[waltpermenter]

When making arguments on climb vs conventional milling it helps to know the difference between the two.
walt

[LeeWay]

Debating? Maybe. I showed evidence one way or other. I never said my conclusions were fact. In fact I questioned them myself. Ray answered me. Now I know. Lets continue on.
That wasn't even the first time today that I have been wrong.

[Peter Neill]

Lines from side milling and swirls from surface milling are just one of the reasons why surface grinders were invented
These finishes are perfectly normal for anyone who has ever operated a manual mill, and why we always kept a little box of toolroom stones and a bit of emery paper handy. I must admit to having a little giggle over some of the testing suggested here.

As it's aluminium, just a little rub with some wet & dry paper and a little lubricant and she'll be sweet.

[cobrakai]

Lines from side milling and swirls from surface milling are just one of the reasons why surface grinders were invented
These finishes are perfectly normal for anyone who has ever operated a manual mill, and why we always kept a little box of toolroom stones and a bit of emery paper handy. I must admit to having a little giggle over some of the testing suggested here.

As it's aluminium, just a little rub with some wet & dry paper and a little lubricant and she'll be sweet.

That's usually what I end up doing. A little scotch-brite, sandpaper, and it's good to go. But that's messy, and a fair amount of extra work. If we can avoid it, why not at least get closer to the end goal in one step?

[chevydyl]

I seen that same type of finish on a motor mount I had made out alum for my mill build. The machine used to make the part is a Southwestern Industries PROTOTrak K4 with AGE2 control servo driven. Big knee mill. I've seen that machine make parts with a perfectly smooth finish. But I told the operator not to spend a lot of time on the finish and that's what I got. Except for the hub recess for the motor you can feel and see the ripples. I had said that was the most important part.

[cobrakai]

k I made some interesting observations tonight.

I used the same long 3/8" tool as mentioned in the OP and set up a chunk of aluminum in the vise. After a cleanup pass, I ran it through several tests all at .375 DOC, .005 WOC while varying spindle speed and feedrate.

So there are 2 issues at play. Harmonics related to spindle or stepper speed induces a ripple into the part that is about .040" apart. Then, there is another ripple which is a normal artifact your speed/feed/tool stiffness.

First, the harmonics were almost completely gone at 3500 RPM. I ran two tests at the same chipload just different RPMs and got different results. One has a noticeable ripple, one does not

If I try cutting a chip that is too thin, I get rubbing. The finish is very smooth, but this is not good for the tool. There is a matte finish to the cut. Sometimes, I can see this matte finish only at the bottom portion of the cut indicating that the tool is flexing outwards away from the metal and rubbing. If I speed up the cut enough to avoid this type of finish, the roughness of the cut becomes more evident simply because the chips are so much thicker.

The underlying issue here is that it seems that a .005" finishing WOC just seems too small. Unless you have a razor sharp cutter, it's very easy to get rubbing. My best surface finish I could get without rubbing was 3500RPM and 45IPM feedrate. This equates to a .004" chip. At 40IPM I got rubbing.

Basically...
For finishing pass - reduce RPM to ~3500, take a deeper .010 WOC, and aim for a .001-.003 chip thickness. Oh, and the general rule of use the shortest, fattest, sharpest tool you can also applies!

Rubbing:


Best finish I could get with this tool at these parameters


Hope this helps someone

[BAMCNC.COM]

Attachment 254698

This was crappy code on my Tormach, you can see by the vertical entry lines on the right. Do you guys ever check your gibs?

[Steve Seebold]

Anyone have insight as to what causes this? The ripples I am referring to are on the flat side of the part. But now that I look at it, the chamfer has a ripple as well.

I roughed with a 1/2" 3Fl carbide 5000 RPM .25" DOC .1" WOC and left .005" for finishing pass. Finished with a 3/8" 3Fl, .375" DOC, .005" WOC, and I used GWizard to calculate the feed for a .001" chip thickness. I think it was something like 5000RPM and 38IPM.

My first thought is I need to go slower. But after some searching, people say this is rubbing and I actually need to go faster and aim for a thicker chip.

