Jump to content

m00dawg's sammichSID *FIXED*


m00dawg
 Share

Recommended Posts

So my build was going perfect (everything is straight, flush against the board, and purty!) until I started on the regulator section. I think I may have made a mistake in using conducting heatsinking compound. When I hooked the regulators up, I ended up having a short with the GND and +5V pins of the 7805. I looked at other parts of the board and didn't see anything of not. I also looked, of course, heavily at the heatsink section for any compound that might be touching +5V. I didn't see any but I thought I would check to see if anyone had similar issues and had a suggestion with where to start?

My plan is to remove the 7805 (once I get another solder-sucker, goddamn your shitty blue ones, Radio Shack!!) to see if the short still exists. I'm going to replace the 7805 just in case it was the regulator. Then I was going to look at shorts where there is a +5V and GND pin close to each other.

Any other thoughts?

Link to comment
Share on other sites

m00dawg: good plan...

I recommend taking out both regulators and replacing them.... perhaps first you can try bending them both up, taking off the heatsink and cleaning the lot, and see if that really was the cause of the short. It's possible to bend them back down and attach the heatsink, although maybe this won't result in as good a contact with the heatsinks, but good enough.

Link to comment
Share on other sites

I don't think that should be a problem, unless you smeared it all over the pcb? I'd be looking elsewhere.

You used the white stuff? Or is yours a black grey / silver grease?

Regards

Mike

Grey/Black. It's a tube of Arctic Silver 3 I've had for ages. Seemed to be probably more than necessary for this project, but it saved me from having to go out and pick some up :)

Link to comment
Share on other sites

Ditch the solder-sucker. This will make your removal much easier:

Actually I have a braid but it didn't seem to do the trick. It goes some of the stuff but not near enough. I've had really good luck with my solder sucker. I've also been told that adding a small notch to the side of the sucker really helps since you can put your iron through the notch. I haven't tried that yet but plan on doing so when I get the new sucker, probably by sometime today.

Link to comment
Share on other sites

m00dawg: good plan...

I recommend taking out both regulators and replacing them.... perhaps first you can try bending them both up, taking off the heatsink and cleaning the lot, and see if that really was the cause of the short. It's possible to bend them back down and attach the heatsink, although maybe this won't result in as good a contact with the heatsinks, but good enough.

My thoughts exactly! In fact, I did that already with the 7805 and I *thought* I got it to its unshorted glory but I think I was testing the wrong pins (doh!). I think I'll just desolder the whole mess, clean it, check for shorts, and then try again. I've got a lot of solder on the 7805 pins and I may have even scratched into the ground plane while trying to desolder (before I found out my solder sucker was busted). I doubt that will be a problem but it means the underside of my board is no longer pristine :P Good thing no one sees that side I suppose...

Link to comment
Share on other sites

So my build was going perfect (everything is straight, flush against the board, and purty!) until I started on the regulator section. I think I may have made a mistake in using conducting heatsinking compound. When I hooked the regulators up, I ended up having a short with the GND and +5V pins of the 7805.

Dude, I did the same thing just now. But I cleaned up some of the heatsinking compound and everything is working great!

Link to comment
Share on other sites

Well, the AS3 probably would have been ok if I wasn't so mess about it. I think the problem started there but is now well beyond just that. First, I pulled up pads while trying to desolder the 7805. I also have some cuts into the ground plane that I made unknowingly last night while trying to separate a big blob of solder that connected between two of the regulator pins. While these are the input and GND pins (not GND and +5V, which is where my short is somewhere) I may have done plenty of damage.

Even if the disaster with the pads can be worked-around (the pads on top aren't in terrible shape), I still can't find the short between GND and +5V. Been looking at the board with the magnifying glass for over an hour and the only thing I can point my finger to is the pads.

Ugly photos attached. Sad but hopefully it doesn't discourage anyone else :) Just sending out fare warning to watch the heat of your iron! I think I was probably fine until I had the brilliant idea of turning my iron past medium to medium-high.

As Wilba says, check twice, solder once. Boy it bums me out though because the remainder of the board is a thing of beauty. I was overly OCD when putting in all those components. Everything is straight and beautiful.

So, for the aftermath:

post-4914-126024050732_thumb.jpg

post-4914-126024051722_thumb.jpg

Chuckles and laughs welcome :) So is advice, and if that's ends up meaning I should just buy another sammichSID kit, so be it :P

Link to comment
Share on other sites

I should also add, it's probably wise to do continuity tests between each build-stage. I made the mistake of not doing those until the regulators were in place. In hindsight, if I did these tests before, I could have isolated the regulators as the cause. Hopefully someone finds my poor decisions helpful at least.

Link to comment
Share on other sites

Chuckles and laughs welcome :) So is advice, and if that's ends up meaning I should just buy another SammichSID kit, so be it :P

Doesn't look too bad there... and no one will see it when it's finished anyway. Get the braid onto that mess on the ground plane.

Finding shorts between 5V and GND is a pain. There are drastic measures for narrowing it down (like cutting 5V supply tracks to isolate which half of the track is shorting). However it is more likely to be around this big mess than anywhere else.

The 7805 output and the 5V pin of the 100nF cap near it both look a bit messy, you should clean them up a bit and recheck.

Just out of interest, is 5V or GND connected to the isolation plane? (the exposed plane under the heatsink) This is not connected to ground unless it conducts from the heatsinks (it doesn't matter if it does, but it might help isolate where the short is).

Link to comment
Share on other sites

Wow the topic splitting feature of this forum is nice! Thanks for moving it, by the way. I should have just posted a new topic in the first place.

