# Mosfet type to drive a 100A DC motor using pwm

I have a bunch of custom made DC motors. They are supposed to work in a range between 80Vdc and 120Vdc. Nominal power is around 800-1000W (they're custom made for a custom application, so specs are somewhat foggy...).

Now they start just by switching a clean contact, I'm requested to give them a ramp up and be able to choose the rotating speed (not the sense of rotation).

I'm planning to drive the motors using PWM and some industrial microcontroller or PLC (although I'll be prototyping using an Arduino), but I'm having the worst time finding a suitable MOSFET to drive such a bulky motor. Any recommendation on that matter is greatly appreciated.

UPDATE: So, I ended up using an IRFB4115PbF mosfet, and it works great.

• Are you planning on running 10 of these motors from one 100A Mosfet? At that Vdc and nominal wattage each single motor only takes about 10A. Even if you want to run them all from a single control signal you might find more 10A-20A parts as opposed to a 100A part, (then you could use one Mosfet on each motor). Cost of a 100A part (if it exists) compared to the lower amp rated parts will be a big factor too. – Nedd Feb 6 '15 at 7:09
• Well, it seems they do exist at not such an outrageous price: uk.farnell.com/international-rectifier/irfp4668pbf/… . . . . uk.farnell.com/infineon/ipb072n15n3-g/… – Nedd Feb 6 '15 at 7:18
• @Nedd nope, it's an automotive application, I'm planning to equip each motor with its own electronics – Miguel El Merendero Feb 6 '15 at 7:33
• So are you expecting a 100A surge or stall current on a 10 motor? – Nedd Feb 6 '15 at 7:39
• Last measurements showed a "spike" at about 90A for less than a second while starting up, and a constant consumption of about 10 to 20Amps in normal operation, (load on the motor is variable). No data on the motor stalling, it's never actually stalling because of a mechanical clutch. – Miguel El Merendero Feb 6 '15 at 7:46

You may want an IGBT for this; the way to use them is very similar to a MOSFET but they tend to be much more sturdy devices (for instance, all IGBTs rated at >100A seem to have voltage ratings >250V as well!). This is important because of inductive kick that is quite easy to get with motor loads. MOSFETs are in theory protected by their body diodes, but in practice need a snubber circuit as well for reliability.

An example of a suitable IGBT is AOK50B60D1, a 100A, 600V part for $2.13 @ 1k. It is rated at 300W, but would likely dissipate quite a bit less in your case. If you do want a MOSFET, try something like IRFB4115PBF, 150V, 104A,$1.87; or IPP048N12N3.

EDIT

A bit more detail: IGBT vs MOSFET: Choose Wisely, by IRF

IGBTs have been the preferred device under these conditions:

• Low duty cycle
• Low frequency (<20kHz)
• Narrow or small line or load variations
• High-voltage applications (>1000V)
• Operation at high junction temperature is allowed (>100 ° C)
• >5kW output power

Typical IGBT applications include:

• Motor control: Frequency <20kHz, short circuit/in-rush limit protection
• Uninterruptible power supply (UPS): Constant load, typically low frequency
• Welding: High average current, low frequency (<50kHz), ZVS circuitry
• Low-power lighting: Low frequency (<100kHz)

MOSFETs are preferred in:

• High frequency applications (>200kHz)
• Wide line or load variations
• Long duty cycles
• Low-voltage applications (<250V)
• < 500W output power

Typical MOSFET applications include:

• Switch mode power supplies (SMPS): Hard switching above 200kHz
• Switch mode power supplies (SMPS): ZVS below 1000 watts
• Battery charging
• The igbt is a great piece of advice. Dumb me didn't even think about that. I see all of these components do not accept 5V logic levels as input, may I ask you to point me to a simple coupler or interface ic? – Miguel El Merendero Feb 6 '15 at 7:35
• IGBT devices are wonderful for high-voltage electronics. However, they have significantly higher Vce on voltage than low-voltage MOSFETs. Searching for "IGBT" on Digikey and looking at individual devices (as opposed to arrays) shows that the lowest Vce ON voltage of any device that Digikey has in stock is about 1.15V. It gets worse from there. – Dwayne Reid Feb 7 '15 at 0:16

Large MOSFETs are surprisingly inexpensive. I just did a quick search at Digikey: voltage greater than 200V, current greater than greater than 100A. The least expensive device is a IXTH110N25T-ND: 250V, 110A continuous, 24 milli Ohms (which is a little high). But there is a couple of dozen other choices available, ranging all the way up to 230 Amps.

Obviously, you want to use PWM to ramp them to whatever speed your target is. Depending upon what you want, this can be as simple as a 555 timer or small microcontroller. However, you will want to use a proper FET driver to feed your FET.

• 110A * 24 milliOhms = 2.6V. This is almost exactly tied with the Vce of the IGBT I suggested :) This is at room temperature: the situation changes at higher temperature in favor of the IGBT, I'm afraid... – Alex I Feb 7 '15 at 1:00
• @Alex I: yep - I choose to show the least expensive MOSFET that was at least 200V and 100A. For a few pennies more, we get the IRFP4668PBF-ND: 200V, 130A continuous, 9.7 milli Ohms. My answer wasn't intended to specify which part the OP should use, rather, to show that other choices are available and that he / she should see what is readily available and choose an appropriate device. – Dwayne Reid Feb 7 '15 at 1:12

Please, just take servo. Any project of this kind will kill you, unless you have time money and knowledge. On the other hand there are servo available for like \$300 that do anything you need. Only develop your own, if you are going to produce high volume series.