Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: cncnovice2 on December 21, 2008, 02:10:49 AM
-
I'm having a problem with my X-Axis losing steps Or is it microsteps and not returning to zero. The mach3 says it is back to zero but I lose about .0005 or so each time I run this program:
F20
G20G90
M98P1234L3
M30
O1234
G1X4
G1X0
M99
System consists of the following... Computer 1.83Ghz Athlon, 1Gig memory, PMDX 120 Breakout board, Geckodrive G212, Homeshopcnc 1200 oz.in. Stepper motor, 1:2 timing belt, 5 TPI ballscrew on 9x42" table.
When I run the above program with a dial test indicator zeroed on a block fixed to the table, it reads a lower number at the end of the program run, ie. the machine is slipping toward the positive direction. I don't know what is causing this as changing the feedrate, acceleration, step pulse timing, etc. has not helped. I have gone through the Windows Optimization procedure, and that didn't help. What should I try? I'm not even sure it is software or hardware related... Thanks for any help or suggestions you might have, Don
-
What does the driver test look like?
Have you tried changing the active sate for Dir?
Have you got the correct current set resistor in the Geckos for your motors?
Is it just the X axis? If so have you tried swapping drives? have you checked there is no binding in the axis? Pulleys etc all tight?
Hood
-
Just echo what Hood has posted. Easy to get to some tolerance then the "system " comes into play with
each part, mechanical and electrical, even the indicator,possibly having an influence.
Be meticulous.
RICH
-
First a clarification. The x Axis is slipping to the positive direction only. If I se the program to go negative it behaves and repeats perfectly. The Y-Axis is also doing it only it only goes toward the negative direction... The MPG when set to .001 inc. repeats perfectly on both axes, but when set to .0001 inc. doesnt repeat in X. Again, it loses steps to slip toward the positive in X. Making repeated back and forth moves yields mor and more error. The indicator is mag base mounted on the head and touching the fixed jaw on the mill vise.
Step is active low for all axes. Direction is active low only for the X Axis. Could this be a mechanical problem? I will try to post pictures of my setup. This may take another post since my card reader is finiky. Thanks, Don
-
Do you have a Breakout board with optos on it? Could be the optos are too slow. Also what are your steps per unit for the problem axis. Try setting the active state the other way and then you will have to reverse the axis direction in Homing and limits to correct for that.
Hood
-
Hi, again. Breakout board is a PMDX 120. Steps per unit is set to 10000. Kernal speed is 45Khz. I should be able to post/attach a picture of the control setup now. the cable in front of the BOB is the E-stop sense input from the BIG RED BUTTON. Computer is off image to left. Gecko powersupply is upper left. 800VA 70V. The Current set resistors are set per Geckodrives calculation formula and the motor spec. The signal cables to the G212s are 4 conductor Shielded Com cable. The shield is grounded to the aluminum plate at the Geckos only. The SJOOW wires at the bottom are the outputs to the motors. I've been using this machin for about 2+ years now and just recently reinstalled the Windows, and Mach. Yes, It has been "optimized" per the Mach Support instructions. Thanks, Don
-
I reversed the X axis in Homing/Limits then ran the mirrored version of the program with the result that the table is still slipping to the left. Sounds like something other than the computer? Don
-
OK your steps per unit are 10,000 and I am presuming you are in Inches, if so then that is a resolution of 0.0001 inch. As part of that is micro stepping then you are never going to get things that accurate, it shouldnt really be cumulative though.
Have you tried dropping the acceleration a bit to check that its not just skipping a beat every now and then.
Hood
-
Yes, I have tried accel values as low as .1 (ramps up very slowly) and got the same results as accel. 4 which is what I'm using now. Don
-
Yes sorry just re-read your opening post and saw that. Dont see if you have tried changing the active state of the Dir pins but I tend to not read things properly so apologies if you have :)
Hood
-
.0005 is a small amount(;-) it could be backlash, Screw torque, bearing preload, bed gibbs need adjusting, Vibration and THat long arm on your indicator, gravity and that large vise on one end. ANY and all could add up to .0005. I bet you can lean on the mill and get that amount.
