Creating a sin wave from a square wave

[msmit250]

I'm working on a project that involves creating synchronization waveforms to be used outside of an NI compact Rio. I can guarantee that my cRIO clock times are synchronized (using TSN/802.1AS in the cRIO 9035s). Based on an example I found, I have labview generating a square wave based on switching the signal every time period N (using the clock as the reference). However, I'm having trouble finding a way to convert this signal into a sin wave. What is the best way to convert my square wave of frequency F into a sin wave of frequency F? 

[GerdW]

Hi msmit,

when F is known you can use the sin generator in your FPGA…

Or use a good low-pass filter at the output of your square wave! 😄

[wiebe@CARYA]

Another option would be to create a look up table with a (quarter of a) sin wave. Then use time as an index in the table. Calculate the proper index and negate the output if you stored a quarter of the sin wave to save FPGA memory. Very old school, but it still works!

[Henrik_Volkers]

just wire a VCO.vi and and PLL.vi   ...   with solder 😄

[msmit250]

Thanks all. I think I'll just end up using a low pass filter.

[Bob_Schor · Accepted solution]

While I haven't bench-marked it, I suspect you have chosen the worst method to do this.  The fastest, certainly, is the simple look-up (or just compute using the Sin function from the Math Palette).  Direct, basically one operation per time point, no "approximation", etc.

Bob "IMHO" Schor

[wiebe@CARYA]

---

@Bob_Schor wrote:

While I haven't bench-marked it, I suspect you have chosen the worst method to do this.  The fastest, certainly, is the simple look-up (or just compute using the Sin function from the Math Palette).  Direct, basically one operation per time point, no "approximation", etc.

Bob "IMHO" Schor

---

I don't have a FPGA at hand, but does the sine function work on FPGA? I'd think it doesn't, that's why I recommended a LUT.

Some sine wave approximation algorithm (google) might work too.

[Bob_Schor]

---

wiebe@CARYA wrote:

---

@Bob_Schor I don't have a FPGA at hand, but does the sine function work on FPGA? I'd think it doesn't, that's why I recommended a LUT.

Some sine wave approximation algorithm (google) might work too.

---

Good call.  There is a Sin function on FPGA, but not on a RT Platform.  I agree that the LUT would be simple and fast -- I used it a few decades ago to generate a Sum-of-Sines "unpredictable" waveform on a PDP-11 ...

Bob Schor

[wiebe@CARYA]

---

@Bob_Schor wrote:

---

wiebe@CARYA wrote:

---

@Bob_Schor I don't have a FPGA at hand, but does the sine function work on FPGA? I'd think it doesn't, that's why I recommended a LUT.

Some sine wave approximation algorithm (google) might work too.

---

Good call.  There is a Sin function on FPGA, but not on a RT Platform.  I agree that the LUT would be simple and fast -- I used it a few decades ago to generate a Sum-of-Sines "unpredictable" waveform on a PDP-11 ...

Bob Schor

---

I'd say on RT, you can use the normal Sine function(s) from Mathematics palette?

I used sine LUT's a lot back in the 90's to make C64 demo's\intros (this one is not mine, just the first I could find)!