Pots question to RG and others

[axg20202]

I needed a 100K reverse log pot, but couldn't find one for love nor money. However, 500K reverse log pots are available, so I thought I'd try and play with the total resistance by adding parallel resistance. Perhaps unsurprisingly, putting a 120K resistor across the outer lugs to get it down to approx. 100K messes with the resistance curve enough to give me a hump somewhere in the middle of the taper - turning the pot up gives a steady increase in value, but it eventually exceeds the predicted value at some point, and then with continued turning the resitance value goes down again to the approx 100k I was aiming for by adding the parallel resistor. It's unusable like this. Is there anything clever I can do to get a rough approximation of a 100k reverse log pot using a 500K reverse log pot?

Any help much appreciated.

[smallbearelec]

Hi--

My SKU 1007 and click the drop-down to select:

http://www.smallbearelec.com/Detail.bok?no=116

[axg20202]

Thanks Steve. I should have been clear in my original post that the pot I'm working with is part of a dual-concentric CTS pot - I have been swapping wafers with single gang CTS pots to create the values I need. A 500K RA was the closest value I could find made by CTS. If you know of a source of 100k RA CTS pots (I don't even know if CTS makes them), that would be great.

[smallbearelec]

I wondered why you were having trouble; now I begin to understand. I'm fairly sure that CTS would make a 100K "C" taper pot for you, but minimum order quantity is usually 1,000 pieces. If your project would justify that many, e-mail me at smallbearelec@ix.netcom.com. I may be able to help.

[axg20202]

Alas, no, it's for a personal project not a commercial venture. Looking at RG's "Secret life of Pots" article, it appears that I might be able to take a linear 100K and use 3 resistors in a parallel network to give an approximation of reverse log. I need to take the time to read this and do the maths.

Would someone with more of a maths head than me care to help me out with this based on the equation given?

http://www.geofex.com/Article_Folders/potsecrets/potscret.htm


Thanks.

[Paul Perry (Frostwave)]

Trying to make one from a linear pot, I think you would need to start from a 500K linear, at least something higher than the 100K target.
On the bright side, if you can live with a bit of cramping, a linear is closer to a reverse log, than a 'normal' log is...

[axg20202]

Thanks Paul. That makes sense. I suppose a linear taper is getting me half way there. As you say, bunching could be a problem, particularly as the pot is for a classic 'vibrato' circuit (i.e. a tremolo to those of us that, unlike Fender, know how to use the correct terminology!).  I'll probably test a few pots out with the circuit I'm building (a redesign of a Fender Rhodes preamp and stereo trem circuit) once it is built to see if I can get away with not using the exact type specified. If I can't get the circuit to work then the taper of the pots will be a minor concern! 

Thanks all.

[axg20202]

To RG:

I've read your very useful "Secret life of Pots" article. If I understand this properly, to turn a 100K linear pot into a 100K reverse log pot, I would have to add a 25k resistor from the CCW lug to wiper, assuming I want the 'b=4' curve. Can you confirm that this is correct? I guess I'm getting confused by the article because on one hand you describe 'b' as a fraction and then later show it as an integer. There also seems to be some differences in how b is defined. It is first defined as b = R3/R but then later the 'taper resistor' (which I'm assuming is also R3) is defined as R/b. Perhaps the first instance where you define b should actually read b = R/R3? Are these errors or am I just being thick?

Thanks,

Andy.

[Pedal love]

http://www.upscalemusic.com/proddetail.php?prod=POT-100KRLGC

[axg20202]

Thanks for the link. It's not a CTS pot, which is really what I need, but worth knowing.

RG: care to weigh in here? I'd like to learn this stuff for future reference so I'd greatly appreciate clarification on my points above relating to your pots article.

Cheers,

Andy.

[slacker]

I've read your very useful "Secret life of Pots" article. If I understand this properly, to turn a 100K linear pot into a 100K reverse log pot, I would have to add a 25k resistor from the CCW lug to wiper, assuming I want the 'b=4' curve. Can you confirm that this is correct?

That's right if you wire the pot as a voltage divider a 100k pot and 25k resistor gives you the reverse log taper. The only possible problem with this is as R.G. mentions in the article the total resistance of the pot + resistor varies as you turn it, with the pot all the way up it's resistance is 100k but when it's turned all the way down it's total resistance is only 20k. Depending on what you're using it for this might cause problems.
If you wire the pot as a variable resistor then if you use a 500k pot and a 120k resistor that gives you roughly the same taper as a 100k reverse log pot.

[axg20202]

I've read your very useful "Secret life of Pots" article. If I understand this properly, to turn a 100K linear pot into a 100K reverse log pot, I would have to add a 25k resistor from the CCW lug to wiper, assuming I want the 'b=4' curve. Can you confirm that this is correct?

That's right if you wire the pot as a voltage divider a 100k pot and 25k resistor gives you the reverse log taper. The only possible problem with this is as R.G. mentions in the article the total resistance of the pot + resistor varies as you turn it, with the pot all the way up it's resistance is 100k but when it's turned all the way down it's total resistance is only 20k. Depending on what you're using it for this might cause problems.
If you wire the pot as a variable resistor then if you use a 500k pot and a 120k resistor that gives you roughly the same taper as a 100k reverse log pot.

Cool. Checking my circuit, it needs to be wired as a variable resistor. So are you saying I should take a 500K linear pot and connect 120K across the outer lugs or should the 120k be connected to one lug and the wiper?

[slacker]

It should be wired like in the bottom picture in R.G's article.