Xband + frequency to voltage converter

Koori

Hello!

I'm working with the X-band motion detector for a school project. We had the idea to use a frequency to voltage converter (IC LM2907) in order to use our microcontroller's ADC to read the voltage level instead of reading the frequency directly with the MCU's timers.

We tested the LM2907 with a function generator and it worked just fine. However when we connect the sensor's output to the converter's input, there is absolutely no output from the converter. We have checked the converter's output directly using an oscilloscope.

Me and my teammate suppose that the converter isn't able to convert because the sensor's frequency output varies so much. Can anyone confirm this?

Also, are there any other alternatives to convert the sensor's frequency to voltage level? We don't have enough time to change the code to read the frequency.

Thank you all in advance!

Phil Pilgrim (PhiPi)

Have you tried a simple low-pass filter (resistor + cap)? That was a stupid answer. Sorry: I was thinking PWM, not frequency.

-Phil

Chris Savage

This reply doesn't really answer your question but I did wonder why you wanted to add more hardware and switch to the analog realm? By doing this you increase the complexity of the circuit. You now also have to scale your voltage to get a reasonable scale and accuracy will be difficult as well as calibration. Most microcontrollers have an internal CTC that you can use to measure the frequency which is directly proportional to speed. That way you'd get the exact data without additional hardware, scaling or conversion. Just a thought. I didn't know if there was a reason you chose not to go this way.

Koori

Chris Savage wrote: »

This reply doesn't really answer your question but I did wonder why you wanted to add more hardware and switch to the analog realm? By doing this you increase the complexity of the circuit. You now also have to scale your voltage to get a reasonable scale and accuracy will be difficult as well as calibration. Most microcontrollers have an internal CTC that you can use to measure the frequency which is directly proportional to speed. That way you'd get the exact data without additional hardware, scaling or conversion. Just a thought. I didn't know if there was a reason you chose not to go this way.

Thanks a lot for your reply. I am currently studying electronic engineering. The microcontroller we used in class is pretty cheap, and after checking the reference manual, datasheet, and programming manual, it does not seem to have a CTC.

As I previously stated, I believe we could've also measured the frequency directly using the timers, but we didn't have enough time to remake the code.

Anyways, it's fine now, the teacher allowed us to present our project that way, with the annotations of why it didn't work with the frequency -> voltage converter.

I also appreciate your advice regarding scaling, accuracy, calibration, and measuring frequency with CTCs (if available), I had not considered it before. I will take that into account for future projects, thanks!

Chris Savage

Koori wrote: »

The microcontroller we used in class is pretty cheap, and after checking the reference manual, datasheet, and programming manual, it does not seem to have a CTC.

I also appreciate your advice regarding scaling, accuracy, calibration, and measuring frequency with CTCs (if available), I had not considered it before. I will take that into account for future projects, thanks!

I apologize for not spelling that out. CTC = Counter-Timer Chip/Circuit

Thanks for your reply though.