Holes moving

[mats_asson]

One of our guys have had problems with drill points suddenly moving. He have been making some hole operations that worked. After that he moved on and made some other stuff in the file and suddenly when backplotting the holes had moved. Anyone noticed anything similar? I have never noticed it myself but I always use circle entities as drill points and he uses points. Could that be the reason? Anyway, he has been doing it the same way as he done it in v9. Strange phenomenon I think. Or is he doing it the wrong way maybe?

 

Mats

[Guest]

I'll bet he is not picking up the center of the holes. If you are not careful with your clicking or have some other autocursor snap on you can easily pick the side of the hole or even an arbritary point within the hole.

 

When I am doing holes I make sure to set my autocursor to center.

[Jim at Gentex]

I agree w/ John. It's very easy to pick the endpoint or midpoint of an arc, especially if you don't zoom in on every hole. It's also possible to have the auto-cursor 'slip off' the point just as you click it resulting in a sketched point rather than a fixed one. I always set my auto cursor to 'center' just to avoid that particular mistake.

[Flycut]

I've worked alot with fanuc controles and one really really weird thing that I have noticed and no one beleives me until I show them is...

If you use POSITIVE tool comp on a tool and the next tool is a drilling cycle; the hole locations WILL be affected!!!!

The bigger your positive diameter offset the bigger the problem.

I first noticed this when I had to offset a 3.00" facemill to replace a 4.00" with +.500" toolcomp the following centerdrill was about .010" off location.

I know it's hard to believe but I've seen it happen in more then 1 shop on a variety of machines.

This might not be what this thread is about but I thought I'd mention it.

[Guest]

flycut,

 

It sounds to me like cutter comp was never turned off. I have worked on MANY fanucs, old 2's and 3's, 6A 6B's 10's, 16's, 18's, 21's, I have never seen this as you described.

 

The only thing I can see might cause such a condition is cutter comp not turned off.

[Flycut]

I hear you John but I clearly remember trying to extend the comp cancel line because I thought the line was to short to allow all the offset to be removed

 

I hardly ever see positive comp beeing used but even in cases where the offset was minor like (.010") it's minor but still there.

+ the drilling cycle has to be the very next operation.

In 10 years I've only seen it happen twice (other then the times where I did it on purpose to prove my point).

Try it and let me know.

Very currious.

[mcpgmr]

I've seen someting similar but not in a drilling cycle. It was after a tool change and g40 was never invoked. The active comp value was added to the new comp value. Doesn't hurt to have g40 on z retracts between machined features even while working with the same tool. As long as it's being turned back on again before machining the next feature it should work fine as long as their is enough lead in distance to turn comp back on again. JMHO.

[mats_asson]

He actually showed me an earlier copy of the file where everything was correct. And I know of the problem with picking points. That's why I always use entities instead. But his holes actually moved to a totally different position. But if nobody have noticed this he is probably doing something wrong.

 

I agree with the cutter comp not being turned off. I've seen very strange things with fanucs.

 

Mats

[robk]

Just grasping @ straws here, but was the WCS moved or rotated somehow? Since you picked entities for your points; were the views combined afterwards? Again I'm just making guesses here, but I personally have never seen geometry move by itself...

[mats_asson]

It wasn't me having problems, it was one of my workers. I'll check the wcs thing with him tomorrow.

 

Mats