Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: WillD on July 05, 2016, 12:46:05 PM
-
Hi Guys,
I'm trying to set this up for the best performance, it's tricky as reading through pages here on the forum with conflicting info.
So far the settings seems to make the motors really hot, I know this can be normal. An issue is over 200ipm rapids it stalls, also around 100ipm it resonates and stalls.
I'm sure I can get higher rapids and cutting speeds!
My goal is:
Cut Speed: 150ipm
Repeatability 0.015” (+/-)
I have 2M542 drivers with Nema 24 steppers. 48V 7.3A power supply (including another Power supply for the Breakout board w/ UC100 motion controller & Mach3.)
I have a 4th axis and 2 steppers to drive the Y axis, 5 identical motors for all.
Steppers Nema 24:
Voltage: 4.2V
Current: 3.5A
Inductance: 1.6mH
resistance: 1.2
Settings I have at the moment are 2.37A for the drivers, Full current ON : SW4 5 Microstep res, 1000 step/rev
I'm not sure if these drivers can be calculated the same as Geckos for voltage, etc...? Max is 50VDC for them, I'm supplying 48v.
Should I push up to 3.31A for them...? Is my power supply 7.3A enough? Do I calculate 5 motors, @3.31A = 16.55A meaning I need a 20A power supply?
-
I don't think those drives will like you getting too close to the peak voltage rating of 50v, due to back emf which may kill them (all depends whos drives they are, as there seem to be a few copies about).
In order to remove (reduce) resonance... increase microstepping resolution.
Why did you choose 5 microstep? You talk about ipm... which normally means you may be running imperial pitch leadscrews.... what pitch are your leadscrews?
Why do you want 200ipm rapids?
What cutting feedrate are you intending to run?
I'd suggest making your drives use 1/2 current when stopped.... and not full current or you will need a larger power supply.... holding current is not normally a problem... as current is more used when the cutter is moving.
It's more about acceleration than high rapid speeds, as if the acceleration isn't there.... rapids mean nothing.
What acceleration are you running.... how did you tune the setup?
Suggest you try using a test gcode file, how big is your table travel? Set the acceleration lowish, and find the maximum rapid speed, back it off to maybe 75%, then increase the acceleration till it stalls, and back of to 75% and you should be OK to start off with.
Power supply..... leadshine and gecko provide PDFs on power supply selection, do a search for those terms and see what comes up.
What do you intend to cut (material), and with what (cutter material, size and flutes?)
-
Thanks for your reply Rob! They are JMC-Motion drives.
I don't remember the pitch by heart, I'll check it in my docs. They were already set at 5 microsteps.
Trying to keep up with orders and squeezing to the max of my machine. 150 to 200 rapids would be nice to cut down on program time.
Ideally I would like to be cutting between 120 to 150ipm. I cut HDU, Acrylic & Polycarbonate, v-carve cedar and occasionally hardwoods / plywood.
My acceleration is 10 at the moment, I've been tweaking that. But I see I should work on Velocity first!
My travel is 40" x 48", 5" Z travel (but I'm happy with my 30 to 45ipm for it). It's really X-Y I'm working on. I will switch to half current.
I was recommended to put fans on the steppers as they run awfully hot.
-
Here are PDF's of the drivers and motors.
-
The induction on those drives is quite high, have a look at these similar rated drives.
http://cnc4you.co.uk/resources/Stepper%20Motor%20Nema%2023%2060BYGH401-03%204Nm.pdf
No I don't work 4 the company..... just an example to consider.
You need to consider the parallel wiring arrangement.
I don't know how much you know about induction, but it is basically best to think of it as resistance to current flow that is proportional to motor speed..... faster the motor turns.... the higher the resistance..... a bit like wind resistance in a car..... but sometimes you are driving a sports car, and sometimes a bus.... except both have the same horsepower..... but they won't go as fast as each other.
The technical answer is XL = 2*pi*F*L, where XL is the inductive resistance (in ohms), pi is 22/7, F is the frequency of the input voltage to the drive, and L is the inductance.
Can you provide your steps per unit settings from within Mach, or the leadscrew pitch to consider too, thanks.
-
when I spoke about copy drives.... here is a very similar drive to yours with a comprehensive selection manual. Whilst it is not the same as your drives I would suggest a read as some of the info is still applicable on power supply selections and considerations.
http://www.leadshine.com/UploadFile/Down/M542V2m.pdf
Within this manual is some useful information and considerations of power supply selection which may be worth a read to expand you knowledge in this area:
http://www.leadshine.com/UploadFile/Down/DMShm_P.pdf
and the well known and referenced Gecko information is here:
http://www.geckodrive.com/support/step-motor-basics/power-supply-basics.html
-
Thanks Rob, I'll check those out.
