Question about Mosfet IRF44N

Hello masters,

I need advice about mosfet B input treshold voltage,

how do we create linear increase and decrease in voltage upon introducing electricity to the circuit?

This circuit will create soft on/off on car headlight bulbs.

I have Tried capacitors but they are not efficient..

Maybe PWM may help?

The aim is to increase/decrease gate voltage between 0-5 volts linearly.

Best Regards
Sami

A 0 to 5 volt linear increase of gate voltage will not create a linear source to drain current. That Mosfet won't even begin to conduct between the source/drain until 3 volts gate/source voltage is reached and the rate at which it's current rises is exponential not linear.

It also doesn't reach it's low Ohmic region for it's full rated current until slightly above 5 volts.

You need to clarify your goals.

That mosfet turns fully on with a 10V gate-source voltage. The threshold voltage is when it first starts to conduct. Headlights are relatively high current devices. I can't guess why you apparently want them to come one and go off slowly, but if that is what you want, use >10V gate-source voltage and PWM, I hope you know it will be much easier to do it as a "low-side" configuration.

John

Hi,

If you want a linear current controlled by a MOSFET you have to linearize the response of the MOSFET. That is because the MOSFET response to a gate voltage is not really linear. There will be a range where it might be almost linear though, but that could even change with temperature.

The best way is to control the MOSFET using an active control method. This means measuring the current and providing feedback to the drive circuit which then adjusts the current. This isnt that hard to do, but you need a current sensor in line with the load that can provide the feedback, and an op amp to control the driver. The op amp gets feedback from the current sensor and adjusts the voltage to the MOSFET gate, and that controls the current to follow a nice linear transfer function. The linear part of the transfer function should only take a second or so and then saturate to provide full current to the bulb.

If you dont want to bother with a current sensor you might get away with voltage sensing even though that would not make the current itself linear. The voltage could be fed to the op amp circuit without any special sensors.

Many bulbs fail due to loss of filament material and vibration. The filament material migrates to the surface of the glass enclosure and thus makes it thinner over time. A little vibration then breaks it or a little thermal shock but by that time it's ready to go anyway due to vibration. Thus a slow turn on may not help that much anyway. It would depend on the type of bulb how much this actually helps the bulb life.

I would use PWM, not a linear type circuit.

The headlight bulbs have a very low initial resistance - this increases when they are at full brightness. That will be a fair amount of current for a mosfet to control in its linear region when they are cold.

Have you considered starting the headlights in series then running them in parallel when they have "heated up" ? I've not personally tried that but it may work ?

Hi,

PWM isnt as good as linear for this application, believe it or not. That's because with a bulb in it's low resistance region you still have to bang the filament with high current pulses with PWM, until the filament starts to heat up. That's not better than turning it on fully. To get PWM to work in this application there would have to be an energy storage element as well, namely an inductor. The inductor would provide a linear current to the bulb while the PWM bang bangs from 0v to 12v.

A linearly controlled MOSFET for this application is not a bad idea. For many applications we would never want to control a MOSFET or other transistor in the linear region, but for this app we have an operating time that is much much shorter than most applications. It only has to control the MOSFET for a second or so not for say 3 hours. So the transistor is in it's linear region for 1 second and fully off or fully saturated for all other times. Since the average power dissipation highly depends on the operating duty cycle, the MOSFET would have plenty of time to cool down. Perhaps a small heat sink. If the lights have to blink on and off then the heat sink would have to be larger, but for normal operation it should be just fine.

For example, say the lamp draws 10 amps at full power with 12 volts. That puts the max power of the transistor at around 30 watts (max power point). But lets assume even higher, lets assume it is 60 watts. Normally a heat sink for a 60 watt device would have to be large or cooled with forced air. But for 1 second of operation at 60 watts (and that power is not present the whole time anyway) even a smaller heat sink would not heat up that much.

