Inaccurate frequency measurements with cFP-CTR-500

[jhathaway]

I am trying to measure frequencies using a cFP-CTR-500.  The possible ranges for my frequencies are relatively low, say between 20 and 200 Hz.  I used the low-frequency measurement example found at http://forums.ni.com/ni/board/message?board.id=110&message.id=3619&view=by_date_ascending&page=1

The specific file from the forum I used is:  http://forums.ni.com/attachments/ni/110/3631/1/Low_Frequency_Measurement_(CTR).zip

I set everything up in MAX just like the instructions indicate, and when I try to measure the frequency, it returns what looks like either a random number or a measurement of some line noise of varying frequency (the highest number I've seen so far is 3000 Hz).  The frequency reading seems random a lot of the time, but mysteriously seems to land on 500 and 1000 Hz a lot.  I attached an image from my scope showing an example waveform of the frequency I'm trying to measure and an FFT spectrum from that waveform.  It doesn't look to me like there is sufficient noise on the line to cause this much chaos.

The sensor I'm using is a proximity switch, and the frequency is coming from a rotating motor shaft with a pickup point on it.  I've wired a 12 volt power supply to the connector block and I'm powering the sensor from the Vsup and Com terminals as indicated in the wiring diagram for the counter. 

I'm hoping that someone has some advice on how I can get reliable measurements this way.  I don't know if I did something wrong, am cursed, or simply have an inappropriate signal. 

Thanks a lot,

     -Jim

[Hal_L]

Hi Jim,

I took a look at your signal, and I agree with you in that the noise should not cause the problem you are seeing. The signal needs to be TTL compliant in order to work with this example.

A great way to double check your connections and to make sure that everything is configured properly is to simply use another one of the counters on the FieldPoint module to output a pulse train. Then you can measure the frequency of the pulse train with the example. Please let me know if this works properly for you. Also, which version of the FieldPoint driver do you have?

Regards,
Hal L.

[jhathaway]

Thanks for the response.

I'm using FieldPoint 5.0 and LabVIEW 8.0.

I've got a busy week this week, so I most likely won't be able to test my connections with the pulse train until Friday. 

I'll see if I can figure a way to get the signal down into the TTL range (0-5 volts, right?) - but according to the manual, the inputs on a cFP-CTR-500 don't respond to 6 volts and lower. 

Thanks again,

     -Jim

[Hal_L]

Hi Jim,

You are correct about the module not being able to read a 0-5V TTL signal. Anything less than 6 V will be read as low and anything greaten than 10 V will be read as high. Please let me know what you find out when you try to read the frequency of the pulse train with the example.

Regards,
Hal L.

[jhathaway]

Okay, I set up counter input 2 and output 1 to output a pulse train to counter input 1.  I used a 1000 kHz reference to count on counter 2 and set the terminal count to 50.  I then set up output 1 to toggle, reset on and its output source to counter 2.  I also tried setting the output mode to on pulse.  Everything else is set up per the instructions in the example.  None of these worked.  I changed the terminal count to 1000 so I could watch the lights on the module switch on and off, and it is working properly, so I don't think there is a wiring problem. 

Thoughts? 

Thanks a lot,

     -Jim

[Hal_L]

Hey Jim,

Thanks for the update. Right now, let's make sure that the counter input are reading simple event counting properly. So, generate the pulse train with one of the counter outputs as you did before and then just read the count number and make sure that functionality is working correctly. It is probably best to use a shipping example for your pulse output if you are not doing so already.

Once we get this part working, then try the same thing with your external signal to make sure the module is reading those pulses. One final thing: could you please let me know what revision of the FieldPoint backplane you have? Please let me know if you are able to successfully perform these operations and I will continue to work with you on this.

Regards,
Hal L.

[jhathaway]

Hi Hal -

I tried counting a generated pulse, and it worked fine.  I couldn't find a shipping example, but the previous pulse method I used seemed to work fine.  Counting my actual signal also worked well. 

Regarding the revision level, I wasn't sure what you were looking for.  I'm using a cFP-1808 backplane, but if you're looking for a revision level deeper than just the type of backplane, I'm not certain where that information would be located. 

Thanks again,

     -Jim

[Hal_L]

Hey Jim,

Thanks for the update. It sounds like we are starting to make some good progress towards getting the frequency measurement working. As for the backplane revision: if you look on the bottom, you should see a part # and serial #. The part number should be a series of numbers and letters with a -01 or something similar at the end. The last letter before the - is the revision of the backplane.

Now that the counting part of the application is working, have you had a chance to configure everything again for the frequency measurement? Let me know which revisions of the backplane you have and try the frequency measurement with the pulse still being generated from the FieldPoint Counter module. Please continue to keep me posted.

Regards,
Hal L.

[jhathaway]

Hi Hal -

Sorry for the long break - I've been quite busy and haven't had too much of a chance to work on this project.

I've been looking into this, and I wasn't looking any farther in this than trying to get just one channel to measure frequency properly.  After looking, I realized the maximum number of individual frequency measurements I can make on one CTR-500 module with this low frequency example is 4, unless I'm mistaken.  Is that the case?  Each frequency measurement requires its own gating channel.  If that is the case, this method will not work for me at all.  I need as many frequency measurement channels as possible.  For now, I'm going to have to put this counter thing on hold.  I'm looking into building a frequency-to-voltage conversion board that I can put in my system and use voltage analog input modules to measure that.  I've got a few sample chips on the way, we'll see how that goes.  If you have any advice or experience with how well those work or any pitfalls, I would welcome it.   

Thanks for the help, Hal.  I appreciate it a lot.