[LeeWay]

I had to order some emergency collets for this. My 3/8" plastic rod is actually .3773". That is within it's rated tolerances though. That is too tight in a 3/8" without even being loaded up. That may have been why I was having trouble with the 5/8" brass as well. It was fairly tight. I might be able to get the E collets to be sized where the par puller will work reliably without a groove. I am going to mike all my stock and see what I have. E collets will be good to have on hand anyway. I let this all ride for awhile Because of Mr. Jackson's passing. He will be missed.

I was also able to figure out that the 15L wants Gcode in diameter mode. I did get Dolphin to work and output code that did not get any errors.

It just will not run it right yet.
Still working on it.

[LeeWay]

Anyone any good at editing a Post Processor?
Mine in Dolphin is set for radius mode.
I need it to be diameter mode.
I have looked at it for a while and cannot see any obvious place to change that.
Thanks, Guys.

[dammachines]

Anyone any good at editing a Post Processor?

I've never edited a post processor and have never seen Dolphin, but I tried to Google just for fun...

Turns out on the Dolphin website there is a Mach 3 post for Diameter mode and Mach 3 post for Radius mode.
There are a few differences between the files, but I'd say the one that's important to what you're looking for is under the heading "AXES:"

The Radius version contains "YSCALE 1" where as the Diameter has "YSCALE 2"

You may want to check out the files yourself just to see if there are any other differences, but that sounds promising.


Dave

[LeeWay]

Thanks, Dave. Turns out that I did have those PP's in my folder as well. I had Dolphin modify my older post to update it. It is labeled for version 13. Most of the other posts in the folder are labelled for version 11. Those two posts were not marked with a version.
I gave it a try and the code looks good.

One thing I noticed is the initial string is different from the Conversational on the lathe.
I wonder if just adding the same string that conversational outputs to the top of the Dolphin code will fix my next issue?

What it is doing is after homing and setting G30, hit cycle start and it keeps traveling toward the spindle past G30. Enough for crashes. I have Dolphin set for offset #1, which is G54. It is going past that point.
Worth giving it a shot I think.

[LeeWay]

As it turns out, all of my normal material is oversized almost exactly by the tolerances mentioned in this thread. So I will need to use all three E collets to make this work right. The 3/4" may have enough play, but the 9/16" and the 3/8" stuff needs a bigger collet than the standard one in the kit I received from Tormach. Thats kind of a bummer, but it is part of the process. Not really that big of a deal. One time job I think. With the E collets, if I do get material in the future that is smaller, I think it will still take up the slack between it and the standard collets.
Just something else for you guys to consider when ordering. I did think about getting one when I ordered initially. It would not have helped too much. Getting a metric set of collets might help too, However my 3/8" stuff oversized would still be 9.58 mm. Non standard in that case.

[MichaelHenry]

You can buy collets in 1/64-in increments off the rack, so to speak. There are lots of vendors, but Mari Tool is well regarded:

https://www.maritool.com/Collets-5C-...153/index.html

Hardinge will sell you a collet in 0.001" increments but the cost is probably pretty substantial.

I used to turn CPVC or PVC fairly often and those were both way over the nominal diameter. I used either a Jacobs Rubberflex or Pratt-Burnerd collet chuck for those since each collet covered a range of +/- 1/8-in but neither of those are very amenable to bar pullers.

[LeeWay]

Thanks, Michael. I did consider that, but thought 25/64 would be too big. Some of it did measure .378". 25/64ths is .39. However, I have not messed with these collets enough to know how much extra clamping room one might actually have. Since they are round though, I suspect not much.
I checked the Tormach collets too and the 3/8" and the 9/16" are actually a bit undersized. That probably has a lot to do with it. The 3/4" is right on though. I have never turned an e collet, so I will scratch that off my bucket list..
Very little holding pressure is actually needed on my plastic parts. They get peck drilled and parted. That doesn't take much holding force. As tight as it is now, I could turn the parts with the collet open.

I did find that I had to modify the chip door.
My parting tool was within an 1/8" of the door when parting about 3/8" from the collet nose. So, first real modification. No going back now.
Here is a picture of about what I cut off. I just used an angle grinder. Piece of cake.

Would not have been an issue with the full enclosure, but then the turret would likely be installed.

[LeeWay]

I did finally figure out the issue with the code snippet not working for the puller and why it was complaining with an error.
I was trying to do a Feed move on the puller. It says it will not do a feed move without rotating the spindle. Well we an't have the bar puller in use with the spindle rotating.

It doesn't complain when I do a G00 move instead. However, that is way too fast. You might think that 60 IPM isn't very fast, but when you have the puller rapid in and then the closer slam open, puller clamp closed and rapids back to zero then the closer closes and puller opens all within about a second, you will know what I mean. It''s almost too much happening at once to take in. Putting in some dwells may help, but the retract move still needs to be slower than rapid.
Thoughts on what I can use there?

Here is the code that I have at the end.

( Bar Puller )
M06 T0505
G00 X0
G00 Z-0.630
M8
M64 P2
G00 Z0
M65 P2
M9
M05 M30

[tmarks11]

Provide feedback on Tormach's PP bug reporting site about this issue?

