|
Valen
Experienced Roboteer
Joined: 07 Jul 2004
Posts: 4436
Location: Sydney
|
lower PWM frequency is better for the fets but worse for the motor.
Given the size of the motor your driving you might get away with something in the 500Hz range (more inductance means lower ripple), The switching period Vs the "charge" period is such that the "stiffness" of the power supply is largely irrelevant.
Your also going to want to add some nifty diodes across the motor to help with the freewheeling current when the switches are off (rather than passing it through the fet body diodes) High speed and low forward voltage is what your after but the current doesn't matter that much, you can probably overrate them by double or more.
In terms of fet selection your main interest is in RDS on. The lower that is the less heat the thing is going to generate. Thats why there are 300A fets in 50A packages.
I have a design for a 6x 1405 setup thats about as big as your thumb, its been working on the bench (well it was a while ago), I haven't run it as a low side switch but it should be good for 200-300A or so if you can get the heat out adequately. Thats with 3khz PWM. It just needs a brain, power supply and the software written for it, but that should be pretty straight forward. Whole thing will probably look a bunch like our current bolt switch, 2x ring terminals with some LED's and a servo plug.
I'll see how I'm going later in the week, work is a bit full on at the moment.
A thing to look for would be the power supply chip, I think something in the way of an automotive grade voltage regulator, around 100-300ma output at 14 volts or so.
The hard part will be finding one thats happy with a maximum input voltage of around 28-36 volts and an absolute max of 50.
Otherwise we'll just get a high voltage regulator and roll some protection circuitry in front of it. _________________ Mechanical engineers build weapons, civil engineers build targets
|
Tue Jun 24, 2008 10:52 pm |
|
|
|
|
|
|
|
|
|
|