Dual Rev Log Pots at Digi-Key

[JohnRoberts]

I don't recall ever using or wishing I could find a "dual" reverse audio taper pot. The single reverse audio taper is useful for typical mic pre gain circuits, but even there it's hard to get good gain law around wiper "hop off" region (without secondary process tricks). Perhaps useful in a stereo gain circuit, but tracking could be a concern.

For state variable designs, I prefer using forward log/audio taper but connected potentiometrically (is that a word? I mean as a voltage divider only). Further I don't use end or range limit resistors in series, but connect the pot to ground and parallel with a real resistor so wiper is grounded at low frequency limit and frequency is set by the more precision discrete resistor in shunt.

Pots are notoriously wide tolerance for their actual resistance (typically 20%). They focus their accuracy on taper conformity. You will get much more repeatable and accurate frequency control using the pot in it's strength (as a divider) with higher precision external components to set actual resistances. Note: you want to keep pot impedances low wrt external circuit impedances to reduce impact of tolerance and loading at intermediate frequencies.

With such a topology, the pot is out of the circuit at both extremes so your tolerance is purely the precision of the external parts used. Tolerance between the endpoints is the external component tolerance and precision of the taper which is much better than the 20% bulk resistance spec.

I've been messing with SVF since '70s and it took me a while to figure this out, so you won't see this in my earliest designs, but I've used it in several later designs with good results.

JR

 

[pstamler]

[quote author="RogerFoote"]Buss gets used alternately for bus in power distribution as well as audio..Even here in the US of A.[/quote]

And it used to mean "kiss", which is probably how it got into the spell-checkers' whitelist.

Peace,
Paul

 

[JohnRoberts]

Indeed 'buss' is a real word as Paul points out, so not flagged by spell checkers.

There are many individuals and even some professional audio companies that should know better, using the (IMO) wrong word, very much in error.

This appears to be mostly an audio industry affectation, and there is some confusion with "Buss" a brand of fuses. The computer and as far as I can tell, rest of the engineering community gets it right.

JR

 

[JohnRoberts]

[quote author="mediatechnology"]The PEC are really smooth feeling. As smooth as CP.

As a mic gain pot they were pretty good too. I think they're a 10% rev log which is the more standard flavor.

In that particular app (mic preamp) 5% or 2.5% rev log would be better but that wasn't the pot's fault. They just feel professional and are rugged. You won't find 5% or 2.5% rev log anywhere in stock.

I'm glad Digi-Key has someone that knows these are in-demand.[/quote]

Back at my old day job we made several custom pots, to give customers gain laws that they like. To keep a customer happy, give them good nominal gain setting at 12 o'clock, and as much gain as they would ever need when WFO, without sounding scratchy or jumpy.

We would routinely require up to three separate screening overlays to get a smooth transition in hop off region with good nominal gain law. My recollection is something like 25-30k Ohm total, but with smooth low ohm adjust ability at hop off. The individual screening overlays would use different (lower) bulk resistance inks, in just the high gain (low resistance) end. Obviously more expensive due to the extra manufacturing steps, but customers can be surprisingly picky, and incremental costs in high volume manufacturing was on the order of tens of cents.

JR

 

[JohnRoberts]

[quote author="RogerFoote"][quote author="JohnRoberts"]and as far as I can tell, rest of the engineering community gets it right.

JR[/quote]

Lol, that you know of... The engineering community is far larger than consumer electronics. :green:[/quote]

Huh? consumer :?:

I've been researching this one for some 25 years and have yet to find a credible justification or proper reference for spelling it 'buss' in any engineering discipline.

One British editor (STUDIO SOUND) I discussed this with years ago found an obscure Scottish root for 'buss' that meant a cup or vessel, but even he felt omnibus was the proper derivation and 'bus' the proper spelling.

I looked in places like the IEEE S-100 Dictionary-Standard of technical terms. Not a consumer standard but intended to cover all (electrical) engineering.

Frankly in the grand scheme of things, this is nowhere near other common errors like saying phase when you mean polarity and such. One is not likely to be confused or make a mistake due to use of the wrong word/spelling for bus but to me it is a tell about the person's discipline with language and terms. I don't waste time and effort trying to change the behavior of others (I bring it up here to inform), but I am truly interested in knowing what is correct for my own use, and base my opinion on much investigation.

Counting hits on google is amusing, and no doubt if 99% of the population prefers 'buss' then 'buss' it is, but please indulge this tired old man, what pray tell is the root or derivation of 'buss' that you consider justification, or only that you've seen others write it that way?

JR

PS: Perhaps I get motivated to attempt precision in writing from doing a lot of technical writing. Over the years, seeing some of my earlier efforts really bothered me when I see my rookie mistakes preserved in ones and zeros.

 

[JohnRoberts]

The few pot series I've seen that were capable of mix and match elements were pretty sloppy wrt wiper tracking from section to section.

On the non-customizable series, one might be able to swap out a rear section but the front is usually captured by a mechanical swage attachment between pot shaft and rear wiper assembly.

Since the front element is captured any attempt to mix resistive elements means sacrificing two pots to get one. Also since these are designed to be machine assembled, hand assembly may introduce front to back tracking errors.

