A death in the family.

dehuszar · Post by **dehuszar** » Mon Sep 06, 2004 12:40 pm

I have just had a terrible morning.

I plugged in my Magma chassis looking forward to delving deeper into Flexor and the P100 this morning and my Magma let out a thunderclap and would not power on.

I took it apart to see if there was a fuse or circuit breaker that could be repaired, but alas the power supply was blown and would not turn on again.

Then I noticed it.

There was a capacitor lodged in the corner of the chassis. I took out my CW cards and discovered that it had come off of my Scope/SP card.

As I have no other place to test out my Pulsar II card at the moment, it is quite possible that the whole caboodle (my Magma chassis, Scope/SP, and my Pulsar II) has shat the bed.

My laptop, which was not directly connected to the Magma yet, but was plugged into the same Surge Protector, has not suffered any noticible damage. But the LCD looked like a black and white lava lamp until I rebooted. It seems to be working fine for the moment. Hopefully that will not change any time soon.

I know a few people in this forum have had to ship their cards off for repair. I'm curious as to what their repair policy/process involves (beyond waiting). Was there warranty coverage, or did you have to pay for it? Or am I doomed to buy a new one (which I would do, BTW, even if I had to eat bologna for the next year).

If I have to ship it to Germany and it takes 3-6 months, I'll be happy to wait (well not happy, but you know what I mean), but right now I'm really feeling turbulant as I won't be able to get any details until the USA is once again open for business tomorrow (Labor Day and all).

Any information or experiences that you could share in this realm would comfort me a great deal. I'm not afraid of bad news, I just can't stand waiting so I know what to do.

Incidentally, the capacitor reads

47
6s
OM7

and was attached in the center of the far end of the card, opposite the IO plate. It looks like it could be soldered, but I'm not going to touch it until I'm told that I'll have to just buy a new card.

Who knows what the extent of the damage could be. Unfortunately, electronics don't often wear their symptoms very visibly.

Sam

[ This Message was edited by: dehuszar on 2004-09-06 13:41 ]

snoopy4ever · Post by **snoopy4ever** » Mon Sep 06, 2004 12:46 pm

Very sad to hear Sam..

I hope everything be less bad than it "sounds" .., but I think you have your feet on the ground.

Best of luck

Snoop

[ This Message was edited by: snoopy4ever on 2004-09-06 13:46 ]

Post by **garyb** » Mon Sep 06, 2004 12:51 pm

there is warranty coverage, though it's dependant on the age of the card and probably won't include the death of the magma supply(if that's what killed it). the cap could be from the cards power supply, in which case the card might still work if a new cap was installed properly. probably, it'd be a good idea to send it in for service anyway, as there may be other problems that aren't fatal yet, but if power was connected.....

you need to find a freind with a pc that you can put the pulsar in to test it, good chance that it still works......

cwa does offer service, it just takes a little time. give Ali a call tues....(actually, he may be there today, i don't think this is a holiday in Canada...)

dehuszar · Post by **dehuszar** » Mon Sep 06, 2004 1:02 pm

Well, the power supply is of no consequence, and I can pick up a cheap Magma on eBay if I have to, but the 7PCI slot monsters that are always on sale are a bit less portable than I had become accostomed to. But whatever gets me back on track, so to speak.

I'll call Ali right now... thanks for the tip.

Sam

dehuszar · Post by **dehuszar** » Mon Sep 06, 2004 3:46 pm

Well... mildly good news.

The Pulsar II and the Magma still work. The Power Supply on the Magma is still toast, but I was able to cannabalise an oldie from my Dad's not-in-service Pentium 133 and MacGyver the whole thing together. So I can at least power up and fiddle with a few things.

Not so much on all the plugins registered to the Scope though. Maybe CW will have pity on my and allow me to temporarily transfer them over until I am able to get the Scope repaired (assuming the warranty is still in order). If the Scope is over a year old, it's not much older. Anyone know what the expiration date is on these things?

In the meantime, I've got a lovely opportunity to reacquaint myself with the stock synths and plugs. I never really learned Prisma and Lightwave, so it should be a fun project getting back into the swing of things. At the very least it will dull the pain.

Sam

Post by **Guest** » Mon Sep 06, 2004 4:58 pm

I solder's mine a year ago and still work fine.

You have to be carefull as not to apply too much heat.

Post by **garyb** » Mon Sep 06, 2004 5:24 pm

you can try replacing the cap, but take care not to install it backwards....the card IS fixable.

dehuszar · Post by **dehuszar** » Mon Sep 06, 2004 6:00 pm

Good to know, thank you. I'll probably wait until CW says I'm hosed before attempting surgery. If there IS still a warranty, I'd surely void it tinkering with the soldering iron.

Anyone know how long the warranty lasts?

