PWM motor circuit and diode

[Chris Smith]

The advantage of one FET driving another FET is that the first tiny FET can use the medium level logic from your signal, and turn it into a completely unloaded or light load “snap on” decision with no ramping effect from a transistor, or saturation problems, which then drives the second FET gate properly because it can handle this gate with out effect or loss. <p>The first FET can use up to a 12 volts source and back to ground via a dummy load to make up for any deficiencies in switching voltage drop, to drive the gate of the second FET. <p>Its a signal booster for the voltage level in the signal to the main FET. <p>A transistor has many draw backs, It has resistance, saturation etc. where as the FET is “voltage present” at the gate, to drive another simple gate which then snaps on and off. <p>It acts like a darlington to boost the voltage value to the second gate. <p>Transistors act on current, and fets need volatge.

[positronicle]

--Edited by Positronicle--

[Jesse]

Thanks guys, your electronics experience is certainly appreciated.<p>Jesse

[Jesse]

By the way, I hope all of this will lead of to my simple problem of a HOT diode, and being able to use a more "normal" size one. Worse case: I'll just heatsink a TO-220 package diode.<p>I have a darlington array that I'm going to experiment with now with PWM. Should be able to switch fast enough. We'll see what happens...<p>Jesse

[upsmaster]

try a fast recovery diode or increase the piv voltage rating.<p>
joe

[Chris Smith]

Darlingtons switch slower. <p>They must saturate all of the gain stages to operate, which makes the sensitivity increase, but the switching times slower. <p>Darlington FET arrays have a better response time because of less capacitance VS more saturation time.

[Jesse]

Thanks Joe. The diodes I'm using, now that I look at them, are "general purpose" rectifier diodes (these particular ones are from Radio Shack). Usually I get my stuff from Digi-Key or others, but I didn't have any on hand large enough, so had to make due. I do have some fast recovery diodes on their way, so that very well may make a huge difference.<p>By the way, my darlington array idea is dead, as it requires a reverse voltage (it would ground out my Gate pin on the MOSFET rather than sending it a good +5v).<p>Jesse

[positronicle]

--Edited by Positronicle--

[rshayes]

"Hot" means different things to different people.<p>Semiconductor manufacturers want to make their ratings as high as possible. That 3 amp diode is probably rated based on a high junction temperature (probably 150 centigrade and possibly higher). With lead mounted diodes, they usually specify some heat sink on the leads, usually a certain area of copper very close to the package with an ambient temperature of 25 centigrade.<p>When you put your finger on that package, it is within 1/8 inch of an area that is above the temperature of boiling water.<p>With 2 amps of current, the diode is probably disipating about 1-1/2 watts. Some heat sink may be needed at this level.<p>Diodes are cheap. Useing a 5 or 10 amp diode is probably the best solution. A TO-220 package with a small heat sink will probably run reasonably cool.

[Robert Reed]

" Lets try this again. Feed the raw pwm signal coming from your micro, to the base of the npn, connect the emitter to ground"<p>
A base emittert junction presents a very low impedance.
Won't this pull a lot of current and hog voltage?

[positronicle]

--Edited by Positronicle--

[positronicle]

--Edited by Positronicle--

[Chris Smith]

A small signal FET as a signal driver can be run from 12 volts, on through the FET to a dropping resistor of high value, and the signal is tapped before the resistor for a VOLTAGE signal to the second FET gate. <p>This way the first FET actually acts as a "GATE Signal" only, and doesn’t draw anything per se in the way of amperage. And its fast. <p>The tap merely forms a signal voltage for the second FET.<p> Tiny FET, tiny draw,... from the original signal input, amplified [voltage] for the second FET gate.<p>[ August 31, 2005: Message edited by: Chris Smith ]</p>

[Ron H]

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by positronicle:
Adding a base resistor of 2.2k or so would probably be a good idea.
Here's an updated schematic.
[URL=<a href="http://imageshack.us" target="_blank">:D<p>[ September 01, 2005: Message edited by: RonH ]</p>

[rstofer]

You might consider the Intelligent Power Switch http://tinyurl.com/b9oom They're a little pricey at $2.52 (Digikey) but all of the MOSFET driving nonsense has been dealt with.<p>You can see other devices in the Automotive section at www.irf.com