Unstable sine PWM output

[HenrikOlsson]

Hi,
It might be the DIV32, we'll have to get Darrel to answer that or have someone dig into the ASM code and figure out if the PBP registers used by DIV32 are being saved by DT-Ints or not (I still think they are).

My guess is that it isn't actually DIV32 that is causing it but something else with the way the frequency is changed. Why not try constantly ramping the frequency up and down in your main loop? Don't use DIV32, just "sweep" the reload value up and down between your MIN and MAX values. Does it still glitch? If so we know that it's not DIV32 that's causing it but something else.

If it turns out that it IS DIV32 then we can look at other solutions. The DIV32 thing was in response to your request of directly calculating the reload value based on the desired output frequency scaled in Hz. Simply rescaling to some arbitrary "unit" might allow the calculations to be done without DIV32 but it comes with a couple of commitments which I won't go into until we know that that DIV32 is the problem.

Can you explain to me where the values 34696 and 32 comes from?

/Henrik.

[pxidr84]

Hi,
It might be the DIV32, we'll have to get Darrel to answer that or have someone dig into the ASM code and figure out if the PBP registers used by DIV32 are being saved by DT-Ints or not (I still think they are).

My guess is that it isn't actually DIV32 that is causing it but something else with the way the frequency is changed. Why not try constantly ramping the frequency up and down in your main loop? Don't use DIV32, just "sweep" the reload value up and down between your MIN and MAX values. Does it still glitch? If so we know that it's not DIV32 that's causing it but something else.

If it turns out that it IS DIV32 then we can look at other solutions. The DIV32 thing was in response to your request of directly calculating the reload value based on the desired output frequency scaled in Hz. Simply rescaling to some arbitrary "unit" might allow the calculations to be done without DIV32 but it comes with a couple of commitments which I won't go into until we know that that DIV32 is the problem.

Can you explain to me where the values 34696 and 32 comes from?

/Henrik.

I've done like you said a simple ramp in the main loop :

reload=reload-1
if reload<40000 then reload=64600

The sine frequency ramped down, and after many cycles, no glitch appeared on the oscilloscope. During this time I've played with the potentiometer to vary the amplitude and also I navigate into the inverter menu, the sine wave was not disturbed at all.

So the problem is clearly DIV32. Apparently this command disturbs DT Instant Interrupts.

I've already changed the "34696" and "32" dummy values to "1120" and "1000", it's far better to understand.

[HenrikOlsson]

So the problem is clearly DIV32. Apparently this command disturbs DT Instant Interrupts.

Or the other way around, or there's a problem with the actual values being calculated like overflow/ underflow/whatever. I'm still not convinced but it seems you are so lets go with that for now.

OK, you're using an analog input to set the desired output frequency, the ADC have a resolution of 10 bits giving you a value of 0-1023. If we're going to try to come up with an alternative way to calculate the reload value (not using DIV32) can you tell us:

A) What is your minimum output frequency, ie what frequency should an ADC value of 0 give and what TMR1 reload value does that frequency require?

B) What is your maximum output frequency, ie what frequency should an ADC value of 1023 give and what TMR1 reload value does that frequency require?

/Henrik.

[pxidr84]

Maybe it isn't DIV32... I don't know, but when I remove it, the glitches disapppears. I've also already tried to change the dummy values.

My min. frequency is 1Hz (freq=10), reload value=about 0
and my max. frequency is 120Hz (freq=1200), reload value=about 64600

And I don't directly use the 0-1023 potentiometer value, first I convert it to the "freq" variable :

potsense=(potsense<<6)**maxfreq

"maxfreq" correspond to the maximum frequency defined by the user. Here, "maxfreq"=1200. So here "freq" variable varies between 0 and 1200. Note I'm using an additional number to have the 0.1Hz accuracy (for exemple, 50.2Hz frequency equals freq=502).

I have to said, your timer reload routine with DIV32 was pretty accurate and provides very good results. In terms of accuracy, I really had a 0.1Hz precision, both in low and high frequencies.

[HenrikOlsson]

Hi,
OK, so when Freq is 10 you want Reload to be 0 and when Freq is 1200 you want Reload to be 64600. That's 64600 over 1190 or a reload value of 54.286 per unit Freq.

Reload = (Freq */13897) - 542

When Freq is 10 you'll get (10*13897/256)-542 = 0
When Freq is 1200 you'll get (1200*13897/256)-542 = 64600
Do not allow Freq to be less than 10, you'll end up with a negative value.

This is not perfect but it might work, at least you'll get rid of DIV32. If you use the exact reload values for Freq=10 and Freq=1200 instead of the aproximate ones it might get a bit better in between the two extreme ends.

By the way, the lowest possible interrupt frequency, when running at 40Mhz and without changing the prescaler is 152.588Hz which equals an output frequency of 1.69Hz so Freq should really only be allowed to range from 17 to 1200 for 1.7 to 120Hz.

/Henrik.

[pxidr84]

Unfortunately, it doesn't work as expected but there is no glitch when I played with the potentiometer.

For example, at 100.0Hz I obtain an 9.5Hz output,
at 110.0Hz -> 17.6 Hz
at 90.0Hz -> 6.5Hz...

And there is no way to use this? http://www.picbasic.co.uk/forum/showthread.php?t=12433

or this?
http://melabs.com/resources/fp.htm

(just a suggestion)

[HenrikOlsson]

Hi,
Aaargh, sorry about that, temporary brain-fart on my behalf.

At the moment I'm afraid I can't come up with an alternative solution. I'm sure there is one though and hopefully someone else will jump in.

I've never used the floating point routines and I don't really see any reason to do it here. N-Bit math might be an option but I currently don't have the time to do the experiments for you so you're going to have to try it yourself. Darrels examples in the thread is pretty self explainatory.