IMO perhaps best to stick to standard parts wherever possible. Maybe roll your own for one-off proof of concept prototypes but performance and reliability may be shaky even for DIY.

Note: I haven't looked at the subject pots so my comments are based on sundry older series.

JR

 

[Svart]

You guys are missing the obvious point that they are 30$ a piece.. :roll:

I just ordered dual rev logs from OMEG and they were 2.50$ each.. :thumb:

For 30$ I would buy two dual deck rotary switches and make some nice attenuators..

 

[adamasd]

Whatever happened to those claims of various VCAs being used for VC resistors? Doug Curtis had that as a feature on the datasheet of the 3360 dual VCA way back when. I never tried that application, but if it worked and you could trim zero and span, it should be the ultimate multi element pot. At least it sounds that way.

A few years back there was an article on EDN.com using the LM13700 as a VCR, it worked quite well, can't seem to find it though.

Seems like if the loose tolerance/tracking of multipul ganged pots is acceptable LRDs would also work. Or at least be easy enough to match to the standards of pots and alot simpler then the VCA/OTA methods. Never get scratchy and the slow response ignores any scratchiness in cheap control pots.

 

[JohnRoberts]

It typically comes down to the Benjamins... To get equivalent linearity to a cheap pot requires a much more expensive VCA and increased component complexity. I used a dual OTA in a state variable configuration to make the sine wave generator for the old Loftech test set. The more than 1000:1 osc range in a single sweep worked well with voltage control and the linearity was acceptable in the context of the application.

JR

 

[deveng]

John,

Speaking of Loftech, I have one of the original units. I've been wanting to add an extra pot for "fine" frequency adjust for years. Just haven't got around to it. Do you know or remember what the mod is to do this? I don't have schematics for the unit.

sorry for the side track in this thread!

Jeff

 

[Slenderchap]

We stock a range of OMEG P16's and P20's and ship anywhere ... quite a few single and dual rev log values.

http://www.audiomaintenance.com/acatalog/potentiometers.html

We also do some Sfernice and Radiohm.

Colin
www.audiomaintenance.com

 

[Svart]

Svart, that's exactly what I am trying to get away from ($2.50 pots)

That doesn't mean I am just trying to throw money away, but after nearly a half century of doing audio electronics, cheap ass pots are not acceptable. It means I might wait to finish something because the panel controls are $20 instead of $7.50, but hey.

True, but does cost really determine the reliability? We all use pots in equipment that use carbon tracks and loose tolerances dating back to the 70's and before. the problem lies in lubrication and DC. A lot of folks just use the pot forever and fail to clean and lubricate them or allow excessive DC to wear at the carbon. This IS an electromechanical device and we all know that without lubrication the mechanical portion will wear out. I personally consider this maintainence just like changing the oil in the car.

I know that some track materials will last longer than others but sometimes careful design will easily overcome some potential problems. However I'm sure that your experience does show that there is signifigance to the pot design and track material, please don't think that I am in any way trying to negate that. I also understand that industrial usage for pots just needs the next level in reliability.

I just needed 10 pots to finish what I am building and OMEG was the easiest to deal with especially since I could verify the quality before ordering unlike the few other places to source a dual revlog pot..(smallbear)

BTW, those 2.50$ pots are plastic track, metal case and metal thread. the all plastic ones are even cheaper.

I once contacted Alpha to see if they would make a pot with a centertap. they wanted 30$ each for 1000.. The same pot without the centertap is 2$ from mouser. Does the price mean that the custom pot would be better somehow?

svart: Omeg only list one distrubutor in the US for their product. I went there and if you click on the picture to link to pots it's broken. Seems like Omeg need a Mouser or Digi-Key. Where did you order yours from?

Directly from OMEG via email. They are making them custom for me, as they do with all their orders. 2.50$ each is a great deal since they are technically doing "custom" work. the only sore point with me is that they don't do (center) tapping anymore. I suppose someone could do a blanket order from OMEG and stock pots that people use regularly too but again you have to specify what you want when you order, they don't stock anything as far as I know.

As for me, I designed a motor control that used a digital pot that controlled the error voltage for a PWM IC. It worked very well although the digipot was rather expensive compared to the rest of the circuit. The digipot was controlled by a pair of Quantum's touchsensor ICs. (I also adapted the unit to work with a regular pot too as well as a rotary encoder.) My idea was that the digipot could be switched from manual control over to MCU control and back on the fly so that a remote computer could control it as well as a person standing in front of it and still have a discrete board that could be used in standalone mode at any time. It worked out really well but got shelved because the company simply couldn't sell it for what they wanted to sell it for, they were afraid their customers would take a look inside the box and see the 20$ worth of parts after paying 5k$ and be pissed. At least I got to keep the prototype..



Slenderchap: thanks

 

[Svart]

Thinking about it, maybe a rotary encoder driven digipot could be used for audio.. A lot of the digipots are really just CMOS stepped attenuators internally. I've never tried to put audio through one, maybe the distortion would be high though.

 

[Svart]

By DC lets clear up that you mean DC riding on AC...

yes, of course, I should have cleared that up but I was thinking faster than I could type!