Precision wirewound pot wiper bounce?

Funny, I recently discovered that in my old Heathkit oscilloscope. The = horizontal position knob is loose for almost all of its travel: on = closer inspection, the contact broke away from the conductive plastic.

The funny part is, you turn it and it does nothing, except at certain = points. When it's open, the voltage just hangs there (held in place = with bypass cap; the input uses a FET follower, so leakage is = negligible). So as you turn it back and forth, trying to find the right = position, you pass it over the "active" areas, and it slews a little = bit, so you keep working it, ohh, too far, go back the other way...

I've seen old wirewounds that are scratchy. I'm a little surprised a = new one would be.

Tim

--=20 Deep Friar: a very philosophical monk. Website:

one would be.

Maybe it's just my luck....Back in the early 70ties when "build your own audio amp" was popular in my neck of the woods, I built this harmonic distortion meter and the circuit called for wirewound pots. Worked fine except for the wirewound pots. Had to wipe then back and forth a few times for everything to work smoothly even when they were new.

I bought this recent wirewound pot with great reluctance based on that previous experience.

On Jul 13, 10:53=A0am, "Tim Williams" wrote:

With wire wound, it can be the other way around.

The wire faces that the wiper hits are initially more rounded, providing a smaller tangential surface than is there after wear has taken place.

You should stick with then name brands, as it were, for multi-turn pots. There are only about three or four.

Bourns is pretty common, and should have addressed such a problem decades ago. Another thing you can do is put a pot across one leg of your 'bumpy' pot to give a fine resolution adjustment of that leg, which should make the 'bumps' less prevalent.

Can you think about why they are only one turn devices?

Also, on the other, it should not matter about the bumpy ride on the other scenario if the final position of the pot is all that matters. It would matter if it is a setting where the pot is going to be used constantly, like in a power supply voltage setting knob or an audio amp volume control.

I have seen hybrid pots, multiturn wirewounds that have a cp coating over the wire element. That smooths out the discrete wire steps and sort of lubricates the motion.

But to the op's question, a wirewound pot wiper will show electrical steps as you cross turns, but it shouldn't go open at any point. Wiper contact resistance will vary a lot, so don't load the wiper.

John

I'm am more than a little surprised, T. Moron Williams, that you are too goddamned stubborn to set your line length correctly.

More likely that you are too goddamned lazy to find out how to do it in your pristine Outhouse Unexpressive top notch news client. Even more utter stupidity on your part, that choice was.

You are about as retarded as it gets, Williams. Are you Keith's brother?

We use a lot of the Bourns 3590 sereis ten turn pots. These are wire wound, but I've never seen any problems. If you look real close you can see little steps in the output, presumablly as the wiper moves from one wire to the next.

George H.

That is actually a rule.

Think about it.

There must not be an open wiper circumstance at ANY time during the operation of the pot... ever.

Yes. There are actually several wiper "elements" and more than one turn in contact at any one time.

The "bumps" are spikes in resistance as one wiper is walking off the top of a wire, and the next wiper is walking onto the top of another.

The old way was to make a single wiper span more than one wire top. They later arrived at the multi-fingered approach to make the transitions as 'smooth' as possible, without setting up losses within the pot itself at the junction node (wiper contact 'point').

Another situation where the ideal circuit diagram and the real part do not match exactly.

Make a single wire finger, and get transition issues. Wipe enough fingers to kill the transition issues, and you get contact node issues.

I think the happy medium is three fingers and always falling off one and riding up on another as the center follows along on whichever wire top it happens to be on.

Another example of this is an optical chopper wheel,where they double the wheel's resolution by staggering in a second sensor one half hole pitch distance from the first, resulting in two pulses per hole pitch lineal traverse.

No, you? ;) BTW, Bourns makes a Hybritron=AE Element 10-turn pot, what exactly is that, some sort of hybrid I presume? They seem to be used in game controllers.

Agreed, my experience so far seems to be that they perform poorly while >

Well, Bourns does give a wiper bounce time of .1ms for their precision wirewound and hybritron pots. Not certain whether that's the norm or coincides with a vibration/shock condition. Whatever, all indications are that the wiper of the pot in question is losing contact with the element. I'll probably request a replacment but my hopes aren't all that high.

I'm using the pot to vary the voltage to the VCO input of a function generator. A 1-turn pot is a bit coarse and I estimate that a 10-turn pot is going to be too fine. The 3-turn pot is just right except for the mentioned issue. I'll try replacing that first. Thanks!

Ah, this is a low-voltage potentiometer? Switches and pot wipers get very reliable at 100V and up, but fail a lot in the 1V range (because the thin film of dust and dirt is enough to insulate the low voltage). That's why a lot of low-voltage circuitry design avoids wiper contacts.

I take exception to the advice to 'not draw wiper current'. Sometimes you can fix dirty pots in-situ by doing exactly that. It carbonizes the oily film... Same with switches.

These are 3 turn conductive plastic - 10k and 5k

What is it, a small helix that the wiper climbs?

Looks that way.

Grant.

Thanks! That's good to know. C>

Decided to monitor wiper voltage using a storage oscilloscope. With

Drop in voltage is around 4V and occurs at around 6V output. Worst case wiper bounce duration is around 2ms. Here's the puzzle, why is the waveform so square as though a solid state device is involved?