Hi, i am looking to drive a LVDT (Transformer) with a 20kHz square wave (between two pins - A high B low then A low B high). Is it possible and how would i do this ?
Interestingway to fry your device. Use a H-Bridge to drive the inductance 
zwieblum:
Interestingway to fry your device. Use a H-Bridge to drive the inductance
Not necessarily.
A long time ago that technic was used to create 220 volt from the 12 volt car system using a 12 / 220 volt transformer and a few transistors. A free running oscillator made it work.
Well i tested the LVDT with my Stero at 20kHz and it pulled only 15mA so i figured the Arduino could provide that ?
The problem is not if the Arduino can provide it, the problem is what happens the other way round, can the arduino handle overvoltage?
@OP. How did You measure to read those 15mA? Use the advice given in #1! Arduinos are not useful as power sources.
@zwieblum I agree
Post a link to the product page or data sheet for the LVDT.
Thanks for the advice, the current and voltage was measured using a multimeter so not sure if reliable. However the Resistance of the coil is 32 Ohms so i trusted it. I don´t have a datasheet unfortunately. Why wouldn´t it be a good idea if you stay within the Current limits of the arduino?
'cause you can't.
Measuring square wave currents or voltages calls for speciell instruments. Standard, ordinary multimeters will show "something". Oscilloscopes would be the best.
If the resistance is 32 Ohm or more and the controller output is 5 volt You ought to be safe.
You should not draw from or sink over about 20 mA through an Arduino pin. That means for a 5V output, the minimum resistance to ground is 250 Ohms.
The inductance of the LVDT will reduce that for an AC signal, but you don't know the inductance.
Connecting a 32 Ohm LVDT between two outputs as you propose will probably destroy them, but perhaps not immediately.
Oops. I lost a decimal.
For a Arduino Uno (or other board from the AVR family), this library creates those signals: teckel12 / Arduino toneAC / wiki / Home — Bitbucket.
The Arduino Uno can not drive that directly !
Since the Arduino Uno can handle a lot on its pins, it might work for some time. But you should definitely not do that (as jremington already wrote). You need a few resistors and 4 transistors to drive the LVDT, or a H-Bridge.
When you buy a H-Bridge, I suggest to use one that can handle a frequency up to 1MHz, since a small transformer might perform best at a high frequency. Or is the 20kHz defined as the frequency that it must work with. If that is so, perhaps it should be a sine wave ? Are you sure that a square wave is allowed ?
Hi,
thanks again for the input, H-Bridge it is. As for what is allowed and so on, i have no clue or datasheet. And later on i would like to use some coils that i already wound myself so all bets are off. But thanks to you i should now have the right direction to work with.
I once built a prototype for a large European car manufatorer. That device would tell the position of the brake pedal, supposed to be used in an ABS system.
The higher frequency the higher output. I think I used some 100 - 500 kHz. The coil was some simple winded coil.
Get some current limited current driver, can be a pwr supply, a resistor and a transistor.
Hi again, i implemented the H-bridge design an it´s working :). I was however wondering how you would rectify such a signal. I thought about using something like this precision rectifier Single Supply Precision Rectifier Design and Optimization - YouTube but i am a little puzzled as you don´t have ac in the sense that one wire ist grounded and one switches between negative and positive. I tried to gather some information and it seems using a second (not connected) power source with the imput floating is not a good idea. Maybe you could enlighten me once again. Cheers
Oh special thanks Koepel, that library is really nice
Edit: Would battery powering the rectifier circuit be a good idea ? then pulling one side of the signal to the battery ground via resistor ?
To measure the AC output signal, you can use a voltage divider to bring up the average voltage so that it swings about 2.5V (for a 5V Arduino, for example). As shown here.
I think you misunderstood me. I don´t have a signal that ist switching between + and - but rather two pins that are switching A:+5v B:0v to A:0v B:+5v thus giving "AC". I was wondering if it is possible to use a precision rectifier (using an opamp) which uses the same ground. I guess not but a lot of people here understand more about electronics than me. The signal of course doesn´t stay at 5v. it drops depending on probe position (otherwise there would be no need to measure it :P)
Edit: to explain: So far i measured the ac with my multimeter wich is fine, but i would like to have additional resolution. So i thought about using an analog to digital converter which can measure differential dc input.
Please explain more clearly what you want to do.
As I understand it, the signal you want to measure is the output of the LVDT, which is magnetically coupled to the input signal and is therefore completely isolated from and floating with respect to the Arduino (signal generator) ground. If that is the case, the solution linked in reply #16 is fine.
A precision rectifier is also an option, and again, the ground reference is not an issue.
Hi, i have different types of lvdts, some of the conventional type which i wound myself (3 coils) -> yes, not connected so easy rectification. Some are only 2 coils with common center tab like here http://alliancesensors.com/sites/default/files/images/ge08.png. Here you apply ac over the outer tabs -> input and output connected
