Pot wiring. Pot questions.

[Diabolical Artificer]

This one's been nagging at me for a while. When Wiring a pot or preset as a variable resistor, not as a potential divider there are two ways to do it. 1) One connection to either end and one to the wiper. 2) The other is the same but join wiper and one end. I struggle to "see" how this works. But it does. Thinking about it the connected bit act's like a variable R as the distance from one end and wiper in/de-creases. Still it also makes sense that effectively your connecting across both ends so should get the full pot value, not variable.

Why is the wiper and one end connected? Both #1 and #2 work the same, I think on a linear pot. Is there a difference electrically?

Secondly on log pots surely the pot has to be connected the right way, but you never see pots marked for correct connection, why? Surely the logarithmic taper works one way only, IE from 0 degrees to 270 degrees. Connect it backwards 270 deg - 0 deg and it's antilog, no?

Andy.

[Craig Sawyers]

Originally Posted by Diabolical Artificer

Secondly on log pots surely the pot has to be connected the right way, but you never see pots marked for correct connection, why? Surely the logarithmic taper works one way only, IE from 0 degrees to 270 degrees. Connect it backwards 270 deg - 0 deg and it's antilog, no?

Andy.

The convention is that turning the pot clockwise is correct to get log operation. You just have to think about that carefully when wiring one up. And yes - I have got it wrong before even when thinking (overthinking?) it carefully!

Craig

[M0FYA Andy]

I've always assumed that connecting the 'unused' end of the track to the wiper is good practice to accommodate the, admittedly unlikely, possibility that the wiper lifts off the track intermittently. So the rest of the circuit sees a resistance within range rather than open circuit.
That's probably my background coming into play again, always asking 'how can things fail?'
Andy

[trh01uk]

Andy

Good question. Note that what you are talking about is strictly called a "variable resistor". While a "pot" (which is short for potentiometer) is the thing you are not talking about - a potential divider.

So getting back to your variable resistor with the wipe connected or not connected to one end of the track. Electrically when everything is working fine there is no difference between the two arrangements. But variable resistors have a nasty habit of the wiper losing contact with the track. And if that happens when only one end of the track and the wiper are connected you suddenly get an open circuit - i.e. infinite ohms.

If you connect the wiper to one end of the track and the wiper loses contact with the track you then get just the full resistance of the unit instead of infinite ohms. In some circuits that difference may be critical. A typical circuit where its critical is in the adjustment of the output voltage of a voltage regulator. In some arrangements a variable resistor that goes open circuit can then tell the regulator to put out maximum volts - and you may then find whatever its powering is fried.

Your description of the log/antilog operation of pots is certainly the one I have too. Its based on the convention of increasing volume when you turn a knob clockwise. Given that convention, I think you can only wire them one way to get the correct log effect. No doubt one of the audio guys will confirm.


Richard

[Dave Moll]

Certainly, what has been said so far aligns with my understanding of logarithmic variable resistance tracks - owing to the logarithmic nature of sound volume (i.e. as sound volume increases linearly, the actual energy increases logarithmically). Less commonly, I have in the past also come across reverse-log tracks - usually in combination with a log track in a double unit for balance control on a stereo amplifier, one track increases the volume of one channel while the other track decreases that of the other channel.

[paulsherwin]

Some amps do use a log/antilog pot as a balance control, but it isn't very common because the subjective operation is no better than a dual linear pot, and it's easy for tolerance errors to produce a skewed stereo image with the control at mid position.

[ColinTheAmpMan1]

The most common use of antilog pots that I know of is on the "tremolo" controls of some Fender amplifiers of the valve era. In front of me I have a layout for a Fender Deluxe Reverb AB763, i.e. the classic "blackface" type. The tremolo intensity and speed controls both have antilog law, although being American, they are referred to as "reverse audio". Interestingly, the speed control is wired as a variable resistor, rather than a potentiometer, i.e. wiper connected to fully-anticlockwise terminal; they then go via a 100k resistor to ground. You will often find that Fender amplifiers share the same front ends, differing only in the output stage. So "AB763" is often seen relating to amps other than the Deluxe Reverb.

As an aside, Fender called this effect "tremolo" mistakenly; it is actually "vibrato", but then Fender consistently used these terms almost interchangeably and incorrectly. For those of you that don't know, "tremolo" is a musical wobbling of pitch, while "vibrato" is a wobbling of volume. Very few instances of true "tremolo" exist in guitar amplifiers (Magnatone & Vox, excepted).

Colin.

[joebog1]

HHMM Colin the ampman!!
I think you have the effects backwards
Vibrato is when you move your finger up and down the neck on a violin, OR change the pitch slightly
Tremolo is almost exclusively used on electronic/electric music instruments.

Joe

[Diabolical Artificer]

Thanks all, that makes sense.

Re logarithmic pots. If you have the pot shaft facing you, terminal's down is the terminal first on the left ground or sig in? You get more confusion if the terminal's are facing up, IE rotated by 180 degrees.

Andy.

[Craig Sawyers]

It is like one of those IQ questions where you have to visualise something and say which figure is similar.

First on left is ground with shaft facing you and terminals down. Just visualise what is happening to the wiper.

Craig

[mhennessy]

Hi Andy,

It's simple - you just have to pretend you have X-ray vision

In other words, I use my imagination to picture what's inside the case, and can "see" where the wiper is as you turn the shaft.

What helps is that when I was a kid, I took literally everything apart, so in practice, I'm just re-living happy childhood memories!

Often, there's a gap in the case where the terminals emerge, and with the right light you can see the wiper come into view as you get near the track ends.

