Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Fastest1 on December 15, 2009, 10:07:28 PM
-
Anybody understand this? The wording seems like it would be obvious but is it? I swear I have run threads both with and without it checked with similar if not better results without it checked. If it is checked, I uncheck spindle speed averaging as they would seem to conflict in the threading scheme, again I could have some flawed understanding of what is necessary. And today in a lapse where I had been using the manual speed control, I forgot to switch it back to automatic speed control on my spindle, hit cycle start and somehow the tool still hit the right place on each return pass at least 5 times before I realized my error. You know if I had a list of all the things to reset, change, plug back in, I couldnt see the machine. ;D
-
Well I am extrememly surprised you got a good thread with it unchecked. Normally you would see a stepped like leading edge to the thread.
Hood
-
Like Hood, I am also suprised. You should use always have spindle feedback checked. That provides for monitoring of the rpm while doing actual threading. Spindle speed averaging took the rpms and adjusts the next pass based on an average of the monitored rpms. If there was not much change in the rpm then you may get away with it. Don't fool yourself, as the spindle rpm changes even when not under load, and you don't have the equipment to measure it.
Not having those checked you will find that you have lead error or run into problems. If you do find you have lead tolerance issues it's your lathe system and not MACH. Testing shows the software can meet the criteria shown in figure 4.4.5 in threading on the lathe write up. The accuracy of the software exceeded my measuring capabilty.
RICH
-
I agree in theory. I was shocked at the results and as usual it was an accident. I was merely changing settings trying to figure out why speeds didnt match indicated speeds and proceeded before correcting the setting. That is exactly why I posted this. I might try to replicate it but damage usually occurs when I do things on purpose. So for correct threading I can see where spindle feedback would be a necessity however is speed averaging also to be checked? Seems like averaging would defeat the compensation that the feedback was provided for. What is up with the first pass? It starts in segments and then smooths out over the next few passes. Very interesting to hear and watch and to see a decent thread being cut. I am hypnotized watching it perform. The mathematical idea that you could match linear and rotational travel to a single point at 400-500 rpm is incredible.
-
What is up with the first pass? It starts in segments and then smooths out over the next few passes.Very interesting to hear and watch and to see a decent thread being cut. I am hypnotized watching it perform. The mathematical idea that you could match linear and rotational travel to a single point at 400-500 rpm is incredible.
Not sure what you are meaning by the first pass bit.
Heres a vid of my lathe doing the first op of the pullstuds I made for my Beaver Mill, not sure you will hear much though ;D http://www.youtube.com/watch?v=8J9q0NqoHhg
The material is EN24T and the thread was 16mm x 2mm pitch.
I also have a video posted of the full operation of the pullstud with the Computurn lathe and Bridgeport but its at much slower speeds as it was before I got the smoothstepper.
The pic below has the original pullstud for the mill on the right, the left is the ones I made and below is one of the ones I made for Brett (Chaoticone)
Hood
-
Hood, great work as usual. When I first start the simple threading wizard, the very first pass does a 3-4 segmented pass. I hear the steppers slow down and speed up. It does this only on the first pass and each pass after that smooths out and speeds up til S value is acheived. Usually 3-4 passes til up at full speed. The threads are very evenly spaced and appear very good. I have no knowledge or tools to verify quality of thread. I do know the nut fit better with less play than the bolt I copied the measurements from. I bought the bolt for .67 cents, I told my wife mine was better so I am going to charge .68 cents and claim it is American made. I do have some surface issues that might be the aluminum (home depot) I am using. I will have to try some different stock.
-
Hood, 2 more questions. What is a pullstud for? (I am guessing it is for a drawbar). Second now that I see your toolchanger, I have a new idea for my rotary table that is sitting on the mill table LOL. On my little lathe that could probably work as well as anything else I have seen. Just a round plate to bolt to the table with slots cut or broached with set screws. Too easy. Well I say easy but the rotary table (Sherline too) is bigger than the entire lathe table but I am going to look. It might be a good idea for my next lathe.
-
Your first pass doesnt sound right at all, the motion should be constant and the speed should be up to its max before the thread starts. How much of s start are you putting on the Z? If your steppers have fairly slow acceleration you will maybe need to increase the Z distance so that it has time to get to speed before the cut starts.
Yes the pullstud is for the mill, instead of a screw drawbar like on a manual mill mine has a hydraulic operated drawbar that has jaws on the lower end, they open as the bar is pushed down, go over the pullstud and then the drawbar raises again and the jaws close round the pullstud and pull the collet holder into the spindle taper.
You will need some form of brake or pawl on the changer to make sure there is no movement when the tool is in position as even on a small lathe the cutting forces will be fairly large, especially in threading. I also have a front toolpost on the lathe, you should see it in the video, I am about to replace with another turret that I got, looking forward to it, just need to find the time.
Hood
-
"So for correct threading I can see where spindle feedback would be a necessity however is speed averaging also to be checked?"
YES ( checking it should do no harm and i need to go back and find some specific info on it . There is only 450 pages of stuff on threading, so don't hold you breath while i look ).
"What is up with the first pass?"Do you mean what happens during the first pass? That is explained in section 2.1 of the write-up.
RICH
-
Rich, I always just thought averaging was used only for the display in Mach so that you dont see that jumping about if you have variations such as you may get with a vee belt that is unevenly worn but maybe not.