If you can ref home then do so and run a program. Then do a verify and see how far it is off from the home position.
Just a thought, (;-)
-
no problem. I've tried changing the active state of the Dir pins and the result is that when I start the program that moves + first, When it gets done... It still moved positive. It seems to be about .0005" each time I run the repeatability test. Yes it is cumulative. Running the test with the accel at .1 just now after 6 runs, I've accumulated about .0032" error. This agrees with the Dial on the right end of the leadscrew. It is a standard Bridgeport type. This is a very frustrating anomaly. I haven't had this happen before until just this last week. Could adding the 4th Axis possibly be related? It was added about 2 weeks ago. Pretty straight forward. Thanks for the help, Don
-
To Vmax549, Thanks for your input. See the above post. The machine is pretty tight, but smooth. We rescraped the ways and refitted the gibs about 2 years ago when w got it and prepared it for the retrofit. There is about .0005" mechanical backlash in each axis. If it was staying within that and not creeping off in one direction I would be fine. I haven't hooked up any REF switches yet... To lazy. BTW, the small bronze handwheels were installed to prevent accidents, shuch as bruized legs, blown parts, etc. They were raw castings I turned. Sorry about the clutter on the table. Don
-
Don,
Just so you know, i am are trying to help and understand frustration. So bear with me on information requests.
As i said before, it could be any part of the system or combination of the components. Sometimes easy to fiind
and sometimes just elusive as a fox. Little things can be additive and as you go deaper into the each and every
item it can even become un-nerving.
What kind / accuracy grade of ball screws are you using?
Are the results the same along the axis? (say try three different points along the screw travel, if 10 inches of screw travel,
then say at the 2-3" / 5-6" / 8-9")
Can you check or verify the steps / unit over a longer distance of travel? ( say 10", 5" movements )
But before you aggrevate yourself, or we drive you crazy, try the following as it is easy to do.
The only way I know of isolating / checking the steps sent to a stepper without electronic gismos would be to
to put a index mark ( fine line) on the motor pulley ( can increase the size via paper circle), send a G0 move for say 1 rev of motor, set the pointer on the index mark,
do the same move,when it completes the move the indexed point should be right a pointer. Over and over again. if you do the same move for 15 or 50 times does it come off the mark? The system has been isolated down to software + sending devices+motor. Thus it eliminates the rest of mechanical items in checkout. You can work your way down the mechanical train and look for changes in movement. You can also choose to not microstep.
Poor mans encoder readout. ;)
RICH
-
Hi, Rich. Thanks for the suggestions and encouragement. I'm also posting in the Yahoo! Gecko forum and Maris has just suggested the pointer on the motor too. he also suggests a couple of other tests for repeatability at different distances to see if it is drifting relative to the amount of travel. I will try those steps soon and report back. He also suggests trying "no auto standby" on the option header as another test. Thanks, Don
-
Hi, again. below is a description of the tests Mariss had me try and my results.
Mariss, I just ran the tests you suggested. 1" and back and lost
about .0002" each repitition. Running the 10 reps lost .0020" and
100times lost .020" . The mark on the stepper pulley was off a
corresponding amount as well as the dial on the ballscrew end
opposite the drive. Then I set the No Auto-Standby mode and it had
no effect. What next? Thanks, Don
--- In geckodrive@yahoogroups.com, "Mariss Freimanis"
<mariss92705@...> wrote:
>
> 1) Set up a test program where you move some distance (1"?) and
back.
> Repeat it 1, 10 and 100 times. Does the error grow a repeatable
> amount; 1, 10 and 100 times bigger?
>
> 2) Take a Sharpie pen or a scribe. Mark a line from the motor
shaft to
> the motor body. Does the mark line-up after 1, 10 or 100 tries of
(1)?
