I would like to use an analog meter movement to display a "window" reading. For example if I want to monitor the output of a 120V isolation transformer whose voltage I would like to have within +/- 5% of 120V , a meter reading 0-130V would not provide the level of resolution that I really care about. I would like to have the meter read within a window of 110V to 130V, with the desired 120V right in the middle, so that you can tell from a quick glance if the needle isn't pointing straight up, there is a problem. And further, you will be able to tell quickly if it's off by 2V or 10V, without having to stare at the meter through a magnifying glass and calculate minor divisions on the scale.

I'm guessing a bit of circuitry will be required in excess of the typical trim pot and series resistor. Maybe some Zener diodes, maybe an op-amp or two. What will it take? Does anyone have a canned circuit for this? As usual, I've already googled before asking, and I'm googling while I wait for answers, but as of yet I haven't found anything.

 

I would like to use an analog meter movement to display a "window" reading. For example if I want to monitor the output of a 120V isolation transformer whose voltage I would like to have within +/- 5% of 120V , a meter reading 0-130V would not provide the level of resolution that I really care about. I would like to have the meter read within a window of 110V to 130V, with the desired 120V right in the middle, so that you can tell from a quick glance if the needle isn't pointing straight up, there is a problem. And further, you will be able to tell quickly if it's off by 2V or 10V, without having to stare at the meter through a magnifying glass and calculate minor divisions on the scale.

I'm guessing a bit of circuitry will be required in excess of the typical trim pot and series resistor. Maybe some Zener diodes, maybe an op-amp or two. What will it take? Does anyone have a canned circuit for this? As usual, I've already googled before asking, and I'm googling while I wait for answers, but as of yet I haven't found anything.

This project was in every other hobby magazine around the 70s, mostly for car batteries because anything below about 10V means calling out a recovery truck.

Normally you'd add a Zener in series with the meter and multiplier resistor, I've seen Zeners as high as 180V - but its unlikely to be anywhere near the accuracy you want.

A TL431 is adjustable and pretty stable, but only goes up to about 37V, and it needs at least 1mA standing bias. Your best bet is bridge the movement between the TL431 derived Vref and a sample of your voltage via a resistor divider.

 

Stab in the dark here; rectify it to DC, use a voltage divider to get it to a workable range, then use opamps (or a purpose built IC) to compare to a reference voltage source.

 

Do you have a meter selected for this readout?
Or is it arbitrary at this point?

 

This project was in every other hobby magazine around the 70s, mostly for car batteries because anything below about 10V means calling out a recovery truck.

Normally you'd add a Zener in series with the meter and multiplier resistor, I've seen Zeners as high as 180V - but its unlikely to be anywhere near the accuracy you want.

A TL431 is adjustable and pretty stable, but only goes up to about 37V, and it needs at least 1mA standing bias. Your best bet is bridge the movement between the TL431 derived Vref and a sample of your voltage via a resistor divider.

Thanks for that idea. Googling "car battery voltage analog meter circuit" I found this. It's a start; something I can investigate further.

EDIT: another video from the same guy.

 

Do you have a meter selected for this readout?
Or is it arbitrary at this point?

I've made an offer on these, but I'm thinking that the physically centered needle will complicate things. I might retract my offer if possible. Consider it arbitrary.


EDIT: I can't retract my offer. Hopefully the seller declines. I might be stuck with these

 

Your best bet is bridge the movement between the TL431 derived Vref and a sample of your voltage via a resistor divider.

Something like this?

With the components shown there's about 27V across U1.

 

Used to be referred to as an expanded scale meter (I think).

http://www.thebackshed.com/windmill/articles/ExpandedScaleVoltMeter.asp

http://www.oeiss.org/docs/691_Voltmeters.pdf

http://www.qsl.net/kh6grt/page4/expscale/expscale.htm

Regards, Dana.

 

Stab in the dark here; rectify it to DC, use a voltage divider to get it to a workable range, then use opamps (or a purpose built IC) to compare to a reference voltage source.

Here's my take on your idea, strictly conceptual - all of the values would need to be re-evaluated, but the basic concept seems to be working.

This sim uses a step parameter to run AC input from 110-130VAC in 5V increments, and you can see that the DC output of the instrumentation amp moves from 2-10V in 2V increments.

 

Stab in the dark here; rectify it to DC, use a voltage divider to get it to a workable range, then use opamps (or a purpose built IC) to compare to a reference voltage source.

And here's one more re-interpretation, using a transformer to drop most of the voltage. I have no idea if performance would be better or worse, but it makes component selection much simpler... and is probably better in terms of this forum's terms of service as well...