Problems with carbon composite in a HV string.

Hmm, I'm not sure how you know that, but OK. I don't know about SF6, but in general a heavier gas has less thermal conductivity. (each atom carriers the same energy 3kT, but heavier atoms move slower... shorter mean free path, so less energy gets transferred per unit of time. What I din't know about SF6 is that it's a molecule*. It'll have more than 3kT of energy per molecule.. but I don't know how much more.) This may sound like a research project, but you might stick a little TC temperature sensor of the resistor and see if it's any different in air vs SF6.

And then as other's have said it may be the voltage rating of the resistor. Got a part number/ spec sheet?

Good luck, George H.

*molecules are to hard for physicists to deal with so we like to pawn those off to the chemists. :^)

Carbon comps are the only though hole resistor with non-magnetic leads, (that I know of) so we still use a few. They also have great 1/f noise. If you like that sort of thing.

George h.

Seriously, look at the Caddock parts.

It's got to be expensive to replace those strings of carbon comp resistors.

--

John Larkin                  Highland Technology Inc 
www.highlandtechnology.com   jlarkin at highlandtechnology dot com    

Precision electronic instrumentation

I had a look at your file and I really don't see anything there that fits our application?

As for the cost of the resistors, we just got a batch of 1000 with a 5 year old date code with tight coefficient specs, 5 bucks each.

I am not the one footing the bill for this so I don't care :)

We did get a return correspondence from a resistor manufacture and apparently they are more than willing to make us a batch of 5k for a very reasonable price.

Jamie

What? I heard they were the only type to exhibit excess noise.

Tim

--
Seven Transistor Labs 
Electrical Engineering Consultation 
Website: http://seventransistorlabs.com

That series of resistors is a serpentine thick film on ceramic. They make high voltage versions. They are *very* fast, with less shunt capacitance than a carbon comp would have. They won't have the degradation mechanisms that a comp resistor probably has.

$5 each for carbon comps?

Labor cost will still dominate replacing the resistors, perhaps often.

--

John Larkin         Highland Technology, Inc 

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Sorry that's what I meant. If you want to study 1/f noise then they are the "classic" example.

George H.

Err.. well, depends on what is "great" for a given purpose, after all. :o)

Tim

--
Seven Transistor Labs 
Electrical Engineering Consultation 
Website: http://seventransistorlabs.com

On Mon, 5 May 2014 17:14:08 -0700 (PDT), Bill Sloman Gave us:

How does an idiot like you even know what his set-up is experiencing?

Films OPEN, idiot. BULK FORM carbon composition devices ARE indeed the correct choice for this application. Ask anyone who has performed HV design work.

The current limit resistor in nearly all, if not ALL HV supplies is of this very type. The exception are the devices JF posted a link to. They are NOT cheap, and they are physically much larger, and a much larger device for using them would have to be fabricated.

EVERY Geiger Counter in current use by the Navy, sold to them by SAIC, was made by us, and they ALL have bulk resistors in them. When the package is smaller than a long pack of cigarettes, one is not going to place one of those ceramic substrate type parts in that package... EVER.

Impulse absorption is the same way.

Getting around the effects of the HV is difficult, but switching to another type of device is not going to work without significant device design change and cost.

On Mon, 5 May 2014 17:14:08 -0700 (PDT), Bill Sloman Gave us:

"MV"? Where?

And the fused lattice issues crop up over time, not immediately.

On Tue, 6 May 2014 18:45:41 -0500, "Tim Williams" Gave us:

Shot noise? Johnson noise?

Think about it. As the resistance value gets higher, the composition matrix gets looser and looser from a conduction/lattice POV.

More collisions, and more circuitous path to get from one end to the other when incomplete lattice work is involved.

On Tue, 06 May 2014 17:30:13 -0700, John Larkin Gave us:

But they are NOT meant for impulse absorption.

So.

Nice presumption, there. A carbon matrix is a carbon matrix. There is no reason that ANY such device would not "age" with so much stress introduced into it.