My understanding is that a finishing pass chip should have have a .001 for smooth finish. Roughing is .003 to .005 for our machines. Is this sounding right?

I'll tell you exactly what's wrong. You need to leave about .015 for your finish cut. 5000 RPM is OK to ruff. Then you need to slow down to 3,500 RPM to finish. I would ruff at 5,000 RPM, 30 IPM .15 to .25 deep depending on how much stock you're removing from the outside of the part. Then slow your spine to 3,500 RPM to finish and you'll see a marked improvement in the finished part.

I did make one change to my machine. I replaced the drive belt with a Gates Belt. THAT alone made a huge difference.

[twocik]

Here's my theory on the ripples/gallops. I've spent a bit of time trying to figure this out with Mike Sr. All of our machines have natural backlash of .0007" - .0013" in X & Y that can't be adjusted out. So the cutter going in for the finish with that kind of play could be the cause. After balancing the fan, adding the power twist belt, adjusting the hell out of the machine I was able to get in closer for a finer finishing pass. This dramatically reduced the gallops, and couldn't be happier. At 7x here's what the finish looks like for straight passes, diagonal passes is another animal....



Tormach


Mori Seiki

[Don Clement]

I don't try for the perfect finish right out of the Tormach. What I do is to run my parts in a vibratory deburring machine using a medium grade conical plastic media for a consistent matte finish on all aluminum parts. I really don't want a mirror finish anyhow as I have the parts anodized satin black which includes a pre-anodizing chemical etching process for the satin black finish. I have been using a Burr King SX200 for well over 10 years with excellent results. Models 110, 150 & 200 Vibra KING Bench Top Bowls - Product Details In addition to removing machine marks the vibratory deburring machine puts a very small radius on all sharp edges.

Don Clement

[Steve Seebold]

I have both a 110 and a 150 Burr King Vibratory tumbler. In the 150 I use 1/4 inch diameter ceramic media and in the 110 I use crushed walnut shells with jewelers rouge. The ceramic media takes the sharp edged off and the walnut shells provide a bright finish.

I have tried the plastic media, and didn't like the matt finish it produced. I guess that's why we're in different businesses.

I can get a pretty good finish right off the machine.

At 5,100 RPM, the finish sucks so I have learned to ruff at 5100 RPM at 40 to 50 IPM, then slow down to 3500 RPM at 30 IPM for the finish cut.

Another thing I did was I took the drive belt off the machine and took it to a local auto parts store and replaced it with a Gates belt. WHAT A DIFFERENCE. Way better.

[BAMCNC.COM]

I'd be interested to see surface finish from same parameters on a 770 - since the spindle is "balanced to 10k rpm"

[TurboStep]

I'll tell you exactly what's wrong .... you need to slow down to 3,500 RPM to finish.

I'm with you there Steve. I didn't buy a 5000 RPM machine to have to finish at 3500 RPM

Here's my theory on the ripples/gallops. I've spent a bit of time trying to figure this out with Mike Sr. All of our machines have natural backlash of .0007" - .0013" in X & Y that can't be adjusted out. So the cutter going in for the finish with that kind of play could be the cause.

Yet another theory? Sorry but I don't buy it for 2 reasons:
1. If your gallops are anything like those of the OP (as I'm sure they are judging from images I've seen of your finished parts) then the "gallop" frequency would be at the spindle frequency / 5 and very constant. If your mill was "galloping" you would more likely either see a frequency corresponding to the spindle frequency or a much more random pattern.
2. The table on your mill weighs in at over 150lb. It would be rather difficult to gallop the table at this frequency, especially if the machine is only taking a finishing pass.

I feel like I'm trying to convince everyone that the world is indeed flat! Any more theories?

... There is a matte finish to the cut. Sometimes, I can see this matte finish only at the bottom portion of the cut indicating that the tool is flexing outwards away from the metal and rubbing....

