Large Kearney & Trecker HMC conversion.

[bostosh]

Cool restore,
I made a lot of parts on those Mickey-Matics good precise workhorse.
punched cards to punched tape.
The high tech was the random tool selection with the coded rings on the holders. The low tech was the lead hammer to wack the tool changer arm when it hung up.

Now if you can find the 5-axis one with the dual contouring rotary tables.?

[samco]

I have seen those in pictures.. Not in person. Those where the mickey-matic II ? We used the old punch tape control up until a few years ago. Worked ok - but very basic. Emc is going to be unreal compared to it.

http://www.electronicsam.com/images/...controller.JPG
(I made the 'tape emulator' using an old 386 and the printer port)

sam

Cool restore,
I made a lot of parts on those Mickey-Matics good precise workhorse.
punched cards to punched tape.
The high tech was the random tool selection with the coded rings on the holders. The low tech was the lead hammer to wack the tool changer arm when it hung up.

Now if you can find the 5-axis one with the dual contouring rotary tables.?

[samco]

actually did some servo tuning. I am pretty new to tuning but I am happy so far. (can only get better)

This is a .1" move at 30ipm. you can see the blips at the begining and end - that is acc/deccel. the peak error is .00017". I have done a little machining with the following error set to .001. No ferror trips yet

http://www.electronicsam.com/images/...ing/tuning.png

so far so good.

[samco]

Last major hardware mod (except for a control panel)

This is the encoder for the spindle.. This will allow for rigid tapping.

Heating up the timing gear
http://www.electronicsam.com/images/...timinggear.JPG

installed
http://www.electronicsam.com/images/...timinggear.JPG

Now just need to come up with a bracket (and a belt guide on the encoder pully)
http://www.electronicsam.com/images/...dleencoder.JPG

sam

[ebrewste]

the peak error is .00017".

Was that with a realistic acceleration / velocity rate? Those are good numbers you should be quite happy with!

[samco]

IIRC that was a .1 inch move at 25ipm (so it got into the cruise phase) with an accel of 20in/s^2. I have the max velocity set to 200ipm as that was the original control did and we know it was designed for it. the acc is already higher than the orig control

I plan to revisit the acc and tuning at a later date

sam

Was that with a realistic acceleration / velocity rate? Those are good numbers you should be quite happy with!

[samco]

Here is the 3rd and 4th rigid tapping attempt. (all successful)

we have to play with the vfd for braking and acceleration. it was setup quite a long time ago and I don't remember how it was setup. it has braking resistors but the transmission has quite a bit of rotational mass..

1" 16 tap
60rpm
[nomedia="http://www.youtube.com/watch?v=LLcztAbgEcU"]YouTube - kearney and trecker rigid tapping 1 inch hole.[/nomedia]

400rpm
[nomedia="http://www.youtube.com/watch?v=2E3PMooch1k"]YouTube - kearney and trecker rigid tapping at 400rpm[/nomedia]

shot of the tool changer swapping out the 7/8" drill with the tap drill.
[nomedia="http://www.youtube.com/watch?v=q6S-QfmWcsI"]YouTube - another tool change with a 1 inch tap and 7/8 inch drill[/nomedia]

sam

[samco]

Oh - this is the spindle encoder mount - showing the collet unclamping also.

[nomedia="http://www.youtube.com/watch?v=5vvP4L_hr90"]YouTube - K&T spindle encoder with collet unclamp/clamp[/nomedia]

sam

[beezerlm]

The 400 rpm tap was impressive. looks like you went a lot deeper into the part on that one. Was it difficult to sync the spindle to the z axis feed rate in EMC2? Does all you encoder feedback go through EMC2?

[samco]

Yes - everything goes through emc2. The rigid tapping is done with G33.1 which syncs the axis(s) with the spindle. (just works ) That was the same program between the 2 runs - just changed the spindle speed. The reason it went though further is that the z axis has to follow perfectly while the spindle decelerates and reverses. Our spindle has a pretty weak drive and has quite a bit of rotational mass so it takes time to slow down and stop - so at a faster rpm - it takes a longer distance to stop and start. I hope to tweek it more - and if there is issues we might add a brake of some time.

thanks for looking!
sam

The 400 rpm tap was impressive. looks like you went a lot deeper into the part on that one. Was it difficult to sync the spindle to the z axis feed rate in EMC2? Does all you encoder feedback go through EMC2?

[samco]

so - here is a video with it all coming together. This is a short gcode
program that spots, drills, taps a hole. At this time emc pauses motion
when the tool prep is happening - so I modified my ladder so it sends
the tool prepared bit instantly but the actual tool change is inhibited
until the ladder is done with the tool prep. (if that made sense.) The
program is short enough that for a few of the tools the tool change
sequence has to wait for the prep.

[nomedia="http://www.youtube.com/watch?v=39q6kvrSBSk"]YouTube - Kearney and Trecker CNC running simple program. (tool changing and everything)[/nomedia]

After this video was done we played with the spindle vfd and got the
acc/dec twice as fast.

sam

ps - yes the 7/8" drill is bent. (best enco had 15 years ago )

[beezerlm]

Very nice! I hope my K&T runs like that someday!

[advt001]

Awesome!

[samco]

2 random videos... (nothing too exciting)

[nomedia="http://www.youtube.com/watch?v=1a0YpKfSGE8"]YouTube - 6 inch carbide facing mill[/nomedia]

[nomedia="http://www.youtube.com/watch?v=Kh4YD0d071c"]YouTube - another tool change - large shell mill / slitting saw[/nomedia]

Over the last week I have been using it for a bit of r&d. All I can say is - awesome!. I have a mocked up control pannel right now - pretty basic as I am using emc2 axis interface. I have 2 rotory swithed - feed hold and a jog wheel.

one rotory switch selects what the jog wheel does. feed override, maximim velocity, spindle override, x, y, z.

the second rotory switch selects the jog increment. (.001, .01, .1)

I really like it. Maximum velocity works nice as it just clamps the machine at a max feed rate. (doesn't scale it like feed override does) works great for proofing a program while not slowing down the cutting rates. but say clamping the maximum the machine will move at 20ipm or less.

Cutting some aluminum with a slitting saw. I had it plug up on me once so I was checking it every so often - click feed hold, run the spindle override to 0, check cutter. awesome.

this was what happened when it plugged. (of course it didn't break a sweat )
http://electronicsam.com/images/Kand...idaluminum.JPG

sam

[beezerlm]

Looks good Sam. Thats a big face mill!

Is the project almost complete? What do you have left to do?

[samco]

The big things are done. A few things that I want to work on is

-closed loop spindle control (I have the vfd tracking pretty close but it isn't perfect - also using +/-10v)
-finalize the control panel and machine panel
-remote pendand? (not a priority)
-revisit axis tuning. (I have min following error set to .001 and have not had it trip - but I can get the z axis to buzz quietly sometime while jogging.
-flood coolant hooked back up.
-other odds and ends

ran it quite a bit this weekend - no issues. worked awesome.

sam

[samco]

This is the next thing to get hooked into emc...



cannot wait (seems to be a renishaw knock off)

sam

[beezerlm]

This will be interesting!

[samco]

[beezerlm]

Nice. I have seen a lot of people probing in the machine with EMC2. Have you checked a ring gauge to check the accuracy? That would be very handy if it was a reliable check.:rainfro: