OPT3101EVM: Sensors forum

John Petro

Part Number: OPT3101EVM
Other Parts Discussed in Thread: OPT3101

We have run a slower sample rates per the advice that the evaluation board does not support 4000 samples per second. We reduced the samples to 7 per second with no change in data problems, phase still shifts with temperature and distance steps about +/- 6 to 7 mm from the actual distance (12 to 14 mm total). So the first proposed solution to our issue did not work.

We looked for a way to do the phase calibration which was also suggested as a solution and could not find a way to do this with the software that comes with the evaluation board.

It appears that the phase is sensitive to shifts in temperature of less than 1 degree C. As an engineer I find this hard to understand or believe since I cannot imagine a part that is so sensitive. We ran a graph with the phase and displaying the internal 8 bit chip temperature and the on the board evaluation kit temperature sensor. I would attach that graph but do not see how to do that in this thread.

The last response did not cover my question of using an external oscillator rather than the internal one to help improve this situation.

Please help us understand what is going on here.

Regards,

John Petro

650 526 8129

over 5 years ago

0 Alex Bhandari-Young over 5 years ago

TI__Mastermind 25895 points

Hi John,

I will get back to you on Monday.

Best,

Alex

0 Alex Bhandari-Young over 5 years ago in reply to Alex Bhandari-Young

TI__Mastermind 25895 points

Hi John,

The linked post refers to using the SDK with OPT3101 EVM. As stated there, the EVM has calibrations that will be loaded to the device when using Latte GUI and cancel the temp effects. These will not be loaded using the SDK. If temp calibrations are not loaded then performance will not be good.

We have a number of resources for designing with OPT3101 outlined here https://e2e.ti.com/support/sensors/f/1023/t/815766 there are a number of things to be aware of and covered in the docs. The how to calibrate doc: http://www.ti.com/lit/pdf/slau791 covers temp calibration. I am not sure what you mean by using an external oscillator, but this would not be expected to help with temperature dependence. However, the device temp calibration feature is there for this.

Time of flight devices are very sensitive due to the speed of light and the nature of the technology. We have calibrations on the device for each of these sensitivities that need to be addressed.

Best,

Alex

0 John Petro over 5 years ago in reply to Alex Bhandari-Young

Prodigy 20 points

Alex,

Thanks for your response to our issue with the 3101 part.

We are using the Latte program to run the evaluation kit. The document you referred us to on calibration actually talks about using the SDK program to implement calibration. However your email suggests using the Latte program.

What we are seeing is a variation of about 10 to 12 mm in position with no change in target position. This change occurs with temperature changes. The baseline of the distance does not move, as the variation is always between 35 to 45 mm in measured distance. Your data sheet specifies sub mm resolution which is true for one measurement but measuring the same distance over time the actual resolution is 10 to 12 mm due to this temperature drift.

We have a screen capture of the actions of the Latte program as it starts up but again I do not see a way to attach this file. There are several lines in the program about calibration but we do not know if the correct calibration is being invoked.

A basic question. Using this optical distance measurement on a static target, what is your expectation of the drift in distance measurement over time and temperature?

Regards,

John

0 Alex Bhandari-Young over 5 years ago in reply to John Petro

TI__Mastermind 25895 points

Hi John,

Can you send me a csv export from the Latte GUI showing the drift? The EVM users guide covers the "write data to csv" process from the plot screen (there is a check box for saving data).

The 10 to 12 mm drift is over how much temperature?

This depends on the amplitude level you have. I would expect less than 2cm drift on a typical EVM over the temp range with 20k amplitude or higher. EVM is also optimized for mid-long range so depending on the distance target this drift could be better than EVM.

OPT3101 will not achieve sub mm accuracy. The AFE supports down to 0.22mm resolution, but other effects such as crosstalk need to be considered that make it a few mm in accuracy. This is explained in the intro to system design doc listed in the collateral overview I linked to above.

Best,

Alex