Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

PWM effect on heaters.. ?

Status
Not open for further replies.

Wp100

Well-Known Member
Hi,

Just looking for some real world experience regarding driving mains voltage heaters such as radiant wire heater or ceramic plate heaters like those used in reptile cages.

Seems there are several methods of controlling such devices to keep a more stable temperature.

There is simple switching on and off at full power, basic continuous PWM pulses/variable duty cycle and pulsed proportional.

The question is, do any of these methods have an adverse affect on the heaters life ?
Is it better to have lots of short pulses or less longer full power events ?

thanks.
 
I think that for a heater with a long thermal time constant such as the ones you cited, any of the method you're considering would work equally well, and none would have an adverse effect on heater life as far as I know.
 
I agree with OBWO549, just use a 0 crossing switch and keep your time constant long enough for for a full cycle of your power source.
 
Hi,

Just looking for some real world experience regarding driving mains voltage heaters such as radiant wire heater or ceramic plate heaters like those used in reptile cages.

Seems there are several methods of controlling such devices to keep a more stable temperature.

There is simple switching on and off at full power, basic continuous PWM pulses/variable duty cycle and pulsed proportional.

The question is, do any of these methods have an adverse affect on the heaters life ?
Is it better to have lots of short pulses or less longer full power events ?

thanks.


I would assume that PWM applied to the heater would mean the heater is never getting to the full temp level as a simple thermostat. Keeping the unit at a lower temp will significantly extend the life of the heater.

On the other hand, if your unit buzzes significantly as it heats up, this noise comes from the fact that the resistance of the heating element goes up as the unit gets warm (even red hot - depending on heater type). That means you may hear the heater buzz the whole time the unit is running if it never gets up to temperature.
 
I have used triac + reduce firing angle to reduce the power of an on-demand water heater (actually many on-demand water heaters have this as a feature). It is just an open loop reduction in duty cycle. Works fine as long, as Obw0549 mentioned, the system is slow to change temperature with respect to the pulse rate. Here it is 50 Hz and I'm sure the heating element is very slow by comparison.

I had two identical such on-demand heaters, one for each shower. The one without the power reducer lasted 5 years - the length of the warranty. The other one (knock on wood) has lasted 11 years and counting.
 
Normally Triac control with burst firing is used.
There is a replication of the mechanical/Thermal bi-metal controller on the Picmicro site using a PIC and Triac control in burst mode rather than phase angle control which is not needed on Electric element control.
Max..
.
 
In summary, if you are going to reduce the average power into a heater, do it by delivering full half-wave cycles (or even better, full-cycles) of the AC power waveform. Vary the number of missing cycles to control the average power.
 
You can use the simple relay method, its used a lot in industry, so long as the heater is rugged, as the inrush current every time the relay switches gives the heater a electrical and mechanical thump.
Phase angle control is a good way, just be sure to use adequate noise filtering.
 
That is the advantage of Triac burst firing over phase angle, suppression is usually not required as the switching is at zero crossing.;)
Max.
 
Last edited:
I worked with industrial heaters for 40 years. I have worked with all type of controllers heaters act like a DC load pulses have no bad effect on the heater.

If you have 2 heaters of equal wattage & equal voltage with 2 different lengths the longer heater will out last the short heater. Example 1500w 120v 10" long heater VS 1500w 120v 20" long heater, the 20" long heater will out last the 10" heater. WHY = the 10" heater has 150w per inch while the 20" heater has 75w per inch. 75w per inch is less damaging than 150w per inch.

Connections were always the #1 problem with heaters going bad. If your heater is factory made and plugs into the wall there is not much you an do except make sure the plug makes very good connection with the outlet all the time. If you have wires connections to a screw on the heater it will soon become loose and arc burning up the screw connections after 1000 temperature changes. Solder stranded wire then wrap them around the screws check them regular to make sure screws are tight. If you have 2 wires to connect do not use crimp on or twist on wire nuts on the wire after 1000 temperature changes it will become loose and arc. Solder your wires together then screw on a wire nut.

Lead has a higher expansion rate than, copper, aluminum, steel, when the connection gets warmer lead expands making a tighter connection.
 
