0
\$\begingroup\$

I'm currently involved in a project with the aim of measuring the tremor of mice limbs (displacements of about 100 um). The sensor my group has chosen is TCRT5000 (Vishay), but we are facing several problems:

  1. We don't know which is the best configuration to convert the collector current (output) into a voltage signal that can be firstly conditioned and then sent to a microcontroller (a PIC for instance).

  2. We don't know which resistance we have to consider in order to calculate the thermal noise and thus the sensor's resolution.

Thanks in advance. Below you have the link to the datasheet.

https://www.vishay.com/docs/83760/tcrt5000.pdf

\$\endgroup\$
  • 7
    \$\begingroup\$ Fix the title please. No need to shout... \$\endgroup\$ – Chupacabras Dec 6 '19 at 16:54
  • \$\begingroup\$ I think this sensor will give very low SNR. You need a much smaller target emitter , 8 deg LED or laser diode with heatshrink tube for small aperture and same for 5mm detector into R-RC comparator. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Dec 6 '19 at 17:09
  • \$\begingroup\$ Did you want to measure displacement , frequency or both. then define this and how it is measured, excited \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Dec 6 '19 at 17:24
  • 1
    \$\begingroup\$ I'd be more concerned about whether a reflective sensor can give you the info you need at all. Personally, I think you'd have more luck with an accelerometer. \$\endgroup\$ – Scott Seidman Dec 6 '19 at 18:23
0
\$\begingroup\$

The electrical interference from all sources ( line E field, stimulator, SMPS noise, AM RF signals ) is expected. So use twisted pairs for everything.

A small RF cap used to get rid of AM RF is optional as Photo Diodes (PD) already have this capacitance.

The Load R determines signal gain and rise time could be 10k to 1M depending on stray light blocking ( optical apertures), proximity and reflection coefficient.

I estimated 2% reflection in my design which is unknown without all physical experiment assumptions given in question.

This design differentiates the signal into a comparator by taking the difference of an integrating RC filter. So the actual DC voltage does not matter (0~3V) as long as there is the reflection pulse and not crosstalk from the stimulus pulse. OK?

I do not recommend this sensor and instead use the Vishay 5mm smallest angle emitter (<10') and Photodiode with black filter.

OK any questions? Here is the sim.

I simulated steady DC LED light with R1 to indicate reflected signal only !! Reduce R2 if voltage >3V. It can sense near 0V down to 1mVpp ok with care on noise.

enter image description here

Use heatshrink tubing over 5mm LEDs to block all stray light and focus only on target then bundle together Emitter/Detector as a pair and separate,shield RX wires from noise.

====> <==== small aperture <3mm with heat shrink tubing on 5mm LED , PD.

EMitters OP297A or equiv PIN PD = SFH213FA {Mouser} or equiv

| improve this answer | |
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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