I am building a heater controller and would like to ask for someone's expert opinion. Is it better to use a Zero Crossing SSR or a Random
Turn SSR when controlling a heater? At the most I will be switching a 20A 230Vac load. The SSR will be controlled by a DC output of a PLC. Any comments or suggestions would be helpful.

Thanks
Steve

 

Random fired units are more versatile for application of proportional on/off control as the total switching interval is 1/2 cycle, or 8.3ms. The virtue of zero crossing SSR's is low noise due to switching at zero load, but at the expense of less control over the duty cycle or extending the switching interval.

Most electrically heated systems have slow enough thermal response time that an extremely short switching interval isn't necessary, and a zero crossing unit may be satisfactory. Don't let the switching period be too long, though, as the thermal cycling also contributes to wear in the heating elements. Also, media flowing through a heater will change temperature much more quickly than when you are heating a fixed mass of material, like liquid in a tank.

So the answer is to balance the sensitivity of your application to electrical noise, heater wear, and control accuracy, while considering the rate at which your heating element and load are capable of changing temperature.

If you are not controlling temperature, and the heater is 100% on until the process is stopped, you may as well use a zero cross unit.

SMyres@ Mindspring.com

 

I have designed several similar heating systems for electric furnaces and used the zero crossing version due to the lower electrical noise (with 250+ solid state relays in use, that could be significant). I used a one-second time proportional control, which gave me about 2% resolution of the ON-OFF time each second. It also allowed for the thermo-cycling (as mentioned in an earlier reply to this post) to be fairly minimal.

It was implemented in a PLC-5. Each second the PID loop updated and I took the percent output and compared it to a free running 1.00 second timer. If the percent output was higher than the accumulated timer value I turned the output on. Be sure to accommodate a zero % output in your logic (using a GRE vs. GRT will make a difference).

Regards,
Russ Kinner
AVCA Corporation
Maumee, OH USA

 

Zero crossing is better because it generates lower inrush current less electromagnetic noise etc. Watch for the cycling time on the SSR. If you cycle them too often they fail. See the spec sheet for details. Also remember that SSR's can fail in the short condition so you will need some
other safety to be able to remove the power in that case.

James Bouchard

 

I did something similar a few times, but never with a one second time base. I generally used 10-20 seconds. But this was for water, not air, maybe it makes a difference.

 

With the 1 second timebase, you could see the elements brightness pulse if the output was between about 25-75%. Much below that and the elements didn't glow at all and much about that range the elements stayed bright all the time.

The application was to heat automotive glass and as the blanks traveled at 50-80 mm/sec, a longer timebase would have caused an unacceptable variance in heating from one piece to the next. The elements were commercially available 1/2" wide nichrome ribbon and worked on 240 V. This worked well and we've had no failures of SSRs or elements in several years except when a lightning storm took a few SSRs out. BTW, all SSRs are protected by fast acting, rectifier type fuses (80A rating for a 65A resistive load).

Russ Kinner

 

We did it for heated rolls on a 1 second time base but later had to move to 2 seconds because it was killing the SSR's.

James Bouchard