Is there an engineer in the house? Blown capacitor while refurb'ing a 1050...

[SlagOMatic]

So I've got a handful of 1050's in various states. I've managed to successfully refurb and upgrade two of them. I fully disassemble them, give the boards a bath in an ultrasonic cleaner filled with electronics cleaner, rinse with distilled water, rinse again with 99% IPO, dry with air blower, replace all capacitors with kit from Console5, replace the TP110, 5v, and 12v regulator, apply new thermal compound, reassemble, and test. If testing is successful then I install a Happy 1050 upgrade, test again, and if successful I do a little happy dance and move on to the next one.

 

I worked on one tonight and though I did all of the above, this one needed some extra attention. This one happens to be the very first 1050 I ever owned and it's seen some shit. Once upon a time I installed a US Doubler upgrade in it which has worked without incident, but I wanted to change that to a Happy upgrade. I also didn't want to actually destroy the US Doubler upgrade (which has two 6810's soldered together) so I had to order a replacement 6810, which I did. (I actually ordered five of them, all clearly used/recovered from old hardware, not new-old stock.) I installed the replacement 6810 along with the Happy upgrade. The drive select switch was damaged (permanently jumpered to D1) so I got a new switch from Best Electronics and installed that too. I ran the multimeter over the bits that I worked on and everything seemed fine. I plugged it in without the drive mechanism attached and again, it seemed fine. Power light went on and the computer saw it on the SIO chain. It was receiving power for no more than about 10-15 seconds. I turned it off, connected the drive mechanism, and turned it on again. It got power and the drive initialized; everything still looked good. This time I left it turned on while I connected the SIO cables to the computer. Just as I was about to turn the computer on, BOOM! The capacitor at C67 exploded like a firecracker. I immediately shut everything off and unplugged it.

 

The capacitor in question is 4700uF 25v and I have a few extras, though not many. As I'm not an electrical engineer (I'm good with a soldering iron and basically understand a multimeter, but reading a schematic and troubleshooting a circuit is beyond my skills) I'm wondering where I should start to troubleshoot this. It could be something as simple as that capacitor actually being bad. OTOH, I replaced 11 capacitors, two regulators, the drive select switch, the TP110, the 6810, AND installed the Happy upgrade so there's a lot of variables at play. I do have two known-good Happy-upgraded 1050s that I can test against but I also don't want to risk frying one of the 1050s that I've managed to successfully resurrect. As my limited understanding of schematics allow, I looked at the field service manual for the 1050 and C67 appears to be part of the 5v power rail. The wisdom of the internet tells me that capacitors can blow by being overheated (which is likely the case here, given the puff of smoke that came out after it blew) and I'm guessing the reason for the overheating is a bad short somewhere.

 

Uneducated instinct is telling me to double-check all of the other capacitors I replaced to see if any of them are shorted, replace them if so, replace the blown one at C67, and then start substituting some known-good parts from one of my good 1050s. I'm reasonably sure I can write off the drive select switch as being a potential cause, and if I'm reading the schematic correctly I can write off the 12v regulator and any capacitor that isn't on the 5v circuit. Assuming none of the other capacitors on the 5v circuit are damaged/shorted that leaves me with the Happy upgrade, the replacement 6810, and the TP110. I can test the Happy and 6810 using parts from my known-good, and I have extra TP110s that I can swap in (assuming the TP110 could conceivably be the problem).

 

Any advice/insights/prayers would be appreciated. Thanks.

[_The Doctor__]

did you use the insulator sheets and spacers on the regulators?

did you make sure the orientation of capacitors etc were correct?

did you inspect for shorts/bridges splatter flecks and flakes?

did you clean up flux and whiskers?

when you plugged the drives mech in did you put the connectors in incorrectly? not putting them as noted in many a thread can and will kill sensors and sometime blow crap up. etc etc.

Edited February 22, 2023 by _The Doctor__

[BillC]

42 minutes ago, SlagOMatic said:

Just as I was about to turn the computer on, BOOM! The capacitor at C67 exploded like a firecracker.

Perhaps it was installed with the polarity reversed!

 

[macsonny]

2 hours ago, BillC said:

Perhaps it was installed with the polarity reversed!

 

That's the best way to have an exploding cap!

[TGB1718]

6 hours ago, BillC said:

Perhaps it was installed with the polarity reversed!

Pretty much guaranteed with electrolytics, shorts etc will damage other components, but unlikely

to damage capacitors (other than over voltage).

[Kenshi]

That's the smoothing capacitor just after the rectifier. Yes, electrolytics will go pop in no time if you put them in backwards. Pay attention to the stripe.

 

Lucky for you, it probably didn't take out anything else in the process, but I'd check diodes CR17 to CR20 to be sure.

[SlagOMatic]

On 2/21/2023 at 10:10 PM, _The Doctor__ said:

did you use the insulator sheets and spacers on the regulators?

did you make sure the orientation of capacitors etc were correct?

did you inspect for shorts/bridges splatter flecks and flakes?

did you clean up flux and whiskers?

when you plugged the drives mech in did you put the connectors in incorrectly? not putting them as noted in many a thread can and will kill sensors and sometime blow crap up. etc etc.

Only the TP110 has an insulator and spacer, and they're both in place and doing their job. The 5v and 12v rectifiers have neither; they are supposed to be grounded to the heat sink.

