New DIY BLDC controller

Hi,

I designed new BLDC controller based on ATMega8 microcontroller, intended to use in robots. Diagram available here:

formatting link
Any comments are welcome.

Regards, Michal Goralczyk

Reply to
Michal Goralczyk
Loading thread data ...

Your FET drivers seem to be redundant - you already have 5 volt drive from the micro, and FETs dont need any current. Looks like redundant series resistors on you input filters, too.

Luhan

Reply to
Luhan

"Luhan" wrote in news: snipped-for-privacy@75g2000cwc.googlegroups.com:

Luhan,

I hope you said that with tongue-in-cheek. Do you think that all of these power-MOSFET driver IC's were created as brother-in-law components? ;)

Seriously, yes, you can drive a some MOSFET's directly off a PIC or micro. A 2N7000 is great in this. Power MOSFET's, however, do require a lot of gate current, in an inital short pulse to charge and discharge the gate-source capacitance as the MOSFET is turned on or off. If there isn't a good source/sink of this high current (micro's have very limited output currents) the power MOSFET will spend a lot of time heating up in it's linear region before it turns completely on or off. Not good in driving high-current loads.

Goggle "MOSFET driver".

Here's a good place to start:

formatting link

Good learning,

Ken

Reply to
Ken Moffett

Ken,

I'm a bit new in this area. I have noticed some limitations of driving MOSFETs from PICs directly. It depends on the application. For smaller motors, it seems to work just fine. In other cases, using Darlingtons instead of FETs eliminates the input capacitance issue.

I'm not sure of the OPs motor size and current requirements. Also, some of those multiple resistors-in-series dont seem to serve any function.

Luhan

Reply to
Luhan

"Luhan" wrote in news:1154829040.937063.223550@

75g2000cwc.googlegroups.com:

Luhan,

I think I was caught by your blanket statement "and FETs dont need any current". :) The resistor pairs look like the OP was trying for specific precision values (i.e. 100K+15K=115K and 22K+22K=44K), but I don't know why. Maybe Michal will come back and enlighten us.

Ken

Reply to
Ken Moffett

Ken,

My blanket statement was incorrect. However...

In this circuit, which appears to be feeding a 3 phase motor, it looked like the switching speed from phase to phase was around 100 Hz or so. In that case, you could spend a few hundred microseconds transitioning the FETs without them heating up. Also, some of the current spike would get eaten up by motor inductance.

Here is the only actual (working) circuit using a MOSFET that I have done to date...

formatting link

Luhan

Reply to
Luhan

"Luhan" wrote in news: snipped-for-privacy@i3g2000cwc.googlegroups.com:

Luhan,

Slick! Somehow being the target of that makes "shooting victim" sound like a good thing. ;)

Can you accept a couple of 1 meg images of circuits where I've used MOSFET's?

Ken

Reply to
Ken Moffett

Ken,

No problem, I got DSL. Use the email address on my site - GIF image near bottom of home page (click on logo for home page).

Luhan

Reply to
Luhan

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.