trimpot reliability

Thread Starter

shortbus

Joined Sep 30, 2009
6,890
On a PCB, is the use of a high quality trimpot acceptable for use as a voltage divider to set reference voltages for comparators? By high quality I mean like one made by Bourns, not one from China. It will be a set and forget type of a thing, to get a voltage level not available by a normal voltage divider. I've seen them used in other things, but have heard from members here that they will die after a time in a circuit. If not being used like a volume or tone control that is constantly being changed, are they reliable?
 

Brevor

Joined Apr 9, 2011
297
Yes they are reliable, They are not made to be constantly adjusted, but for once in a while use they are fine. Bourns lists in their datasheet the number of operations they should be good for.
 

dl324

Joined Mar 30, 2015
8,769
IMO, a trimpot will be more reliable in an application where it isn't changed frequently and will fail sooner if they're being changed "constantly". Bourns specs their trimpots for 20-200 cycles...
 

joeyd999

Joined Jun 6, 2011
4,231
I haven't use trimpots in 15 years -- because they are unreliable. As mechanical devices, they are subject to instability due to wear and vibration. The wiper contact is susceptible to the effects of humidity and atmospheric contamination. They generally have awful temperature coefficients. The "precision" multi-turn versions are the worst: the mechanical linkage between the adjustment screw and the wiper is susceptible to backlash errors -- both during adjustment and afterwards due to vibration.

These days I do *everything* digitally. Transducer -> A/D -> signal processing. Any calibration or scaling is done with math -- which is completely immune to the above stated effects.

If I require precision analog, I'll use the best voltage reference and low Vos opamps I can afford and 0.1% precision fixed resistors (which are relatively cheap these days).
 

#12

Joined Nov 30, 2010
18,076
I have spent decades in the analog industry where 10 turn trimpots were used for calibration. As a confession, most of them were protected by hermetically sealed containers as soon as I was done calibrating them. Under those conditions, they were good enough for industrial metering, military designs, and laser accuracy adjustments.

ps, if you whack them a little bit with the butt of your tiny screwdriver, the backlash will settle.
 

AnalogKid

Joined Aug 1, 2013
8,081
A quality, fully sealed part should serve well. Also, Bourns and others make trimpots with extra resistors in series for scaled voltage dividers. Excellent temperature tracking between the pot element and the fixed resistors. Not many values available, but if they have one that meets your needs, the temp tracking is way better than separate parts.

ak
 
Last edited:

Thread Starter

shortbus

Joined Sep 30, 2009
6,890
Thanks guys. I know many "old time" transistor circuits used trimpots. After setting the output, they usually had a 'dot' of colored varnish or enamel put on the adjustment screw to body, to lock the position. Using a trimpot will make my life easier.
 

cmartinez

Joined Jan 17, 2007
6,380
I've been using trimpots with some success for a while. And although they are susceptible to temperature variations, the fact that the whole trimpot is a single piece cancels that error when it¡s used as a voltage divider (but not when it's used as a rheostat)
But I think everyone here agrees that they're designed to be used infrequently. When I plan a circuit of which several others need to be made, I normally make a "master circuit" that includes trimpots, and then I measure their final setting. Then I make the rest of my circuits using precision resistors matching that setting. That not only makes for a more reliable circuit, but also a cheaper one, since trimpots are (for my taste) expesive devices.
 

#12

Joined Nov 30, 2010
18,076
My experience in production line QC confirms that the same settings happen over and over when the circuit would have been equally served with a little more time in the design stage. It is only when you are adjusting un-predictables like op-amp offset that the settings are always different.
 

cmartinez

Joined Jan 17, 2007
6,380
My experience in production line QC confirms that the same settings happen over and over when the circuit would have been equally served with a little more time in the design stage. It is only when you are adjusting un-predictables like op-amp offset that the settings are always different.
Very true... I once bought a very expensive trimpot (by Vishay, about $20 dlls) for adjusting the gain of a high precision Instrumentation Amplifier. It was expensive because it was designed with an extremely small temp coefficient.... guess it all boils down to how much precision you want in your application
 

PackratKing

Joined Jul 13, 2008
843
Many times past, I have cobbled up a trimpot, with a specific resistor in parallel, so to dial in a particular resistance value not available in standard... Once I tune that circuit in, I coat the slider / resistance band in a drop of epoxy so no further adjustment is possible...
 

tindel

Joined Sep 16, 2012
659
When I worked in aerospace trim pots were outlawed. This makes sense due to the very large vibration and shock loads experienced during launch, thruster, and pyro firings. Not being able to trim a variable made the worst-case conditions tough in some applications even with 0.1% resistors.
 

Thread Starter

shortbus

Joined Sep 30, 2009
6,890
Thanks every one for your thoughts on this. The trimpots will only be used to set voltage reference levels for some 339 comparators. The resistor values I was coming up with aren't available so thats why the use of trimpots. Once those levels are set they won't need to be changed.
 

strantor

Joined Oct 3, 2010
5,025
I had my fill of trimpots working in the wire & cable plant. All the equipment was vintage and full of trimpots. The DC drives, the analog PID tension control boards, the eddy current clutch controllers. There were enough trimpot issues to keep food on the table indefinitely. The worst were the PID tension controls, so sensitive. I would get called out, "the dancer (tension feedback arm) is acting up again..." Go tweak those PID trimpots until you get it settled down, and move on to the next. You'll be back tomorrow, or next month, or not for a long time, but for sure you'll be back at some point. The common theory was that the pot developed a "bad spot" if it sat too long in one spot. Most times, turning the pot a quarter turn and then back to the original position would straighten it out; it's as if they were getting a carbon buildup between the wiper and the [...whatever the other thing is called...]. But it's a sealed pot, I didn't/don't get it.
BTW these were "high quality" Bourns trimpots mostly.
 

#12

Joined Nov 30, 2010
18,076
Another alternative is nichrome wire, cut to length.
I've done that. :oops:
I cut 1&1/2 turns out of a 200 ohm, wire wound pot when I needed 1.3 ohms in a battery eliminator circuit. What the heck. It worked. :D
 

strantor

Joined Oct 3, 2010
5,025
Actually, nichrome's temp coefficient is so high that measuring its resistance is often used as a way of controlling its temperature in some sophisticated hot wire cutting machines.
Come to think of it, I believe I did read that before. IIRC Roman Black (MrRB or The_RB, or whatever his avatar was) posted some details about his hot wire foam cutter on this forum some years ago. I believe that's how he controlled the temp. He powered it with pulsed high current DC and read the resistance in between pulses to determine the pulse width.





EDIT:
No!, I found that old thread, and though RB participated in that discussion, it wasn't him, it was Joey (already participating in this thread).
I can be done, and I've done it myself.

I've used a PIC micro to drive a platinum wire to glowing red at an adjustable temperature within 1 degree C.

The PWM drove the power (4-6A) into the wire (freq. about 20kHz, on-time ~4uS). During the "off" time (~46uS), a precision current source engaged to measure the resistance. That was fed back into a *very* fast PID to compute a new PWM value for the next cycle.

Worked like a charm, and off 4 'AA' NiCads!
 
Last edited:
Top