Do we have more than 25 turns trimmers ?
What is the resolution, one can really expect with those tiny screw 25 turns ? I would say 1/8 turn
Are they all the same among popular brands ? In other word is there a best about accuracy setting and stability.
My problem: A +/- 2mV offset that I wish to trim at 1uV accuracy.
What is the resolution, one can really expect with those tiny screw 25 turns ? I would say 1/8 turn
Are they all the same among popular brands ? In other word is there a best about accuracy setting and stability.
My problem: A +/- 2mV offset that I wish to trim at 1uV accuracy.
For trimmers that is pretty much the maximum (well, 26 actually.)
Considerably better than that - Bourns has very good datasheets and application notes that go into details and specifications on this.
Mostly, but Bourns is probably the most popular and they all have differing grades of products.
If you need the full 2mv range of adjustment the best way is to just use two trimmers in series - one larger value for course adjustment and a smaller one for fine adjustment (for example, a 10K for course and a 250R or even smaller for the fine.)
A standard trick is to invent or create a combined circuit containing two potentiometers, one named "COARSE" and the other named "FINE".
- Set the FINE knob to its 50% point.
- Dial the COARSE knob until you get very very close to the desired result.
- Dial the FINE knob to slowly creep up to the desired result.
- Victory. Add a drop of nailpolish or mechanic's ThreadLocker to fix each knob in position permanently.
One consideration though - trying to get a uV or better of resolution on the lower end will bring in serious considerations of the temperature coefficient of the parts - the change over temperature of the larger value trimmer can easily swamp the adjustment of the smaller. Keeping the power dispersion of both parts small as well as keeping them as close to the same temperature (and with the same tempco) as possible is highly desirable.
Vishay makes a line of low tempco parts but they are a little pricey but, as always, you pretty much get what you pay for.
Vishay makes a line of low tempco parts but they are a little pricey but, as always, you pretty much get what you pay for.
While not directly addressing what you are doing, the Keithley "Low Level measurements Handbook" hits on many of the issues that need to be considered for that level of adjustment, and is available for free here:
https://www.tek.com/en/documents/pr...low-level-measurements-handbook---7th-edition
https://www.tek.com/en/documents/pr...low-level-measurements-handbook---7th-edition
There is another consideration that has not been mentioned so far.
The circuit you are trying to trim to the micro-volt level may have drift well in excess of tens of micro-volts. Drift occurs when power supply voltages change and when each part's temperature changes (even slightly). If the circuit has not been designed to be micro-volt stable, it is likely that it is not.
You can verify this by setting the offset and recording the value. Then leave the unit on, and record the offset value regularly over a period of time. The time intervals can range from every hour to every day; the intervals do not have to be exactly the same. The changes in the offset will reveal the stability of the circuit.
A circuit that is not inherently micro-volt stable can be controlled, but it will require active offset trim (via an opamp providing DC feedback) rather than a trim pot. An active trim continuously adjusts the output to have near zero offset, compensating for changes in the circuit being controlled.
-Morty
The circuit you are trying to trim to the micro-volt level may have drift well in excess of tens of micro-volts. Drift occurs when power supply voltages change and when each part's temperature changes (even slightly). If the circuit has not been designed to be micro-volt stable, it is likely that it is not.
You can verify this by setting the offset and recording the value. Then leave the unit on, and record the offset value regularly over a period of time. The time intervals can range from every hour to every day; the intervals do not have to be exactly the same. The changes in the offset will reveal the stability of the circuit.
A circuit that is not inherently micro-volt stable can be controlled, but it will require active offset trim (via an opamp providing DC feedback) rather than a trim pot. An active trim continuously adjusts the output to have near zero offset, compensating for changes in the circuit being controlled.
-Morty
@mtarr is exactly correct - I was assuming the OP was already trying to work on the reference side of the circuit but that was certainly never stated. uV levels of accuracy are very easy to get theoretically but very difficult to achieve in practice. The active trim @mtarr mentions is, in practice, a very complicated and difficult thing to design and have work correctly.
Commutating autozero opamps achieve one-microvolt input offset with no trimming. OPA2388 (USD 3.66 / 0.25 uV / GBWP= 10 MHz) , TLV2387 (USD 1.93 / 1.0 uV / GBWP= 5.7 MHz) , TSZ122 (USD 1.54 / 1.0 uV / GBWP= 0.4 MHz) are a small sampling of what's available and on the shelf today.
Yes, and there are and have been several options. The point was not that you couldn't adjust something to 1 uV, it was that it is very difficult to get it to stay there. As you correctly point out the components are fairly available individually; it's when you start putting together the required system that will deliver that level of accuracy over time that the error budget becomes much more difficult to manage. Again, it can certainly be done on a DIY level but it requires the consideration of quite a few factors, many of which are not always obvious at first. Which all circles back around to the point that you (I believe) were aiming for originally: the OP is not going to achieve any kind of uV accuracy with a $3 trimmer, however many turns that it has.
Thanks for all your infos.
The 1uV stability of the other stuff is another challenge. All being differential and balanced.
A microphone preamplifier using Demrow-Cohen topology ( redone by Samuel Groner under the name Montegeneroso ) where I want no capacitor but the two film capacitors for phantom power isolation.
I build four of such preamps long ago, so quite well aware of the issues.
I am more, now; On the idea of using coarse trimming by commutating resistors DAC like and fine tuning by a trimpot.
The 1uV stability of the other stuff is another challenge. All being differential and balanced.
A microphone preamplifier using Demrow-Cohen topology ( redone by Samuel Groner under the name Montegeneroso ) where I want no capacitor but the two film capacitors for phantom power isolation.
I build four of such preamps long ago, so quite well aware of the issues.
I am more, now; On the idea of using coarse trimming by commutating resistors DAC like and fine tuning by a trimpot.