> If not, is the angle error 1, 10 and a 100 times greater when you
run (1)?
>
> 3) Try setting the drive to "No Auto-Standby" on the option header.
>
> Mariss
-
In your pics of the drives I think I see caps connected to them, which terminals are they between and are there meant to be caps there? I have only used 201 and 202 drives and I cant recall caps being needed there.
Hood
-
Hood,
You should have caps on the 201 drives as noted in the liiterature from Geco.
That cap can save you from blowing a drive.
Rich
-
should it be in the positions these drives have them? Its been so long since I had the 201's fitted I cant remember but seemed to remember they were on the Voltage Inputs from the PSU.
Hood
-
Here is a similar topic. http://www.machsupport.com/forum/index.php/topic,7364.msg47127.html#msg47127
I had a similar problem that was resolved by increasing the PULSE width.
Here is a clip.
RC
-
Hood,
In reply to your post #18 and my post #17 info from the Gecko site.
"If the cable run from the power supply to the drive exceeds 18” or if a fuse is used, place a 470uF
100 VDC capacitor across each drive’s power terminals 1 and 2. Make sure the capacitor’s “+”
lead goes to terminal 2 and that its “-“ lead goes to terminal 1. The capacitor is necessary
because the drive draws power supply current in 20 kHz pulses and needs “flywheel” to work
properly."
RICH
-
yes Rich, thats what I was meaning regarding their location. In the pics that were posted it looks to me like the caps are on the Step or Dir terminals. It may not even be caps but certainly looks like it. I would imagine the G212 would be based on the G202s and as such would have the power caps built in but possibly not, in any case I am wondering about the caps I think I see in the photos.
Hood
-
:)Well Hello Hood,
This may be the only difference between the two. Didn't see anything else.
RC
-
Just saw this
-
RC, this is what I am seeing, looks like a Cap but cant determine where it is actually fitted to, just thinking if its in the Step or Dir connections then that could well be an issue.
Hood
-
I see what you seeing Hood and if it's a cap it dosen't belong there.
RICH
-
Hood, I saw that as well but couldn't make it out. I assumed that was what you were referring to.
Looked through the manuals, like Rich said, nothing like that mentioned.
RC
-
Hi, the mystery stuff is not a cap. The red is a collar on the grounding lug underneath the terminal block to the mounting screw for the shield on the cable.
The black tape on the end of the cable keeps the shield foil and the 4th, unused conductor held back tidy. The current set resistors are made up of 2 resistors in series with red shrink tubing on the outer ends to prevent smoke... Z axis has 3 resistors due to it being a different current value. The Manual does not require additional Capacitance at the terminals like the earlier units did. The case is wider to accomodate the built in unit. Hope this helps dispell any Red Herrings ;)
Thanks,Don
-
LOL, bet you have been reading this for hours and kept saying I will wait a bit longer before I post and tell them ;D
Hood
-
Hood, Naa. I just got the chance to see the postings. I know the wiring looks a little rough, but it is actually safe and clean. It has worked for over 2 years.
Overloaded, I just tried raising the step pulse to 5 and had no change. Had to exit Mach to get the number to persist, as in the other thread. Earier version of Mach3...
Thanks, Don P.S. I'll be going to work soon... I Owe, I Owe, So Off to Work I Go...
-
Wiring looks pretty good to me :)
Hood
-
Don,
When you get home, the testing of the stepper awaits you.
