[Attempt at] Repairing a PKR251

How to get to vacuum, what the classes are, and what is needed for what job.

[Attempt at] Repairing a PKR251

Postby chrismb » Sun Oct 16, 2011 8:44 am

My PKR stopped working during an experiment with an unstable plasma. My PSU blew a power output transistor. It is unclear which came first.

The PKR251 is a very useful gauge, this being an 'active, full-range' gauge, meaning it has both Pirani and magnetron gauges built in which automatically switch over at some defined pressure, and that therefore puts out a given voltage versus pressure for atmosphere down to 10^-9. Therefore, though I just scored two magnetron gauges off ebay for a song (and they seem to work too, always a bonus!!! But the chambers need a clean) the PKR251 is a king of gauges and so I am keen to fix it.

Here is the gauge before my disassembly, for those not familiar with these (the wiring is, obviously, my own as I do not have the specific connector for the pins):


I removed the chamber off the PKR251, which is a straightforward operation for routine maintenance. You are then left with a cap full of electronics. This is a 'non-repairable' item.. but as we know, anything is repairable with enough will!!

However, 'non-repairable' means 'one-way plastic clips' and it looked like I had to break the plastic frame up to get the bits out:


There are 3 layers of round boards that open out.


The part I suspect is indicated below. Reads 2.5 ohms, or so. Seems to be an ohmic resistance, no rectification or non-linearities. I can't quite make out the letters. Maybe that indictes it has overheated a bit?


Well, I'm looking to you chaps to tell me what you think, else I will be forced to use the 'brute' technique - apply the volts and let the high current flow until one of the components feels like it is getting hot!

Looks like it is that component, a cap and a voltage regulator parallel to that 2.5 ohms. I would presume the voltage regulator is hard to 48V for a few seconds, but the cap is likely vulnerable. Make your guess, and I'll desolder whichever is the odd-on for being faulty.
Last edited by chrismb on Sun Oct 16, 2011 11:07 am, edited 1 time in total.
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: [Attempt at] Repairing a PKR251

Postby chrismb » Sun Oct 16, 2011 10:34 am


I did some searching and I reckon this is a 33A TVS. I found a line across one side of the package, which I am bound to conclude is 'cathode'. I can't see what else it'd be.

Anyhow, I figured of the components I could remove, only that one was easy enough to put back again anyway. Took it off, and it reads 2.5 ohm. It is a dead TVS. I shall buy another and it will be replaced in short order. (I hope that is as far as the fault condition got, but makes sense wrt the simultaneous transistor failure in the PSU - an excess current draw outside the SOA of the transistor during a pulse event, or an unregulated output of excess volts that the TVS successfully clamped. Either way, it makes sense.)
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: [Attempt at] Repairing a PKR251

Postby Doug Coulter » Sun Oct 16, 2011 5:52 pm

Best of luck with that. Sure it's not a diode? The TVS's I've seen don't have a polarity. At least, we think if it's motorola, it's not a capacitor since they don't make them (AFAIK). A little tracing back to see if it's across the power supply, or perhaps in series with it might pay off quite nicely...Always wondered what these guys have inside (I have two here myself, and yes, they're very nice to have working correctly). I'd guess the power supply went first - mine have seen some serious nearby lightning, and still work. Perhaps a huge impulse picked up on a ground loop did both in.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
Posts: 3023
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: [Attempt at] Repairing a PKR251

Postby chrismb » Mon Oct 17, 2011 1:38 am

Whilst looking for replacements, I note that On Semicondctor (was Motorola semi deivision) mark their SOD128 packaged 31-34V/1mA unidirectional TVS with '33A'. It has a clamping voltage of 45V and 4W, so with the power supply trying to push out 48V, something was going to give, in a few fractions of a second!

Perhaps a local discharge in the plasma pulled the PKR251 to a place where the TVS saw a much bigger voltage to ground, and gave in, thereby overloading the PSU. Or either the TVS or PSU BJT could have given up for no particular reason.

If this fixes the PKR251, then I guess we shall never really know as 'normalty' would be restored... and we put it behind us and move on to the next misadventure!...
Posts: 620
Joined: Thu Aug 05, 2010 6:32 pm

Re: [Attempt at] Repairing a PKR251

Postby johnf » Mon Oct 17, 2011 12:35 pm

I suppose this is why I keep telling people to use TVS (transil) diodes to protect their equipment.
Normally you use the bidirectional ones on AC lines and unidirectional on DC lines. The unidirectional are more hardy to large HV discharges which make them ideal for self biasing electron suppression apertures. I also use them to divide off enough potential from terminal vols to supply the arc volts for the ion source ( ie if you use permanent magnets then you dont need an isolated supply for the ion source)
Posts: 381
Joined: Sun Aug 08, 2010 3:51 pm
Location: Wellington New Zealand

Return to Vacuum Technique

Who is online

Users browsing this forum: No registered users and 1 guest