Hood
-
That is what I would have though too regarding the initial pass. I am not sure if it actually slows because the toolpath on the rod is very symetrical. It just seems to sound that way. Since there are so many variables in my set up, loose gibs and screws I will spend an hour or so adjusting before my next real passes. Also I will check all of my settings and possibly record it. I have been thinking about another youtube account for these videos. I know how much I enjoy other videos when I need to see new set ups or ideas. Of course mine would be more the crash and burn type but maybe someone could learn from my mistakes! I learn better by making my own mistakes, seems like I must verify the results for myself. Somebody elses testing procedure could have been different or flawed. Btw I start the thread .1 away from the piece. I do notice in the turn mode my accel is different and I really havent had the time to sort those issues out yet. I have played with the PID settings in an attempt to change it. In my mill profile the accel on the spindle is quick but not on the lathe. Much more of an incremental accel, same motor just different profiles so I am guessing it is a settings change. I dont get much over run either direction regarding rpms. Seems to stay within 3-4 rpms while cutting and 1-2 if no load.
-
Rich, I always just thought averaging was used only for the display in Mach so that you dont see that jumping about if you have variations such as you may get with a vee belt that is unevenly worn but maybe not.
Hood
Interesting, so maybe averaging helps quell the erratic rpms I see occasionally? Btw the first pass segmenting I was referring to even occurs with an air cut. I sometimes turn off my controller to reset my tool by manually turning the stepper when it is off (could be the problem why I have issues at startup LOL). I have read there are ways to reset the code or tool offset instead of the way I am doing it. I just havent learned them yet. Btw did I tell you how much abuse a little Sherline can take waiting for an idiot to hit the Estop? What an excellent learning tool.
-
Hood,
I may be confused with something else.....that's why i need to go back and look through the testing information.
It's one of those "the light bulb comes on but it's kind of dim". :D
RICH
-
"I must verify the results for myself. Somebody elses testing procedure could have been different or flawed."
Convince yourself and if you come up with an easier or better one than is in the write-up let me know.
"Btw I start the thread .1 away from the piece."
3 to 5 pitch diameters away is a general rule of thumb, but, it depends on your system.
"in the turn mode my accel is different"
The accel will vary and is dependant on how far you are away from the start point to the thread and how fast the spindle is turning. Make note of the accel required for a particular threading operation in the wzard.
"Seems to stay within 3-4 rpms while cutting and 1-2 if no load"
Waite until you try doing larger threads, or threading material other than Al.
That's were experimenting helps. Also consider doing flank cutting.
"Btw did I tell you how much abuse a little Sherline can take waiting for an idiot to hit the Estop? What an excellent learning tool. "
I sure had fun with mine and at least it will stop if there is a problem. You get a bigger lathe / horsepower it will do damage.
RICH
-
Fastest
It is often the case that wee machines can take more abuse than bigger machines. This is because they dont have much weight or power behind the tools so damage is less. My lathe is big and heavy and fairly fast, the saddle, crosslide and turret/toolpost probably weigh in excess of 400KG (880lbs approx) so if they smash into something at 10m/min the something has to be fairly solid to withstand it. Then the problem would be all that force has to go somewhere so my turret would take the brunt of it. Fortunately I have only had minor crashes during the learning curve of CNC lathes and just smashed up a few tools, the worst was the threading tool that is an upside down type and costs about £120 :( thankfully I only did that once ;D
Tool offsets and work offsets are all dependant on how you have things set up and how repeatable your replacing of tools can be. With a toolchanger you can set upp the tools to have the correct offsets in relation to your master tool, then all you need to do is find where that tool is in relation to your stock. Zero position on the X is always the centre line of the bed and Z is usually the end of the stock or at least the part you are making if the stock will need facing. With accurate homing you can have all your tools set up so that if you command a 20mm dia move it will go there no matter which tool you have in the slot but obviously that needs a means of placing the tools in the slot accurately and be repeatable.
Now bacjk to your first pass, is the spindle stable when the first pass is happening? if not then it is likely the axis is reacting to the varying spindle.
Hood
-
Rich and Hood thank you for helping in my understanding of this topic. I have truly enjoyed the thread. I was very aware that I would try to learn as much as I could on a machine that wasnt so destructive to itself before going bigger. I also could have bought a plug and play system but what would I have learned? This is fun, challenges the brain and isnt filled with groupies like GOLF. Rich I definitely wouldnt even want to tackle such a technical topic, nor could I. If I ever do have a suggestion that I think would help, I will definitely throw it out there. So far your doing great. "I must verify the results for myself. Somebody elses testing procedure could have been different or flawed." this comment was a statement on my stubborness/ignorance at others peoples experiences applying to my life. Been a problem since birth, I am 48 now and just beginning to realize it! I guess that might be better than never realizing it. Does age make you give in and conform? or make you realize you werent as smart as you think you were?
-
Ok my bad,
Spindle speed averaging was for use when timing was available. It would average the speed of multiple slotted discs
as compared to a single index over a few rpm. Now multi slotted discs are not provided for in the latest version of threading / MACH3 Turn. Timing should be disabled in configuration. Amazing what you find in the manuals.
BUT......I still need to look in the volume of info on threading as that dim light bulb above my head is still flickering for some reason. :D Maybe it's an indication that the brain is still active! ;D
RICH
-
Maybe it's an indication that the brain is still active! ;D
RICH
No probs with that Rich, easy enough to fix, go to the Operator menu then down to Brain Control and you can switch it off totally ;D
Hood
-
I switch it off in a different way but requires an additonal knowlegable control person. ;D ::)
Timing is everything when threading. >:D
RICH
-
;D