Oscilloscope Fluke PM3082 powersupply problem?

On Sun, 13 Jun 2010 14:16:36 -0700, powerampfreak ??o??:

Flames out the back from an incorrect measurement set up? That would have to have been one hell of a mistake.

=8D:

e l

I appreciate your wise input! What I was thinking about, was if he in some way measured on mains live net for some reason. Like if he forgot to use the isolation transformer.

I might suggest that you install a fuse that fits [the correct current rating if possible, but not essential] and then do a 'Hi-pot' test between the two mains prongs and the chassis/earth prong. It is quite possible that in his test set-up that the earth prong was isolated from earth and the 'scope chassis was raised way too far above earth and caused a preakdown of the insulation betwixt the mains wiring and earth internal to the 'scope. I would look particularly at the 'Y' caps of the line filter since thay may have 'healed' after the fault but could certainly have caused an explosion of the mains fuse if there was sufficient power applied to the chassis from the test set- up. An example could be measuring 550V or 600V non-isolated 3 phase power where you might get >1000V between the chassis and 240V mains power. If such a thing happened, replacing the 'Y' caps [and fuse] might be all that's required.

Neil S.

On Sun, 13 Jun 2010 15:10:58 -0700, powerampfreak ??o??:

One would assume that this would be an unlikely scenario but not that far removed from reality. Anything is possible I suppose and your knowledge and diligence will no doubt find an answer. I would try to stick with the cold hard facts and not let speculation muddy the waters.

When looking for a manual, also look for the Philips PM3082 scope. It's a nice 100Mhz 2 trace scope probably worth fixing.

Ask the retired engineer if there was more smoke than flame. My guess(tm) is the traditional Philips fried electrolytic capacitors. Flames mean that something caught fire, which is usually a resistor. I've seen an amazingly large flame come out of power resistor, leaving just a small burn hole.

Well, if it's not the primary, it's probably something shorted in the output section of the power supply. My guess (again) are the electrolytics. Do you have an ESR (equivalent series resistance) test meter?

Can you disconnect the output of the power supply section by unplugging something or cutting a wire/jumper? Take the load off and isolate the power supply. Extra credit if you can separate the low voltage and high voltage (CRT) power supply sections.

You might also want to use an AC isolation xformer, in case it really was an internal AC power line fault, to prevent a repeat performance.

Sure. Measuring something with an AC power line ground fault will cause plenty of current to flow through the probes, case, and possibly the internal circuitry. However, that usually fries the scope probe and/or vertical input circuitry which is mostly near the bottom or front of the scope. The back is inhabited by the power supply section.

Not. I'm familiar with earlier Philips scopes (PM3214 etc), all of which have electrolytic capacitor issues. Replacing all the power supply caps and some of the PCB electrolytics, usually revive them.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

"powerampfreak"

** For a glass fuse to shatter and blow like that, there has to be an arc from end to end - which would explain the flame out the back. For there to be an arc inside the fuse, the AC current has to be very high ( say 50 amps ) which means there was a virtual dead short on the other side of the fuse.

That short may have blown open during the event, typically a copper track vaporises or a semi blows open.

It may also be that a EMI suppression cap across the AC supply ( or supply to chassis) shorted and then blew open.

Keep looking and checking things and you will spot something - like a pair of wires poking out the PCB that connect to nothing.

.... Phil

o

the

y

a pair

Great advice everyone! I'll continue investigating the powersupply deeply! I will let you know when I found out!

powerampfreak schrieb:

...

Once the SPS of a notebook computer made the mains circuit breaker (16Amps) interrupt, while I connected it to an outlet. Big bang, arcs and stuff.

It turned out, that one, maybe two, diodes in the rectifier was open afterwards. The SPS is still working on 220V, on 110 probably not.

HTH, Falk

IF you are sure there's nothing wrong on the primary, check for shorted diodes and caps on the secondary side of the switching transformer. You mention that you checked the "switch". Do you mean the power transistor(s) that drives the primary side of the switching transformer?

As Jeff said, isolate the power supply from the rest of the scope if possible. You should be able to use a lower rated fuse at this point to help minimize any damage if you don't have any way to current limit the AC input.

claims to have a service manual for $10.

Might also want to search for free manuals for PM3084, PM3092, or PM3094. They may share the same power supply.

