Hello!
This is my first time using a mosfet driver for my project that's why I want to see if what I'm doing is right.

I just need to turn on and off the mosfets from my microcontroller and it will just short it directly to the ground for a high current result. I will be using this for my spot welder project.

Here is the schematic:

Sorry if the schematic is not that clear.
I uploaded the datasheets of the driver and the mosfets. Thanks

 

hi,
The IRFP2907 requires a Gate driver voltage of ~12V in order to ensure that its switched ON 'hard'.
The MOSFET's also require a low value balancing Source resistors to ensure that the 3 FET's currents are 'balanced', ie: ideally each carrying a third of the total load current.

What load current does the spot welder draw.?
E.

 

hi Leo,
This LTSpice simulation shows the effect of using a Vgs over the range of +5V thru +12V.

Note: plotted total current and the dissipation in M2 MOSFET.

Used a IRF2907 as substitute for IRFP2907.
E

 

Agreed with ericgibbs,

The driver might need 12V supply as your mosfet is not a logic-level mosfet.

Some kind of a series-resistor might also needed for each gate of the mosfet, to avoid the driver being overloaded by the capacitive load.

 

Ooh I guess I misunderstood the datasheet. I thought this driver can compensate for the lack of voltage of uC to drive non-logic mosfets with just 5V. Maybe by changing the supply voltage of the driver between 12V-18V(which is the driver's maximum supply voltage) I can properly drive the mosfet?

I find it hard to implement the source resistance needed since I needed to produce as much current as I could for a good weld and looking for a small resistance with a high enough wattage would be hard?
I'm hoping to produce about 500A(though I can't really compute it exactly since most of the resistance will be coming from the imperfections of the connections) from the capacitor and handle it through 4 parallel mosfets rated at 209A so I guess a little unbalance will just be alright?

Thank you for your responses

 

The ON resistance of MOSFETs has a positive temperature coefficient so they tend to automatically share current. The one carrying more current will heat up, increasing its resistance, which then lowers its share of the load. I think you will be okay with no sharing resistors.

 

As Carl points out, the MOSFET's will try to self balance, but be sure to try and keep all the FET Source pins external resistive paths to 0V as equal as possible, to a 'star' connection point.

You will require a 'big' heatsink, refer to the simulation which shows the approx dissipation of each FET.
E.

 

Noted!
Just one last question regarding the fully turning on of the mosfet using the driver. Since it's not a logic mosfet, I would be needing at least 12Vgs to fully turn it on. Can I make the output of the driver reach 12V by supplying it with 12V? Does its output voltage equals its supply voltage? With that given, can the input from uC drive it properly? If all of that is not true then how?

I read the datasheet and that's what I concluded. I just don't know if it is right.

 

hi leo,
This clip from the d/s shows that the Vout will be close too the
[Vsupply - 0.025V] , so a Vsup of +12V will be OK.

The 4424Z will accept an input of Vhigh as low as +2.4V, so thats also OK, when the MCU is powered by 5V.

E.

 

YIPEEE...

The most knowledgeable electronics person I know has decided to start helping here

God Bless you Eric. You are a good person. Just watch AAC and the Technical knowledge base here shoot through the roof.

You are the cherry on the top.

Your Friend,
tvtech

 

Thank you very much for all of your help especially Eric! I've also tried breadboarding the circuit a while ago and also obtained the desired result!