 

Yes, orientation is correct.

 

Yes, no shorts/bridges/splatter/whiskers/other bad things. I also tested all the circuits that I touched before plugging it in.

 

Yes, connectors are in correctly. I marked their order and orientation before I unplugged them.

[thorfdbg]

The prime reason for exploding caps is really putting them in backwards. One can not overstate this, and I would bet that this is exactly what happened. Unfortunately, the way of how the polarity is indicated is really pretty much dependend on the manufacturer, so you really need to double-check with that and the drive schematics. Also, if you put in the diodes reverse way, that would certainly lead to major damage, including the caps.

[SlagOMatic]

Okay, I've got closure.

 

I'm left with little choice but to say this was just a bad capacitor. I removed the destroyed cap, tested a new one, installed the new one, then tested all along the circuit and couldn't find a single fault. I powered it up and let it sit for about five minutes with no fireworks, then connected it to the computer and ran the Happy 1050 diagnostic tests where it passed everything. I'm now going to take the mechanism out for cleaning and lubricating and a new drive belt, but otherwise this drive works fine now.

[TZJB]

17 hours ago, SlagOMatic said:

Okay, I've got closure.

 

I'm left with little choice but to say this was just a bad capacitor. I removed the destroyed cap, tested a new one, installed the new one, then tested all along the circuit and couldn't find a single fault. I powered it up and let it sit for about five minutes with no fireworks, then connected it to the computer and ran the Happy 1050 diagnostic tests where it passed everything. I'm now going to take the mechanism out for cleaning and lubricating and a new drive belt, but otherwise this drive works fine now.

C67 is shown as 4700uF 35V in the field service manual so maybe the voltage is near the rated value of the capacitor?

[Kenshi]

1 hour ago, TZJB said:

C67 is shown as 4700uF 35V in the field service manual so maybe the voltage is near the rated value of the capacitor?

The manual says the plug feeds in 9 VAC. Do a little math and the most voltage the capacitor should ever see is around 24.5 V. I don't know that a 25V capacitor is going to last for decades under such conditions, but it shouldn't blow after a few minutes of operation. It seems more likely it was put in backwards or was just bad.

 

I'm a little concerned about C68 though. Between it and the rectifier is a voltage doubler, so it's going to see at least 45 V accounting for some losses. Why does the manual say it's a 35 V capacitor? And did Slag replace it with a 25 V as well?

[_The Doctor__]

I would contact the vendor to correct the kit to use what Atari used. Running at full capacity is a bad idea, fluctuations and inductive kick or inrush current is real thing.

[Kenshi]

I just realized I multiplied by 2 an extra time to get the numbers above. C67 and C71 will see about 12 V at most and C68 about 22 V. Personally, I'd prefer at least a 30 V capacitor for C68, but the 25 V for C67 is far more than enough.

[SlagOMatic]

19 hours ago, TZJB said:

C67 is shown as 4700uF 35V in the field service manual so maybe the voltage is near the rated value of the capacitor?

I can't find that spec in my copy of the FSM, but 100% of the C67 caps from all of my 1050's are 4700uF 25V, as are all of the replacements I've installed/will be installing. The two caps at C68 and C71 are both 6800uF 35v.

[SlagOMatic]

16 hours ago, _The Doctor__ said:

I would contact the vendor to correct the kit to use what Atari used. Running at full capacity is a bad idea, fluctuations and inductive kick or inrush current is real thing.

They ARE the same spec as what Atari used. OEM C67 is 4700uF 25V, as is the replacement cap.

[SlagOMatic]

13 hours ago, Kenshi said:

I just realized I multiplied by 2 an extra time to get the numbers above. C67 and C71 will see about 12 V at most and C68 about 22 V. Personally, I'd prefer at least a 30 V capacitor for C68, but the 25 V for C67 is far more than enough.

C67 is 4700uF 25V, while C68 and C71 are both 6800uF 35V. That's both OEM and the replacement cap kit I'm using.

[TZJB]

3 hours ago, SlagOMatic said:

C67 is 4700uF 25V, while C68 and C71 are both 6800uF 35V. That's both OEM and the replacement cap kit I'm using.

That is weird, but seems correct. My copy of the diagram must be wrong.

 

[Kenshi]

3 hours ago, SlagOMatic said:

C67 is 4700uF 25V, while C68 and C71 are both 6800uF 35V. That's both OEM and the replacement cap kit I'm using.

This is the manual I was looking at and probably the same as what everyone else is as well.

 

http://www.atarimania.com/documents/atari-1050-field-service-manual.pdf

 

On page 5-5, it shows C67 to be 35 V. As I said above, it'll only see about half of the 25 V rating of the capacitor you put in it so I think what you used is fine, but this manual is why everyone is saying that it should be 35 V.

[SlagOMatic]

27 minutes ago, Kenshi said:

This is the manual I was looking at and probably the same as what everyone else is as well.

 

http://www.atarimania.com/documents/atari-1050-field-service-manual.pdf

 

On page 5-5, it shows C67 to be 35 V. As I said above, it'll only see about half of the 25 V rating of the capacitor you put in it so I think what you used is fine, but this manual is why everyone is saying that it should be 35 V.

Yup, that's what I have too. I was looking at a later page that has an (incomplete) list of components. Didn't occur to me to check the diagram. But yeah, that makes sense. Thanks.