Surface Finish Issues on New Machine

Hello,

I recently purchased a brand new VMC from a company that I would like to keep anonymous for the time being. Let's just say it's a company that has been around for a long time, but they just started selling this type of machine. The important thing is that the machine's weight is about 8600lbs and has X-Y travel of 30 x 20 inches. I was the first shop to take delivery of this machine in the whole country.

I have owned my own shop for about 3 years now and this was the first new machine I purchased. Before this I was using (and still own) a Fryer MB-14. I still work my day job where we have Okuma, Fadal, and Haas machining centers. I have never seen the issue I'm about to describe except for maybe on the Fadal and it's not even half as bad as what happens on my new machine. Not to mention that the Fadal is worth less than 25% of what I paid for the new machine.

The new machine I purchased has a serious issue with "coining" during simple, 2D toolpaths...but honestly it's more like gouging. It occurs when the machine is moving in both X and Y. If I start my feed outside of the material and feed in with a straight line, no coining occurs. If there is a radius on my lead-in then coining occurs at the beginning of the cut. Simple 2D pockets look horrendous.

The machine leaves beautiful finishes when doing 3D surfacing work.

The machine builder has had people out to see the machine from their service team, applications team, and yesterday they had the controller manufacturer come out and take a look. They told me that the problem is the floor in my shop, but I'm not buying it.

The floor in my shop is anywhere from 4-6in thick concrete. They feel that the floor is no good because if you put an indicator on the spindle of the machine and stomp on the floor, the dial jumps around less than 0.0005". Here's the thing though...same thing on my Fryer, but it leaves better surface finishes. Additionally, I put an indicator on a Haas at work and did the same thing and the needle moved more than 0.001"! The Haas leaves perfect finishes.

I feel like the tuning of the machine is just too aggressive and it's "jerking" around when it's changing directions. The machine builder ran a program of mine on one of their showroom machines and it was better, but still not acceptable for the price of the machine.

I'm wondering what you guys think...thanks

Pictures would help. I have seen a lot of things get blamed on the machine that were just process issues.

8600lbs isn't very heavy for a 30 x 20 VMC though. The weight/size ratio might betray a level of quality.

I have seen a semi drive by my shop before on the indicator and I have over 12" under all my machines. I don't think your concrete is the issue.

Agreed, the forklift passing along outside in the limestone parking lot can be felt in the shop, on 9" of crete, 40 ft. away. Not the concrete.
Maybe ask your sales person if he thinks you'd get better advise on your specific make and model of machine from, the practicalmachinist, reddit, Facebook, Instagram and cnc zone communities that you are an active participant in (weather or not you are).
If this is a new entry into a very competitive market I am sure they don't want the first and only hit people get when searching for info on their new baby to be you and your issues.
Remind him of what's at stake.

wonder how thick the concrete is in the showroom???????

Pics of issue?

Hi guys, thanks for the responses.

I am going to do some testing tomorrow and Friday on multiple machines with the same cutter and holder.

Here is a picture of the issue. I was running a 3/8, 3FL Helical flat end mill at 12k rpm, 108 IPM (.003 ipt) in a side lock holder. The holder and end mill were brand new. The holder is balanced to 15k rpm per the manufacturer (Erikson, Kennemetal’s cheap brand). This is an “area mill” tool path in MasterCAM with 35% step over and 0.005” depth of cut. The “coining” can be felt pretty easily with your finger and finger nail.

boosted said:

Pictures would help. I have seen a lot of things get blamed on the machine that were just process issues.

8600lbs isn't very heavy for a 30 x 20 VMC though. The weight/size ratio might betray a level of quality.

Click to expand...

A Haas VM2 (this machines direct competitor weighs almost the exact same.

barbter said:

What's the finish like for finishing a pocket (the bottom) - if you have a 5 thou cut in ally and run at 2x metre ish feedrate?
Is it like this?
Fanuc oimc HSM G05.1 / G08

Edit - is the controller Fanuc?
Did they connect a laptop into your control via the PCMCIA or USB to check parameters?

Click to expand...

The controller is a Siemens and it looks much worst than your issue, but it is the same thing.

Almost looks like a bue on the cutter, but I'm sure you ruled that out.

Might as well spill the beans on the mtb...it doesn't sound like they really care...yet

I would very carefully inspect the cutter before blaming the machine.
And give the holding fixture a tap-tap looking for less than a solid holding.
I have seen brand new machines with something not tight from the factory..and brand new cutters that were off.

Has that part run better/OK on another machine? Does a tap tap on the fixtured part seem solid?

Might as well spill the beans on the material, thickness, and size of the part also

So immediately when you said diagonal my thoughts went to servo tuning. Then you mention the Siemens.

That is what I would put my bets on.

Rcgiovannani said:

The controller is a Siemens and it looks much worst than your issue, but it is the same thing.

Click to expand...

Does changing your cycle 832 settings have any effect?

Maybe dumb question: how's the subgrade underneath the slab? I have seen awful surface finishes at high feed rates because of a void, which makes the slab act like a cantilever. Sounding the slab with a 2x4 should give you enough info to rule it out quickly, it will sound hollow if the soil has fallen away.

