PDA

View Full Version : Pretend 4 Axis Cutting (LinuxCNC using indexed 4 axis to cheat not having G93)



subnoize
08-31-2013, 05:50 PM
[REDACTED]

subnoize
09-02-2013, 06:55 PM
OK, I finally got this silly thing to work. The results are pretty good considering I'm a complete noob at this stuff!

The idea was to use 4 axis index milling to machine 360 degrees around the model. Since madCAM doesn't support G93 "Inverse Timing" I can't cut using 4 axis continuous milling at this time but I wanted to try anyways. I figured that instead I could do the milling as index cuts and go back and fix the A axis speed by hand.

Well, it worked! I used planes to bisect the model allowing me to do 4 cuts, each 90 degrees of each other and ended up with a full 360 result. I could never get the cutting planes to work using the index mode. They seemed to work just fine in 3 axis mode. Any insight here as to why this didn't work would be great. Using planes to stop the machining was hard to do right.

Does anyone else have any cool ideas on how to do this in an easier way?

Thanks for any advice and insight!


snz

Gregore
09-05-2013, 01:16 AM
I too also use surface plane to stop the tool paths , you could try making 2 planes 90 degrees to each other and then joined together through the center of x and also make a drive surface that covers the top area of this wedge shape then just rotate the drive surface and the stop plane for each tool pathing.

Have you tried to do a continuous rotary to see if it works? Mine works fine with a controller that needs g93 . Now granted I am cutting wax for jewelry which has a very wide range of chip loading and still keeping a good surface finish. Also you can do all your cutting back and forth in x and that will lessen the effect of not having g93 to keep the feed rate at the set value.


P.s. I did email joakim to ask for g93

subnoize
09-05-2013, 02:13 PM
Also you can do all your cutting back and forth in x and that will lessen the effect of not having g93 to keep the feed rate at the set value

Ha! I did not think about that, using the X axis. I was cutting along the Y so of course it takes forever when traveling at 15 degrees a minute as oppose to 15 inches per minute. Cutting in the X axis will mean that the Y axis motion will be limited to the step-over distance which will never be much over 0.0225" for the work I'm on right now.

Right now I am still cutting machinable wax because the mill is in my dining room. I'm worried if I start cutting metal then the swarf will get all through the house and my kids will get it in their feet and stuff. From what I hear, steel swarf is nasty and needs to be in a place that can be easily cleaned up.

So, thanks for the ideas! I will certain explore them more while we wait for G93.

Gregore
09-05-2013, 02:32 PM
I think that if you are cutting in wax that you can set the radial offset (in the controller) close to the size of the part and you will have no issues with simultaneous 4th axis cutting .

subnoize
09-05-2013, 04:19 PM
I think that if you are cutting in wax that you can set the radial offset (in the controller) close to the size of the part and you will have no issues with simultaneous 4th axis cutting .

That is how the Mach 3 folk compensate for the lack of G93 in 4 or 5 axis milling. It works great as a work around for those like yourself in the jewelry industry. The models are smaller, round and mostly even surfaced. The stock material is very soft. Also, the Mach 3 folk set the A axis max to never be more than the maximum surface speed / chip rate at any radial.

For the moment this is, in fact how I have done all of my 4 axis continuous milling. It just get real darn slow when you want to mill something where the radius is from 0.0625" to 2".

Gregore
09-05-2013, 04:51 PM
I have managed to mill stuff with extreme differences in size on the same part, with no ill effects. could be due to the type of controller I am using ( DynoMotion kflop) or it could be as you say the soft materials. I do know that we did not set a max surface speed / chip rate at any radial. we set the radial to max so we had no skipping steps on the stepper motor when running the 80 to 1 harmonic drive. I always thought the mach 3 handled 4th and 5th with out g93/ with out cheating it.

subnoize
09-05-2013, 06:46 PM
I have managed to mill stuff with extreme differences in size on the same part, with no ill effects.... or it could be as you say the soft materials. I always thought the mach 3 handled 4th and 5th with out g93/ with out cheating it.

Wow! Yeah, wax is fun! Machining metal is were the feed rates become very, very important. The feed rates are also a safety issue, not just a factor in surface quality. When these end mills shatter they fly off with a great deal of force.

Also, at $40 to $400 per end mill, I can't afford to break them. Stock is expensive as well, especially the Titanium and Aluminum I will be working with.

Mach3 forums is where I started learning about RS274/NGC in the first place. I don't think they would implement an out of date standard but they could have. I found most people using Mach3 use BobCAD/CAM, which fully supports RS274/NGC.

Gregore
09-05-2013, 08:23 PM
I assume RS274/NGC is a type of gcode

I some how do not think you will be doing titanium without g93 ! :eek: (are you listening Joakim?)

My spindle 150 watt nsk ( upgrading soon) sort of keeps me to very light step overs so with the little bit of metal I have done I have not had too many problems. though i am sure I will break my fair share of bits as I mill more metal parts.

I sure am lucky to have a friend that sort of oozes this type of knowlege ( the head machinist at Dupont santa barbara) I would be totally lost with out his help. From the time I got the mill to making usable castable parts was about 1 week and no bits broken , just a few pieces of wasted wax. And the finish is as good as the best I have seen in my industry (after a little rhino settings help from joakim) .

what are you hoping to make?

subnoize
09-05-2013, 09:21 PM
I assume RS274/NGC is a type of gcode

A. RS274/NGC is an interpreter which reads numerical control code and produces calls to a set of canonical machining functions. NIST is the standards body for the USA. You can find more details here if you are extremely bored. NIST Manuscript Publication Search (http://www.nist.gov/manuscript-publication-search.cfm?pub_id=823374).

NIST is the National Institute for Standards and Technology. They defined our telephone system, railway gauge and time (via the atomic clocks). The RS274NGC was to simplify CNC machines off-loading processing to the CAM software and therfore dropping the price of CNC machines. NGC stands for NextGeneration Controller.





I some how do not think you will be doing titanium without g93 ! :eek: (are you listening Joakim?)

Titanium is actually milled with a very slow spindle speed which surprised me. If you mill it really fast you have to bathe it in coolant.


what are you hoping to make?

Well, to make it brief and to the point, I am a member of EAA.org and I am building my first aircraft. The design is John Monnett's Sonerai IILS, a low wing VW powered plane that cruises at 140 mph and tops out at 170 mph without redlining. You can find out more here; Monnett Sonerai - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/Monnett_Sonerai) or here Sonerai.net - Home of the Sonerai Community - Index (http://sonerai.net).

So, not only have I gone from computer programmer to machinist / CNC programmer but I am also learning to weld 4130 tubing on top of building my own engine from the block up! Welcome to Great Plains Aircraft! (http://greatplainsas.com/)

So, the CNC thing is kind of part of a larger picture....

svenakela
09-06-2013, 03:51 AM
If you're making an airplane, get the hell out of the living room and get yourself at least a garage!

Dan B
09-06-2013, 05:50 AM
Titanium is actually milled with a very slow spindle speed which surprised me. If you mill it really fast you have to bathe it in coolant.

Do a little research into cryogenic machining: For example:

Video: Cryogenic Machining of Titanium : Modern Machine Shop (http://www.mmsonline.com/videos/video-cryogenic-machining-of-titanium)

I know this is beyond what you can do at home, but I've seen this in action, and it's the ultimate method for machining titanium.

Dan

subnoize
09-06-2013, 08:02 AM
If you're making an airplane, get the hell out of the living room and get yourself at least a garage!

That is the plan! My lease is up in December and I'm looking for a place with a nice two car garage...