If I didn't know better I'd guess (and this is only a guess) that your end mill has seen better days. If I ever got a finish like that the first thin I'd do would be to inspect the flutes using a loupe.
Step

[BAMCNC.COM]

Still nobody realizes that going with CONVENTIONAL for the finish pass will help take out the backlash ripples... Think of how the endmill pushes on the work, and what that does with the table with *loose gibs* (which is another major factor).

This will help, but like Turbo said, there are still other maters at play. It's an *affordable* CHINA made Tormach.. What do you expect?

Steve, what is the number on your Gates belt, I think I need one.

Anyone ever thought of filling the hollow casting with some epoxy granite?

[popspipes]

I use climb because I think the finish is better/smoother.
I havent checked to see if the ripples are better one way or the other, I more or less solved my wavy finish problems by balancing the spindle and drive and changing to a fenner segmented belt, it was noisy when new but as it wore in it quieted down considerably.

The cutter marks have always been there and I think always will be as long we use milling cutters, they are less obvious with a high spindle speed, more flutes on the cutter, and lower feedrate.
I am also a big fan of the shortest fattest 4 flute cutters for finishing thatI have.

this is the way I do it and it seems to work for me...........

[TurboStep]

The cutter marks have always been there and I think always will be as long we use milling cutters, they are less obvious with a high spindle speed, more flutes on the cutter....

Hello Mike
That's another interesting point. More flutes on the cutter, Hmm. The question is, have you tested this or are you convinced that it must be that way because simply everyone says that it must be that way?
Do you get the impression that I may be about to declare that the world really is flat
Step

[popspipes]

I have always had better luck finishing with 4 flute cutters than two flute, the 4 flute effectively doubles the rpm for a given feedrate.
I also think the 4 flute is more rigid than a 2 flute cutter as well.

It may be from the old Bridgeport days, I would always crank up the rpm and take a small finish cut at a lower feedrate to make the part look nice.

I also like the shortest cutter for the job, less deflection.

Just my personal opinion and everyone has one of those ha!

Dont tell me the world is flat, who knows I may fall off the edge ha!

[twocik]

"Yet another theory? Sorry but I don't buy it for 2 reasons:
1. If your gallops are anything like those of the OP (as I'm sure they are judging from images I've seen of your finished parts) then the "gallop" frequency would be at the spindle frequency / 5 and very constant. If your mill was "galloping" you would more likely either see a frequency corresponding to the spindle frequency or a much more random pattern.
2. The table on your mill weighs in at over 150lb. It would be rather difficult to gallop the table at this frequency, especially if the machine is only taking a finishing pass.

I feel like I'm trying to convince everyone that the world is indeed flat! Any more theories?"



Take a deep breath bud ! I saw a frequency pattern before the mods, now it's all good. All I was saying was I'm happy with the finish now vs before. Sh*t the world is flat lol Please tell us the secret to unlocking this riddle Turbo. Diagonal cuts leave a horrible finish no matter what I try, any ideas ?




Great looking parts Don!!

I run a 3 step wet/dry vibratory tumbler setup and anodize my own parts too. Believe me I've tried everything before these mods and nothing gave me the finish I was after (anodizing will show everything). Most of the parts I cut are square with the idea of knocking off the corners in the tumblers.

[popspipes]

Chris,
The diagonals wont look as good as a straight x or y cut, I think it has to do with the digital stepping effect, maybe servos would be better in that respect? I have noticed this on my parts as well, there are two steppers involved to get the diagonal.
My model parts need the machined finish as thats what the guys that buy them want to see.

I am for the most part, well satisfied with my machine finishes now.

[Zetopan]

There is a good geometric reason why diagonals must look different. Imagine taking a step in only the X or Y axis directions.
The cutter to work piece movement in this case is the step size. If you take a simultaneous step in both X and Y the distance
becomes square root of 2 larger (i.e. about 1.41X as large). And likewise for 3 axis movements, the smallest simultaneous
movements of all 3 axes cause the cutter to work piece distance to change by the square root of 3 (i.e. about 1.73X larger).
Hence, any 2D or 3D diagonal movements are always coarser resolution than only moving in a single axis.