View Single Post
Old 03-02-2011, 02:03 AM   #4426 (permalink)
MPaulHolmes
PaulH
 
MPaulHolmes's Avatar
 
Join Date: Feb 2008
Location: Maricopa, AZ (sort of. Actually outside of town)
Posts: 3,832

Michael's Electric Beetle - '71 Volkswagen Superbeetle 500000
Thanks: 1,362
Thanked 1,202 Times in 765 Posts
That's pretty, Harlequin2...

My Mill broke today! Well, the stupid motor starting capacitor shorted out. I opened the case, and flipped the switch, and it looked like some sort of faraday cage experiment:


I ordered a new capacitor today and hopefully the other capacitor is still good. crap. I didn't think to check that one too. I'm an idiot.


Since I couldn't do no milling today, gal dern it, I decided to finally test the melexis HB sensor. I drilled mounting holes in 1/4" x 3/4" bus bar, and also the big 1000amp controller 3/8" x 3/4" bus bar. Only had time to test on the 1000amp bus bar. Results were 1000 amps corresponds to an increase in voltage of 1.74v. The linear range is 2v, so it's well inside that. The spacing was 0.125" above the bus bar. I made a pcb that will be sent in soon, since the melexis HB requires capacitors that the low field version didn't require. This made it difficult to make on the mill:


One of the 3 heat spreaders, with the 4 big holes drilled, and the top milled nice and flat:


B+ and B- Bus bars. Notice that they will be mounted so the skinny side is in contact with the board. This should reduce the stray inductance, since it cut out around 3/8" from the top, and 3/8" from the bottom path length:


M- bus bar:


Right in the dang middle of the stupid drilling!!! The stupid motor start capacitor broke!:


The other 2 copper heat spreader bars:


Steel bar:


Those boards should be here maybe in a couple more days! And um, I can't remember what else I was going to say. dang it.
__________________
kits and boards

Last edited by MPaulHolmes; 03-02-2011 at 03:37 PM.. Reason: they were dang heat spreader bars, not bus bars.
  Reply With Quote