Sam

Immanuel · Post by **Immanuel** » Mon Sep 06, 2004 9:09 pm

In Europe 2 years - but I think that is due to European law more than CWA policy.

astroman · Post by **astroman** » Mon Sep 06, 2004 9:23 pm

to me it reads as if the cap just plopped of the board - it didn't bust in smoke or so (?)
or is there any visual trace of heat at it's original position ?

if the material inside the cap caused a short circuit (the only scenario I can imagine to 'melt' it off it's place), the stabilizer/rectifier of the bord's power circuitry is probably damaged too.

The cap serves as a filtering element in the power lines - the board would work without it.

otherwise it would read like a badly soldered board - not funny if the metal of the cap short-circuited the Magma's power lines

good luck with the board, Tom

[ This Message was edited by: astroman on 2004-09-06 22:23 ]

deejaysly · Post by **deejaysly** » Mon Sep 06, 2004 10:43 pm

I believe what Tom is saying to be true.

47uF 6-Volt capacitors very frequently serve as bulk decoupling (filtering) of a power rail, in this case probably the 5-Volt supply, with digital circuits.

If the capacitor was still visually intact, and there is no sign (i.e. blackening, fused or frayed tracks) of any other damage on the board in the same vicinity, then I would be tempted to plug it back in somewhere and see if it still works but then that is me...

It should work without this component provided there is no other damage (fused tracks) inside the board - me thinking these boards are multi-layer construction.

BTW, I feel for you. I would be gutted if anything happened to my Scope cards.

dehuszar · Post by **dehuszar** » Tue Sep 07, 2004 12:20 am

I did smell "that smell", but no -- there was no carbon scoring or melted plastic or discoloration, nor any smoke for that matter. The smell was probably emanating from the power supply, not the card.

The cap came off pretty cleanly, it's metal connectors came out of the solder (and whatever brown waxy stuff they use on top of the solder - I've seen it before but I can't recall it's purpose). It still fits back in place perfectly, but doesn't stay -obviously.

The cap is in a place where it could've gotten knocked a few times, as I've tried swapping the PCI slot order in the Magma to see if it aided in bus efficiency. Then factor in that it's thrown in a padded laptop bag and travels alot (as per its design)...

If my deductions are correct, I think the cap was loose to begin with and when I plugged it in, the cap blew off the board, not melted. Hence the thunderclap once it got juice. This would cause the Power supply to catch a surge, which then -probably by design- killed itself. Nifty defense mechanism if true.

If that IS true, then soldering it back into place would probably fix it. The Pulsar II and the core components of the Magma still functioning leads me to believe that it was the Scope starting that little conflict, and that capacitor is the only visible sign of damage.

But I'm certainly not going to do a job myself at the risk of further destroying the card in addition to voiding any warranty that might keep me from a card that I spent $2000 on. Granted soldering is not a difficult task, but unless I'm already out of options it'd be a foolish place to start.

If I did have to start again, I wouldn't have to buy the /SP this time around as I own the software already, but even $1200 is more than I have, so I'm going to leave it alone and find ways to love my Pulsar II and get better at the plugins I have access to and get better at my cello while CW fixes my Scope (assuming that's an option).

Thanks for all the responses, BTW. This is a very valuable community and it's been very reassuring to hear from everyone. Just hearing some people who have survived such horrors keeps me from panicing outright. I think I owe the forum a song.

I'll let everyone know what I find out tomorrow.

Thanks
Sam

[ This Message was edited by: dehuszar on 2004-09-07 01:25 ]

blazesboylan · Post by **blazesboylan** » Tue Sep 07, 2004 12:34 am

Hey Sam, do you have access to a multimeter? It's safe to prod around just to see if anything on your card short-circuited. Set the multimeter to measure resistance and try different components on the board. If there's no direct link between the 2 points (i.e. current has to go through a capacitor or chip etc) then you should get infinite resistance. And on either side of a resistor you should get some kind of finite resistance.

This won't tell you whether the chips still work, but it could point out other potential pitfalls.

As for the brown wax: that's PCB guck. It could be that the person who originally soldered it melted the plastic a bit (it's very easy to do, especially on an assembly line). It could also be a sign of melting.

Nevertheless, capacitor legs are very flimsy and easy to break. I would guess that, as you suggested, the cap just broke clean off during transport. Transformers tend to be much more guilty of blowing up and producing "that smell". And I've never heard of a capacitor melting a circuit board...?!?

Tom, Deejaysly, et al, do you all know whether it's a polarized capacitor? I don't know much about these things, but as GaryB points out, you can't just put a cap back in place unless either it's non-polar or you line it up in the right direction.

Cheers,

Johann

Edit: BTW, it IS a holiday here in Canada. But Ali works 24/7, so it's generally pretty safe to call him even on statutory holidays!