I actually did try using the braid. Maybe my tip is too small or something but it hardly picks up anything. I could be Doing It Wrong, but I checked the video jbdiver posted and I think I'm doing it right. I'll keep at it :)

I will need to check again, but I'm almost certain the ground plan under the heatsinks is shorted to 5V :/ I did check the banksticks, however, and I can't find any shorts for the adjacent pins. I will check those again though and aso clean up around the 100nF and 7805 pins as you suggest.

The part that confuses me is that the short is actually the top 2 pins on the underside photo (bottom 2 on the topside). I can't see any obvious places where there might be problems with those two pins. I think the problem is there though since that's where most of the damage is, though it's a bit unnerving not to have anything directly.

I actually got a good bit of detail from the photos I took of The Aftermath (it's so bad is deserves a proper noun now :). I think I am going to try that same approach for the board as a whole and see if I can end up with better detail by doing it that way over staring through a magnifying glass. At least if I take photos I can take them to work to stare at :)

Link to comment
Share on other sites

Just out of interest,

When you measure between the 5v & ground, are you using a multimeter set to resistance, or a buzzer?

The Capacitors across the +5 & ground would for a short instance, make it seem as if their was a short between the two (While the cap is charging)

Can you measure the resistance? if it comes up as something like 0 - 10 ohms, then thats a short.

Just thinking aloud.

Mike

Edited by Futureman
Link to comment
Share on other sites

Just out of interest,

When you measure between the 5v & ground, are you using a multimeter set to resistance, or a buzzer?

The Capacitors across the +5 & ground would for a short instance, make it seem as if their was a short between the two (While the cap is charging)

Can you measure the resistance? if it comes up as something like 0 - 10 ohms, then thats a short.

Just thinking aloud.

Mike

When I was building my MB-6582 I had two of the regulators swapped and had to desolder them. The solder sucker and soldering braid didn't work very well. I ended up using some Chip Qwik (I got it from www.curious-inventor.com) and it worked a treat. For those of you who have not seen it before it is basically low-melting point solder. The idea is you heat up your part to be desoldered, then apply some Chip Qwik to the joint which mixes in and makes the solder stay soft longer. You then have enough time to quickly heat all 3 pins on the regulator and pull it out. Chip-Qwik is about $13 US for a small kit of it but worth having around for those tough jobs. The soldering video on Curious Inventor shows it being used IIRC.

Link to comment
Share on other sites

Just out of interest,

When you measure between the 5v & ground, are you using a multimeter set to resistance, or a buzzer?

The Capacitors across the +5 & ground would for a short instance, make it seem as if their was a short between the two (While the cap is charging)

Can you measure the resistance? if it comes up as something like 0 - 10 ohms, then thats a short.

Just thinking aloud.

Mike

I had it on beep mode, but it does show resistance as well. I didn't look at it all the time but when I did it was measuring like .01 or something really low. Thinking about it though, I should probably check again on other parts of the board and see how that does.

@fuzzylizard

Thanks for the input! I just may do that. I wonder if it will help the wicking. The problem is I can't use a solder sucker easily on those scratches because the solder just fills into the gaps. Same reason why the wick doesn't work better I would imagine but it seems to do a slightly better job than the sucker.

Thanks for the info everyone! I guess we'll see what happens later this afternoon...

Link to comment
Share on other sites

Found at least the area where the shorts are. I had to end up cutting tracks, but Wilba was really helpful in providing suggestions about which tracks to cut. It appears to be somewhere in the BankStick section, which also makes the most sense since that is where there are tons of +5 and GND pins adjacent to one another. I scoured that area visually multiple times and haven't seen anything but I'm going to mark the adjacent pins and work on those. I might just reflow the solder that is there, try it, then go from there.

Fortunately, while I had to cut 4 tracks it seems like it won't be terrible to fix. Just not very pretty, but I think I lost aesthetics when I lifted the pads from the 7805 anyway...

Will keep everyone posted. Just figured I'd give other people hope that might be having trouble with the Sammich :)

Link to comment
Share on other sites

It does suck yes but there's a good bit of hope left. I'm going to try to fix the short hopefully this evening and then repair the tracks. I can't see any clear indications of a short but I have a few suspicions of where it is. My suggestions are to do continuity tests during each build stage and to also take special care around those bank-sticks since there's a plethora of GND and +5V pins next to each other. Had I known the short was there I would not have messed up the area around the 7805. Oh also it is probably a good idea not to use a conductive heat-sinking compound :)

Link to comment
Share on other sites

For a project like this though? I think using a slightly worse thermal compound that's non-conuctive probably doesn't hurt anything. I used a silver-based one because it was all I had and I was a bit impatient to have to wait to order some :) If I had a choice, though, I would have gone with a non-conductive compound. *shrug*

I agree with you though, if people opt to use it, it should be a tiny bit. A bit more than when using on a CPU heatsink though since the heatsinks I have run into for these projects are lapped to a mirror finish or anything. The method of using Arctic Silver if I remember correct (which, to be fair, I didn't follow as well as I should have when building the Sammich) is to rub the compound on the heatsink in a circular motion. A tiny amount can be placed on the other surface and rubbed in a similar way. In some cases, you should even rub some off so that isn't too much on either surface which could create air gaps.

I guess the takeaway there is to take your time and, if you haven't used a silver-based compound in a while, maybe practice a little :)

Link to comment
Share on other sites

YAY!!

I was able to find the short! I never actually was able to find it visually, but after flowing all the +5 and GND pins in the BankStick section the problem went away. The area around the 7805, while ugly, appears to be good to go. I've also repaired the cut traces. All's looking well! In fact, I'm about halfway done with soldering the headers on.

Thanks for everyone's help, advice, and patience!

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...