Yes Max of course.
Control is different than phase angle however zero switching reduces Noise & all kinds of issues.
Moc 3040 I think it is, and opto isolator for firing a triac with built in zero x, used some for a disco light, a long time ago.
 
Correct, burst or zero switching is the less noisy and some explanation and suitable opto's is shown in the Fairchild app note AN-3006.
Max.
 
I disagree with the zero effect on lifetime. We made custom tantalum heaters (~30 V) for our own process for use in a vacuum. They were generally heated with phase angle fired low voltage AC. When the heater power was changed to DC (my change), the lifetime of the heater went up significantly.

The change allowed up to compute power easier and resulted in less rack space. We no longer had to make "ersatz" analog metering panels measuring 3 RU (Rack Units) high and could get by with a 1 RU space. We would have liked a power meter on the power supply. The 1500 W DC supplies were about $1500 USD 10 years ago.
 
Tantalum, you mean a semiconductor heater?
Chances are they'll be completely different than resistive wire heaters.
I've seen semiconductor heaters before, tantalum ones are new to me.
 
Tantalum, you mean a semiconductor heater?
Chances are they'll be completely different than resistive wire heaters.
I've seen semiconductor heaters before, tantalum ones are new to me.

They are known as refractory Heating elements: https://www.plansee.com/en/products/components/furnace-construction/heating-elements.html

https://www.thermcraftinc.com/wp-content/uploads/2018/01/Heating_Element_Seminar_Nov_2_2016.pdf

We made our elements from tantalum wire or tantalum foil. They were used in a Boron Nitride Form (machineable until fired) https://en.wikipedia.org/wiki/Boron_nitride Stable to 1400 C in vacuum.

On such heater was used to evaporate copper (1981 F/1083 C). So, it wasn't your average run of the mill heater.

We made one of our high voltage radiant heaters (about 120 V, 200 W) with a cut piece of tantalum foil sandwiched between 2 pieces of Vycor glass. https://www.pgo-online.com/intl/vycor.html (Service temp up to 900 C). We needed to heat our substrates about 2" away to about 200 C in vacuum.
 
Last edited:
Maintaining extruders I mess with heaters & thermal control a lot, but nothing exotic, apart from them being made for the job.
 
I disagree with the zero effect on lifetime. We made custom tantalum heaters (~30 V) for our own process for use in a vacuum. They were generally heated with phase angle fired low voltage AC. When the heater power was changed to DC (my change), the lifetime of the heater went up significantly.

The change allowed up to compute power easier and resulted in less rack space. We no longer had to make "ersatz" analog metering panels measuring 3 RU (Rack Units) high and could get by with a 1 RU space. We would have liked a power meter on the power supply. The 1500 W DC supplies were about $1500 USD 10 years ago.

Any ideas as to the mechanism? Or could it just be the rms of DC is the DC level whereas to get the same heating from AC was peakier which the heater didn't like.
 
Any ideas as to the mechanism? Or could it just be the rms of DC is the DC level whereas to get the same heating from AC was peakier which the heater didn't like.

No idea. The heaters were phase angle fired. I don't remember the numbers, but it was significant.
 
camerart,

Remember that the same heater element at 120 will have 4 times the wattage at 240 ... SO simply adding a diode at 240 to cut the wave in half will only reduce the wattage from x4 to x2.

Just make sure that the diode AND heater can handle the current and wattage ....

At work I designed an industrial heater application for a hydraulic press where the machine can accept 120 or 240, detecting and switching automatically ... The heater is rated for 1300 Watts at 120. On the 240, essentially a 30 Amp Diode is used in series with the heater element, however on the 120, with additional circuitry, we convert the 120 and push it to 145 so that the element ends up heating faster. <-- Actually to "keep up" with the 240 mode of operation. ... i.e. at 240 the heater might heat to target in 3 min ... at 120 the heater might take 7 min ... at 145 the heater takes about 4 min which is an acceptable time range for our application. Keep in mind at 240 the heater is only seeing 1/2 wave ... at 120 and 145 the heater element sees a full wave.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top