Did one ever beat you?

[Craig]

The "High Efficiency" power supply in 7000-series Tektronix scopes. Nightmare to fix. To test them out of the frame, you have to make up a dummy load for each of the output lines (+/-15V, +5V -50V, +130V) because it does not start if unloaded - you need to pull at least 50W for it to start, and preferably 100W. And the controller chip looks for faults in the mainframe and CRT, so you have to connect those lines in a way that fools the device into accepting that everything is OK. At that point you can start looking for the fault(s).

Even so, these early '70's switchers are darned difficult to fault find. There are two of them on the top shelf that have completely defeated me. The 500MHz 7904 version has the added grief that the CRT HT circuit is buried right in the bowels of the switcher. You need to disassemble the switcher to get to it. So if you suspect that the fault in with the HT, you replace suspect components, reassemble the whole darned thing and see if it works - repeat as necessary. Every other Tek scope has the HT on a separate board in the mainframe.

[Murphyv310]

Hi.
I'd say that there isn't an engineer that hasn't been beaten in some way or another.
Working for big rental companies and having a set on the bench for two weeks would never be allowed, if you are self employed again a set on the bench for two weeks makes absolutely no economic sense.
Of course if its your own set or there is no manager or customer breathing down your neck, then you can get into the thick of the obscure fault, persistence, spares, skill and a manual "should" effect a fix, assuming you have the patience !!!!

[AlanBeckett]

You've frightened me now as my 7704A has suddenly gone completely dead, and it's not the Plug Fuse.
Still, I'll know where to come for advice

Alan

[Murphyv310]

Never take the advice of the bin man Alan

[Mark Hennessy]

I've yet to work on a 7000-series. We have one at work, and there's every chance I could inherit it when we eventually decide to throw it away, but I don't think I will (although I'll probably pass it on via the forums rather than see it skipped). As lovely as it obviously is, it just doesn't appeal. not least because of the amount of custom Tek ICs contained within. The thing about curating old gear like this is that it has to be sustainable, so the fewer custom parts, the better. Specific CRTs and PSU transformers are unavoidable, as are switches and potentiometers in many cases, but if the electronics could be as "vanilla" as possible, so much the better

HF PSU transformers tend to fail - the types used in Philips PM3215/7 and similar can fail with a short between 2 windings - that's because there's nearly 2kV between them. The same failure mode can affect many other 'scopes - I have a Tek 576 curve tracer at work that I think probably has the same fault. Odds of finding a replacement? Pretty slim...

I've got a good reputation for fixing switched-mode power supplies, and I run a 4 day course at work telling people how they work, and how to fix them. However, that reputation is partly because I've learnt which ones to try, and which ones to walk away from. Modern supplies tend to "wear out", with capacitors and start-up faults being the main problem - and that's thanks to designers (of ICs and switching transistors, as well as the overall supply) getting good, meaning that "blow ups" are much rarer than they used to be. But even so, if a PSU has blown up, you've basically got to be prepared to replace every bit of silicon and probably many other parts as well. Worth it sometimes; but often not. The trend to putting everything inside a single 8 (or 7!) pin DIP makes that easier for the smaller units...

[Craig]

Yes - some Tek HT transformers do indeed develop winding shorts.  Some seem to be more prone than others - the 7834 multi-mode storage scope seems to be pretty vulnerable.  I have a copy of Tek's winding spec - and it is not at all trivial.  Someone on the Tekscopes list managed to wind one to resurrect his 7834, and it took a while!  The other thing that goes is the HT multiplier, which is something that it is also possible to repair, but the guts need to be de-potted and so is rather messy.  If that goes, the trace just disappears.  But the HT is around 18kV to 24kV, so it is kind of expected that every once in a decade or four it might have issues!

Alan - a specific issue with the 7704A is around the HT box at the rear of the CRT, right side as you face the screen.  Take the lid off.  Now Tek went through about five or six revisions of the board - because the earlier ones had a discrete resistor divider chain made of carbon composition resistors.  Which in the fullness of time burns the board - and we're talking full charcoal here.  Which causes the HT to fail because of the conductivity of the charcoal.  I fixed one of mine by grinding out the charcoal with a dremel and conformal coating.  Then I wired up the chain with new resistors dead bug in the air.

Just for laughs, I attach the 7834's HT transformer spec

[AlanBeckett]

Thanks Craig. I'll get to it shortly.

Alan

[tiv48]

Well here is one thats got me beaten.
I wanted to be able to check a draw full of crystals so i got one of the kits from china £6.91 http://eu.banggood.com/Wholesale-Warehou...59592.html
spent 30 mins building it but on plugging in a test crystal 8.667238 the reading i got was 11.116 ?
checked another two and again the reading bares no relation to the marked value.
this device can also be used as a frequency counter this part works well (see pic ) 
the circuit only shows basic connections to the PIC chip .
on the good side it does show if the crystal is dead or drifts .
but there we are it has beaten me 
Rob T

[Nick]

I've had a LOT of Tek kit over the years, but my 7904A eventually got the "click of death" - the classic 7000-series SMPS failure mode.

I gave it to someone with more time on their hands

[Craig]

The 7904A is an entirely different animal to the 7904.  It is pretty rare and of higher performance, and does not have the HT buried in the SMPS.  Although formally specced at 500MHz, if you use a 7A29 vertical amp it is close to 800MHz.  Whoever you gifted it to, even with a tick of death power supply, is a lucky person indeed.

I've kind of taken a test gear collection a bit to extremes - partly because things I used in the 70's and 80's that were major league investments for the University and companies I worked for are now low in price - and partly because the build standard is such a thing of electronic beauty.

[attachment=14553]