ICP accelerometer

[M.Charbonneau]

Hi everyone,

I have an ICP accelerometer who need a constant current from 2 to 20mA. I decided to built my own conditionner with a current source of 24V/1A, a diode and a capacitor.

I want to know what's could happen if I provide a current of 1A directly at the accelerometer, whithout decrease it to 4mA. Will it destroys the captor?

Some people told me that the current intensity has no importance because the captor will utilise only the current he needs...

What do you think about that?

Thanks in advance,

[Henrik_Volkers · Accepted solution]

The diode you mentioned should be a current limiting diode.

And YES I assume that the ICP sensor get damaged (or migth change it's parameters) if sourced with a current higher than the SPEC.

And NO the capacitor usually will not be destroyed ( assuming a appropriate voltage rating)

You can see the ICP sensor as a voltage (zehner) diode, whos voltage is about 10V +- <sensitivity>*a . So the (max rated) power consumption of the ICP P_icp=10V*20mA = 200mW 

Now replace the 20mA with 1A and think of a 10W ligth bulb 😄

So if you want to build your own conditioner, but mix up some basic stuff, it couldn't be for a proffessional use. (buy will be cheaper than building with your experience) 

You can use the power supply, but need to add a current limiting circuit (current diode, or jFET with resistor). The bias cutting capacitor should have a 35V rating .....

Spoiler

Why I'm so shure, that you already 'tested' with 1A and now hope that the capacitor has blown instead of the sensor?

[M.Charbonneau]

Thanks for your advices !

... Effectively, I've tried to power the ICP sensor with the 1A current (divided by 3 because it's a 3 dimensional accelerometer). Now, I'm in trouble because professionnal in electricity said to me that it doesn't matter to give 1A to a sensor... Because the sensor will pick up only what he needs.

Finally, it's not true, and I hope that I didn't damage the sensor...

I'm ok with the voltage rating and I'll buy a real current limiting diode 😉

Have a good day!

[Henrik_Volkers]

So my guess is that one of the three migth be gone. (the one with the lowest bias got most of the current)

You can do a quick sanity test: Add a 4k7R (4700Ohm resistor) in series to your 24V power supply  and measure the voltage (with a DMM) at the sensor output. It should give a reading of 10V-12V (bias, see spec)

That resistor act as a current limiter (not constant : U=R*I)  so a current between ~5mA (in case of shortcut in the sensor)  and ~2mA (at 12V bias , sensor looks OK but who knows??)

THIS IS FOR A QUICK CHECK, FOR SERIOUS MEASUREMNTS A CONSTANT CURRENT SOURCE IS NEEDED.

Next check would be a recalibration (or at least: using a quite constant vibrating source, rotate the sensor mounting and check for consistent readings, so remount every axis at least twice). This of course with a signal conditioner 😉

[M.Charbonneau]

Ok now I'm sure, the sensor is unusuable and I'm really angry to have been misinformed.

Thanks again for your answers !