Hey all,

I've got a few 5 kohm potentiometers that I'm messing around with, but their behavior is really strange. If I start at 0 ohms and slowly increase the potentiometer resistance, the resistance goes up just fine until about 100 ohms, then it jumps drastically up to about 1.5 kohm, then increases fine after that. I have a few of these pots and all behave the same way, so I'm pretty sure it's not just a bad pot.

I'm looking at using these in voltage divider circuits to provide analog inputs for various uC projects, but if I'm going to want some sort of precision (I always do ), that jump from 100 to 1000 ohms really bothers me. For example, if I'm creating a voltage divider circuit with an output (across the pot) that sweeps from 0-5V with a 15V input and 10kohm resistor as the other element, my voltage would jump from approximately 0.15V to 1.36V across the potentiometer when the pot is barely turned. Not good enough for me

I'm wondering if there's a way to compensate for this abrupt change and get a more "linear" voltage sweep. I can't think of anything right now besides lookup tables in the uC, but that would require setting the lookup tables for each pot individually. Maybe a resistive voltage divider isn't the way I should be looking at it, but I'm not sure. Any thoughts?

 

I'm wondering if there's a way to compensate for this abrupt change

I don't you should go this way.
If these are normal potentiometers they are either dirty inside oder defective.
Strange thing is, you say they all behave in exactly the same way...

Are these new potentiometers? What type?

 

What are the 5K pots from? My first thought is that they have some kind of weird custom track resistance configuration.

 

Are these multiturn pots? If so, I've learned long ago that "precision" multiturn pots are not stable.

They suffer from hysteresis (due to lag in the wiper assembly) and instability due to mechanical shock.

IMHO, only single turn pots should ever be used for precision adjustments. The circuit needs to be optimized to provide the best setability while providing enough range to accommodate all the tolerances which you are trying to adjust out.

I uploaded some software last week to assist in designing optimal resistor dividers using pots:

http://forum.allaboutcircuits.com/showthread.php?t=56960

 

Thanks for the quick responses. To try to answer everyone's questions, I am not sure what these came out of or what they were used for. I tried using one for my DC motor PWM controller project, but the abrupt change in resistance didn't provide good control. They're actually 4-pin potentiometers, and from what I've measured out, the fourth pin is a center tap halfway between the ends of the internal "resistor".

I thought they were pretty slick because I could start it at the center and get (almost) zero resistance between the center tap and the wiper and then turn it either direction and get the same resistance both ways. It looks like what I thought was "slick" is turning out to be garbage . The only marking on them is COB70024, which only told me they were a Clarostat product (via Google search).

I like the "bi-directional" feature that it seemed like I could get out of them. I'm not entirely sure what I'd use it for yet, but I liked the concept.

As far as the turns go, they don't even make one full revolution from what I can tell. They're not trimmers. They're the big beefy kind with a shaft.

 

If they are not physically damaged it may help to spray contact spray (for electronics) if there is an opening somewhere. Then turn it a few times. If the problem is dirt you should get rid of it this way...

If not, as you said : garbage

 

I'm chalking them up to garbage...oh well. Thanks for the input

 

There are 2 types of pots Linear taper B type. Then theres Log taper A type (these are used in audio aps). You should be using linear in your aplication.

 

There are 2 types of pots Linear taper B type. Then theres Log taper A type (these are used in audio aps). You should be using linear in your aplication.

of course! how did I forget that, modern "logarithmic" pots which you can buy in electronics shops tend not to be acctually logarithmic, they have a low resistance bit of track and them a high resistance section (or vice versa) usually they join together at the same resistance value.
How far around the turn does the erratic jump in resistance occur? If it's right at the start I'd say dodgy pot, if it's further along, I'd say crappy modern "logarithmic" pots.

 

the resistance goes up just fine until about 100 ohms, then it jumps drastically up to about 1.5 kohm, then increases fine after that.

Doesn't really sound like a logarithmic potentiometer.

 

It might be a custom Clarostat 70 series Mod Pot designed to go from 1k to 5k with the first 15° of rotation devoted to actuating a switch. If so, the 50% tap would be at 3k. Obviously, no switch was installed and it's too weird to be useful to you.
http://sensing.honeywell.com/index.cfm/ci_id/140805/la_id/1/document/1/re_id/0

 

It might be a custom Clarostat 70 series Mod Pot designed to go from 1k to 5k with the first 15° of rotation devoted to actuating a switch. If so, the 50% tap would be at 3k. Obviously, no switch was installed and it's too weird to be useful to you.
http://sensing.honeywell.com/index.cfm/ci_id/140805/la_id/1/document/1/re_id/0

That's interesting. What kind of application would that be useful in? And how would rotating the pot 15° actuate a switch? Not sure I'll ever do that, but I like learning.

 

Is it possible that these are fader pots? Like from a car radio that lets you change sound level from front to back or left to right.

 

That's interesting. What kind of application would that be useful in? And how would rotating the pot 15° actuate a switch? Not sure I'll ever do that, but I like learning.

That series is configurable into multiple pots on the same shaft, switches, with and without center detents, etc. Yours doesn't have a switch but may have been intended for assembly with one. The common type clicks on in the first 15° of CW rotation from the CCW stop like old car radios that used the volume control as the power switch. If your pot was intended for such an assembly, the resistance in the first 15° of rotation would be ignorable since there'd be no power to the circuit until the switch was closed.

 

@KJ6EAD If it was a model like you said, could it have the described effect? Increasing slowly, then jumps to 1.5k then continues to increase?

 

In case anyone's interested, I played around a bit more with those potentiometers (I really don't enjoy throwing stuff away, especially when I don't understand it ). When I was using them previously, I was making a voltage divider with an external resistor and trying to adjust the voltage that way. Once I actually drew up a schematic symbol for the pot and looked closer, I realized that I could essentially just make a voltage divider out of itself (sorta). So, I connected it like the attached image, and as it turns out, you can start the pot in the middle (wiper at the tap) and turn it either direction to sweep from 0 volts to whatever your input is (almost perfectly linear).

Not sure what I'll use them for yet, but it looks like I'll be holding on to these for a little while longer

 

Not sure what I'll use them for yet, but it looks like I'll be holding on to these for a little while longer

Ok, but can you explain where the "jumping" effect from your first post does come from? 0 to 100 linear then it jumps to 1.5k then linear again?

 

Ok, but can you explain where the "jumping" effect from your first post does come from? 0 to 100 linear then it jumps to 1.5k then linear again?

I don't understand it either. It doesn't jump like I described when wired the way I showed in the image. Your guess is as good as mine...

 

I don't understand it either. It doesn't jump like I described when wired the way I showed in the image. Your guess is as good as mine...

If you were really curious, you could open one of them and find out. but after that it may be really garbage...

 

Not curious enough to destroy them, especially after I figured out a way to use them productively