Resistive Heater - VFD

[nhcf]

Has anyone used a VFD to control power output to a resistive heating system? The system is currently ON/OFF control (with a contactor), approx. 50 KW, 480 VAC.

A client has purchased a VFD to allow varying control to the resistive heating element. I am more familiar with SCR control in this application, have never seen/used a VFD for heater control. It would seem that on paper the VFD could vary power to the heaters, but would present issues with EMI/RFI that would be less with an SCR.

 

[moltenmetal]

Varying the frequency without varying the VOLTAGE will have no effect on the power output of a resistive heater! A VFD controls the power of a synchronous motor by regulating its SPEED, not the voltage fed to the motor.

An SCR controller varies the duty cycle of power fed to the heater, by varying either the cycle frequency or the on-time. You can definitely buy them to control 50 kW (and larger) heaters, but if the heater is already divided up into separate circuits, you can use a number of smaller SCRs or solid state relays on each circuit.

 

[LionelHutz]

But a VFD does control both the voltage and the frequency...

And yes, you might have radiated EMI/RFI problems. Also possibly noise problems due to the heater elements vibrating at the carrier or PWM frequency.

 

[moltenmetal]

Dunno- to me it seems like using a hammer to drive screws...

 

[nhcf]

Thanks for the replies.

I understand how the VFD works. I understand frequency won't change the heater output. The VFD controls to a V/Hz curve, and in theory as the frequency is varied the voltage is varied, so I would think the setup would inherently allow variation of voltage to the heater.

I also agree it is the wrong tool for the job and have suggested they procure an SCR controller. But was curious if anyone had rigged up something like this before using the VFD.

 

[GroovyGuy]

Well, I guess someone was going to do this eventually....
I don't see the (output) harmonics being any less with a VFD, as opposed to a SCR controller.
The VFD relies on the motor's inductance to smooth out the current waveform. Without that inductance the heater's current waveform will be as choppy as the voltage waveform. This is not good wrt radiated EMF.

Also, if the VFD is a typical 6-pulse type of front-end, I don't see there being a significant reduction in harmonics on the incoming either ( although a 6-pulse VFD will have some inductance/capacitance on the front end that will knock out some of the higher-order harmonics).

Why anyone would use a VFD in this application is a mystery to me. Perhaps, it was kicking around the plant as a spare?
Other than the harmonic issue, I believe that a VFD will work in this application. This assumes that V/f can be kept near constant.

"I have not failed. I've just found 10,000 ways that won't work." Thomas Alva Edison (1847-1931)

 

[nhcf]

Thanks,

Just to confirm - the VFD is not being proposed to alleviate any sort of harmonics or noise issues. I believe it is being proposed because someone specified the wrong solution and now that is what they have on site. I agree (or assume) the EMI/RFI will be worse with the VFD than with an SCR, for the reasons mentioned above.

I have recommended they not use the VFD and that they purchase the right controller. But checking with others to get their experiences. Sounds like others are in agreement with what I was thinking.

 

[moltenmetal]

Sell the VFD and buy the (MUCH cheaper) SCR controller that is right for the job.

 

[jraef]

A VFD in scalar mode (V/Hz only, no vector control) will work, but not well. The VFD changes the voltage and frequency to a MOTOR using DC pulses because it is feeding an INDUCTION circuit, and it is the inductive time constant of that circuit that allows the PWM to approximate a sine wave. If you have a resistive heater, there is virtually no inductive circuit. Frequency control means nothing to a resistive heater, what you want is voltage control and without the induction in the circuit, you don't have much control of it with a VFD. I have used resistive heaters to test the VFD as a general load just to make sure the VFD worked, but the control of that heating load was irrelevant for that. As an experiment once (after a similar discussion in this forum over a decade ago), I tried it again and it was extremely inaccurate. I could make a big change in my output command and it had little effect on changing the heating effect. But by increasing the carrier frequency to 16kHz as suggested (by I think Gunnar), that did improve the control performance of the heating effect somewhat, probably because the higher switching rate made the entire circuit look more inductive. But that requires de-rating the VFD current capacity by about 50% because of the added switching losses in the transistors. All in all, it's a waste of money, because a simple SCR controller does the trick for 1/5th the cost, maybe less once you de-rate the drive.

That said, if it is an INDUCTION heater, that's a different story, PWM is used for that all the time. I have also wondered if you used a transformer between the drive and resistive heater, if that would give you more control. Probably would (because of the inductance of the transformer) but there are potential issues with using transformers with variable frequency, so you would have to try to modify the VFD output to change voltage only, and few VFDs are going to allow that. It's not what VFDs are designed to do...


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[itsmoked]

I'd bet the VFD in this commodity world IS cheaper than a Phase controller. I'd try it just to see what happens.


On the other hand a hundred dollar PID controller and some 70Amp solid state relays are probably cheaper than both a phase controller or a VFD and will actually do a good job of temperature control.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

The transformer also wants to see the proper Volts per Hertz ratio.
I have seen transformers used between VFDs and motors.
They were driving submersible pumps in a SAGD (Steam Assisted Gravity Drainage) application. There were a number of wells on the pad. Some had filters after the VFDs, some had transformers between the VFD and the motor.
The transformer would probably work for a heater as well.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[nhcf]

thanks for all of the feedback. Good information and discussion.

 

[waross]

On the other hand a hundred dollar PID controller and some 70Amp solid state relays are probably cheaper than both a phase controller or a VFD and will actually do a good job of temperature control.

That was my first thought as I started reading this thread.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Shozza]

I understand this thread is a couple of months old but thought a late response could still be beneficial for future applications.

Some VSD manufactures offer VSDs with an adjustable voltage operating mode designed for such applications. It enables frequency to remain fixed while allowing variable voltage output to control the heater. Works well for varying the field strength on an electromagnet too.

Powerflex 700VC and the more recent Powerflex 753 are two candidates.
I'm sure other manufacturers may also have this functionality.

 

[waross]

The electromagnet is highly inductive.
Varying the voltage without varying the frequency may not address the issue of low or no inductance in the heater.
Why change one VFD for another VFD when the optimum solution is a controller at about 20% of the cost?

Bill
--------------------
"Why not the best?"
Jimmy Carter