Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: lcluff2000 on November 12, 2011, 09:57:43 PM
-
I've been using my mill at lower speeds for awhile and decided to see if I can increase the speeds. The problem I am having is that I can go up to 45 IPM when I use the keyboard without issue, but when I hit GOTO Z's, it will stall the motor (as it ramps up to the desired speed). I've tried several acceleration numbers that seem to make no difference. I can lower the IPM back down to <20, but that's boring! Anyone have any idea why GOTO Z's would be different than keyboard pendant control and what I could do to potentially get this working at 45 IPM? Thanks for any help!
-
When you are jogging are you only moving one axis at a time? If so then could be your power supply is not quite up to it when moving multiple axis.
Hood
-
I'm only using one axis when jogging. The same goes with when I use the goto z's. I've only moved it away from zero on one axis.
-
Press the shift key and hold it then Jog and see if its ok.
Hood
-
I finally had a little time to try that out. When holding the shift key, jogging the axis has no problems. It runs easily up to 45 IPM. Any other ideas as to why the goto z is different?
-
If you attach your xml I will see if I can replicate here and maybe find the issue.
Hood
-
My xml is attached. I'm very curious to see what you find!
-
Try using the demo version. The one you are using is not supported. >:(
RICH
-
Dont see anything wrong in your xml from first look but will check it out more thoroughly tonight.
Hood
-
Had another look and xml looks fine, just wondering however if it has anything to do withyour A Axis, can you disable the A in Ports and Pins and see if there is still a problem.
Hood
-
what are you using for motion control?
does the machine run smooth or does it have resonance?
I had this problem with my old xylotex drive The solution was a dampener.
-
I tried disabling the a-axis, but the results were the same.
Rich, this is the latest released version on the website (vR3.043.022 -- lockdown) is what is installed; is that not supported?
-
I think the problem is that version R3.043.022 is not compatible with the Pirate license file you are using (please correct me if I am wrong).
Tweakie.
-
Tweakie
It is a genuine licence, not sure if its the original owner but looks likely :)
Hood
-
lcluff2000
Cant see why you would be having issues unless your computer is just on the edge, have you run the driver test at 35KHz to see what it looks like?
Hood
-
Tweakie
It is a genuine licence, not sure if its the original owner but looks likely :)
Hood
If that is the case then my sincere apologies.
(Software4u was used as the license name on one of the pirate copies I looked at over a year ago).
Tweakie.
-
I had my doubts about it as well but Scott confirmed it was a genuine licence and was from a long time ago. Cant be sure its the original licensee but it is a genuine licence name.
Hood
-
Not sure what the problem is.
Are you using the default Mach screen?
Are all the steppers the same? Have info on them ? Post model number, in-oz, size.....
What drives are you using?
--------------------------------------------------------------------------
Since the x & y are working properly,try swapping the cables to the motors by using the x or y cable for the Z. What happens?
If the Z stalls then lower the velocity to 15 for the x or y and try again. What happens?
Now use only the x or y cable to the Z but at 45IPM. What Happens?
RICH
-
Thanks Hood.
My apologies to all concerned.
Tweakie.
-
No problem Tweakie. Yes, I did own a company named Software4u and paid for mach probably ~6-7yrs ago.
The motors are 23H276-28-4B XYLOTEX 269 oz-in on 3 axis. The 4th axis is a Moon 17HD4063-01N (but I disabled that axis yesterday and the problem still exists).
The drive is XS-3525/8S-4. The vref is set to ~3.2v (on 3 axis) which translates to ~2.2A, the 4th axis is set to 1.2V (0.83A). I'm using a 24V power supply.
The problem occurs on all 3 axis'. All 3 axis work at 45IPM (independently) when I jog, but when I zero (independent axis') or use program control the motors stall.
I'm using the default mach screen
I'm running out the door right now, but I'll try the following experiments tonight:
(1) driver test at 35khz
(2) swapping cables to motors
-
Can you check the voltage supplied to your drives whilst running the three axis independently and then when running together, could be your supply is on the weak side.
Hood
-
Hood - can you clear something up for me that's bugged me for years? I've seen this mentioned loads of times but just don't get it. How does running one axis use less current from the supply than when you run (say) all three? (and hence potentially cause a voltage drop). I can understand how that's the case with servos but with steppers I've always thought they "drew" the same current when stationary as when moving (in fact slightly more when stationary if we want to get picky). Maybe I'm wrong but if so I'm darned if I can understand why.
Ian
-
http://www.youtube.com/fixittt#p/u/70/8Am69_F93yA
http://www.youtube.com/fixittt#p/u/71/ivDQTeGuduk
http://www.youtube.com/fixittt#p/u/72/9RoOZs9LrYo
Take a look at these, your problem is the drive i bet you. Find a way to make a dampener for testing. you will be surprised.
Here is a good thread on dampeners. Resonance is horrid in the xylotex. The vibration is going into the motor and builds up. If you lower your accel way down and jog the motor so that it ramps up to speed, you will see that as it reaches full speed, it will stall. Any setting in mach will not solve it
another thing to keep in mind, you are limited to a 24v power supply, which means that if you have say, the 269 ox steppers, you are running them at less then half of their rated power.
-
Ian,
I dont know a great deal about steppers, well I should say steppers are in a long list of things ;)
I was actually thinking more along the lines of not enough capacitance but really I dont even know if that would be the case.
