Haas VF2 questions ans still loking for a used VF2

viper · Aug 29, 2007

First question - We have a 1998 VF2 that, when contouring, sounds rough and has trouble keeping up with the commanded feed rate. We are trying to contour at 200ipm. You can feel the chassis shake. The axis are smooth otherwise. This is NOT isolated to one part. It would seem that it does not like contouring. Is there some programming or even Haas upgrades that can be done to help with this? Our program consists of thousands of point to point moves and you can feel every one. Our new SS with high speed machining is smooth as silk.

Next question - We are still trying to find a used VF2 like what we have. 1997-1999. Can anyone offer any leads? The ones we are finding are WAY over priced. Thanks

Probe Help · Aug 30, 2007

Well contouring at 200ipm with a 10 year old Haas just might not work. You might have to slow it down.

You might want to run a Ballbar test on the machine and see how it looks. It will give you quite a bit of information in a short period of time. Have you looked into it?

www.ncsprobing.com

G-Auto · Aug 30, 2007

viper - what's an ss?

viper · Aug 30, 2007

We have never performed a ballbar test and would like to learn how. The SS refers to a new VF2SS, sorry.

D. Rey · Aug 30, 2007

You may be asking too much from your machine to cut small line segments at 200 ipm.

I have a brand new one that does the same thing at 120 ipm if it is cutting a helix of small lines.

Will your CAD/CAM system output arcs instead of lines? That solved the problem for me.

Dave

viper · Aug 30, 2007

Well, that is kinda what I am getting at. I am trying to determine if this is a programming issue or a machine issue. It would sure seem that the machine is more than capable of going that fast but interpreting all these little lines is hard and probably hard on the machine for no reason.

can someone tell me about this ballbar test? Thanks guys

izzoe46 · Aug 31, 2007

It might be a programming and machine issue. Like D rey said, if you can convert to arcs, you should. Also, At 200 ipm Whats your doc and woc?
That might be something to look at.

If your going 200 ipm that means that your rpm's must be high. So, you have to look at how balanced your holders are. And, how level your machine is.

BTW I would slow it down a bit>>

huskermcdoogle · Aug 31, 2007

Your SS probably has the High Speed Option so that you can read many many block ahead and it calculates its next move before it gets there. Whereas your old VF2 does not have that option. I don't think it even was an option, though it might have been. Someone probably knows how to trick it out.

Husker

vegard · Aug 31, 2007

Ive had similar problems with older Fanucs. The answer was to filter the code into arcs in mastercam. Most large surfacing programs will run smoothly with just an occasional hickup after the filtering. I usually kick up the tolerance in Mastercam, and filter for arcs in XZ and YZ as well as XY.

- Vegard

cncprogman · Aug 31, 2007

BALLBAR

http://www.renishaw.com/en/6813.aspx

micro · Aug 31, 2007

Is this something you rent? Hire someone to come and do the test? I’ll bet it is a bit too pricy to justify buying the amount of times a small guy might use it.

Was it me or was that a heidenhain control on the mill in the videos? Heh.

3t3d · Aug 31, 2007

HOW can the ballbar test "know" where the machine is at as it sweeps a circle?
Can it tie into the machines controller? Will that be an extra cost option?
I don't see any encoder for the angular position, just the extneded length of the ballbar.

It doesn't just "guess" based on the jumps in length (arc error) at the four corneres does it?

Enquiring minds want to know.

D. Rey · Aug 31, 2007

So how much is one of those ball bar kits? I'll take a stab. 10k. Am I close?

Dave

HuFlungDung · Aug 31, 2007

Call me skeptical I guess, but I doubt that the 3d contours are accurate at 200 ipm.

A 4 flute endmill, @ 12,000rpm and 200ipm only contacts the part every .004"

My point is, that there is little use in running a highly interpolated and supposedly more accurate path when the tool itself moves farther than the length of the interpolated segment. So increase the toolpath tolerance, reduce the amount of code and give your controller a chance to keep up.

Shoot me down, I like it

D · Aug 31, 2007

I agree with HU. We do alot of 3D profiling and have learned over the years that we get as good or better finish by opening up the tolerance in cad/cam when programming. We have 2 VF-2SS machines and feed at 275IPM with no problem. But if you try to program with a .0005 tol. on the regular VF's they do feel jerky and makes for a huge program that is not needed. If it is something that absolutly has to be w/in .001 on the contour (doubtfull) I'm afraid your looking at 50-75IPM maybe on a none high speed control.

viper · Aug 31, 2007

We understand that it is not completely accurate but is works for us right now. We only have to hold +/- .010 and have a guy DA the surface when done.

Our machine, on the back option sheet, has an option (not selected)for 1000 blocks/sec. Not sure if that was what they called high speed in th day or what. Would this still be worth having then?

HU - We are working with Al and have never tried a 4FL ballnose for fear of loading up. I know this will help accuracy if we can use them. Would you have any idea with Al??

HuFlungDung · Aug 31, 2007

I was only theorizing about a 4 flute, as that typically might the maximum number of flutes, so with fewer flutes, the distance travelled between chips becomes even longer.

I'm not sure what CAM you are using, but is the toolpath surface tolerance in line with your allowed tolerance? .01" would allow a definite facetted appearance to the surface.

Probe Help · Aug 31, 2007

Close stab D-Rey.. the complete Ballbar kit is list price under $8k. Real easy to set up and run.

It doesn't actually have to know 'where' the machine is. It's hard to explain in a post but I'll give it a stab.

You bascially set a magnetic sphere on the table and another in a tool holder. The Ballbar is a LVDT transducer that fits between the two. Then the machine is programmed in two clockwise and two counter-clockwise circles. When doing the circular interpolation the LVDT measures any of the 'in and out' moves. It can from that determine 13 different errors including backlash, reversal spikes, servo mismatch, squareness, lateral play, cyclic error, straighness, scaling error, etc.

Does that explain it at all? There was a link in this post and I have a .pdf if anyone wants pretty pictures. Email me and I'll get you a copy.

www.ncsprobing.com

viper · Aug 31, 2007

Probe, sounds good but maybe not practical for checking a cheaper mill. could we possibly just cut a 4in bore with an endmill and check with a bore gauge for out of round??

HU, I was hoping you would tell me the 4FL is the way to go. It would sure make more sense if we can do it. Might be worth a try. As far as this programming, we did not program this and got it from another source. Sounds like we need to reprogram the whole thing. We are using .0005 ridge height right now so we may increase that. I will have to look for these filters or ways to program arcs instead of lines.

cncprogman · Aug 31, 2007

There is a machinery repair service we used before that had the ballbar. He is in the DFW area, don't know if he would travel to AZ or not, but here is his website.

http://www.fixmycnc.com/

Haas VF2 questions ans still loking for a used VF2

Titanium

Aluminum

Hot Rolled

Titanium

Titanium

Titanium

Cast Iron

Stainless

Aluminum

Cast Iron

Stainless

Diamond

Titanium

Diamond

Hot Rolled

Titanium

Diamond

Aluminum

Titanium

Cast Iron

Similar threads