[LeeWay]

I emailed tech support last week on the 13th and received a reply. I just replied to it with my latest snippet and findings today.
Here is my latest code to test.

G30
M06 T0505 ( Bar Puller )
G00 X0
G00 Z-0.630 (Depth of Grab)
M8
M64 P2 (Open)
G04 P500 (Dwell)
G98
Z0 (Pull)
M65 P2 (Close)
M9
G30
M05 M30

I will eventually have to add the repeat codes at the end with the number of repeats parameter.

Since this is not a subroutine, I am not sure what code to use. I know M98 and M99 would call and dismiss the subroutine number. I may wind up having to do it that way.
Not a big deal either way. Just different codes added.

I don't think it is a flaw in Path Pilot. Most lathes will not feed without the spindle turning. I am going to try the G98 here. Not sure how that even works, but we will see. Another instance where the spindle would not move during a feed move is when you are using live tooling. I am not sure if that is implemented yet.

[rowbare]

That seems like an odd limitation. How do they do live tool stuff. Maybe it would work with feed set to 0?

bob

[LeeWay]

Ahh. Tormach fixed me up. I received emails from two guys up there and both had the answer.

It seems my parting tool had put it into G95 mode to sync the spindle with the axis movement.
Just adding a G94 and feed rate put it back in the correct mode for this.
Seems to be working pretty well on a few dry runs. I still have the collet issue, so can't run parts yet.
I'll remedy that this weekend.

Here is the code I am using now.

M06 T0505 ( Bar Puller )
G00 X0
G00 Z-0.630 (Depth of Grab)
M8 (delete)
M64 P2 (Open)
G94
G01 F10 Z0.00 (Pull)
M65 P2 (Close)
M9 (delete)
G04 P.5 (allow puller to fully open)
G00 X0.00
G30
M05 M30


This will work for a bar puller that only uses Z axis. An X axis puller needs a bit more code for the X moves, but should still be about the same.

I can also eliminate the coolant codes and just use the collet closer switch for the puller too. They both happen simultaneously anyway.
That actually answers a few questions and simplifies things.
I'll get some new video up soon.

Thanks a bunch to Tormach support.

[LeeWay]

The bar puller hasn't missed a pull yet. No groove and it doesn't leave much of a mark at all on the delrin. I still need to reduce the pull just a hair. It is biting too much with the face cut. I did make a very poor video. First video with a go pro. Light in poor. Camera mount in a poor spot. Had itset to wide angle apparently. I promise to do better in the future.
I did get the E collet turned out okay. Pretty simple really.
So, now how do I tell it to do multiple parts? I guess I will have to make it a subroutine? That doesn't sound right. Apparently what I do not know about Gcode could fill a book. Somewhere.

Also have a package coming from Tormach next week. That has got to be my Monitor stand.

Video should be up soon.

YouTube

[dammachines]

Fusion 360 is still long way away, 3 weeks from lathe it might end up like the last up date 7 weeks

Did you see? Turning will be included in the update coming tomorrow! Hopefully the post won't require too many changes to work with my Ah-ha controller...

Dave

[daniellyall]

yep that`s going to be good if it works they still have other thing`s to fix I will spend Monday (my time Monday, Sunday your time zone) trying to kill it the pp post is well all most there it`s pretty much the same as inventors looking on the HSM forum

[LeeWay]

That is good news. Dolphin is just not yet ready to run. That or my knowledge of it is not sufficient to handle it. I did manage to get it to work somewhat, but it will need quite a bit of rewrite on the PP. That isn't something that I can handle at this point. And Fusion 360 it ain't.

Also probably a safety issue worth a note.
Shut down procedure when you have a collet closer like this should start by first turning off the air and letting it bleed down. Otherwise the lever actuates back and forth during shut down. It is powerful enough to break bones.
Likewise during start up, the air valve should be turned on last.

This is the valve that I amusing.
McMaster-Carr

[LeeWay]

I did get some help to figure out how the G code will repeat a certain number of times by wtopace in this thread.
http://www.cnczone.com/forums/tormac...ml#post1737036

It is also in the manual as was the part about the G94 and G95 modes.
As I said before, the manual is an excellent read. If you read it. I did actually. Just didn't correlate what I was reading to my problem.

It was really simple.

o103 repeat [22]

right under the date at the top of the G code.
Then

o103 endrepeat

at the end right before M30.

I haven't yet figured out if that count work in that position or if I need it at the end of M30.
I'll get that figured out. It's all coming together pretty well.

[daniellyall]

that sound good if you could just do a sub that will make life easier it would be good if you could attach a alarm to it so when it`s out of rod it lets you know

all the path pilot pp are almost there for fusion

[LeeWay]

I would still have to edit the Gcode to put the sub info in there. It can't get any easier for me than to add those lines at the top and bottom of a program. Want a different amount, change the number at the top of the code.

It works great just like I mentioned. With the bottom line just before the M30. It stops precisely on the amount of parts you have for the repeat. That's it, Guys. I'm there. Now I need Fusion to follow me here. Right now it is down for Maint. That usually happens just before the new release, correct?

[daniellyall]

yep 3 more hours