HP 1630A goes BANG

[Tim T]

Several years ago, I obtained a HP 1630A logic analyzer thinking it might be useful one day, and other than an initial powerup check it has sat unused. Well, that day has arrived, and all appeared to be fine over several short sessions of maybe 30 minutes to an hour or so just to learn how to use the thing. Then, when about 3 hours into a session for pulling out some bus data – BANG. I killed the power within a few seconds and if memory is correct the fan and CRT continued to work over that interval.

The interesting part is no smoke or smell, and in tearing down the 1630a I have yet to find any obvious damage. I replaced the RIFA safety cap that is exposed on the power supply board before starting to use the analyzer.

Is there a non-destructive way to check the two X2 and two Y caps within the sealed EMI filter?

Thanks,

[TheMG]

Take it out of circuit and measure capacitance between L-N, L-G, and N-G.

Capacitance values may be labelled on the EMI filter, or should be able to find them on a datasheet. If not, L-G and N-G readings should be about equal, and less than the N-L reading.

[Tim T]

Right, a photo would have helped. 

I read N->GND at 6.12 nF
Same for GND->L
N->L is 290 nF

N->L resistance is 10.31M

Inductance across L is 2.88 mH
Same for N

Thanks for help. For some reason I was thinking it would be more complex.

So it looks like I need to dig deeper to find the fault.

[coppercone2]

well thats a pleasant story for the holidays

I heard some kind of high pitched like almost whistling noise from something like a fluid was leaving from something last week but I can't find where it was lol.. this is why I put extreme effort into grounding robustness for all equipment I have.

[srb1954]

Right, a photo would have helped. 

I read N->GND at 6.12 nF
Same for GND->L
N->L is 290 nF

N->L resistance is 10.31M

Inductance across L is 2.88 mH
Same for N

Thanks for help. For some reason I was thinking it would be more complex.

So it looks like I need to dig deeper to find the fault.

Measure the L-G and N-G resistance, preferably with a proper insulation resistance tester that applies 500VDC to the capacitor under test. Any resistance under 10M indicates a failing Y capacitor.

Also test the L-N insulation resistance test with the insulation resistance tester to see if the insulation resistance drops below the 10.31M previously measured. Any significant drop in insulation resistance indicates a failing X capacitor.

[oz2cpu]

if something said BANG, there must be some kind of visual damage to find,
I would also expect some part of the circuit is not working uptimal now,
you need to keep looking

[DC1MC]

I had once a 300pF cap fully disintegrating, with a loud bang, but no smoke or traces, only the (short) legs remained, was difficult to found, especially since the basic functionality of the device was not visibly affected.

[Tim T]

Yea, the lack of smoke or obvious damage is why I was thinking it may have been the sealed EMI filter.

One nice thing about the 1630 series is the CPU and acquisition boards can be easily removed from the motherboard, as well as ability to disconnect individual ribbon cables for CRT and keyboard from the CPU board, making narrowing down the problem area a bit easier in principle. So far, the power supply alone looks to work fine, though unloaded, the -2.4 rail is running at -3.4. Did learn that this SMPS does not like to start when using a variac to bring up voltage, so it was a bit of a "grit your teeth and duck" moment when first powering it up at full mains.

Found something odd with the -5.2v path on the motherboard showing just 17 ohms to ground. The only things on the MB are connectors and branches off the -5.2 to four resistor networks (each with a 10 nF ceramic cap to ground) that appear to be voltage dividers which if my thinking/math is correct will supply -3.0 volts to the acquisition boards, but this is a guess. The otherwise very comprehensive service manual doesn't say much about the MB. I lifted each cap just to check and still had the 17 ohms. Anyhow, I am getting -5.2 v at the ps test point, with no over-current warning LED.

Next step is to add back the cpu board.