PWM controllers with built in FETs?

[eeka chu]

I want to power a 100W+ thermoelectric element and figured using a linear regulator would get the supply baking hot at lower settings on the element it's self.

Are there any PWM IC's you can think of that incorporate a FET? I'm dealing with around 12VDC and 9A.

I'm not a genius when it comes to digital gear, so it'd be great if I could control the pulse width through a simple divider setup; particularly since the thermistors monitoring the temperature will be resistive to start with.

One with a built in FET that could handle those voltages and currents would be excellent! I want to keep it to the absolute minimum number of separate parts.

If all else fails, a driver would be good. And then I'll go for the 555 if that fails.

With it powering a TEM, and not any form of magnetics, I don't want a high frequency at all; I'll just end up with loss and noise problems, 100Hz or something around that would probably be fine.

I know they do a lot of these for SMPS for LEDs and the like, but they'll need separate gates and I think they're mainly geared towards high frequencies to keep the magnetics tiny.

Thanks for your time!
John

 

[runebivrin]

The first question is why you'd want to have a variable regulated supply in the first place. Normally you'd just turn on power, wait for the thermistor to measure target temperature, switch of power, wait until temperature drops below a specific limit, turn on power again. Repeat as necessary.

 

[eeka chu]

I want to be able to leave the system alone at a specific temperature gradient without having to check back on it.

If I just hit the juice, it'll rocket down until I switch it off, then bounce back up since the element has such little thermal momentum; hence the PWM supply. I want to be able to select specific temperature gradients rather than just go straight to it's maximum.

I certainly don't want to have to stand beside to flicking the power on and off by hand. A 555 would solve that. I'm essentially asking for a 555 with an internal FET that'll handle those voltages and currents.

Normally you'd just turn on power, wait for the thermistor to measure target temperature, switch of power, wait until temperature drops below a specific limit, turn on power again. Repeat as necessary.

You seem to have entirely missed the point of a closed loop PWM supply given that these elements will establish their maximum Delta T in a matter of seconds. You're recommending I stand beside it and flick the power switch every second or more, by hand, for a few hours? Is that a joke?

 

[star882]

Just use a common PWM controller and a MOSFET.

 

[runebivrin]

I want to be able to leave the system alone at a specific temperature gradient without having to check back on it.

If I just hit the juice, it'll rocket down until I switch it off, then bounce back up since the element has such little thermal momentum; hence the PWM supply. I want to be able to select specific temperature gradients rather than just go straight to it's maximum.

I certainly don't want to have to stand beside to flicking the power on and off by hand. A 555 would solve that. I'm essentially asking for a 555 with an internal FET that'll handle those voltages and currents.



You seem to have entirely missed the point of a closed loop PWM supply given that these elements will establish their maximum Delta T in a matter of seconds. You're recommending I stand beside it and flick the power switch every second or more, by hand, for a few hours? Is that a joke?

Well, the idea was to build electronics to do the switching. I certainly didn't mean to suggest that you do the switching manually. But if you need to apply a lower voltage than the maximum supply then certainly a PWM solution is a good idea. And I think a 555 and a FET or IGBT would do nicely.

 

[runebivrin]

Late addition:

However, I think even 100 Hz is unnecessarily high for the application. Thermal inertia would probably make something like 10Hz sufficient. The question is if you want to regulate the temperature of the TEM or of that which it cools?