Yes, with the shaft facing you and the terminals down, the left one is ground. If in doubt, turn the shaft fully anti-clockwise, and measure to see which end has continuity with the wiper.

Another tip: to tell if a pot is linear or log, rotate the shaft to half-way round and measure from wiper to an end. For linear, you'll get close to 50% of the stated value, and for log you'll get typically 10% or 90%.

Now, if you know which end of the track is ground, you can tell if it's log or anti-log. Between ground and wiper, 10% means log, 90% means anti-log.

With all that in mind, if you know for sure that your pot is log, then that's another way to determine ground. Again, put the shaft in the mechanical centre of the rotation, and if your 10k pot measures 1k between wiper and an end, that end is ground

Once you've got the ground end sorted, a small mark with a black "Sharpie" doesn't go amiss...

Cheers,

Mark

[Diabolical Artificer]

Smashing! Thanks both.

Andy.

[ColinTheAmpMan1]

Originally Posted by joebog1

HHMM Colin the ampman!!
I think you have the effects backwards
Vibrato is when you move your finger up and down the neck on a violin, OR change the pitch slightly
Tremolo is almost exclusively used on electronic/electric music instruments.

Joe

You are quite right, Joe. I stand corrected and will throw my self into the same pit of ignorance that I have consigned the Fender people to! The shame of it! My only excuse is that what is sometimes called a "whammy-bar" has been inaccurately called a "tremolo arm" for years and I was probably thinking along those lines.
Mind you, "tremolo picking" is a rapid up-and-down picking technique that is used commonly in mandolin playing, so tremolo is only almost exclusive to electronic instruments.
Colin.

[G6Tanuki]

Pragmatically, are we talking about a 'potentiometer' used for audio signals at the input of an amplifier, or are we talking about a 'potentiometer' used to set the bias, balance or drive-levels of an amplifier?

In the latter case - having designed and built a few hundred serious RF amplifiers over the years - I'm conditioned to wire the two track-ends to the ends of the bias-supply, but also to fit a couple of fixed resistors, one from the 'hot' end of the bias-supply to the wiper, and another from the wiper to the 'cold' bias-supply.

That way, if the wiper ever loses contact with the potentiometer-track you at least have *some* degree of bias applied to the output-valve [see http://www.cpii.com/docs/datasheets/...500A7-8877.pdf and realise why I don't want to have customers melting the grids...]

[Herald1360]

In the latter case - having designed and built a few hundred serious RF amplifiers over the years - I'm conditioned to wire the two track-ends to the ends of the bias-supply, but also to fit a couple of fixed resistors, one from the 'hot' end of the bias-supply to the wiper, and another from the wiper to the 'cold' bias-supply.

Interesting idea- by picking the resistors you can determine the default bias level, as long as you can live with the nonlinearity over the adjustment range that they introduce.

Anyone else notice the misplaced comma in the 3CX1500 datasheet?

[Diabolical Artificer]

Q1 was pots in general. Q2 refered to log pots as used in amps to control gain.

Having difficulty deciphering terms as used in your explanation G6T. Ends of the bias supply? Are we talking fixed bias here, if so ends of bias supply means ground and negative Xvolts (or pos Xv) ?? The hot and cold confused me a bit also, but presume you mean again ground and tother end of fixed bias PSU.

So in effect your putting resistance in parallel with the pot in case the wiper lifts.

A.

[Skywave]

Potentiometers, often (but not always) with a nominal 6 mm. shaft diameter and typically from the far east, often have a letter, A or B associated with their nominal Ohmic value. Usually - but not always - an A indicates a logarithmic law (mnemonic: A for audio); a B a linear law. I do not know what letter is often assigned to a reverse-log. pot.

On the topic of reverse log pots (also known as anti-log pots, I believe), they are commonly used in Eddystone valved communications receivers in the position of R.F. or I.F. gain controls, located in the cathode cct. of an amplifying valve cct. So when such control does need replacement, doing that can get a bit difficult, especially as some of the older Eddys. used wire-wound pots. in those positions. Substituting a linear gives a very cramped 'scale'; a log. pot. will suffice, but to increase the gain of the associated stage, its shaft needs to be rotated anti-clockwise, of course!

Finally, it has been my experience that log. pots that obey a perfect logarithmic law are virtually impossible to obtain. (Not that that matters in most applications). The track is often constructed 'piece-wise' to approximate a logarithmic law.

Al.

[kalee20]

Definitely, connectiung the wiper to the redundant end terminal is sound engineering practice. It makes no difference when things are working OK, but could stop a drama turning into a crisis if the potentiometer starts to wear out and the wiper loses contact with the track.

Originally Posted by Skywave

Finally, it has been my experience that log. pots that obey a perfect logarithmic law are virtually impossible to obtain. (Not that that matters in most applications). The track is often constructed 'piece-wise' to approximate a logarithmic law.

Not just virtually impossible, totally impossible!

If the logarithm of resistance is to be proportional to the angle of rotation, then, for a 270 degree 1M control with 10% resistance at half-scale (a common convention), you'd see 100k at 135 degrees. And at zero degrees you'd have to see 10k. Because every 135 degrees gives you a change of x10. So you'd never be able to get zero resistance!

[Station X]

Plotting the resistance against rotation angle for a log or anti-log law pot generally yields two straight lines with different slopes.

Not just virtually impossible, totally impossible!

Some wire wound log pots used a tapered former for the wire, these can approach log near enough as damn it.

This http://www.geofex.com/Article_Folder...s/potscret.htm explains all.