On Monday 14 December 2015 14:54:54 Gene Heskett wrote: > On Sunday 13 December 2015 16:39:16 andy pugh wrote: > > On 13 December 2015 at 11:01, Gene Heskett <[email protected]> wrote: > > >> Feed the cap bank through a resistor. > > >> Use a (cheap $5) solid state relay to bypass that resistor when > > >> LinuxCNC is turned on. > > > > > > I had largely discounted that idea because the first cycle surge > > > is likely high enough to fuse a 40 amp Crydom. It trips a 20 amp > > > service breaker instantly when its master switch is closed. That > > > would also be complicated by needing a high side drive for it. > > > > The relay won't see the surge, because what the relay does is bypass > > the surge absorbing resistor. > > > > The relay would go on the AC input side, so you would only need one. > > > > You don't need a "high side driver", you need a 5V signal from > > LinuxCNC. > > I got that grokked finally Andy, so I just ordered 4 of them, 40 a, > 480 volt from MPJones, a tad over a $50 bill by USPS. Now to find a > pair of uncommitted BoB pins. I'll keep one for a spare, but the third > one will replace the ice cube I am controlling the vacuum cleaner > with. I'll use one to switch the line, one to bypass the resistor in a > few seconds, and one to switch the vacuum. > > Cheers, Gene Heskett
Progress report of sorts. I gave up trying to find the correct baseing for the 2n2222a's I had so I drove up to Bridgeport and stopped at our surviving radio shack and made the usual comments about the place looking like a good place to open a radio shack, heck, even the sign on the front door looks good. Wandered back to where they had moved the parts drawers, and lo, and behold even, their zero fill was at work, so there was a bag of assorted small signal npn transistors which I transferred to my hand, and over in another drawer was some IRF511's, so I emptied that divider too, coming home with 2 of those, all for less than 9 bucks. So this morning, I rip out the 2n2222a wireing and bugs, and wire up the IRF511's (yeah, its 5000x overkill, but that is what Rome had). I had originally thought I should be using Schottky diodes for the rectifiers, and I am borderline right, it works as long as the function generator is making 4.85 volts p-p or more, but drops out at about 4.7 volts. These things need about +3.75 on the gate to begin to turn them on. I have 1n914's in there now, si types, but the Schottky ought to make it work at 4 volts p-p. However the teeny psu is making around 36 volts, so the SSR's are getting 35+ when the p-p is 5 volts or more, and thats above the rated 32 volt input voltage. So, do I halve the voltage by moving the tranny lead to only use half its 25.2, or is this within the ratings of one of these Chinese 40 amp Crydom look-a-likes? According to the curves, that would raise the threshold voltage to turn on by a few millivolts and I'd probably need the Schottky's to work well. Or I'd have to up the load r on the cF capacitor, currently 1 megohm now. Shutoff time is a bit quicker than on time. I left it running with one of them on, but no loading, and if in an hour, the red led on the SSR is still on, I believe I'll wire it into the psu later this afternoon. The hal file is already fixed to run it. Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Genes Web page <http://geneslinuxbox.net:6309/gene> ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
