MOSFET Question...

[adamey]

If I use a IRFZ44 to control a circuit using PWM (lamp dimmer), but want to use a relay for full on, are there any issues I need to be aware of?

Specifically, will the MOSFET behave oddly if full voltage is presented to the load from a relay even though it's still providing a PWM output to the same load?

 

[Chaerl]

What do you mean by this? the load (lamp), contact of the relay and FET (DS) is in series Or the FET is driving the relay? it would be helpful if you have schematics.

 

[adamey]

The FET and the relay are connectd in parallel. Normally the FET drives the load (lamp). Occasionally for full power the relay will drive the lamp.

This is for daytime running lights on a car. The FET powers the lights for reduced output through PWM. If the driver flashes the high beams then the load will receive 12V direct from a relay to bring the lights to full power.

I want to know if the FET is OK with being in parallel with a relay and having full voltage applied to the load from another source (in this case a relay).

 

[Chaerl]

With my previous company, we have issues a on that circuit. I'm trying to find a copy of that lessons learned document. Though, it was not a relay, but a mechanical switch, but I think its the same effect. Instead of FET, use BJT.

I can't explain why now. It's in the tip of the mind but I totally forgot the reason.

 

[adamey]

Can I use a SPDT relay where the load gets switched between the MOSFET and 12V? Would this cause problems for the MOSFET if it gets dis-connected from the load while it's being driven at the gate?

 

[Leftyretro]

The details depend on if your using high side switching or low side switching, but in general there should be a way for relay contacts to bypass the MOSFET switch and allow full on without damaging the MOSFET. Show us a circuit drawing and I'm sure you can get better quality comments.

Lefty

 

[crutschow]

Another factor is whether the PWM circuit has an smoothing inductor in the circuit or not and where the relay is connected to bypass the PWM circuit.

 

[Boncuk]

Why bother with an RDSON of 22mΩ?

If you use 100% PWM (continuous max gate control voltage) the difference between using the IRFZ44 and a relay will be negligable.

Relays also have a certain contactor resistance depending on quality.

Boncuk

 

[adamey]

^ Because most newer vehicles use a relay to turn on the lights. I want to add daytime running lights (DRL's) by using PWM on the high beams to operate them at reduced power. When the driver requests high beams, I don't want to have problems with my DRL circuit.

Some cars run the high beams in series to do this, but that's a lot of extra wiring.

 

[Chaerl]

This is true. For high current loads, such as lamps (H/L, DRL, Turn Lamps), normally, relays are used. This is due to the thermal consideration if semiconductor is used (though some smartFETs can do the job). Macro point of view for cost analysis, it would be cheaper for the manufacturer to used relays.

Back to the topic, I just realized that your load is a lamp. I know I said earlier about issues from my previous company. The load was actually inductive, it was for the Horn in the car. The PWM is for the anti-thieft alarm and the switch is for the driver (horn switch). It was more on the stored energy in the coil that destroys the FET. When the switch is closed, a very low resistance is paralleled with the DS and the FET is off. I forgot what phenomenum was that. I don't know it it would occur if the load is a lamp.

But if it is not, then, you need to take of the inrush current which is very high (40-60A) for few milliseconds. I would suggest to use a smartFET with a built-in inrush protection. Ideally, it adjust the limits by reducing the RDSon during start-up at which inrush occurs. Try to see infineon products. They are good at this.

About using SPDT relay, I don't see any problem. But still consider the inrush current.