Re: Power MOSFETs for Essisted bike.

[jcmassey@netspace.net.au]

At 11:02 PM 30/05/01 -0400, you wrote:
>James-
>
>I have a part number or two to try. I have used the International 
>Rectifier IRFZ48V at 60V and 72A. But it's new and hard to find. 
>Newark lists stock on the 1010E, a 60V  and 71 or 82 amp part 
>(depends on the datasheet I read). It has a 0.012 ohm RDSon. I have 
>never used it but it seems more common. And $3.18US each. But I have 
>to tell you, I am not an EE, and so my knowledge on the selection of 
>these or any FET is quite weak. You get what you pay for with free 
>advice :)
>
>Seth

Farnell (who I think own Newark) have IRFZ48N [not -V] (55V, 53A, 0.016
ohm) A$18.68 (about US$9.70)
They also have a Siliconix device that looks like it would be the best
device for the Essisted bike - 60V, 75A 0.008 ohm, SUP75N06-08, Au$27.06ea
(about US$14.00ea). 
Best device for $ at this voltage I can find here is 60V, 80A, 0.01 ohm,
STP80NE06-10, Au$10.97 ea (about US$5.70), an ST device.
 
A couple of STP80NE06-10 I think will be on my next Farnell order. Price
doesn't seem too bad. Going to be running at 24V (possibly uprated later to
36V) on a 24V series motor (nominally 3/4HP, 15min. duty). Expect to be
running up to 25Amps (hilly town, hence the possibility of needing to go up
to 36V). Controller design allows for full-on FETs, so two in paralell
should dissapate (25A @ 0.005 ohm = 0.125V, 0.125V x 25A) = 3W!! nice cool
controller. Important, since it'll be in the frame between the riders'
legs. Mind you, when at 50% duty cycle it could be another story, 25A
through the freewheel diode at 0.6V for 50% of the time = 7.5W, plus
unknown heating during switching. Will see how it goes. Once controller is
complete I'll post some performance figures. For thise who have asked
off-list I'll be sending a bunch of JPEG's and some text on how I have done
my bike. Hope to be getting the JPEGs done this weekend (if my friend with
his digital camera ever turns up).

I'm still looking for a device I can get here that would have either 150A
or 200A continuous with a 500-600A pulse, that won't cost more than a 500A
controller by the time I buy the 3 or 4 needed. I'm working out the pro's
and cons' of building my own EV controller for 72 or 96V at 500A. If I
build, I can fix..... big consideration when freighting times in weeks each
way could be incurred. Then again, a good manufactured controller working
well within its limits won't blow up (or do pigs' fly).

James Massey