Hood
-
Bob,
I think your correct on it being resonance. Adding a rotating mass to the motor will change the resonant frequency / speed at which resonance occurs.
It is a combination of the drive electronics and the rotating systems moment of initia of that creates the situation ( simplisticaly said) .
RICH
-
The driver test says "system excellent" up to 65 Khz.
Measuring the voltage showed no difference when running the motors, it stays at 23.5 VDC. I'm not too worried running at half the rated power since the application will have little to no load.
I took the acceleration of the x-axis down to 0.1875 (the minimum it would let me set). That equates to 4 seconds to ramp up to speed. I then jogged the x-axis up to 45 IPM with no problem several times in both directions.
When using goto z's at this slow ramp rate, it usually gets up to speed (not always), but after a little bit at the top speed starts to stall or as it is starting to slow down.
In addition to that test, I tried to squeeze the motor shaft with a cloth while using goto z's (to see if I could change resonance). Still, it stalls.
Note that I've never had trouble jogging with any axis up to 45 IPM. I'm not sure why goto z's is different than jogging. Does anyone have any ideas for why these would be different within mach3? Is there a reason the resonance would be different with either method?
Also, it seems odd to me that the y-axis or z-axis will stall too even though they have different loads due to the design of the taig mill. Shouldn't there be a difference?
I'm curious about the dampers though. Does anyone have a source for purchasing or design? If they are not too expensive (or easy to build) I could give one a try.
Thanks a ton for the input so far. I really want to understand this issue.
-
Hi Larry,
The mid-band resonance issue has been covered here and on the zone quite a lot and looking through a previous thread Gerry said that his Xylotex was practically unusable without the dampners. http://www.machsupport.com/forum/index.php/topic,17670.0.html
Some have had success using just a flywheel but technically they should ‘rattle’ and have the moveable elements which dynamically alter the resonant frequency as soon as the stepper stutters. They are pretty simple to make and from the tests that I have made the one I fitted to my 4th axis makes a big difference to the rapid speeds.
Tweakie.
-
Hi Hood - no probs. I've tried to get a definitive answer to this before with little luck. All I can say is I'm pretty sure that contrary to what seems to be a common belief, steppers actually "draw" the most current from the PS when stationary and this reduces with speed. That's why (very simply put) steppers eventually stall with speed.
lcluff2000
I'm not too worried running at half the rated power since the application will have little to no load.
This is slightly missing the point. Granted, at low speed you won't be "too" far away from the theoretical "optimum" torque (because of your phase current of 2.2A as opposed to the rated value of 2.8A). However as speed increases, your 24V is pitifully low compared to the ideal for your motors of 60V (if my calcs are correct). High speed torque is a direct function of Voltage. Even with "zero" load your motors would stall way below their potential maximum speed.
Also, it seems odd to me that the y-axis or z-axis will stall too even though they have different loads due to the design of the taig mill. Shouldn't there be a difference?
As above, the load isn't the real issue here.
Note that I've never had trouble jogging with any axis up to 45 IPM. I'm not sure why goto z's is different than jogging. Does anyone have any ideas for why these would be different within mach3? Is there a reason the resonance would be different with either method?
This is JMHO but with jogging, the Mach application isn't doing much. It simply tells the driver to jog - the driver's doing all the work independantly. With commanded moves, the Mach app is doing more. This *might* mean that the jitter is marginally worse. In a *good* system, this probably makes no difference. However, given you're "undercurrenting", "undervolting" and sadly your drivers are known to not handle mid band resonance as well as some other drivers, it may be just enough to cause the problems your seeing. Just a guess.
Just a quick comment re "rattlers". They do seem to work well with less than perfect drivers. With decent drivers that handle mid band resonance they (IMHO) bring nothing to the party. Personally I'd swap drivers rather than go the rattler route. As I said - JMHO.
Ian
-
I really want to understand this issue.
Do an internet search for "Jones on Stepping Motors". The book or course is rather technical but certainly worth a read if interested.
RICH
-
Hi lcluff2000,
Here is a basic but very informative article from the GECKO site about stepper motors.
I found it quite interesting.
Russ
:)
-
if you need them, I am sure I can dig out my old rattler dampeners....
-
Thanks everyone for all the information! I've learned a ton.
Fixittt is going to set me up with some rattler dampeners. Cool guy!
I'll try that out and let everyone know how they work.
-
I would also recomend getting some thrust bearings for the motors if you dont have any already. If you put thrust bearings inbetween the motors can and the screw coupler, and the motor can and the dampners it will keep the shaft from being able to move inside the motor can. Which could result in backlash issues. I know I had major problems with the xylotex 269 motors with massive amounts of shaft play inside the motor can.
-
im waiting to hear the results of the dampners
-
I only had a few moments between when the package arrived and when I left for Thanksgiving. On first try with one axis, it didn't seem to make much of a difference. When I get back I will experiment more.
-
dont forget to tune the motors again. You should be able to notice a difference with the accel really low. Jog an axis and listen to it ramp up. Then put a dampner on and jog......
-
I've been experimenting some more. When I realized windows update was still enabled, I disabled it and got to 45 IPM without the dampeners. Using another faster PC, I'm up to >50 IPM. When I have some more time, I plan to see what the upper end is.