Till then.......pay,pay,pay or no play! ;D
RICH
-
Hi, again, and Merry Christmas to you all. I've had a chance to play a little and checked some things with my breakout board... The breakout board is a PMDX-120 ( www.pmdx.com) and it has a feature called Step and Direction Conditioning Modes. Mode "0" is no conditioning. Each of the other modes regenerates or retimes the step pulses or latches the direction signal to either the rising or falling edge of the step pulse. I found that modes 1 and 2 both doubled the error. None of the other modes did anything different than mode "0". Mode 1 provides step regeneration on the rising edge of the incoming step pulse, and No direction latch. Mode 2 latches the direction signal to the rising edge of the incoming step pulse, but does nothing to the step pulses. I am wondering if this means something??? Thanks, Don
-
Hi, again. Brain is still cogitating on this problem and I was reading thru this thread again:
Here is a similar topic. http://www.machsupport.com/forum/index.php/topic,7364.msg47127.html#msg47127
I am wondering a couple of things... Does anyone know what symptoms a bad Buffer IC would exhibit? Reason for My asking is that the PMDX-120 B.O.B. has a step and direction buffer that is socket mounted in case there are any "mishaps". The buffer is a SN74ACT254. I am wondering if replacing it might be worth a try... Might take a while to get one, but if it is a breakout board issue, it sure would be nice to fix. If anyone has any experience with these devices, I would appreciate it, Thanks Don
-
Hi, Again, Guys. I think I've finally solved this one... I downloaded and installed the latest version of Mach3 from the website, and Lo and Behold--- The test programs work perfectly! :) I thought I was in for a defective mobo or BOB but it was just something in the software... Repeated testing in all axes and both directions are showing perfect repeatability. I quess there have been some major improvements in the Lockdown version of Mach3. I guess the moral to the story might be to always use the updated version of the software. Good work, Thanks, Don :)
-
Good for you Don,
For closure one more question. What Mach version was used before and after?
RICH
-
Arrgh! Its Back! Tried to run the repetitive program last night and at first it seemed all right. tried upping the feed from 4ipm to 20ipm but at F8.0 it lost steps again. I tried acceleration set as low as .16667 (default from entering .01) and up to 20 but it loses the same number of steps each time... So, I shortened up the amount of X travel to .100 to save time and adjusted the feedrate until I determined that the lost steps occur at anything over 7.5887ipm. At 7.6ipm feedrate no steps are lost at any acceleration setting. So, I shortened up the X-travel while leaving the acceleration at 2 and the feedrate at 2 and found I was losing steps again. Shortened up the X to .0002" and the machine just vibrates over the .002" of lost steps. Now the wierd part. My machine has a max resolution of 10000 steps per inch. I set line 6 of the program to .000005 - half the minimum travel resolution of my setup. when I ran the loop program, the machine almost smoothly moves positive .002". ??? Is this a software issue? Is this a computer issue? Is this a breakout board issue? I was leaning toward the BOB but theoretically, the machine shouldn't move any where if the commanded movement is smaller than the resolution of the machine and its software settings, shouldn't it? At any rate, new sn74ACT245 buffer should arrive next Monday. I did try this program both in Constant Velocity and Exact Stop modes. It made no difference. When I got done running all these tests, I reset the distance to X .100 and feedrate 2 and now it was losing steps at those settings too. :'( Did something warm up and puke, or was I imagining things at first... Thanks for any help you can provide, Don
-
I remember a long time ago people having problems with similar issues and the Geck Servo drives that had the step multipliers, well at least I think I do but my recollection is hazy. I am sure the cure was to take the chip out and just use as the normal servo drive, Can you do that as a test?
Hood
-
I dont have the other component that plugs into the Gecko mainboard, but still wonder why the computer would send out signals to the drives that are smaller by 1/2 than the resolution limit of the Mach3 Motor tuning setup? Also the Gecko on this axis is now the one I recieved only a few weeks ago. ???
-
Hi,
I am have a terrible time with lost steps. My Machine has been working perfectly for a long time now. But out of the blue - lost steps.
I have been chasing my tail for over a week with little joy until now. I have checked everything printer ports, power supplies, mechanicals, static electricity,wiring, software, gcode - all ok!
The one thing I have changed prior to the problem becoming evident was : I upgraded from mach3 version2.63 to version R3.042.020.
This appears to have caused the missed steps.
I have reverted to R2.63 - no more missed steps!
Still testing!
