Selecting MOSFETs for PIC Halogen Lamp Driver Circuit

[PICpopper]

Looking for some help selecting MOSFETs and R/C values for a PIC lamp control circuit. I'm trying to use a PIC18LF2620 PWM output to drive a Halogen lamp using a lead acid battery for power. The PWM freq I'm guessing to be around 1K Hz, but not sure. It's for a night predator hunting app, and don't want any flicker that coyote eyes can see.

I need to use 3.3V because of the RF tranceiver I selected. I'm not an EE guy and have very limited electronics knowledge let alone MOSFETs, and I'm drownding in all the choices. I think I need a logic level MOSFET (for 3.3V Vss) that can handle maybe up to 20V and 10A (for possible lamp upgrades). I planned to include the MOSFET driver shown in attached circuit. But that's as far as I can figure out. Help would be appreciated!! (You can get a PIC kit pretty easy, but I guess it's the hookup to the real world that'll get ya.)

I have two configurations I want to build, one with a 12V battery powering a 12V 55W Halogen, and the other a 6V battery with 6V 52W Halogen. If I can use the same MOSFET for either that would be great, and if I use a MOSFET that can handle even greater power output, that would leave me room to control bigger and better lamps down the road I guess.

I'm a software guy by trade, but can read schematics a little which gets me into trouble (like with this project , so please bear with my lack of knowledge.

Thanks!

 

[ronsimpson]

1.) Do not use a high side driver like the LTC1157! If anything use a low side driver.
2.) If you keep the PIC at 3 volts use 5 volts supply (or higher) on the FET driver. It is a little hard to find FETs that like 3 volts.
3.) If your PWM is only 1khz then you may not need a FET driver.
24N06L, IRF6607, IRF6618TR

 

[dougy83]

The LTC1157 has an internal boost converter which will supply 8.7V to the MOSFET gate in your design. Therefore you don't need to worry about getting a 3.3v MOSFET.

However, the LTC1157 has a very low gate drive current resulting in a slow turn on speed of the MOSFET. This will increase the power dissipation of the MOSFET (it will heat up more than it needs to).

You could also drive the gate from the high voltage (6-12V) supply, as many MOSFETs can handle up to 20v on the gate. You could use something like the attached circuit, or you could get a different driver chip that does this for you. In fact, you can use a LM555 to drive the MOSFET (see 2nd attachment) so long as the supply voltage doesn't exceed 16V.

 

[ronsimpson]

Dougy83 is correct. The LTC1157 does boost up the gate voltage. I only had a sort description of the part that only said “high side driver” and did not mention “boost up”.

 

[PICpopper]

Guys, thanks for the mosfet and driver circuit suggestions. I'll try a few over the next week and report results back.

Dougy83, I like the one chip driver approach since is simpler for me to lay out. You mentioned problem with the very low LTC1157 gate drive current. Can you tell me how to find the gate drive current from a mosfet driver spec, and what value typically should be good/fast enough?

 

[smanches]

The gate drive current is based on how fast you want it to switch. Not the PWM frequency, but the rise time of the leading edge. You calculate it as follows...

Q = total charge of the mosfet.
t = Switching time period.

I = Q / t

Example: Mosfet with 100nC charge with a 40ns leading edge.

I = 100 / 40 = 2.5A (both the charge and time period are in nano-whatever, so we don't have to convert them to a common denominator.)

 

[dougy83]

Can you tell me how to find the gate drive current from a mosfet driver spec, and what value typically should be good/fast enough?

The spec I'm looking at is: **broken link removed**

On page 2, ton Turn-ON Time for 5.3V out into 1nF gate takes ~240us up to 750us. As you want to switch at 1kHz (period of 1000us), most of the time the MOSFET won't be conducting fully. Depending on the duty cycle, the MOSFET may not actually ever turn on (as the gate voltage may never get a chance to rise past the gate threshold voltage).

The gate drive current is shown in the appropriately labelled graph on page 3 and is always less than 11uA.

 

[PICpopper]

Break out the Datasheets and Calculator...

Doughy83, Smanches, appreciate your help with the examples on calculating the charge, ramp up time and relevance to mosfet duty cycle. OK, the fog lights are starting to come on... starting to make a little sense.

Ordered some parts and what looked to me to be a reasonable mosfet yesterday. Will look into the ramp up time for the other driver I'm looking at and try out Doughy83's circuit and 555 timer when the stuff arrives. Ordered spares in case I make some smoke .