[lcvette]

I think your problem is your are simply refusing help... You keep going on about the Oldham's, they are not your problem.
At what some point you are going to have accept we know what to do and just do it? You say you have done everything but your results are off still. Video your process so we can make sure you are doing it correctly, there's no shame if it's wrong just a possible solution we can help you with.

[onocyclone]

Ok tomorrow I will recalibrate the axis and set the axis calibration to 320 to check everything is as it would be.

I'll then calibrate the axis with ought backlash comp on and once it is good I'll turn the comp back on and measure the backlash.

Once everything is looking good in machine a square block with a circle on and evaluate it again.

I'm not refusing help im just confused as to why I get pretty much perfect results when using a DTI clock. Then referring to a DTI finger and getting the same results.

I'll just start from scratch again and see where I get.

Cheers,

Olly

[lcvette]

If you can film it that would be the most helpful and include the setup for measuring not just the result you are seeing. That's probably the most important.

Believe me I cracked my skull trying to do some of this 8 years ago before someone pointed out a simple mistake I had been making, sure that it all feel into place. So the less you omit, the more likely we'll spot the ah-ha moment.

[treyjugson]

Hi guys,

I've got something interesting to report,

I machined some small parts with a 2mm end mill and these are the results:

Attachment 273460Attachment 273462Attachment 273464

It's quite interesting how internal features are more accurate but all of it is programmed to climb mill !!

Any ideas guys ?

I'm going to investigate the 12mm tool I was using before, see if it is in the collet correctly.

Cheers,

Uhhh, so what's the problem? Your worst is only .003 inch, the inside .0008 inch. You seriously don't think it should be 0 every time? Calipers are a blunt tool anyway. Did you program a small finish pass? You are forever going to be chasing your tail trying for perfection from these hobby machines, .0008-.003 is good enough.

[lcvette]

Those were only one set of results, his others were significantly worse, read the rest of the thread, I would agree otherwise.

[treyjugson]

Those were only one set of results, his others were significantly worse, read the rest of the thread, I would agree otherwise.

No these were his latests results after making corrections, did you read the thread? There comes a time to call it quits and accept the completely acceptable results unless pulling hair out is your goal.

[lcvette]

Yes... I did.... These were internal dimensions not external so there is a variance between the two, I would not call that OK or acceptable. I'm not sure why everyone keeps saying these machines shouldn't be accurate because they are hobby mills. That's just lazy in my opinion but if your OK with the results he lists through out the thread power to ya.

[treyjugson]

Yes... I did.... These were internal dimensions not external so there is a variance between the two, I would not call that OK or acceptable. I'm not sure why everyone keeps saying these machines shouldn't be accurate because they are hobby mills. That's just lazy in my opinion but if your OK with the results he lists through out the thread power to ya.

The 18.08mm measurement was an internal dimension? LMAO Who said they aren't accurate? .0008-.003 is very respectable which is my point to the OP. Trying to achieve much better from these machines is an exercise in futility but keep going if you think different. I only commented about his most recent results, never said anything about his previous results being acceptable now did I. Don't misquote me.

[onocyclone]

We are talking millimeters here (mm) not inches. Metric not imperial.

[lcvette]

"It's quite interesting how internal features are more accurate"

From his post, the one you quoted him on.....

If you go back and find his external results as I mentioned they are much further off, big difference, not trying to get in a pissing match here just trying to point out that you replying on incomplete information.

If you'd like to add to the conversation based on complete info I encourage you too, perhaps you will see something the others and I have missed. But those results aren't accurate for squat compared to what he should be achieving. To many other 0704's out there doing it with the same parts he's using to accept it as good enough IN MY OPINION.

[treyjugson]

We are talking millimeters here (mm) not inches. Metric not imperial.

Yes I know that's why I made a point to say inch, .08mm = .003 inch, .02mm = .0008 inch. I made the conversion for the US audience so they understand how small an error we are talking about. Measure one of your hairs you pull out to compare to see how minute an amount is being worried about here. I'll keep my hair thank you, not enough to waste worrying about error this small.

[treyjugson]

"It's quite interesting how internal features are more accurate"

From his post, the one you quoted him on.....

If you go back and find his external results as I mentioned they are much further off, big difference, not trying to get in a pissing match here just trying to point out that you replying on incomplete information.

If you'd like to add to the conversation based on complete info I encourage you too, perhaps you will see something the others and I have missed. But those results aren't accurate for squat compared to what he should be achieving. To many other 0704's out there doing it with the same parts he's using to accept it as good enough IN MY OPINION.

Have you ever calibrated your calipers? Like I said they are a blunt tool, I've seen some that the inside jaws are .005 inch (.127mm) off from what it should. This could easily explain the difference in his inside and outside error. Please show me the g0704 that gets better than .0008 inch, how's yours?

[lcvette]

He's .003" std out on his bad axis, which is what the world uses as far as tolerance holding, not the best, are you saying you go off the lowest error and call it good?

As far as tools, I use a brown and sharp vernier and it's nuts on using a gauge block to test, you need better tools, .005" out....... Throw your calipers away they are junk!

[treyjugson]

He's .003" std out on his bad axis, which is what the world uses as far as tolerance holding, not the best, are you saying you go off the lowest error and call it good?

As far as tools, I use a brown and sharp vernier and it's nuts on using a gauge block to test, you need better tools, .005" out....... Throw your calipers away they are junk!

I said .0008-.003 is acceptable, again! I didn't say I USE calipers with inside jaws off .005., SEEN them. I use Starrett. So do the parts you make on your g0704 measured with brown&sharps consistently come out better than .0008-.003 inch? His calipers look like cheap junk so HE'S the one that might want to check them.

[lcvette]

Yes the parts I make are generally within .0015" or better. If yours aren't I suppose you don't need them to be, for his parts he said he needed them to be. I don't need it on everything, however sometimes I do, the capability is beneficial. The machines are capable of you need them to be with a little understanding of what your working with.
I would remind that in the end, it boils down too what the OP is trying to accomplish, this is his thread after all so that should be how the topic is replied too. So if you feel you have anything else to add beneficial by all means please do. Otherwise it's pretty clear this is a pointless argument.

[lcvette]

Have you ever calibrated your calipers? Like I said they are a blunt tool, I've seen some that the inside jaws are .005 inch (.127mm) off from what it should. This could easily explain the difference in his inside and outside error. Please show me the g0704 that gets better than .0008 inch, how's yours?

This says nothing to .003", only .0008", and you are pretty rude to start swearing at me. Not sure why you are so riled up, I'm simply replying to your posts.

[treyjugson]

This says nothing to .003", only .0008",

Yes it does, and why is that? Because you have said that my acceptable tolerance of .0008-.003 is not acceptable to you. So if the minimum of .0008 is not acceptable to you Please show me the g0704 that gets better than .0008 inch, how's yours?

[lcvette]

.003" > .0008"

In this discussion, we are talking about machine accuracy, so the best the machine can handle in its current state is .003" off of a target of 0.0000". So maybe I don't understand your perspective certainly, but from my understanding, the larger number here is the important one, it dictates his ability to machine parts too there design specification. Now if the designed tolerance is +0.003"/-0.000 he'd be OK.

He states he needs tighter tolerances than that. What have I missed?

[LeeWay]

The busted end mill could very well explain everything you guys have been...um...discussing.

[treyjugson]

I have done it that way also. Should be set correct. .....

This was your response to the OP's video on how he checked his backlash. Several people have pointed out that this method does NOT do anything to show the backlash in the axis. As you think it did makes for a case of the blind leading the blind. Will you now go on a tirade with them for trying to set the OP straight?