UFO politics Keeps his word 12-12 12 Let the games begin.......

[TinselKoala]

Did this thread go dead?

I see that they are contemplating using Arduino for controlling things. Good luck to them. I'm glad they are getting good expert advice, too.   

Here's a little bit of information that might be useful to someone, someday.

Switching an IRF530 mosfet directly from the Arduino's digital output pin:
http://www.youtube.com/watch?v=wxxPWbDqIcA

Using the PWM output of the Arduino to drive an optoisolator which then drives an IRFP460mosfet switching rectified line voltage into a 700 Watt load:
http://www.youtube.com/watch?v=QuaCp4dc9EM

I just pulled 4 ea. 18n50 mosfets from an old PC ps, and I see that UTSource has IRFB4310 for between 1.50 and 3.00 each.

Just trying to be helpful.......

(The Arduino "motor control shield", not shown here, uses twin L293D H-bridge driver chips which could in turn be used to drive 4 mosfets or sets of mosfets at their full required gate voltages.)
Some good info on using the L293D chip with Arduino:
http://www.me.umn.edu/courses/me2011/arduino/technotes/dcmotors/L293/L293.html

[picowatt]

Did this thread go dead?

I see that they are contemplating using Arduino for controlling things. Good luck to them. I'm glad they are getting good expert advice, too.   

Here's a little bit of information that might be useful to someone, someday.

Switching an IRF530 mosfet directly from the Arduino's digital output pin:
http://www.youtube.com/watch?v=wxxPWbDqIcA

Using the PWM output of the Arduino to drive an optoisolator which then drives an IRFP460mosfet switching rectified line voltage into a 700 Watt load:
http://www.youtube.com/watch?v=QuaCp4dc9EM

I just pulled 4 ea. 18n50 mosfets from an old PC ps, and I see that UTSource has IRFB4310 for between 1.50 and 3.00 each.

Just trying to be helpful.......

(The Arduino "motor control shield", not shown here, uses twin L293D H-bridge driver chips which could in turn be used to drive 4 mosfets or sets of mosfets at their full required gate voltages.)
Some good info on using the L293D chip with Arduino:
http://www.me.umn.edu/courses/me2011/arduino/technotes/dcmotors/L293/L293.html

Does anyone know why UFO wants/needs to spin his motor at 10,000RPM?  She's gonna' blow captain... 

As for driving the MOSFETs, it is all about switching speed and a good HF layout and design.  At 100 amps plus of switching current, the object is to have the MOSFETs only be "on" or "off", with the transitions between those two states as short as possible.  This keeps the MOSFET dissipation to a minimum.  A slower driver allows the MOSFET to operate in a "linear" region for a length of time wherein the dissipation would be very high.  The use of a proper gate driver with switching times well under 100 nanoseconds would minimize MOSFET dissipation.  Although 20-40 nanosecond transitions would be preferred, 100 nanoseconds requires a minimum circuit bandwidth of 50MHz to produce a decent square wave at that speed, with more BW than that preferred.

The buffered 555 circuit discussed over there, while better than just a 555, is not the equal of a high speed gate driver capable of 10 to 20 amps of gate drive.  Also, in order to charge/discharge a large gate capacitance with up to 20 amp driver pulses, requres good HF layout and design to prevent ringing/overshoot at the gates and driver supply.  At the MOSFET drains, commutator noise will be an issue, and protection such as a snubber should be considered.

Now, more complex four phase switching is being considered.  While the use of an Arduino a is a possibility, a proper motor control program should include RPM feedback from the motor and a bit more complexity in code than just a PWM determined by RPM (such as accel/decel ramp, current limit, etc).  Even a simple RPM referenced PWM control will require some degree of hardware or software damping to prevent speed "hunting" within the control loop.  Of course you could also put a human in the control loop and just manually adjust speed with  pot as load and prime mover speed changes.   

If it was known that a PWM circuit was going to be needed to prove OU, it seems it would have been much simpler to do a proof of concept using the smaller motors that were modified, as their current draw was much less than the 100 amps plus being discussed.

I believe I read that these modified motors were not supposed to require additional supply current as the load increased, which does not seem to be the case so far.

Possibly UFO will return to repeating his efficiency tests using his Prony and do a comparison to a non-modified motor.  It seems pointless to move forward with a possibly complex and costly controller design project if the motor itself is not all that efficient.

But then, possibly not...

PW 



   

[MileHigh]

PW:

You are in a completely different world....

... The real world.

I think that you mentioned that you are retired.  It sounds to me like you could easily be a $300/hr consultant.

MileHigh

[picowatt]

I see UFO also spun a bearing during his high speed test.  Again, anyone know why he has decided he needs this kind of speed?

Pushing !0,000RPM without dynamic balance and mechanical strength analysis/testing seems a bit scarey to me.

I thought he needed only 3600RPM for driving the AC gen...


PW

[tinman]

I see UFO also spun a bearing during his high speed test.  Again, anyone know why he has decided he needs this kind of speed?

Pushing !0,000RPM without dynamic balance and mechanical strength analysis/testing seems a bit scarey to me.

I thought he needed only 3600RPM for driving the AC gen...


PW

The one thing you wont see PW is ufo doing the same test he did on his moded motor with an of the shelf motor.
I have done the test ,his way and the correct way.
Useing his method,my tired old 24 volt motor had an efficiency of 130% plus.
Useing the correct method,i came within 4% of the manufacture's spec for that motor-about 74% efficient.
Now useing UFO's own measurments,his motor(when calculated correctly) is about 70.4% efficient.

So im guessing he is now going for the land speed record-or something like that.
I think i would just go and buy one of these motor's if i wanted high RPM-and only $480.00 here in OZ.
http://www.400hertz.net/Products/ME-C500-75WC.htm

In reguards to the motor speed controller,if it is going to be under 20khz on the frequency-would it not be better to use IGBT's?
They handle current much better than a mosfet,and are what most HF tig welders use today.
My little caddy is happy at 140 amps all day,or 170 amps at 60% duty cycle.
Infact it would be quite easy to convert one of these to a motor speed controller.

Maybe UFO is going for speed now to keep the interest there?