I would bet that they DO age similarly. Just not nearly as fast since the path length is much longer.

So, if they did not open up, because they are not meant for that purpose, they would still age downward in value for the same reasons. Just not nearly as quickly. More likely to experience open failure modes. Likely constantly.

On a sunny day (Tue, 6 May 2014 18:12:46 -0400) it happened "Maynard A. Philbrook Jr." wrote in :

mm sounds like 0bama

Some time ago I noticed 1 GOhm several watt resistors for 1$50 cents or something on ebay, film of course. For 1.5 MV and an estimated 10kV per resistor you would need 150, or about 225 $, much less likely in that quantity. But WTFDIC I do not pay for it ;-)

As to 'pulse power capable', get a life, 'pulse' can mean many things to many people, but I think your rise times are so slow that the inductance of the spirals in the resistors has no influence, and if it is some divider that can simply be compensated for. I am with J.F. on this one, fix the problem, use a recent better solution.

There is also the time required to solder 1000 x 22M versus 150 others. And of course the power consumption.

On Wed, 07 May 2014 05:08:21 GMT, Jan Panteltje Gave us:

Carbon comp resistors have no spirals, idiot.

On a sunny day (Tue, 06 May 2014 22:11:25 -0700) it happened DecadentLinuxUserNumeroUno wrote in :

I was talking about the film ones, moroning idiotic flabber poster. Illiterate incable of reading blubber excrementing fool! ;-)

On Wednesday, 7 May 2014 14:52:05 UTC+10, DecadentLinuxUserNumeroUno wrote :

Electron microscopes typically run with up to 30kV from one end of the elec tron column to the other. I once got dragged into a discussion of what we g oing to have to do to satisfy a customer who wanted 100kV. When I was much younger I got peripherally involved with Alex Strojnik (the father, not the son) when he was setting up a 600kV scanning transmission electron microsc ope in Melbourne.

e loads.

Carbon film resistors, like carbon composition resistors, don't necessarily open. Like all negative temperature resistance elements they can form "hot channels" and carry a great deal of current with very little voltage drop.

I've seen a nominally 10k carbon film resistor passing an amp of current at a fraction of volt drop. It measured resistance afterwards was still 10k.

The procedure for getting the hot channel was not gone into - the guy who w orked it up was on the "intrinsically safe" circuit design standards commit tee, and he'd used it to get carbon-film and carbon-composition resistors t otally banned from any device that was going to be labelled intrinsically s afe.

Proper high-voltage-rated resistors aren't cheap. It's a small market.

Their lengths are typically set by "flash-over" distances. Sinking them in SF6 (sulphur hexafluoride) does help. Alex Strojnik sank his in transformer oil .

You wouldn't. The ceramic is typically alumina, and it isn't cheap. Geiger counters aren't spectacularly high voltage devices

says 400V to 600V. Your design may be more old-fashioned.

Which way is that?

ther type of device is not going to work without significant device design change and cost.

If the device you've got at the moment is blowing up, you are stuck with de vice design change anyway. Carbon composition resistors were always cheap r ubbish, and calling me an idiot won't change that.

--
Bill Sloman, Sydney

On Tue, 6 May 2014 22:37:02 -0700 (PDT), Bill Sloman Gave us:

They do when they are being pulsed with HV, dumbass.

On Tue, 6 May 2014 22:37:02 -0700 (PDT), Bill Sloman Gave us:

Which they will do.

Now try it at kilovolt stress levels. You couldn't be more stupid if you tried.

On Tue, 6 May 2014 22:37:02 -0700 (PDT), Bill Sloman Gave us:

You're an idiot.

Consumer grade, just like you are layman grade, AND SLOMAN grade. Slow as f*ck. You still haven't even mentally matured past twelve. How quaint that you think you bring anything valid to this discussion.

These supply 1500 volts at 8mA. They get set at about 1250V. You are a clueless google tard.