…coming back to the 3326A repair, a slight mystery remains – at one time in the past the over-voltage protection circuit must have been triggered, destroying not only the SCR, but also damaging some quite large traces on the board. Typically, the SCR is just switched on for a few milliseconds, before the mains fuse will blow, with no damage of any trace. Typically, these HP devices are well engineered, so there is only a small chance that they didn’t design the system right, to the extend that power would be provided to the board for any length of time even with the over voltage protection circuit triggered.
Now, the 3326A has passed a 24 hour run-in and multiple power cycles, time for the final safety check which I perform on any instrument repaired, like, checking the ground resistance, checking the isolation resistance, checking the mains fuse rating, etc.
THE FUSE. SOMEONE MUST HAVE BEEN FULLY CRAZY, AND PUT IN A 20 AMP FUSE!!!
That’s why the over voltage circuit didn’t work, and presented a fair risk of fire and instrument damage, of an instrument, sold at over USD 10k, 1990 dollars.
The fuse rating – 3 Amp, normal blow.
And a big warning sign, hard to overlook.
But what is a normal blow fuse? In the US, fuses are actually rated by the manufacturers, and standardization is well above the level of fusing time, etc. – two types are typically sold, ‘Slow Blow’, and ‘Fast Acting’. Slow blow fuses are easily idenfied, they either have the wire wound around a ceramic core, or are filled with sand.
Found this document, related to ‘normal blow’ – as it turns out, these fuses are actually identical in rating to the typcial ‘Fast Acting’ fuses that I have in stock here.
Well, there might still be some tiny difference of the original, HP branded ‘normal blow’ fuse, but test shows that a regular 3 Amp AGC ‘Fast Acting’ fuse is working perfectly find. So for the time being, we can just consider ‘Normal Blow’ and ‘Fast Acting’ fuses to be identical. How easy would be the world, if all things were build according to common, well-established standards. At least, for the fuses.
Dear Simon,
I’m trying to repair a 3326A with a similar problem! The SCR ‘protection’ circuit seems to be quite fond of using PCB tracks as fuses rather than the fuse. My fuse was 2 amps, only slightly too large than the specified 1.5A (240V) but at some time in the past the SCR clearly triggered and I have a big earth track burned-out on the A70 board.
My problem is though that my A70 board is marked Revision C, and the service manual I have only has schematics and layout for an earlier version (yours, judging by the photo). Do you know of any source of manual update pages to cover the Rev C board please? Rev C is different in layout (although the main functional blocks are still in the same places) and in circuitry – for example there are two SCRs. It would be good to work out at least which one triggered.
Many thanks if you are able to help at all please. I’ve also asked on eevblog.
Best regards, Bill Woodbridge (UK)
Hi Bill,
I have checked the various manuals I have for this unit, unfortunately, no backdating sheets or info about th rev C board. It may not be very helpful, but would assume that the circuit is rather similar design, and hopefully you can trace back the gate trace to the related trigger circuit. Repair should then be relatively straightforward.
Simon
Thanks so much for looking. Yes, I’ve now got as far as determining that for the Rev C board there are separate SCRs for crowbarring the +5v and +15v lines, the gates presumably triggered by separate comparator outputs upstream as you say, although I haven’t got that far back yet. It was the +15V that triggered (or at least, that’s the one that’s now permanently shorted and is adjacent to the burned-out track). New SCR, board repair and (very) tentative power-up of A70 to follow! Thanks again, Bill
I think replacing the SCR and checking the comparator should fix it. If possible, I try to power the board by a bench power supply first to avoid a loud bang again! Simon