moving coil meter, biasing to bipolar

[timeout]

Hi All,

I have a school demonstration project in which a slowly varying bipolar sine AC voltage (~0.1 Hz) is putting current through a coil.
in order to demonstrate that a magnetic field is generated, two compasses are in close proximity and show the alternation of the polarity by constantly reversing.

In order so illustratively show that the current flow is actually reversed I would like to use a moving coil meter with a custom "scale" showing N and S in red and green. For this the resting position of the needle has to made at half scale - i hoped for the screw adjust to be able to, but after checking three different meters, i came to the conclusion that this really only sets +-10% FS or so - which is reasonable i think.
True bipolar meters (as used on some Kepco BOP models) are quite hard to find around....

The meter i would like to use is 60 mV to full scale. What would be a good strategy of offsetting the meter electronically ?
using my function gen (sine, 100mHz using 30mV offset) to test it did exactly what i want in the application...

Did anybody come across this need ? is it weird ?
Would you use an AC coupled (10uF or so) op amp and bias it accordingly to half scale ? is this over-engineered ?

Thanks in advance!

[iMo]

For example you may bias the coil meter (0..60mV) with two resistors and 1.5V AA battery.
With the generator running from -30mV to +30mV your coil meter will run from 0mV..60mV.
The zero output of the generator will be in the middle of the coil meter (at 30mV).
See the below schematics. You may use 22ohm and an 1k5 potentiometer (10turn) to adjust the middle perfectly.

[BNElecEng]

Do you know much current is flowing at the peak of the sinewave?
I found some cheap moving coil ammeters online which are centred at zero, so they will show the change in direction quite nicely. Here is an example, search for the term"DC 0-250uA Rectangle Analog Panel Ammeter Gauge Amperemeter Class 2.5"
 

[capt bullshot]

These meters measure current, not voltage, though they're often specified for voltages - then just use Ohm's law.
Take this as a starting point: put a current source that delivers half scale current in parallel with the meter. This current and your measured current adds up within the meter and brings you the desired result. No need for fancy buffering and AC coupling.
How to implement this in practice, depends on your circuit.

[Paul Price]

IMO, with your circuit,  when there is -30mV Vin the meter will show 60mV, pegging FS, and with +30 mV Vin it will show zero, and only with 0-V Vin it will show 30 mV. or half full-scale. if you use  a meter of 60mV FS.

That's misrepresenting the generator voltage output polarity! It also feeds current into the generator. Also the meter open circuit will show 0 and only closed-circuit (connected) will it show midscale as zero.

Ideally, what you want is 0-V Vin showing 30mV or half full-scale(60mV), and +30 mv Vin to show 60mV (full scale) and -30mV to show 0V on the meter, and showing 0-V, and do this whether or not the meter input connections to the generator are open or connected to the generator.

The problem with your circuit is that you get a 30mV reading (1/2 FS) only if the input connections to your meter are shorted.

However, it is easy enough to do this with a dual op-amp like a LM358 and a 9V battery and a slight modification of your circuit. It would need a high-resistance pull-down resistor from the + input of one of the op-amps to keep the very sensitive input from floating all over the place and allow a zero Vin reading correct. Hint: one of the op-amps forms a +/- 4.5V supply with a virtual zero-voltage ground as one side of the meter input.

[Gregg]

Zero center analog meters are still available and not too expensive.  ebay item 172557379144 for example

[C]

Zero adjust screws are designed for a small adjustment range.

If you remove the meter face. You will probably find that adjust screw moves a lever a small distance.

Often that lever will allow a center adjustment.

C

[Gyro]

To convert an existing meter to center zero, you need to remove the movement from the case and adjust the lever (smaller than the top one) that holds the bottom hairspring. The fine adjustment at the top then works as normal (to fine adjust the center position).

You obviously need to very careful doing this, it's very delicate. The movement fixing screws are normally underneath the scale plate so this needs to be removed, leaving the pointer very exposed. You need to do it on a brightly lit clean surface, clear of any magnetic particles (swarf). I'd only recommend doing it if you have a particular existing movement in mind, a large scale demonstration one for instance.

Don't forget to set the top, fine adjust lever to mid travel before you adjust the bottom one.

[timeout]

Hi All,

Thank you very much for your detailed replies.

I checked the meter which i have again, and i unfortunately don't see any way of mecahnically setting the bias point appropriately. (@Gyro)

buying a center zero meter is an option, and thanks, i found some offers online, unfortunately all have loooong shipping times to my place.

could you lead me to a circuit which would mimic the behavior of a center zero meter - it does not have to be adequate for precision instrumentation usage, i understand that using a one designed for the purpose in the first place, would be better.

i have +/- 15V rails at hand as well as a low frequency current (~ 100mHz, 200mA) reversing its polarity periodically.
This is already done by a mighty OPA547 - i just cant really figure how to synchronize the meter, in a center zero fashion, with the current.

best regards

[Gyro]

I checked the meter which i have again, and i unfortunately don't see any way of mecahnically setting the bias point appropriately. (@Gyro)

It's the bit that the end of the hairsping is attached to, it normally rotates under a friction wave washer around the pivot - in just the same way as the top one, it just doesn't have a lever. Really best not to mess with it if you're not sure what you're doing.

If you have an the luxury of op-amp and bipolar supplies to hand, then it's easy to do. Simply configure the opamp in a standard inverting amp configuration and bias the non-inverting input to a pot between the +ve and -ve supplies rather than 0V. you'll want to play around a bit with series resistors at each end of the pot so that the adjustment isn't overly coarse.

[timeout]

Thanks agin!

I breadboarded together a small circuit using an NE5532 op amp.

It works exactly as planned. I can set an arbitrary bias point using a multiturn trimmer, and also amplify the AC component of the signal.

Are there any obvious pitfalls in the circuit ?
I routed a PCB which i would otherwise go ahead and etch

R1 and R2 may have different values, especially R2 would have a much lower value, in order to make the bias point adjustments using R4 less harsh.
Using R3 i can amplify the slowly varying AC to taste.
I included C4 just as a matter of course to prevent any oscillations, it may have a larger value, as the meter is only usable  to max ~10 Hz. Is this worth anyway ?
I did not see any other way of "AC coupling" 300mHz without using enormous capacitor values...C3 - may be smaller, but i guess it needs to be several µF anyway...

Regards and Thanks!