Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: ranchak on September 20, 2010, 11:03:41 AM
-
I have a series I ridgid ram that appears to either be loosing steps or has excessive backlash. I'm not too sure where to start. I am machining multiples of the same part, but the inconsistencies vary. One part will be off on the "top" of the part the other on the "bottom". I attached some pics to show the problem. It's hard to see, but in the first picture the bottom is smaller than the top and in the second the opposite, the top is smaller than the bottom.
Here's some further information. I removed the drive belt and installed a dial indicator on the y axis. I then moved the axis and compared the dial indicator reading to the DRO in Mach, typically I had about a .0003 - .0006 difference. But the difference didn't compound, meaning it didn't keep adding up, the axis was always within this difference. Of course I wasn't able to see what happens when the table moves a few inches. I am wondering now if maybe my motors are loosing steps?
-
If it is backlash then really each part should be the same amount off, that is presuming you are starting at the same point each time.
Sounds like your problem is more likely steps being lost or some intermittent mechanical issue such as pulley slipping occasionally.
To test backlash just set a dial on the table and have the point against the quill or head, move the table one way until the diaal moves a bit then stop, zero the DRO in Mach and change to Jog Step mode and move back the other way at 0.001" steps, as soon as you see the needle move read the DRO and subtract one step and you have your backlash for that axis. You can repeat at other points of the axis and also on the other axis.
What type drives do you have? what size motors? Do you have gearing between motors and ballscrews?
Hoodl
-
Hi Hood, my machine is a BP Series I CNC, I am using a PMDX-122 BOB, Gecko 203s and the original BP motors. The axis are belt driven. The pulleys are held by a key and taperlock pulleys. I checked and my lube pump is lubing the ways. I am going to try and reduce my acceleration and see how that affects my overall cut times and if there is a difference in the finished part. I will check to see what my backlash is.
-
It could well be the steppers, they are likely getting tired with the age of them. I have a Rigid ram as well with G202's and PMDX 122 but I fitted new steppers. Mine is a Boss 6.1 which was the UK version of the Boss6 I believe. The gearing on the UK ones were 2.5:1 where I think the USA ones were 2:1 and I think earlier versions may have actually been 1:1. The ballscrews on mine ar 5mm pitch where yours if a USA one will be 0.200"
Hood
-
Here is a picture of a similiar machine. You are right about 2:1 ratio and .200 ballscrew lead. I don't think that my machine is worn, the ways are well lubed and you can still see the scraping. I tend to agree with you that the steppers may be getting tired, but right now all I am doing is taking stabs in the dark. I really don't have any hard evidence to back any of this up. What are you using for motor tuning settings? I was using Velocity = 70, Acceleration = 4, but I spoke with Steve at PMDX and he thought that my Acceleration was a bit too fast. I am now trying my Acceleration = 2. I'll see what time difference this makes.
-
Mine was originally running 2500mm/min with accel of 40mm/s/s so that would equate to 100IPM and 1.57inch/s/s
After a while however I was not so hell bent on rapids but looked at increasing the accel, I dropped the rapids down a bit, think to about 2200mm/min(86 IPM) and was able to get the accel up to I think 120mm/s/s (4.7 I/S/S)
But as said that is with modern steppers of 916ozin.
I havent used the Bridgeport for probably six or more months as since I finished my Beaver NC5 mill going back to the Bridgeport would make me cry ;D
Hood
-
Oh BTW that pic is a series II is yours like that? Mine is a series 1, heres a pic of when I first did it, its dirty now ;D
Hood
-
No Mine is a Series I, I didn't realize that picture was a Series II. Also I have 1:1 belt drive. I removed all of my cabinets and installed a smaller cabinet to the back of the mill. I'll take some pics of it in the future, right now it's a mess and I am going to move it later this week. I am now running some test programs to see if I can replicate my problem. I really need to figure out where to concentrate my efforts. I'm getting mixed opinions about the motors, some people say replace them, others think the problem is elsewhere.
-
Mine was originally running 2500mm/min with accel of 40mm/s/s so that would equate to 100IPM and 1.57inch/s/s
After a while however I was not so hell bent on rapids but looked at increasing the accel, I dropped the rapids down a bit, think to about 2200mm/min(86 IPM) and was able to get the accel up to I think 120mm/s/s (4.7 I/S/S)
But as said that is with modern steppers of 916ozin.
I havent used the Bridgeport for probably six or more months as since I finished my Beaver NC5 mill going back to the Bridgeport would make me cry ;D
Hood
You can ship it over to me if you want, I wouldn't want you to be upset and break down in tears!
-
LOL well you supply the packing case and its yours.
Really though I am keeping it as a backup and maybe a test machine. Have a new back board board made up with AC servo drives etc but not sure whether to go ahead or not.
Hood
-
I just ran a few test programs, on the last program I noticed that when the machine came back to 0,0 it was off by .0008 in the y axis. I am running another test under the same settings to see if this happens again.
-
But did you start and finish going the same direction? 0.0008 is not very much on 30yr old ballscrews if it has been used.
Hood
-
I don't know what you mean by start and finish in the same direction. Well I haven't been able to get my machine to act up again. I know that .0008 isn't much, but that was in a 5 minute program, I thought in a longer program it might compound. I am now trying to run a pocketing program and see if I can get any errors. At this time I haven't made any progress, I think I will slow down my feeds and try running another part and see what happens. I don't have a whole lot to loose at the moment.
Is this forum running slow today?
