Some PWMOUT questions

[ArnieW]

A long time ago now, I did a basic test with PWMout connected to a solid state relay.

The test failed, and I suspect the problem was due to frequency of the PWM signal coupled with the fact I was using zero-crossing switches.

If I understand them correctly, once triggered they will wait until the cross-over point to switch on, same process for switching off. If the minimum frequency of the PWM is around 4000 Hz, then presumably the SSR will never switch on.

Now I am some days away from my test bench, so a question. Has anyone tried PWMOUT with a solid state relay that is not a zero-switching device? I have some new SSRs that handle greater current but are not zero-switching - can they ordinarily handle a frequency of 4000Hz?

A follow up question - if the PWMOUT is capable of operating in the background, does it continue to work during a READTEMP command on an 18X or 08M?

 

[Technical]

The answer to the second question is yes. pwmout operates in the background using the internal microcontroller pwm feature, any other commands (like readadc) can run at the same time.

 

[steirny]

Arnie,

What load and supply are you trying to control?

SSR's are designed to switch large amounts of ac power, usually sinewave at 50 or 60 Hz.

They are best suited to heating applications and motor control. Not ideal for lighting because of flicker.

When inductive loads are switched at current peaks, noise and supply transients are a problem. Zero crossing is ideal, as it reduces this problem.

Normally they are controlled by gating them on for multiples of the supply frequency. Ideally you should avoid multiples which favour either +ve or -ve half cycles to avoid a net dc effect.

PWM is not suitable for SSR control, whether zero crossing or not. Once the triac is on it stays on until the half cycle is complete. It cannot be turned off earlier, so there is no point in PWM. I would go for pulse out mode. The pusle width will determine the number of on cycles.

More info on your application would help.

Edited by - steirny on 02/05/2007 11:23:55

 

[ArnieW]

Thanks Technical and Steirny,

I was hoping it might be ok because of the background running, but I feared as you described Steirny.

I am switching 15A at 240VAC for full load on a heating element, heating water. I am currently doing this with timed pulses, and then at regular intervals doing a temperature read.

I'm happy to continue that way, just PWMOUT would have been neater.

regards, Arnie

 

[premelec]

You might do well with a PULSOUT to the SSR which has its width modulated in accord with the temperature error - depends on thermal mass etc...

 

[hippy]

For true Zero-Crossover SSR's they should turn on just after crossing zero if their enable is asserted and keep on until the end of the current mains cycle. This way you can limit the output to full on, only half cycles ( +Ve or -Ve ), and one or more cycles with one or more missing cyles between them. In general it's considered better to have a balance of +Ve and -Ve half cycles, so rather than just having +Ve or -ve cycles, have one +ve, one -Ve then skip the +Ve and -Ve, repeat.

Zero-Crossover SSR's can usually be thought of as traditional relays, automatically making sure they turn on and off at times which will cause little electrical noise.

Some SSR's may behave like triacs and switch on any time after the cross over as their enable is asserted. This allows partial parts of the cycle through such as for lamp dimming. The earlier the enable is raised after zero crossing the more wave form gets through. This causes more noise than only switching at the zero-crossover and will probably need to be filtered with coils and caps, quite large ones for heavy loads, and 15A is high.

True Zero-Crossing SSR's only need to have the enable asserted when you want them powered up, and they should sort out the rest. For triac-like SSR's you really need to synchronise to the cross-over, then either set/clear the enable quickly ( emulating a zero cross-over SSR ) or at a time later which defines how much oomph gets through.

It's a form of PWM but synchronised to mains and running at mains frequency, not how we're thinking of PWMOUT in context.

It's hard to say exactly how any particular SSR will behave to high fequency PWM or unsynchronised pulsing as that depends upon its specification. I'd expect to see some quite odd effects.