I've examined the primary side of the powersupply very carefully. Absolutely NO signs of damage anywhere. The Y caps measures correct 2nF (2pcs of 1nF) also X cap ok at 0.22uF. A NTC inruch limiter measures 15ohm, correct according to the text on it, primary switch transistor ok, mains rectifier diodes ok all of them... But the curiousity took control over me. I cleaned up after the blown fuse. Put a new one (1.6AT) - connected it to mains - and it works! All channels and functions seems to operate just fine. There are really no sign of flames anywhere. One thing which I haven't paid any attention yet, is the mains filter which is an integrated block of some coil and filtercaps encapsulated. I will change the Y and X caps to make sure they're ok. I've never seen a shorted secondary side make the mains fuse explode. Just primary shorts has the power to do this, in my opinion based on hundreds of smps repairs. Anyway, it appears that I've got another nice looking four channel oscilloscope on my bench! :-)

transistor(s)

AC

PM3094.

What are the primary side HV caps like? I would replace or at least remove and inspect underneath.

I've been doing repair work for over 25 years, but until a few months ago I had never seen one of those fail. Was repairing a Lecroy 9374 scope that would run for a few hours, then blow the line fuse. Turned out to be the chassis mount input filter. You could actually feel it running quite warm.

Glad to hear you got your scope up and running. If you don't already have a operators manual, you can download one from Fluke.

Congrats. I would have used an isolation xformer and variac, but if you have a supply of fuses, that works.

Probably made by Corcom.

They can be blown by a lightning hit, but not much else.

Are there any MOV disks around the AC power jack? If yes, look for a tiny hole where the flames exited.

Now that you mention it, you're right. Secondary shorts usually just cause the switcher to shut down. If there are big fat capacitors on the output lines (not commonly found on PC switchers), they could store enough energy to blow the primary. I've seen secondary shorts blow the primary fuse in "brick" type switching power supplies, but not resulting in the blackened smoking mess you described. That would require a short on the primary side.

However, this begs the question of what caused the flames and smoke? If the fuse was intact, it's unlikely. If it was an AC power line fault of some sort (lightning hit, lost neutral, hi-v distribution line falls on power line drop), it would have destroyed other pieces of test equipment in the lab. Pardon my suspicious nature, but methinks you should run it with the covers off and keep an eye on this scope for a while.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

Though Jeff is almost certainly correct, it's possible that the original "blast" was caused by a screwed-up ground.

I haven't used my scope in a long time because I'm paranoid (probably without reason) about connecting it to a DUT that is also connected to the power line. (Go ahead, start throwing garbage at me for my ignorance/stupidity.)

There would have been some evidence of the fire and smoke if it were a bad ground. I've done it often enough. Fried scope probe leads were common.

Ignorance is curable. Stupidity is not.

Get an isolation xformer and GRI power jack. Use it on whatever you're testing. They have eliminated most of the pyrotechnics for me.

Most of my scopes are dual trace analog scopes which offer A-B differential inputs. If you're not sure about the grounding, or working on the AC mains side of the PS, this can be useful. Do *NOT* connect the ground leads of the scope to the DUT. Instead, connect the B channel input to the DUT common, and the A channel input to whatever you're measuring. If you're not sure what you're dealing with, use an isolated DVM to measure the AC and DC potentials before applying the scope. If you have a large bank account, a battery operated Fluke ScopeMeter is isolated from AC power.

Always remember that you have only one life to give to your chosen profession.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

Took out the mains filter. It's a "konfektronics HGN365-2-F1". It measures 5mH through each line and about 100nF across. This seems ok, of course it can have some intermitent problem. It's a special type with 5x20 fuse holder and internal mains switch which the switch on the front of the scope alters, through a metal wire. I consider letting this to be unchanged for the moment. It's still a mystery what caused the original blast. I'll have the scope up for longtime testing and see if it will re- blast. :-O

on

l

the

The powersupply input has no MOV. I thought this from the begining until I realized it was a inrush limiter NTC. Actually, if looking very close to the NTC body, there are a small small mark in the middle, but not a hole. If this was the source of flame I would suspect more evidence.

It's usually on the fuse holder or input filter. The idea is that the MOV should be the first device the power sees after the fuse, but before any circuitry. For power supplies with AC connectors mounted on the PCB, the MOV will of course be mounted on the PCB.

It's probably not the NTC thermistor. It looks very much like an MOV, but goes between the input bridge and the input DC filter caps.

If it were smoked, the resistance would be quite high and the scope would not work. However, you might consider scraping the hole with something sharp to see if it can be enlarged. If the outer coating on the TNC themistor collapses into an internal cavity, then there's a problem.

Got an IR thermometer? Wave it around the power supply and input section to see if anything is hot and ready to blow.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

I actually got in touch with the industrial engineer who operated the scope when it failed. He said the scope was NOT under any measurements when the flame came out of the rear! It was just standing switched on and suddenly bang and flame...... It sure looks more and more to be the mains filter, though measures fine.