On Saturday 20 April 2019 00:10:13 John Dammeyer wrote:

> I've been using the Tormach TT type tooling.  So far the R8 tool
> holder hasn't slipped.  I really like the fact that I can insert the
> tooling without screwing around with the R8.  Some of the R8's are
> really tight in my spindle.  Others slip in nicely.   Far east
> inexpensive is what I got and I got what I paid for.
>
> Since I eventually want a tool changer the TT stuff looks like it will
> work and I'm not that hard on my machine for aggressive cuts anyway. 
> But seeing the construction of the ISO30 units and that I find the
> bearings on my Canadian House of Tools version of the Grizzly G3616
> noisy it might be an interesting exercise to see if the tool holder
> can be changed.  The quill is pretty big in diameter.
>
> John
>
One item might be related to how you are driving the spindle. Here I 
quite early on junked the manual scr based speed control, building a 
honking big PSU good for about 127 volts DC and about 20 amps, somewhat 
overkill for a 1 hp PMDC motor rated at 90 volts and 9.7 nameplate amps. 
With one of the Pico System pwm servo amplifiers driving the motor from 
that huge PSU, I'm probably pushing that motor to around 2 hp.

I made an optical encoder to replace the disk full of holes the 
old "hangs on the side of the head" controller used to show you the 
speed, with the optical having (IIRC) 68 slots and one of them a long 
slot for a 3rd opto for the index. Fairly decent quadrature, but 
quantization noise was pretty bad, heavily rattling the gears in the 
head even with PGains below 3 in the spindle pid. Made the head sound as 
if the bearings were all square balls, crunchy as hell.

About a year ago, I made a brass, 5mm extension about 12mm long and 
installed it at the top of the motor by drilling and tapping the end of 
the shaft, and put a 1000 line Omrom encoder on top of the motor.  Using 
just the index from the old optical, and some rs485 to ttl pieces 
because the Omron had differential outputs and the 5i25 needed ttl 
inputs to its encoder, I wrote some hal code to determine the scale for 
such a rube goldberg setup by counting the encoder for 100 index pulses, 
which turned out to be a scale setting of something over 7000 in high 
gear, and something over 14000 in low gear. In the mean time I had cut a 
notch in the edge of the gearshift knob and made up a thin sheet metal 
sandwich to hold a couple roller tipped switches adjusted to just close 
one or the other switch when the knob was fully seated in the high or 
low gear positions.  I used a mux4 to switch the scales used, which 
meant I had 2 unused states, so when it wasn't in one gear or the other, 
I fed about a 10 rpm command to the motor making live gearshifts a piece 
of cake, no longer having to grab and turn the spindle till the gears 
meshed.

The end result is that quantization noise is now so low that PGains of 20 
to 40 in the spindle PID are now practical, speed regulation is beyond 
being stiff, I can no longer hear how hard its working until the servo 
amp starts chirping as it currant limits, and with all that noise gone, 
so is the gear rattling, so I figure with that noise gone, the plastic 
gears will last until I finally strip them doing a rigid tap, which I've 
since done up to 10mm in 1/2" alu all the way thru. Jon's servo amp hits 
the 17 amp limit and chirps some as it limits, but it does cut the 
thread.

I generally drive the tap with code that pecks at it, cutting less than 
1/2 turn per peck cycle when the tap is that big or its not a "machine" 
tap meaning the gullets will fill up, so it needs a backout move to blow 
the tap clean and re-anoint it with cutting oil.

If you use one of those servo amps to drive a spindle, let Jon know when 
you order it and he will change the toroids so it runs a bunch cooler in 
continuous duty. Its a great spindle driver for a treadmill motor, which 
I have on TLM. I've retrofitted a 1hp on a 7x12.  That and tapered gibs, 
plus getting rid of the compound in favor of solid cast iron to hold the 
QCTP makes a whole new machine out of one of those pieces of junk. FWIW, 
there's no compound on the Sheldon either, it was broken beyond 
salvation when I bought it. From that and other clues, the Sheldon had 
fallen over in a previous life, with a heavy chuck mounted as the 
spindle was and is yet, bent. I've reground the MT5 and remachined the 
backing plates so it all runs true again 

The point is, the sound of crushed bearings in the G0704 head is gone, 
and I haven't replaced them. Oh, and the max spindle speed goes up from 
a rubbery 2250 to 2950 revs. In either direction...  Till the cows come 
home.

> > -----Original Message-----
> > From: Gene Heskett [mailto:ghesk...@shentel.net]
> > Sent: April-19-19 7:55 PM
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] ER ATC tool idea
> >
> > On Friday 19 April 2019 18:21:19 John Dammeyer wrote:
> > > Gene,
> > > See below.
> > >
> > > > > And if you can create a taper chuck for the lathe is there any
> > > > > reason one couldn't create a new mill spindle with that taper
> > > > > to replace an R8 spindle?
> > > >
> > > > And how then do you fix a draw bar holder?
> > >
> > > In the lathe through the bore.
> >
> > I'm doing that now, for big ER's in that sheldon, but I had the
> > G0704 spindle in mind, I've developed an abiding hate for R8's and
> > their tool slippage.
> >
> > > Or are you talking about making a new
> > > spindle? John
> > >
> > >
> > >
> > >
> > > _______________________________________________
> > > Emc-users mailing list
> > > Emc-users@lists.sourceforge.net
> > > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
> > 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
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> _______________________________________________
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


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
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to