Otherwise I would guess some sort of servo control issue.

michiganbuck said:

I would very carefully inspect the cutter before blaming the machine.
And give the holding fixture a tap-tap looking for less than a solid holding.
I have seen brand new machines with something not tight from the factory..and brand new cutters that were off.

Has that part run better/OK on another machine? Does a tap tap on the fixtured part seem solid?

Might as well spill the beans on the material, thickness, and size of the part also

Click to expand...

We have tried the same cutter, but older. Same results. I have also seen this issue with other cutters too.

michiganbuck said:

I would very carefully inspect the cutter before blaming the machine.
And give the holding fixture a tap-tap looking for less than a solid holding.
I have seen brand new machines with something not tight from the factory..and brand new cutters that were off.

Has that part run better/OK on another machine? Does a tap tap on the fixtured part seem solid?

Might as well spill the beans on the material, thickness, and size of the part also

Click to expand...

I have not run this is exact part (it's just a test piece) on another machine yet.

Here are the specific parameters incase people want to know:

The material is 6061 Aluminum, it's about 1.0" thick. The pocket is 0.100' deep and 4.5" x 3.375".

For fixturing I'm using MMMUS Piranha 6" jaws with .125 step in a Kurt DX6 vice.

The cutter is Helical Solutions 46221, the holder is Erikson CV40EM038250. I've used an old and a new version of this cutter with the same results. The MTB used a Kennametal General Purpose 3 flute 3/8" end mill in their testing and we got the same results on my machine.

Coolant is HoCUT 795B, concentration of 7-8%

EndlessWaltz said:

So immediately when you said diagonal my thoughts went to servo tuning. Then you mention the Siemens.

That is what I would put my bets on.

Click to expand...

They sent someone from Siemens who said it was the floor. Have you had servo issues with Siemens machines? The MTB went from my cutter, to my holder, to my programming, to the floor, and then to servo tuning. Then they brought in Siemens and now they're back on the floor...

trochoidalpath said:

Maybe dumb question: how's the subgrade underneath the slab? I have seen awful surface finishes at high feed rates because of a void, which makes the slab act like a cantilever. Sounding the slab with a 2x4 should give you enough info to rule it out quickly, it will sound hollow if the soil has fallen away.

Otherwise I would guess some sort of servo control issue.

Click to expand...

I will hit the floor tonight. When we were stomping on the floor it sounded pretty solid, but I will check with a piece of wood. Also, I will mention that the Haas we have at my day job, an indicator moves much more when stomping on that floor and we have no surface finish issues there.

When you tap tap the part with a small hammer when it is in its fixturing does the part seem solid? Or can you feel vibrations?

How thick is that part at its thinnest area?
Re: (Part 1.0 thick ,The pocket is 0.100' deep and 4.5" x 3.375".)

A couple .020 shims under the part would prove it is tight down in the vise.

Traveling the machine table under a tenth indicator does it seem smooth?

Helical Solutions 46221, A good cutter if it is not damaged or too long.
(The MTB used a Kennametal General Purpose 3 flute 3/8" end mill in their testing and we got the same results on my machine. so seems the cutter is ruled out.

If in California the one might need re level after each earthquake.

boosted said:

Does changing your cycle 832 settings have any effect?

Click to expand...

No. I've tried it with Cycle 832 off, with advanced surface, and with top surface with multiple different tolerances.

michiganbuck said:

When you tap tap the part with a small hammer when it is in its fixturing does the part seem solid? Or can you feel vibrations?

How thick is that part at its thinnest area?
Re: (Part 1.0 thick ,The pocket is 0.100' deep and 4.5" x 3.375".)

A couple .020 shims under the part would prove it is tight down in the vise.

Traveling the machine table under a tenth indicator does it seem smooth?

Helical Solutions 46221, A good cutter if it is not damaged or too long.
(The MTB used a Kennametal General Purpose 3 flute 3/8" end mill in their testing and we got the same results on my machine. so seems the cutter is ruled out.

Click to expand...

It's about .900" thick at it's thinest. This is just a test piece, it's not a customer part so I'm just cutting a pocket in a chuck of aluminum I had. Also, this is not the only part this has happened on.

I did a "ring" test of the part with my knuckle and a dead blow. Sounds very solid.

barbter said:

Out of interest, did the siemens guru do any servo tuning? Did he do anything with the control?
Blaming the floor, is a total kop-out because you are where you are and need to get the machine going.

If you look here - he tells you how to tune it and it shows what parameters the auto tune changes.
Siemens Auto Servo Tuning AST - YouTube

It looks like the parameters for feed are around the P1460 number.
This manual explains more
https://cache.industry.siemens.com/dl/files/063/48073063/att_43518/v1/828D_LH2_1110_en.pdf

I would post some questions and even your photo, and ask here
Access Denied

Click to expand...

The Siemens Mechatronics engineer said he "de-tuned" it. It looks better now, but its still not acceptable in my opinion. He also ran quite a few traces of the axises with an indicator plugged in as a separate axis (to monitor each axis). They also ran a trace with the indicator on the floor (but it was attached to the machine). That trace had the worst results of all of them, which is why they claim the floor is bad. But the indicator was on the machine while reading on the floor. That doesn't prove that the floor is moving. The floor could be moving or the machine could be moving around because the "jerk" is so bad.

I will check out the video