How to slow down Arduino's PWM' frequency from near 500 Hz down to 5 Hz or so?

Hi guys, I would like to get slow PWM to control my 3 kW 100A oven. For '500 Hz * 100 A' is too noisy for my ears. :o Thank you!

For 5Hz, why not just bit bang PWM?

schabanow:
Hi guys, I would like to get slow PWM to control my 3 kW 100A oven. For '500 Hz * 100 A' is too noisy for my ears. :o Thank you!

Am I to understand that this over is fed from mains electricity, as in AC?

I suggest that the way to control a heater is with PWM over a longer period than a fraction of a second. Assuming that it is an AC supply then to minimise noise generation you should turn the current on just as the voltage passes through zero. To control the power do something like, for example, on for 10 cycles, off for 10 cycles (half power) or on for 5 cycles, off for 15 cycles (quarter power), or, if appropriate, on for 30 seconds, off for 30 seconds. You get the idea.

I'm a bit puzzled by 3kW 100A, that suggests a supply voltage of 30v, is this correct? 100A at 30v doesn't seem right to me.

Hey schabanow, what else does the sketch need to do? If it only provides pwm, you don't even need to BWoD it. What duty cycle? I think 200ms delay will give you 2.5Hz, so you'll want to use 100ms.

A recent thread: 1hz PWM

Thank you guys for massive and sufficient response!

My metal foundry is powered from industrial welding network at our factory (electric power station): 75 VDC, 2 kA. In other words my circumstances were set that way that I have to do virtually nothing to make my private furnace melt an ingot of wootz steel. :o I mean Arduino nano (AtMEGA328P) + HCPL3120 + IRFP4568

Unfortunately, I'm not a coder at all; my realm is power electronics but mc's.

Heat emitting winding over the crucible has 10 turns of 2,5 mm diam. feсhral wire and gives the ohmic resistance of 0,75 Ohm in total. So we have to drive 100A at smooth (pretty small steps, 0 .. 255) power range from 30 W up to 3 kW. Smoothness (mostly at the end of power range) is needed cause the fechral wire has its maximum operational temp. about 1500 C and melting point of about 1600 C, so wootz melt has (+/-). So the aim of the project is to drive heating temperature as smooth and slow as possible just riiiiight to the fechral melting point, in order to be sure that it had succeeded to give all of its warmth through the crucible's walls to the ingot's melt. So we have crucible and heating fechral spiral to be one-time pieces.

Well, this was the brief description of the task.

Thank you very much dougp for the link. It turned out to be extremely helpful! +1

schabanow:
My metal foundry is powered from industrial welding network at our factory (electric power station): 75 VDC, 2 kA.

WOW!

I had no idea such installations existed, thanks for the education. Well outside my experience.