM isn't connected, haven't ever ran into this problem. Any difference if you hold the lid switch down for the longer boot sequence, just curious.

 

If I hold down the lid switch and have a game inserted it goes through the loading procedure and sounds like it's loading, drive spins forever but still pushes me to this screen and I'm unable to continue. I have the GD-ROM screwed down to the chassis, and I partially screw in the whole console every time I test it, using the original PSU etc - I can't seem to figure out why it gets this far, and it's really bugging me that I actually got it booting partially now

*edit
Unfortunately the next logical step is to remove the chip completely and see if anything under the chip is damaged, since I can't visibly see anymore damage on the board and it boots, but seems to have an issue related to the bios... Best way to check is remove the chip and ensure all the pads aren't damaged. FUN INCOMING.

Luckily enough I have some low melt solder I can use to remove the chip with an iron and not mess up my trace repairs by using a heat gun or something

 

The Bios chip needs throwing in the bin, the heat damage is clear as day. Get a new flashed Bios chip and solder it in.

 

There is no "clear heat damage", there's flux (the shiny parts) and the grinding that took place on each of the corners to access pins that were snapped off. I'll be able to better assess the chip once removed but there's no signs of heat damage to the chip itself (even then, the black part of the chip is a protective layer and depending on the material can withstand temperatures 500C+) long story short, it would be very difficult for somebody to "heat damage" this chip unless they have the equipment where reaching these high temperatures is easy and they intended to do it, or they have no idea what they're doing.

I appreciate the help, so sorry if I come off snarky here - I'm just going to continue and remove the chip, give the area a good solid clean and assess further, since I don't actually know if anything under the chip is damaged, but the chip itself seems to be intact (discounting the grinding of the edges I cut away to access the pins)

 

Ah, my mistake. Either way, you don't know what abuse this chip has endured before it made its way into your hands. These chips are designed to be reflowed at 260c in an oven but they are not designed to handle repeated sustained high temps (who knows what the previous 'modded' did to it). As you are removing the chip, it makes sense to simply replace with a fresh flashed chip.

 

In light of removing the chip I believe I've spotted the issue haha.



pad 1 is gone which isn't a problem since the point isn't used, pad 2 is gone and needs to be repaired, pad 5 was actually connected with a very thin single strand of copper (it was actually thinner than the traces on the board) so I removed it, it needs to be repaired, pad 21 has lifted and disconnected from one side of the trace and needs to be repaired, pad 22 is gone, though it is unused so it isn't necessary to restore it, pad 25 is gone, can't find the trace if it's used or not - need to check some reference images, pad 44 is gone but is unused so isn't necessary to repair

So in total:
pad 1 : destroyed - no repair needed
pad 2 : repair needed
pad 5 : repair needed
pad 21 : repair needed
pad 22 : destroyed - no repair needed
pad 25 : ????
pad 44: destroyed - no repair needed


So this chip probably works just fine, I am however going to get a new one for sanity sake and throw this one in a box somewhere.

Also noticed some light scratches on the board that has removed some of the solder mask from the traces, checked them and they seem intact but for safety I'll probably clean with IPA and apply a thin layer of epoxy just to stop any kind of oxidization, but now I'm suck with the repair process...

Basically my idea is, I have some copper foil tape that I can cut to shape, solder to the trace and underbond onto the pcb using epoxy so that it doesn't have movement on it.

I still need to remove the remainder of the low melt solder from the pads and clean up before hand but this is essentially what I'm with right now.

However, if anyone has a better idea that's easier and a more professional approach to repairing the pads, I'd love to give it a try

 

Currently bonding some copper foil to the traces leading to pad 2 and pad 5, soldered the foil to the trace, bent it back carefully and applied the epoxy and pushed the foil down on top so I can come back to this in 24 hours, tin the new "pads" and continuity check them.

Pad 21 I'm happy enough to just restore the connection via a solder bridge, which worked great.
I've left the remaining pads since I can't see any traces that originated from them, if I'm wrong I'd greatly appreciate the heads up before I get this new bios chip


Thanks for everything so far! Hope this works

 

Correct, 1, 22, 25, 44 don't need repair. The foil on pin 2 looks very close to pin 3, this may short under the IC leads when soldered to. I would just run 30 awg wires from the IC pins to the next relevant via or pad on the circuit. Repairing lifted pads with foil can be a PITA. Good luck.

 

Yeah I was a little unhappy with the foil on pin 2, but I'll be able to work around it anyway, I worked around IC502 pins fairly easily without any visual aid (magnifying glass, microscope etc) and the gaps between those pins were about the same as the gap between the foil and the third pad, worst case scenario I'll fix up that foil tomorrow after the epoxy has set, but I just feel this is a more long term, more aesthetically pleasing route than just doing the same kind of repair that I did on IC502 - if it doesn't work out I'll probably go with the wire route

 

We're all rooting for you here.

 

Changed over to a new chip, checked all the points, tidied up some more stuff and STILL it only boots to the date/time screen and won't progress! I'm gonna need to map out some test points from a working VA1 to test where this one might be having issues still. I did actually find another point damaged that I never caught last time, but repairing it didn't seem to affect the boot at all... And so the journey continues!