1
\$\begingroup\$

I am trying to improve the circuit of a coil heater which operates at 230VAC input. Total power to the coil is 2000W. There is a thermostat which measures the coil temperature and there is a relay which will cutoff the 230VAC input to the coil when the optimum temperature is reached, which we can say as 178°C, via a relay.

Can I convert AC voltage to DC voltage and use IGBT or MOSFET with a PWM signal to control the coil heat and maintain it at 178°C?

Will that be efficient? Can I completely avoid the relay and depend only on the MOSFET to maintain constant temperature to the coil. Will my idea work?

Will it be more efficient to use DC voltage to heat up the coil?

If there is any evaluation kit please let me know. Let me try and see.

\$\endgroup\$
2
  • 2
    \$\begingroup\$ moving to DC would move some of the heat to dioide and transistors, making it less effecient. since coils are slow to heat and cool, you could use a triac or zero-cross SSR to switch the AC on and off per-cycle; essentially a 50hz pwm. \$\endgroup\$
    – dandavis
    Jun 14, 2022 at 19:17
  • \$\begingroup\$ What is it heating? \$\endgroup\$ Jun 14, 2022 at 21:24

3 Answers 3

1
\$\begingroup\$

AC power is like a trusty work-horse. DC is like a rhinoceros.

AC power is easy to manage because it reverses polarity 100 times a second, which means you have a zero-crossing for voltage and a zero-crossing for current (not necessarily at the same time). This means AC arcs tend to self-extinguish, and switching hardware has an easy time of it.

Higher voltage DC power has no zero-crossings. Thus, once an arc begins, it cannot be reasoned with, it does not know pity or remorse or fear and absolutely will not stop, ever, until your entire machine is magic smoke.

That was 600V. 230VAC will rectify into 325VAC, more than half that.

It won't make a hill-of-beans difference in heat

Rectified bouncy-ball DC is exactly the same power as regular old AC. As dandavis mentions, you would lose a tiny bit in diodes and devices (but then, heat is your objective).

Making it DC doesn't make it more controllable

As discussed, less controllable. You're thinking "MOSFET" because that's the tech you're familiar with. But AC has its own tech that's just as good if not better -- the triac.

Look at any random $12 AC light dimmer. It is a triac. This is extremely simple and cheap because the triac doesn't need to interrupt AC power - the AC zero crossing will switch the triac off automatically.

The dimmer only turns the triac on.

Since a power half-cycle is 10ms (8.33 in 60Hz land), you can simply wait n milliseconds from zero to turn the triac on, with "n" being the variable. It's not linear - you must think of the power content underneath the sinewave - but it's easy.

enter image description here

Of course, this degrades your power factor, so you may want to think about that.

\$\endgroup\$
0
\$\begingroup\$

As Dandavis said, moving to DC will be less efficient because some power will be lost in heating the rectifiers and other components.

Replacing the relay with a triac or other electronic switch could also be less efficient than a properly chosen relay. It depends on which is greater, power used by the coil of the relay, or power dissipated in the triac.

An electronic switch will allow finer regulation, variation of heating power, PWM, and other nice features, which may justify the efficiency loss.

\$\endgroup\$
0
\$\begingroup\$

Switching AC with semiconductor devices is not fundamentally more complex than switching DC. Usually you just need two mosfets instead of one, for example. Or use a triac – a very robust device. Rectifying a 10A load efficiently requires synchronous rectifiers, so you might as well use the switches that would do the rectification for controlling the AC load directly.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.