-
Well what I mean is if you moved in a positive direction and stopped then zeroed the DROs and then ran the code and the your last move back to the Y zero was in a negative direction then it is likely your 0.0008" was backlash. If it is backlash then it wont increase, if its losing steps it likely will.
Yes forum is getting bogged down, Scotts been working on it, hopefully he will get it fixed soon.
Hood
-
Hmm....
A program with a prep move to take out backlash along with the dro's referenced to zero moving in one direction and then backtracking exactly in reverse you should end up at 0,0,0. For testing over time just make it a subroutine. So you could test linear or circlular movements.
Add the poor mans pulse counter using a disc to rotate along with the motor shaft gives you a phyisical reading if a step is lost. The bigger the disc the more accurate or even a 75 cent protractor ie; approx 2 deg=a step.
The smaller the number the harder it is to isolate what is contributing to it. Heck it could be the indicator, a little backlash, etc and gets difficult to isolate each part. Even timing belt tension can account for lost movement which you may think is an electronic pulse missed.
FWIW,
RICH
-
Well I ran a few different programs to see if I can replicate a lost move, but I failed. I Well I ran a few different programs to see if I can replicate a lost move, but I failed. I am cutting a part at a much slower IPM to see if that will show anything. One thing I am doing is I am keeping track of the temperature of my motors. So far the x and y motors have gotten up to 135 degrees each. The y motor retains heat longer, I had to run an errand and when I came back the toolpath had finished, the x moltor was 109 degrees and the y was 123 degrees. I don't know if maybe there is a breakdown in the one motor when it gets hot? Right now I'm running at 3 IPM, the x motor is 106 degrees and the y is 124 degrees.
-
Well another scrapped part, just like the others, off in the Y axis. I don't know if it will make a difference but I picked up to new steppers today. Tomorrow I will see if this solves the problem, if it doesn't then I'll start to look elsewhere. One thing that I discovered by accident is last night I accidently left my machine on, this morning I checked the temperature of the stepper motors, the x motor was 96 degrees, the y was 126 degrees. I didn't even think of checking the z motor. It only reaches 96 degrees when running.
Hood, what IPM do you cut with on your BP?
-
Hood, what IPM do you cut with on your BP?
All depends on the material and size/flutes of cutter but anywhere from 40mm/min (1.6 IPM) to about 1,000mm/min(40 IPM)
Hood
-
Should be interesting to see what the new motors will do. A motor that dosen't get warm is not doing any work, the more work the motor does the hotter it will get. As long as the motor is operating in a temp range below the rated insulation you should be ok and with age it would be prudent to operate it taking age into consideration. Greater than 140F you can't hold your hand on it very long and can get burnt. The motor case temp you see is influenced by the how well heat is being removed from it so dirty fins or evan different paint thickness and location can influence it. So relatively small temp changes of say 10 deg diff is subjective for any conclusion. I have motors on my mill that are are older than me and work great. don't be suprised if you see a difference in microstepping position as i found different motors required small change in the steps per unit and would attribute that more to the quality of the motor. Let me remark that it's within a roatation or very small moves.
Just some comments on the motor end they are generalizations.
RICH
-
Motors are installed and we are making chips. Hopefully my next report will be of a completed part.
-
Success at last! I gues I should get at least two parts finished before I start celebrating. It appears that the motor change has made a difference. My finished part is not perfect, but it does fall with + - .005 tolerance. I probably need to check the backlash and see how bad that is. I might try to slow my finishing cuts down a little also. Thanks to everybody who has helped and especially to Hood.
-
Should be interesting to see what the new motors will do. A motor that dosen't get warm is not doing any work, the more work the motor does the hotter it will get. As long as the motor is operating in a temp range below the rated insulation you should be ok and with age it would be prudent to operate it taking age into consideration. Greater than 140F you can't hold your hand on it very long and can get burnt. The motor case temp you see is influenced by the how well heat is being removed from it so dirty fins or evan different paint thickness and location can influence it. So relatively small temp changes of say 10 deg diff is subjective for any conclusion. I have motors on my mill that are are older than me and work great. don't be suprised if you see a difference in microstepping position as i found different motors required small change in the steps per unit and would attribute that more to the quality of the motor. Let me remark that it's within a roatation or very small moves.
Just some comments on the motor end they are generalizations.
RICH
I agree with what you said, I think what was happening is there is a very small amount of error that adds up over time. I normally don't run programs that are this long so I never noticed this problem although I am sure it always existed. Now i need to figure out backlash and see if that changes anything.
-
Keeping my fingers crossed for you :)
Hood
-
The motors have made a difference, the machine seems to be smoother and I just cut a large pocket 10" x 9.25", I used a 2 flute .500 end mill, .006 cut, 40 IPM, 50% stepover, the cut was extremely smooth and consistent. Well, up until my Smoothstepper lost communication with Mach. This happened 3 times in a row, so I installed my new parallel port card and could not get any movement. I got out the old PC that I was using a few weeks ago that was giving me some problems. I managed to get through the project, but there where 3 times that I saw steps lost. I was able to get back on track and get the job done, but I had to work until very early in the morning. I need to get my new PC running, I still have one job that I have to get done by Monday. Hopefully I can find the mistake that I made and get back on track. The port address is something weird like cc00. Back to the motors, the new motors did heat up to 126 degrees, but they would cool down. If anybody is interested I got them at www.anaheimautomation.com. I paid $245 for a Nema 42 motor.
-
Also Steve at PMDX was a huge help in trying to determine the source of my problem as well a number of members here on the forum.