It depends entirely on the speed at which the dimming is required. Mosfets handle surge currents quiet well thermally, as long as it's not cycled frequently and the ramp time in it's linear region isn't too long. But you have to be careful as automotive typically has to derate for a much lower and higher temperature range.

Hi Scead,

He wants to turn the bulb on slowly probably to avoid the thermal shock. However, i've read in Engineering books that the major cause of bulb failure is vibration, especially after the bulb has aged so some of the filament material had migrated to the inside surface of the bulb. The main idea was that if it does die with thermal shock then it probably wouldnt be long before it blew out anyway even if all thermal shock was completely avoided throughout it's entire operating life.

MrAl, you're assuming that you know his true design goal based only on a limited knowledge of what he's posted in text and are providing information based on your assumption not his request, in my opinion this should never be done. It's okay to question his goal but not to provide information based on an assumption I see too many threads go down the 'what if' line to a dead thread with no resolution to the poster.

Even if you turn out to be right it's best to deal directly with the information provided from the poster, if you wish to provide information to the poster this is great, but don't assume you know his intentions and speak for him.

Photoguy has NOT responded since the initial request for information and so this is a dead thread until he agrees or disagrees with the statements made so far or otherwise provides more information.

Keep in mind his name... It's 'photoguy'

This strongly suggests this may be for photo applications which have nothing whatsoever to do with longevity of the bulb, he may just want a more controlled light ramp, perhaps for higher speed photography or post processing? (I'm really reaching here) but if you can take a series of photo's in rapid succession (all good cameras now days do that) and you can create a linear light ramp through that succession you can in post processing dynamically adjust exposure levels.

But now I'm the one going off the deep end =>

Lets just wait for him to shoot us down or give us more information.

Hi Scead,

While you may be entirely correct about the final end use of the circuit i dont think it matters because the first post seems very clear to me and i quote from the first post in this thread:
"This circuit will create soft on/off on car headlight bulbs."

It doesnt matter to me if he is using them for actual car headlights or he is trying to light up the dark side of the moon
(Side note: Many East Indian people were fooled into watching the moon one night looking for a laser light show made by Pepsi. Turned out to be just a rumor so they all watched the moon that might for nothing. Some got bothered and went out and bought Coka Cola <chuckle>).

So from what i read into the first post it seems that he wants to provide a soft start/stop to a bulb. That's what i've been talking about all along. If i am wrong about this then he must have written the first post incorrectly or something. BTW car headlight bulbs make awful photo lights.

MrAl said:

PWM isnt as good as linear for this application, believe it or not

Click to expand...

This statement is erroneous, you can't make it because you don't know the application to begin with, you're responding based on an assumption, not information he provided. Linearized FET from feedback or basic PWM the choice can't be made without knowign the specifics of photoguy's true application not just the basic premise he's told us without detail.

It's all guess right now.

Scead,

I've posted a quote of the first post in this thread...

photoguy said:

Hello masters,

I need advice about mosfet B input treshold voltage,

how do we create linear increase and decrease in voltage upon introducing electricity to the circuit?

This circuit will create soft on/off on car headlight bulbs.

I have Tried capacitors but they are not efficient..

Maybe PWM may help?

The aim is to increase/decrease gate voltage between 0-5 volts linearly.

Best Regards
Sami

Click to expand...

I dont have a problem understanding this post and what he appears to need or want. If you do there is nothing i can do about that, but i responded in a way which i think is appropriate for his first post. If he wants to change it later that's up to him too, but for now i dont have any problem understanding what he wants.

I replied about PWM in particular because a common misconception is that PWM can do whatever linear can do except more efficiently, which was also suggested by his words about efficiency. PWM can work better when there are certain other circuit characteristics that are already in place in the circuit or can be easily added. Sure, we can make this a PWM buck with inductor and filter cap, no problem. But he probably doesnt want to add an inductor especially since it will only be active for a tiny fraction of the time as the slow on/off ramp time is so short.

Maybe he forgot he posted here since it was his first post