Re: [Emc-users] Commutating Reactor

[gene h. <ghe...@wd...>]

On Thursday, May 31, 2012 11:28:24 AM John Thornton did opine:

> I'm sure fresh bearings would help the 15hp and it is a heavy duty TEFC
> motor. I might pull it apart today and order some new ones...
> 
> I didn't measure the temperature of the 10hp and only ran a few minutes
> but I'll keep an eye on it.
> 
> I didn't ask for taps on the transformer but it won't hurt to find out
> the price. Input taps for 220 and 240?

Since the existing autoformer has taps for 208, 220 and 240, I'd say an 
extra at 220, so you could use it for a 220 to 240 step up, or conversely 
swap primary & secondary & use it for a 240 to 220 step down would cover 
the options nicely when combined with the autoformers own taps.

The N terminal on the existing autoformer can then be taken back to neutral 
in the service, thereby forcing the secondary side to become neutral 
referenced.  If this tranny comes with N connections on both windings, the 
N connection on whatever winding is used for drive to the existing 
autoformer could also be tied to this neutral run, but that would be more 
for correctness than any real advantage since it's the N terminal of the 
OEM autoformer that counts here.

I'd check my copy of the NEC, but it's either grown legs or belonged to the 
tv station.  In any event its a '98 issue and could be obsolete.  I haven't 
stumbled over it in 3 or 4 years.

> How is the encoder disk coming along?
> 
> John
> 
> On 5/31/2012 7:21 AM, gene heskett wrote:

So-so. I, took that printout out there yesterday, but being somewhat fam 
with the old code, and having a supply of #58 & #60 pcb drills, I massaged 
that code for the smaller bits and tried to make a 60 slot encoder, then, 
trying to get the exec time down (its about 2.75 hours with that small a 
bit) I pushed too hard & broke the last #58, so I slowed down a bit more.  
So I've broken about $35 worth of bits now to get one poorly cut disk.  I 
added a duty cycle fudge value where the original code just divided by 2.0.

Since mulls are faster, the disk I made yesterday now has a mull by 
duty_cycle, with duty cycle at .45.  Next one will be at .40 to equalize 
the opto outputs to 50%, they were about 57/43 when I put it on the lathe.

I have the A/B slot ring about correct if I reduce the duty cycle from .45 
to .40, I should get pretty close to a 50% duty cycle out of the opto 
devices.  I can bend them back and forth perhaps 10 thou to fine tune the 
quadrature timing, or adjust the slot count one way or the other by one 
count.  But I think I am going to have to dig another 5 thou deeper into 
the piece of oil soaked oak used for sacrificial in order to remove the 
last vestiges of the fact that its not an end mill, but a drill bit with 
fairly steep angles. It left a fin at the bottom of the slot.  

Going from 39 to 60 slots, I had to adjust the encoder's scale & negate it 
but that was half expected.  And I have too much gain, the speed is hunting 
about 10%.

Then, I am going to have to separate the slot cutting and inner hole cutout 
from each other in the codes looping so this thinner stuff has less 
opportunity to flex, and after fighting with the index slot, trying to 
maintain its tapered shape, it occurred to me that a simple slot perhaps a 
thou wider than the bit itself is all that needs since I don't think it 
needs a specific timeing reference the A/B slot ring. So that is the next 
change for this code. The .001 off line movement there is about the limit 
of my mill, but would be intended to assure that only one edge of the drill 
is doing the cutting. When both edges cut, the cut isn't as clean as the 
bit flexes.  I ordered another 5 pack of each bit size yesterday. $89.

Another thing that I question is the math that does the tapering of the 
slots, I have a feeling its not quite right, but cannot put my finger on 
it.  That is about a screen full of code in vim all by itself just to make 
the 5 moves that carve the slot at each depth increment, and the last 
movement should be to the origin of the end of the first movement, but 
there is a small gap when you zoom in the backplot to look at slot 0 and 
the index or to any individual slot.  The error is at least consistent 
though so should be fixable.

I haven't started on that yet for today, I need to make a trip out to get 
some blood pressure & anti-sugar pills & some groceries first.  And since 
the temps are decent, I need to fire up the weed eater & finish trimming my 
jungle too.  And when I do get to the shop, from the looks of things in the 
mills head, I need to replace one of those plastic gears, a hub is starting 
to break out and the gear rim is flopping up & down about 1/16" with 
attendant rattling noises.  There was, 10 years ago, a belt drive kit for 
this thing that gave a max rpm of 5000 but that was discoed several years 
ago.  Damnit...

If they ever get my nose cleaned up for good, I'd like to get a G0704 & a 
ball screw kit for it, it would *2 my xy envelope.  Z I have lots of thanks 
to the all new Z drive I built several years ago.

That old saw about wishes, horses and beggars comes to mind. Dreaming is of 
course free. :)

Thanks John.

Cheers, Gene
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
My web page: <http://coyoteden.dyndns-free.com:85/gene>
Most people want either less corruption or more of a chance to
participate in it.