Hope this helps. Or maybe its muddled things up. but I has fixed my problem.
Cheers
Bruce
-
Hi, I just upgraded from Version 2.63 to the latest version and I STILL have the problem... I just reset my Geckos to 10 u/s per Step instead of 5 and the loss of steps on My test program is nown 1/2 what it was. Looks like it isn't the Geckos. Hopefully, when the new Buffer IC for the breakout board shows up I'll have better luck. I doubt the software is at fault unless there is a setting somewhere that got toggled without my knowledge ( Very unlikely). If it was a problem with Mach, I would expect to see a lot more people reporting it. I'm hoping it's the BOB or possibly the PC. The pc is a PC Chips Mobo with an AMD Athlon 2500+ proccessor. Sooner or later I'll find the problem. Good luck to you, Don
-
Hi, Guys. I think I finally got the problem of the losing steps solved. I remembered that the Mach manual says not to use on-board graphics,and realized that while I have been using an AGP video card since I first built the machine, I hadn't checked the BIOS to make sure the on-board graphics adaptor was disabled... Duh! I went to the BIOS and set the primary graphics to AGP ( it had been set to PCI), and the next line states " Assign IRQ to PCI graphics" Yes/No. Of course it was set Yes. I changed it to No and saved the settings and booted. The computer went through some "Found New Hardware" conniptions before telling me to reboot so some of my new hardware would work... I did. When it was booted up, I checked the device manager to verify that the 2 LPT IRQ's had not changed. They had not, so I didn't have to fix anything there. I started Mach 3 mill, ande everything has worked perfectly for 2 days, now. I don't know why I didn't experience problems sooner... or maybe I just never noticed the lost steps with my one-off projects before, but I know it seems to be working correctly now. :) Moral to the story: If you are experiencing lost steps in repeat operations, check that you don't have an onboard graphics adapter hogging an IRQ and inadvertantly sending step pulses to your parallel port. Thanks for your help, Don
-
Hi, can you point on the manual that says not to recommend using Onboard Display card ?
I have PCI express slot but now I'm using onboard graphic card.
I can not find on the manual that saying not to use onboard graphic card.
Hi, Guys. I think I finally got the problem of the losing steps solved. I remembered that the Mach manual says not to use on-board graphics,and realized that while I have been using an AGP video card since I first built the machine, I hadn't checked the BIOS to make sure the on-board graphics adaptor was disabled... Duh! I went to the BIOS and set the primary graphics to AGP ( it had been set to PCI), and the next line states " Assign IRQ to PCI graphics" Yes/No. Of course it was set Yes. I changed it to No and saved the settings and booted. The computer went through some "Found New Hardware" conniptions before telling me to reboot so some of my new hardware would work... I did. When it was booted up, I checked the device manager to verify that the 2 LPT IRQ's had not changed. They had not, so I didn't have to fix anything there. I started Mach 3 mill, ande everything has worked perfectly for 2 days, now. I don't know why I didn't experience problems sooner... or maybe I just never noticed the lost steps with my one-off projects before, but I know it seems to be working correctly now. :) Moral to the story: If you are experiencing lost steps in repeat operations, check that you don't have an onboard graphics adapter hogging an IRQ and inadvertantly sending step pulses to your parallel port. Thanks for your help, Don
-
Dont know about the manual but here is a screenshot of the Mach3 Home page, computer requirements.
If you have an onboard graphics card and its working fine then no problem, some work fine and others dont.
Hood
-
thank you again
I will post this after I borrow a graphic board for test.
-
Don't know if there is anything specific that says don't use your on-board graphics in any of the manuals, but , because of so manny reports over time, it has become an accepted reply as a problem creator.
RICH
-
For an update on the whole problem, please see "losing steps or something..." in the FAQ section of this forum. I finally traced the problem to faulty signals coming out of the parallel port and while I don't know why they were so wrong, I purchased a Smoothstepper from Warp9TD, and have had NO problems since. Don