Fly cutter only cutting on one side.

I have been using a 2" fly cutter on some aluminium and mild steel.

I am taking very light cuts on 0.1 - 0.3mm.

The finish is fine, but it is only cutting on one side.

I have trammed, and re-checked the mill to within 0.02mm over 15".

The mill has a tilting head so adjustment is not an issue.

I was getting some discrepancies to start with, and found that I had to tighten up the top Z axis lock before the bottom one to ensure repeatability.

All axis are locked tight during the tests.

If I cut left to right, only the front edge cuts. Cutting left to right I can see two distinct cuts. The cut on the back edge is just enough to remove the marks from the leading edge. This tells me the table is not trammed properly.

Before any measurements, I have removed any table dinks with my finest diamond stone. All is wiped clean. I am measuring off of parallels. Swapping the parallels makes no difference.

My questions are:

1. What can cause this? The only thing I can think of is a bowed table, but am I missing something? Are there any other checks I can do to locate the cause of this error?

2. If I cannot identify why this is happening, should I deliberately tram in an offset to counteract this?

3. Should I worry? I realise this will introduce a very minor concave cut, but I do not think this will cause me any problems.

A picture would help so we can see exactly what you’re getting.

There are some pictures half way down this page showing what it should look like.

https://www.cnccookbook.com/fly-cutter/

Edited By Vic on 23/12/2020 23:20:23

Hi Vic,

I can't get a photo of the surface finish as the camera refuses to focus, but it is very good. No evidence of a cross hatch pattern though.

I am not sure what other pictures may help.

This is the best I could take:

The surface looks worse in this picture than it is, but it shows the cut is only one side.

That does look to be cutting only on one side but it depends on the size of the workpiece and fly cutter. I often cut small parts with a large cutter so it only cuts in one direction. What are you using to tram your head?

A dial indicator on a 7 1/2" bar held in a jacobs chuck.

The fly cutter is R8, but if my understanding is correct, any runout in the chuck has no effect as the axis of rotation is set by the spindle.

The work piece is 2 3/4" wide. The point of the cutter travels around about 3" diameter.

Edited By IRT on 24/12/2020 00:14:24

Sorry - a bit misleading. The part I am making is 2 3/4" wide. The bit in the picture was a test piece and only 1" x 2 3/4".

Both show the same finish.

Is the work in a vise or secured directly to the table? (securely clamped and no rock is best)

Are the spindle bearings running with no runout? (they should be)

How flexible is the tool in the flycutter / is the tool small in size and sticking out of the flycutter a long way? (tool should be robust as possible and not sticking out too far, ie a 1/4" x 1/4" HSS tool should not stick out more than 1/4" to 5/16" as a rough guide)

All the above factors can affect how flycutters work.

You only need the tram to be off by a very small amount to get a cut on the trailing edge, If it's very small I would not worry about it, get can similar on my X3 but if you know it's there you can work around it. Either taking advantage of the trailing cut as a fine finish to the leading edges roughing cut or simply cut with the lower edge as the leading one. As Jeff says too much tool stick out does not help. the radius of the cutter marks on your work suggest you were swinging a very large radius with quite a small bodied tool.

One other cause can be a less than sharp cutter and a bit of spindle play or machine flex. If not sharp the tool can get pushed up and ride over the work and the the trailing edge then drops and takes a fine cut.

A lot of people prefer the finish shown in your photo to the double hatch that you get when it is perfectly trammed with its almost holographic look, here is one I prepared earlier.

Edited By JasonB on 24/12/2020 07:22:34

If it is not tramming correctly, the finished part will be concave. Check for flatness in the y axis. If all is OK within the specification limits stop worrying. If the surface is so dished as to be out of spec, re-tram until it is within spec.

What you have sounds like the spindle bearings need a bit of adjustment. Do the bearings warm up when the machine is run for something like 30 minutes? When you tram the head everything is positioned purely by gravity. When you are cutting there is a side force on the spindle that will affect the tram if the bearings are loose. I am assuming you are tramming using something mounted in/on the spindle. With your indicator mounted see what happens if you apply side forces to the spindle. Some very small indicator movement will just be flex in the metal of the machinery, large indicator movement will show the need for bearing adjustment or that the clamping of the spindle is not effective.

Martin C

.

Fascinating and useful things ... Moiré patterns

MichaelG.

Photographing shiny metal surfaces is hard! You might try drawing an obvious line across the work with a dark felt-tip. The camera should be able to auto-focus on a sharply contrasting line drawn across the otherwise bland confusing surface. Another approach is to vary the angle between lighting and camera until the camera 'gets it'.