[ This Message was edited by: blazesboylan on 2004-09-07 01:35 ]

astroman · Post by **astroman** » Tue Sep 07, 2004 2:41 pm

those low voltage, high capacity caps are always polarized, opposed to their high voltage, lowcapacity bros in switching power supplies ;
The latter ones sometimes prefers to go up in smoke... and I never forget the technician in our shop writing an important letter to a customer when this happened

He mumbled something like 'sh*t, why has it go up now ???' when smoke curled up from the back of the Mac, but he managed to finish his last sentence and hit 'print' when the screen darkened, some seconds later the printer started to deliver the page and he shut off the malicious machine

my concern with Sam's board is that the dying powersupply of the Magma could have sent some ill level voltage when the regulation broke down - but if the cap has been displaced accidently by mechanical force it could also mean that a connection inside the board got damaged too, so that some parts are without powersupply now.
Otherwise the board WOULD operate even without the cap.

cheers, Tom (fingers crossed)

[ This Message was edited by: astroman on 2004-09-07 15:48 ]

BingoTheClowno · Post by **BingoTheClowno** » Tue Sep 07, 2004 2:51 pm

[quote]
...Magma could have sent some ill level voltage...
[quote]

Usually the regulators offer over voltage protection, unless the regulators failed.

The polarized caps pop typically when a reverse voltage is applied to them or maybe when they dry and short out (I think). Was the temperature in your Magma box maybe to high that led to premature failure of the caps?

[ This Message was edited by: BingoTheClowno on 2004-09-07 15:53 ]

deejaysly · Post by **deejaysly** » Wed Sep 08, 2004 9:30 am

The brown waxy stuff that is used in soldering is called flux. It aids in producing nice clean solder joints. Normally from a manufacturing point of view I am surprised you see any on the board apart from the area on the board where you have other components that have their tab soldered to the board (i.e. regulators, transistors). Normally the board is cleaned of residue such as this before passing to test stage.

That particular capacitor is definately a polarized tantalum capacitor and I have seen a few times a similar thing happen but where the whole board has come off worse as well as the capacitor cracked and with a nice carbon mark on the board.

Blaze's idea with the multimeter sounds good. If you place each probe on each pad of where the capacitor was sitting, this would be a good test to see whether this has done damage of producing a short between 5-Volt and 0-Volt on the board. You test will produce an illusion of a short for maybe a few seconds max as the rest of the decoupling capacitors on the board get charged up by the potential coming from the multimeter to perform the test. This potential does not cause any further damage because it is only low voltage and low current for sensing.

If you try to put the capacitor back on yourself, normally there would be a + sign in the silkscreen (writing) on the board indicating the positive pad for the component.

dehuszar · Post by **dehuszar** » Wed Sep 08, 2004 10:48 am

Well, good news for the moment...

At Ali's recommendation, I resoldered the cap back on and all seems to work for now. Later today, I'm going to do some stress tests. I've already had it streaming radio all night, but I intend to fill up all of the DSPs of both cards and play with some PCI resource intensive plugs and see what happens.

If it was the power supply, it's done it's damage and is dead. But halleluya!! I'm back online. I'll still need to track down some keys from 3rd party developers, as I reinstalled Windows and my audio-soft, but I'm otherwise back in business.

Thanks for all your input, I hope my rather unprofessional soldering job holds up for years to come (or until I've sold the cards off and replaced them for newer Scope/XPs --or some such designation-- with newer, more advanced SHARCs, 1GB of onboard memory and the capacity to run SFP on-chip and can stream directly to and from hard-disks).

Until then, fingers crossed.

Sam

dehuszar · Post by **dehuszar** » Wed Sep 08, 2004 10:53 am

On 2004-09-08 10:30, deejaysly wrote:
I am surprised you see any on the board apart from the area on the board where you have other components that have their tab soldered to the board (i.e. regulators, transistors). Normally the board is cleaned of residue such as this before passing to test stage.

Ah, yes flux!! That's it. Well, the flux is put on totally uniform and consistantly, so perhaps it was done in anticipation of any repairs or some such thing.

Sam

BingoTheClowno · Post by **BingoTheClowno** » Wed Sep 08, 2004 11:01 am

Resin flux is a necesseray ingredient for a clean solder joint. All boards have components that cannot be automaticaly placed on the surface of the board thus must be soldered by hand. That's why you see the flux on the board, it is perfectly normal.

[ This Message was edited by: BingoTheClowno on 2004-09-08 12:01 ]

astroman · Post by **astroman** » Wed Sep 08, 2004 12:07 pm

On 2004-09-08 11:48, dehuszar wrote:
Well, good news for the moment...

At Ali's recommendation, I resoldered the cap back on and all seems to work for now. ...

great

the cap's solder joint probably served as an interconnection of 2 layers of the board, so without it one was separated from power.

good job ! Tom