Greetings;

Yesterday I did the threads, 1/4-28, .200" long in that nipple, this time 
making it out of an inch of 1/2" grade 8 bolt.  That code has a relatively 
huge leadin, .4" before the thread starts, so that end of it was absolutely 
no problem.  I raised the spindle speed from 200 to 250 revs too. For 
something like that, with a fairly critical fit, I usually start at least 
10 thou big and touch off & rerun by a thou at a time till it does fit.

Sloppy threads there, since one end of the thread is exposed to chamber 
pressures, would tend to leak worse, and possibly even gas cut over time & 
many shots.  So I look them over with a very strong magnifier.

I had originally give the -e option a .009" retract space, but hadn't paid 
much attention to the last thread which is up against a shoulder where the 
.494" diameter #209 cup holder, and the originally .253" diameter of the 
shank where these threads are cut into in.  That shoulder is fairly flat, 
but I have added a nominally .005" step at the root of it to function as a 
gas seal at the root of the threads, where the OEM version had no such 
step, so it sealed against the whole steps area with considerably lower per 
unit of area pressure.  My theory is that by concentrating the threads 
available pull on the much lower area of that step should be a Good Thing 
in terms of the gas seal achieved.

But when it was about 90% cut, I noticed that the last thread against the 
shoulder was obviously undersized because the actual retract motion wasn't 
being done as fast as the -e=.009 was giving it.  Thinking on that, it made 
sense to reset that to about -e= to about .020, or slightly over half a 
revolution.  So I did, and while the damage had been done by then, it at 
least didn't make it any worse by the time I had achieved the desired tight 
fit into the breech plug itself.

Now, at 250 revs, a full thread pitch takes 0.24 seconds, and with a 
nominally 10 thou backlash to take up first, I am amazed that it worked as 
well as it did.

That axis is currently setup with a fairly aggressive profile:
[AXIS_0]
TYPE = LINEAR
HOME = 0.0
MAX_VELOCITY = 0.450
MAX_ACCELERATION = 5.50
STEPGEN_MAXACCEL = 11.00
SCALE = 25400
FERROR = 0.05
MIN_FERROR = 0.01
MIN_LIMIT = -5.01
MAX_LIMIT = 5.50
HOME_OFFSET = 0.0
BACKLASH = 0.010

And BASE_THREAD is 23 u-secs.

But given the atom boards limited step rate possible at that base thread, 
I'd have to assume that the next thing I should do is drop the microstep, 
which I can do, in order to reduce that SCALE, effectively increasing the 
available acceleration until the motor stalls in the backlash move.  It is 
atm running at the drivers max current, 8 wire wired in parallel.

That backlash could be lowered, but it would take a ball screw kit to do it 
as the thread on that bolt is the normal v groove profile, 10mm x 1mm 
pitch.  The nut is bronze and effectively non adjustable because when the 
nut is tilted to reduce the backlash, the wear rate explodes until 2 hours 
later it is back to about 10 thou again.  There isn't room to fit a split 
nut that could be drawn together, I've look at that thought already.  Ditto 
for a ball nut, those I've looked at would need lots of the carriage and 
the cross-slide carved away.

So the real question here is:

What would the math look like that would give me an almost valid -e value 
at a given spindle revs?

That is atm, not one of the values my wrapper code calculates, and it 
should be, as should the spindle speed be in the final version.  Basically 
keeping things "inside" the machines performance envelope.

Keeping in mind the encoder is a 50 cycle, 200 edge encoder which 
translates to a minimum revs needed to achieve fairly smooth Z motion.  By 
the time I have slowed the spindle to 50 rpms, Z is actually stepping some 
nominal number of steps per encoder edge going by.  That is such a small 
movement the threads cut will still be good, but listening to it makes me 
want to put a wagon under it to haul away the corn cobs when the wagon is 
full.  ;-)

I've just spent a couple hours searching for what looks like a better deal, 
but if I want a ball screw, it is the nut I'll have to turn by attaching 
the screw to the crossfeed sled, and turn the nut, mounted in bearings much 
like I did the Z drive for my mill.  Not only bulky but adding unwanted 
weight to the rear of the carriage.  Sigh...

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>
Never buy from a rich salesman.
                -- Goldenstern

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to