0
\$\begingroup\$

I have a device that outputs 12.4 +/- 0.1 mV when it is inactive and 12.9 +/- 0.1 mV while it is collecting data. I would like to trigger another device that expects 3.3V high and 0V low for it's start and stop conditions respectively. Because of the variability of the first device it seems that amplification (of around 660x) would not work. How can I set up a circuit to detect the very small change in the first device's output and activate a 3.3V source for the second device? A comparator seems like the way to go but I'm not sure how to minimize the size of the circuit.

edit: My two devices are a Force Platform system and an Inertial Measurement Unit motion tracking system. I am trying to use the signal from one to trigger the other so that they start data collection synchronously. The IMU is capable of sending and receiving triggers, but the FP are not, so my idea was to split the digital signal of the FP and use the voltage change on one of the output's pins as the trigger. Both are black boxes for me, the companies that produce them will not give me further information as they would prefer to sell me an interface device that does this for me, however, I'm on a graduate student budget...

Thank you! -B

\$\endgroup\$
3
  • 3
    \$\begingroup\$ Your title says a 500 uV change but your description indicates a 500 mV change. Which is correct? You say you're not sure how to minimize the size of the circuit. What do you mean by that? \$\endgroup\$
    – Glenn W9IQ
    Jun 23, 2017 at 20:01
  • 4
    \$\begingroup\$ That seems very unusual that you have to sense such a small voltage change - what is this device? Describe the problem in more detail without assuming a solution. \$\endgroup\$ Jun 23, 2017 at 20:03
  • 2
    \$\begingroup\$ @Glenn It would appear that the two voltages are in millivolts so that the change is indeed 500 uV. In that case, a single comparator with a reference voltage of 12.65 mv (halfway between the 2 possible inputs) would be the way to go. However, as Kevin says, the voltages seem unusual: why are they so supposedly so accurate (less than 1%) if they are just indicating status, and why are they so close to each other. More information could lead to a better solution. \$\endgroup\$
    – Barry
    Jun 23, 2017 at 20:47

3 Answers 3

2
\$\begingroup\$

I am not aware of such good comparators, but there are sigma-delta ADC on market that have crazy resolution. So you can use a differential preamplifier with reference 12.7mV and some gain, then a differential ADC and a microcontroller.

But the question is so weird, that i would guess you do something wrong from the beginning. What is that device with it's output? Is it some kind of a sensor?

\$\endgroup\$
1
  • \$\begingroup\$ My two devices are a Force Platform system and an Inertial Measurement Unit motion tracking system. I am trying to use the signal from one to trigger the other to start collecting data so that they start data collection synchronously. The IMU is capable of sending and receiving triggers, but the FP are not, so my idea was to split the digital signal of the FP and use the voltage change on one of the output's pins as the trigger. \$\endgroup\$ Jun 24, 2017 at 19:25
1
\$\begingroup\$

Comparators with the input stability you need are not what I'd call common (that is, I don't know of any), but there are lots of amplifiers you could use which are readily available. You should Google on "zero drift operational amplifier". Analog Devices, for instance, provides a variety of amps with input offsets in the uV range with temperature coefficients to match. Using one of these as a differential amplifier with a gain of 200 and an offset of 12.4 mV would give you a (nominal) 0 to 100 mV output, and pretty much any comparator can handle that successfully.

\$\endgroup\$
0
-1
\$\begingroup\$

You want to detect which state a voltage is in. One state it will be 12.3 to 12.5 V, the other 12.8 to 13.0 V.

The obvious answer is a comparator with a 12.65 V reference. That gives you 150 mV margin to either case. I can't even guess where your 500 µV figure came from.

Added

I didn't notice earlier that your voltages are in mV, not V. The numbers are the same, except divide them by 1000. 500 µV still doesn't make sense since now the margin between the two bands is 150 µV, instead of mV.

With such small signals, it is better to amplify them first, then do the threshold detection. However, the amplifier needs to have very low offset, or at least good stability if you're willing to do a initial calibration. After the gain stage, you can use a ordinary comparator with a mW offset.

\$\endgroup\$
2
  • 4
    \$\begingroup\$ His quoted voltages, if you ready carefully, are in millivolts, not volts. That is why the difference is 500 uV, not 500 mV. \$\endgroup\$
    – Barry
    Jun 23, 2017 at 20:48
  • \$\begingroup\$ @Barry: Oops, I missed the mV. Fixed. However, 500 uV still makes no sense. The band between the two states is only 150 uV. \$\endgroup\$ Jun 24, 2017 at 20:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.