Back to the question, the problem may be your materials are smearing. Ordinary mild-steel doesn't machine to a good finish particularly well - it tends to tear and smear. Many aluminium alloys are even worse - soft. Could be you don't have a problem at all; the cutter removes metal to create a flat surface and then hides it's tool marks by rubbing the ridges into the valleys. Maybe!

Dave

You can get some very nice machining patterns with a milling machine. Some parts made by I guy I met from Finland had an almost wood grain effect to them. These patterns seem to be caused by vibrations due to lack of rigidity of the setup or the machine itself. It is of course difficult to achieve if you’re actually trying to generate the effect on purpose! I think most of the previous posts have covered the double hatching effect and tramming. When I got my first mill I was told to use the longest bar I had to set the head perpendicular to the table. As a point of interest my current setup is a bar about a foot long with a DTI on the end.

As far as I can tell, everything is clamping is rock solid.

The lathe is a gear head model, so is noisy anyway. Hard to know if it is bearing noise.

The cutter is 5/16" and about 1/2" sticking out.

I ground and honed the cutter before making these cuts. The finish is very good. In the past when I have had a blunt tool, the finish suffered so I think it is still sharp.

I have never checked the spindle runout, but would this cause problems? Would it simple be a shift in the axis that tramming would remove?

I can see how play in the bearings could cause issues, but would this be seen in both directions?

(I am thinking out aloud here, and trying to understand more- not questioning the advice).

The bearings do warm up after a while. What does this tell me?

The machine is about 15 months old now so maybe I need to check and adjust the bearings... something else I have no experience of doing.

I didn't measure a significant change in thickness across the part so it is plenty good enough for the purpose, but as I keep reading about the cross hatch pattern so I know something isn't right and I would like to get to the bottom of it.

The part is clamped up now in the middle of the job so I can't remove it, but hopefully I will get out there over the next couple of days and report back.

Bearings should have some pre-load to ensure they run correctly. Under normal running even with no load this preload will cause them to warm up slightly. Often the instructions to set the preload on machines like this is to tighten them and run them for 30 minutes and check how warm the bearings are. They should get obviously warm but not too hot to touch. After 15 months of use I would expect a certain amount of bedding in to have taken place and think the bearings may need a slight snugging down. Some machines make this easier than others so find out what is involved before committing to it. Mine has two locknuts and a tabbed lock washer that are easy to access, yours may be buried deep in the machine. The spindle bearings being set correctly is as important as locking all the other sources of unwanted movement.

Martin C

(Particularly if you fly cut this piece on both sides) if you cannot measure any significant difference in thickness across the width, how do you propose improving anything? (Running a dial gauge across the y axis of the part would show even less)

To get the diffraction effects shown on JB’s pic demonstrates flatness to low tenths of a thou, I would guess (of course that part might be off a lot, on thickness, from end to end, but I doubt that🙂 ).

I think you are probably right not done it yet.

This was option 3 in my original question, but anything like this is always an opportunity to learn something.

I have just been back to check the measurements again.

Both sides have been machined. I find the part is not dished, but actually a wedge.

Measuring the thickness at the rear, centre and front, I find the rear and center are the same, but the front is 0.01mm wider. I suppose this could equate to a dish of 0.01/4 or 0.0025mm over a 2.75" width. (I think my reasoning and maths are correct).

Again well within the tolerances I need.

Pleeese stick to one system of units.🙂 I don’t mind which (I favour SI units - and powers to base 10 of those units - as they are so much more universal these days). Mixed units can make my head ache just sorting which is appropriate at the time or throughout the thread.

For most things a thou, over 3 inches, will not be noticed by many. Only precision kit needs better. The biggest problem is likely squaring stock with the mill.

I hope I can now get far better finishes than even JB’s pic shows - I can now take off tenths of a thou if minded to do so - but I might only surface grind when I want a more pleasing finish - not closer dims - or the same dims for a set of parts.

Just no point in going crazy on dimensions when not needed. Experience will improve your understanding of your particular machine so practice is the name of the game. Most on here do not have tool-room quality machines but many can make perfectly satisfactory working models with what they have.

I also know there is an awful lot I don’t know about surface grinding - but I will not let that bug me too much.

Our drill mill has no easy method of tramming, but is close enough for government work. When running an 80 mm shell mill in one direction of the X axis, there is one sided cutting and then going back, both sets of marks show. The Tom Senior can be trammed, but it is not easy, we took over an hour to get it to 0.01mm in 150mm. Trouble is that it moved every time it was tightened up and we hadn't had enough practice. I would prefer to tilt the work rather than loose the tram unless there is no choice.

If you are getting both types of marks, then that is giving you a choice of finishes depending on the direction of the last cut. That is an advantage for many, and the ammount of deviation from flatness would only be a few microns.

Edited By old mart on 24/12/2020 15:56:52