3-D contouring - gouges

JoBwan · Mar 27, 2009

Good morning everyone out there. I came into work this morning and had an operator come to me with a random problem. Ok, here goes. We are using a 2000 VF-O with (HSM?) option. Randomly, parts are gouging on the roughing and/or finishing pass. It has done both. Example: We have run this 3-D job many times on this machine with no problems. Same (saved) program and type of cutters as always. He ran 4 parts perfect, the 5th is gouged. Ran 3 more good parts, and the next one is gouged in 2 different locations on the part. Ran 1 good and then 1 bad etc.... Gouges are never the same spot on the parts. It could be in the Y or Z axis, but have not tracked down yet. Here is some stats: 1045 merchant grade CRS. T1= TIN coated-carbide inserted ball endmill, .015" DOC, 6000 RPM, and 50 IPM. T2 = another identical endmill, .015" DOC, 6000 RPM, 75 IPM. Questions are: How do I know if it in fact, has HSM option? Has anyone ever seen this type of machine behavior? Can anyone give me a road to the light?
Thanks in advance,
__________________
Just a good ol' boy, never meanin' no harm.
Joe

capecod · Mar 27, 2009

Just a thought...

Do you have really good flushing of chips?

Maybe you occasionally recut a clump of chips?

John

StreetSpeed · Mar 28, 2009

I had a similar problem yesterday as a matter of fact on my 2005 VF-3. I was profiling some aluminum at 250 ipm. On one corner only, it would start to turn before it should have, and leaving just a little bit of an ugly chunk on what needs to be a beautiful finish (after an 80ipm finish pass). So, I turned the feed down from 250 ipm on that one section, to 125 ipm, then sped back up to 250ipm, and the problem was gone. I kinda just wanna blame it on the Haas machine. Are your gouges always in the same place or in random places?

Perry Harrington · Mar 29, 2009

Speaking from what I've read on my 93 era control, HSM is actually an M code to turn off screen updates, it is M76/M77. They say that above 47ipm the axis parameters for the encoder count window need to be widened for continuous motion. At the factory they set each axis for 1000 counts as the window of when the axis has reached it's intended destination. The manual recommends setting this to 2117 for up to 100ipm continuous motion. These are parameters 101-104 on my control. On my VF-0 there are 33867 counts per inch. The factory window is 1000 counts, or .0295 inches, they recommend jumping to .0654. This is essentially the potential inaccuracy of a contour at high speed. The solution to reduce error is to reduce the feed rate. Your control may be newer and faster, and you have AC servos too, but there are limitations.

JoBwan · Mar 30, 2009

Gouges

StreetSpeed said:
Are your gouges always in the same place or in random places?

This was stated in my original post. The gouges are in many different locations on the parts. We ended up moving the job to another machine, but would still like to figure out whats going on with the gouging problem. :willy_nilly:

__________________
Just a good ol' boy, never meanin' no harm.
Joe

Speedie · Mar 30, 2009

I wonder if you have a bad encoder or some carbon tracking in your servos. Typically 3d surfaces @ 50ipm are hard to make right on a older type controller and require a high speed control. What usually happens is the servo waits for the controller to tell it what to do and a couple turns of the spindle later you have gouges.

JoBwan · Apr 1, 2009

Can't find parameters?

Perry Harrington said:
On my VF-0 there are 33867 counts per inch. The factory window is 1000 counts, or .0295 inches, they recommend jumping to .0654. This is essentially the potential inaccuracy of a contour at high speed. The solution to reduce error is to reduce the feed rate. Your control may be newer and faster, and you have AC servos too, but there are limitations.

Can anyone post the procedure for finding the settings? I have looked through the parameters and I don't see anything related to the (HSM) "Look Ahead" settings.

__________________
Just a good ol' boy, never meanin' no harm.
Joe

Perry Harrington · Apr 1, 2009

On your machine Haas recommends adjusting setting 85. They say this is basically the same as the parameters I mentioned. Have a look at the description in the manual; my machine does not have a setting 85.

JoBwan · Apr 24, 2009

Thanks for all the posts. Ended up being a loose "Y" axis ballscrew. Not enough to feel in the table itself, but did however, turned the assembly nut 1 1/2 turns until it was tight. I ended up calling in a technician to do this so that it was done properly. Machine is running once again!!

__________________
Just a good ol' boy, never meanin' no harm.
Joe

JoBwan · May 5, 2009

To All, an update.
Well the technician didn't dig into this far enough, cause it happened again.
After seeing what was going on, I determined myself that the "X" axis cable had a break in it. After inspection of the cable, I was able to sse the exact spot where the break was. Changed the cable and back to running again. The break was on one of the encoder wires, and with the "Y" axis in a specific area, the encoder was not reading properly. Once the "Y" moved a little the "X" would catch up instantly (Rapid) to where the encoder read that it should be. Hope this helps someone out there in the future.
__________________
Just a good ol' boy, never meanin' no harm.
Joe

Edster · May 5, 2009

So was the y axis ballscrew loose too? What did the tech tighten on the ballscrew?

JoBwan · May 5, 2009

Two Different Issues

Edster said:
So was the y axis ballscrew loose too? What did the tech tighten on the ballscrew?

Yes, The "Y" was a little loose. He tightened the nut that holds the ballscrew to the casting. This however wasn't the main culprit. This stumped me at first, but when the "Y" axis moves, it will also move the "X" axis cable. Thus, when the cable re-made contact from 3-D milling, the "X" would jump .001" to .250" depending how far the "X" travelled without the encoder reading it's travel. Could have really caused some major crashing. We were definitly lucky not to have had a crash. :drool5:

__________________
Just a good ol' boy, never meanin' no harm.
Joe

Edster · May 6, 2009

So it was the nut on the ballscrew that tightens the end of the screw against the thrust bearing?

It's good you found the problem. Sometimes stuff like that will drive you crazy for a while before you finally figure it out

viper · May 13, 2009

JoBwan said:
To All, an update.
Well the technician didn't dig into this far enough, cause it happened again.
After seeing what was going on, I determined myself that the "X" axis cable had a break in it. After inspection of the cable, I was able to sse the exact spot where the break was. Changed the cable and back to running again. The break was on one of the encoder wires, and with the "Y" axis in a specific area, the encoder was not reading properly. Once the "Y" moved a little the "X" would catch up instantly (Rapid) to where the encoder read that it should be. Hope this helps someone out there in the future.
__________________
Just a good ol' boy, never meanin' no harm.
Joe

I ma not sure what your tech tighened but if he pulled back the front Y axis way and tightened that nut, he should not be working on machines! That is the floating end of the ballscrew and you could actually remove the nut all together and it would not change anything. They are supposed to be tightened I think 1/8 turn past seated by hand. Something does not sound quite right.

Search

Largest Manufacturing Technology Community on the Web

Search

3-D contouring - gouges

JoBwan

Aluminum

capecod

Plastic

StreetSpeed

Hot Rolled

Perry Harrington

Titanium

JoBwan

Aluminum

Speedie

Stainless

JoBwan

Aluminum

Perry Harrington

Titanium

JoBwan

Aluminum

JoBwan

Aluminum

Edster

Diamond

JoBwan

Aluminum

Edster

Diamond

viper

Titanium

Similar threads