• Please review our updated Terms and Rules here

H88 takes up smoking..

falter

falter

Veteran Member
Joined
Jan 22, 2011
Messages
6,573
Location
Vancouver, BC
I was trying to power up my Apple III and I guess accidentally plugged in my H88 instead. I hear it beep, then next thing I know a cloud of smoke is coming out. Smelled like RIFA, but I'm pretty sure H88s don't have those. Anyway, I pulled the plug, and then found where it appears to have come from.

I'm not exactly sure *what* happened, but it looks like one of the tantalums, formerly blue, burnt up. I'm used to tantalums exploding - I guess they can short and burn too? My concern is why it burned. I don't think it's the PSU going haywire - the other boards seemed fine.

Anyway, can tantalums just burn like that?
 

Attachments

  • 20230222_205032.jpg
    20230222_205032.jpg
    648.1 KB · Views: 49
Definitely crispy.
Isn't burning just exploding but more slowly :)

I have smoked a few tantilums over the years, and exploded a few more. It may depend on how much current can be sourced.
Nowdays I'm more into trying to reform the whole board full first.
 
Strangely, I've also had this happen on an H89 and an H8. Can't remember whereabouts it was in the 89 but the one in the H8 was on the 16V line on the CPU board. Powered up the machine, no display so started to investigate. After 10-20 seconds, the capacitor started smoking. Very different experience to working on Northstar Horizon's where I usually cower below the table in fear of shrapnel when switching on without the lid in place.
 
Yes i should be saying H8.. not H88. I've been warned to replace all the tantalums in my Sol-20 for safety reasons. I remember when a bigger one went off in my Poly 88.. sent shrapnel every which way. I've seen videos of actual fire with tantalums so I'm careful now to fire up machines away from anything that can burn.

I'm hoping this is just a short in the cap and not something else. But I've looked over the other boards.. no damage. Hopefully the components on the CPU board didn't get damaged.
 
Yes, Smurf Grenades can explode, burst into flames or just get warm depending upon how they fail.

@Hugo Holden made an interesting remark (which I checked for accuracy and (of course) he is always accurate) where the amount of power dissipation in a tantalum bead capacitor takes place when the internal resistance is the same as the internal resistance of the power supply unit. Any higher (or lower) results in less power dissipation within the tantalum bead.

Of course, the higher the power dissipation, the more likely the capacitor is to go BANG and explode.

There are then degrees of 'getting hot' from very little to BANG!

The very low resistance is the interesting case. This is where a low resistance is present within the capacitor - so a low power dissipation - but a potentially high current is flowing - and this could cause damage within the PSU itself.

On the other side of the scale, the capacitor could go completely open circuit - so it performs no useful function at all - but you can't detect that.

Dave
 
Thanks Dave. Yes I'm wondering how I should approach this. I don't see obvious signs of damage anywhere but at ground zero. Psu looks ok. Would I be best to remove the boards and power up to see how functional the psu is?
 
That might be a sensible approach.

Can I see signs in your photograph of it potentially damaging another (nearby) smurf grenade?

Dave
 
It is actually a bad case when the Tant goes to a very low resistance and the pcb tracks sometimes vaporize and it is a shame when a pcb gets damaged.

Tants are not the only ones to do it. Recently I had a blue 0.1uF 50v monolithic ceramic capacitor burn up in a unit I built myself some 20 years ago. But they don't do this nearly as often as the old Tants from the 1970's thru 90's era.

Apparently, unless it is just marketing rhetoric, newer generation Tants are supposed to be better, but it might take a few decades to find out.
 
It kind of makes me nervous about all my machines. More the risk of fire than anything else.

Changing them all out entails it's own risks to the machine.. damage to pads and traces, getting something backwards by accident.. etc.
 
I've replaced all the smurfbombs in my H8 with same-value nichicon electrolytics. One on my zero-org card had burned like that. I suspect that the board is fine, just clean the schmoo off with a toothbrush and fix it.

Most, not all, but most of those 2.2uF tants are across the power supply rails. They're not going to take any logic with them when they pop.
 
falter said:
It kind of makes me nervous about all my machines. More the risk of fire than anything else.

Changing them all out entails it's own risks to the machine.. damage to pads and traces, getting something backwards by accident.. etc.
Click to expand...
It is more of a smoke & a stink out the room risk, than a fire risk.

If I replace the old blue Tants, I tend to replace them with modern Tants. The reason the Tans were used in the first place, over Electrolytic Caps, is that they have a much lower ESR, so good for putting on the inputs & outputs of the common garden linear voltage regulators, like the 7805.

Speaking of these, I would like to point out that the linear voltage regulators, like the 7805, the 7812 and their negative regulator cousins, are much more special than most people think.

There is a tendency for modern designers to accuse them of being inefficient, wasting energy as heat, and better replaced by more modern units which use a switching methodology.

But you might want to consider the notion of efficiency; It could be describes as; (wanted form of energy out / energy in) x 100 %. But here is the odd thing, if the energy out you wanted from an electrical appliance was just heat (the typical domestic Bar Heater), it would then be almost 100% efficient except for the red glow. I'm am not against heat production, within reason, if you can gain other benefits.

The genius of the design of the 7805 regulator, and their analog cousins, is the protection from over current and over temperature. They go into a safe thermal shut down mode. So if your Tant caps on the output side at least short out, it is no drama and nothing gets damaged.

Unfortunately most of the supposed better switching regulator replacements don't have this Thermal and over-Current protection.

I would rather have something labelled as inefficient, that was better & safer. Also the linear regulators have much lower noise than the switching ones. They do have interesting noise though. If you want to hear it, you simply couple the regulator's output, via a capacitor into the Phono input of your Stereo amplifier, it sounds a lot like White Niose. It is normal. I discovered this way back in the 1970's working with Audiometers.
 
daver2 said:
That might be a sensible approach.

Can I see signs in your photograph of it potentially damaging another (nearby) smurf grenade?

Dave
Click to expand...
Yes. To my eye it looks like the cap on the left in the picture is the one that went into burn mode.. and then that sprayed toxic goo on the board and the leg of the next cap. I'll replace em. The smoke happened for about 2 secs.. fortunately I'm fairly quick reflexed and I shut it down immediately when I realized what was going on. It was quite a large pervasive cloud and it left me scratching my head because I had recently replaced the x1 cap in the A3 and was wondering if it had somehow blown up again. But then I realized where the power cord was going.
 
You must log in or register to reply here.
Back
Top