Real-Time Measurement and Control

cancel
Showing results for 
Search instead for 
Did you mean: 

How to create a sawtooth PWM with Crio 9074 FPGA ?

Is there anyway you could send me the PWM triangle wave example from ipnet under signal generation in labview 8.5 format? It looks like something that could help me. The application engineer I talked to earlier said she would email it to me, but she may have forgot.

 

Thanks,

Dane Miller

0 Kudos
Message 11 of 19
(2,288 Views)

The main problem I seem to be having now is getting a sawtooth wave to go around 300KHZ. So far the fastest I have seen on my output of the 9263 is 30KHZ sawtooth and PWM output. I need ten times that. Do you think a derived clock could perform that much faster? I tried the SCTL and the Q and R requires more than one tick as well as the module read and writes.

 

Thanks,

Dane Miller

0 Kudos
Message 12 of 19
(2,286 Views)

Dane,

 

Did you receive the example from the other AE?  I was talking with her last night regarding this so if you don't have it I can definitely look into downgrading it.  Derived clock 'may' help with that, but the best thing to do would be to have the AO module (possibly AI as well) in a dedicated loop so it will run faster.  You will not be able to use the SCTL with any of the analog modules, that is correct.  Try to break up things into different loops and use a FIFO between loops to get data from one to the other.

Regards,

Jared Boothe
Staff Hardware Engineer
National Instruments
0 Kudos
Message 13 of 19
(2,274 Views)

Jared,

 

I never did receive it from the AE I talked to last night. Could you downgrade it for me.

 

Thanks,

Dane

0 Kudos
Message 14 of 19
(2,273 Views)

I have tried everything I know how to do to try and make the sawtooth wave go around 300KHz and I can't seem to figure it out. I even tried using a relay with a LUT to see if that would perform faster. Is there anyway you could post me an example of what you had in mind using a determinstic loop with a FIFO?

 

Thanks,

Dane Miller

0 Kudos
Message 15 of 19
(2,263 Views)

I just now received the example from the AE I talked to the other day. Not sure if it will be fast enough, but I will try it. I did attempt putting the triangle wave in a SCTL with a LUT and a FIFO to communicate the data to another loop. It doesn't seem to want me to use a shift register? Any ideas?

 

I attached my most recent attempt to achieve a triangle wave with 3MHz steps and 300kHz period on FPGA for crio 9074:

 

Thanks,

Dane Miller

0 Kudos
Message 16 of 19
(2,259 Views)
The NI 9263 is a 100 KS/s analog output product.  According to Nyquist, the fastest waveform you can produce with this product is 50 kHz.  If you need to produce an analog output waveform of 300 kHz or higher, you'll need to use different hardware.  In this case, you may want to look at the R Series product line.
0 Kudos
Message 17 of 19
(2,253 Views)

Thanks, that makes more since now. In the end I don't need to see the sawtooth and I would like to ouput the PWM signal from the 9401 digital I/O module do you think this would be sufficient for a freq of 300kHz?

 

Thanks,

Dane Miller

0 Kudos
Message 18 of 19
(2,246 Views)
The switching frequency of the output channels on the 9401 will depend on the number of channels used and the load on the output.  You should look at the specifications in the manual to verify the 9401 will satisfy the requirements of your application, but on the surface of things, I wouldn't expect any difficulties achieving 300 kHz PWM output using this module.
0 Kudos
Message 19 of 19
(2,244 Views)