Inconsistent part sizes on Laguna Swift

Post Reply
clintlmartin
Posts: 6
Joined: Wed Apr 29, 2020 5:05 am
Model of CNC Machine: Laguna Swift 4x8

Inconsistent part sizes on Laguna Swift

Post by clintlmartin »

Hello,

I've been running a Laguna Swift 4x8 CNC router for about a year now. recently I've started getting inconsistent behaviors from the machine.

I recently drew up a new set of tool paths for a new product, which is for all intents and purposes is a box. Due to the part sizes I end up cutting the various parts out across 4 sheets. I typically cut the sides from one sheet, the top and bottom from another, etc. in this case the top/bottom both come out identical (from the same sheet) and the Left and Right sides ALSO come out identical (also both from the same sheet/tool paths) however the problem is, even though the overall dimensions of the pieces are the same in vcarve the sides end up being slightly shorter than the top and bottom -- slightly as in 0.040in as close as I can measure. unfortunately 0.040 in is too much slack for the box to fit together properly. I double-checked the generated Gcode using NC Corrector 4.0 -- and it shows that the calculated sizes between the different sheets are identical (though I suspect due to rounding/floating point math the calculated sizes aren't exactly the same as what is in the vcarve model -- they are off by 0.0000195in, which I think is pretty darn close)

the panels are cut in the same orientation, they are cut using the same end mill, and at the same feeds/speeds.


All of my cuts rely on 2-sided operations to get all the features I need on both sides of each sheet. to do this I generally mill a set of 3/8 alignment pins through the work piece into my wasteboard. for the past year the hole tolerance has been fantastic. I use ground steel pins for aligning between flips and generally find the hole fit so tight that I have to use pliers to pull the pins out after I get the vacuum spun up.. Lately however the alignment pin holes have been severely oversized -- 0.384in instead of 0.375 -- which leads to very sloppy and slow setup on the flip operation -- and I daresay some loss in the accuracy of the final product as well.

I've gone through all the variations on tool path options and every verification step I can think of, I am starting to suspect hardware or potentially control issues, but I am somewhat at a loss to figure out which it is. If I had loose connectors or screws/rails or similar in the hardware side I would expect the parts to come out inconsistent between subsequent runs of the same tool paths -- however that is not what is happening.

any suggestions?

GEdward
Vectric Craftsman
Posts: 232
Joined: Wed Jul 05, 2017 9:13 pm
Model of CNC Machine: 24 X 36 3 Axis
Location: Ipswich, South Dakota

Re: Inconsistent part sizes on Laguna Swift

Post by GEdward »

What you are describing for all the world looks to me like a mechanical issue related to wear or looseness of some component or combination of components of the machine. If you have a dial indicator you can easily check each component starting with the spindle. I suspect that if you place the indicator on, or in, the spindle and push it around a bit by hand you will see a significant movement of the needle on the dial of the indicator.

Your holes can also blow out, even with a good solid spindle, if the drill/mill has run out from a worn or damaged collet or dirt and debris impacted in the collet assembly. If you take a .375 end mill and check it with a dial indicator and find it has .004 run out then your hole or slot will be .379.

You could change the direction of the cut out, climb or conventional, and if play in the machine is the culprit, then you should see a difference in the size of your finished product.

If the play is coming from wear in your lead screws or ball screws, you should be able to use your operating system's backlash compensation feature to help compensate.

Ed

clintlmartin
Posts: 6
Joined: Wed Apr 29, 2020 5:05 am
Model of CNC Machine: Laguna Swift 4x8

Re: Inconsistent part sizes on Laguna Swift

Post by clintlmartin »

Thanks for the reply.. I don't have any dial gauges currently, but I've been meaning to get some. as I mentioned, I was leaning towards machine issues as well, though I would be pretty surprised if this is due to wear -- the machine is only 1 year old at this point. I would also think I'd get part-to-part inconsistency as well. I cut 3 sheets of the same parts yesterday and they all came out identical -- identically wrong, but still identical.

I'll look at runout and backlash in the machine, check for loose screws/rails/etc. Thanks again!

GEdward
Vectric Craftsman
Posts: 232
Joined: Wed Jul 05, 2017 9:13 pm
Model of CNC Machine: 24 X 36 3 Axis
Location: Ipswich, South Dakota

Re: Inconsistent part sizes on Laguna Swift

Post by GEdward »

[quote would also think I'd get part-to-part inconsistency as well. I cut 3 sheets of the same parts yesterday and they all came out identical -- identically wrong, but still identical.][/quote]

It is kind of interesting, almost amazing actually, that a machine with an incredible amount of wear can actually repeat well enough to produce good parts. A case in point is that a number of years ago I worked in a production machine shop that had a number of bread and butter products that used dedicated machines for production. Two of the machines were a couple of single operation chucker lathes that turned a short outside diameter of a sleeve and put a snap ring groove on the inside diameter. They were so worn out that you could by hand push the tool posts 3/8 of an inch in both X and Y. Yet, because the tool pressure was always the same, they would repeat within .003 T.I.R. all day long.

That is why I mentioned that by changing the direction that the machine travels, thus the direction of the load on the spindle, during your cut out operation you should get different dimensional results.

Ed

Post Reply