At the moment motor settings are:
5080 steps per / Velocity 80.06 (had to set it down, kept loosing steps) / Acceleration 5 / 15 step pulse / 15 dir pulse
-
Ok, they are 5mm pitch leadscrews, and at 200ipm the drives will be pulsing at 16933 hz, so if you are running mach3 using the parallel port, then set the kernal frequency at 25khz.
I would suggest changing your step and dir pulse settings to 20 and trying it as it will coincide with 25khz kernal speed and still be faster than your 17khz max stepper speed..... nice chunky step
Your acceleration needs to come up in my opinion.... 5"/s/s.... at a cut feedrate of 150"/min .... 2.5"/sec means that your corers will be rounded by up to 0.625" (one axis will slow down on a 90deg corner while the other needs to start accelerating in constant velocity mode..... see one of the other recent posts I've commented on in the last day that has this problem...
-
I'm using a UC100 usb motion controller. Not sure if that makes a difference? I'm still learning this and Mach3.
I'll check out your other posts.
-
It's set to a max step rate of 100kHz for the motion controller
-
The uc100 seems to be a good well supported motion controller, I would suggest if possible to try to reduce the controller frequency from 100khz to maybe 50khz, given even if you changed the ustep resolution to 10ustep, that has a maximum frequency of only about 33khz.
You can try other things like 10ustep resolution although that will double your steps per unit to 5080... that may help with resonance... but I don't think it will help with maximum velocity or acceleration
Also check your lookahead is set to about 200 lines on the general configuration tab.
What version of Mach3 are you using?
0.66 is not favoured....install 0.62 although difference I believe is in the control of mpgs... but worth a shot
Rob
-
Should read .... 5080 x 2 = 10160 steps per unit....
-
I will give those suggestions a try and see how it behaves!
-
I've set the UC100 to 50kHz, works fine. Haven't notice a difference between 100kHz or 50kHz.
For step and dir pulse settings to 20, it doesn't let me set it higher. It reverts back to 15?
-
Micro steps are usually set with DIP switches on the stepper driver module, but it sounds like you're trying to change this in a configuration screen. Try adjusting it on the driver module itself.
Stephen "Highspeed" Kruse
-
It was the pulse width that was being discussed not microsteps
-
Ah, ok. Thanks.
Highspeed
-
I know its an old thread, but my question is kinda same and I dont want to start a new thread.
I use a common PRC 5axis bob, with 25khz, and 2M542 drivers, 48V power supply and a bit smaller motors than the OP - 2phase, 3.08V, 2.8A, resistance 1.1, inductance 3.6, 1.89 Nm. No matter how I tune it in motor tuning - slow like 400 mm/m or 1200 mm/m, and no matter what accel, they stall after a while, loosing steps. I chose half current. I am aware that the 48V power supply isnt the best choice for these motors, but thats what I got.
I did set them to 5 micro steps, leadscrew is 2.5mm, 1000 steps per min, multiplier is 2.5 (belt). I tried all possible combinations in motor tuning with step pulse 1-5 and dir 0-5 (not over 5), but no luck. stalling.
I would like to get the fastest rapids possible, cutting feeds have to be slow. but to wreck every part I tried to make for like 10 days......its frustrating.
Thanks for help
-
Hi,
it sounds like you have experimented enough with the steppers and drivers to be reasonably confident that the fault is not the steppers/drivers. You may have to
revisit that decision but maybe you should consider other possible faults.
You haven't specified that you are using a parallel port, I don't know anything about your 'common PRC BoB', but will presume it has parallel port input. If that is the case
then if the parallel port is stalling or stuttering in any way missed steps and outright stepper motor stalls result.
May I suggest run DriverTest.exe and post the results, 'Good or Excellent' is not an adequate report, we need to see either the graphic line and/or the numbers describing the jitter.
A screen shot is good. Use the 'Additional Options' to the lower left of the text input pane to attach and preference is that you use Paint or similar to scale the image to around 600kB.
Craig
-
dont have the internet in my shop, so I took a pic.
common prc bob, cheap 5axis breakout board that is sold on ebay. I use the same bob, drivers and steppers on my mini lathe, and its under linuxcnc, with 1200 mm/min rapids. on my mill, I cant go over 500 mm/min without the stalling.
-
Hi Sharpshooterser,
I am really not sure that Driver Test provides any useful information but according to section 2.2.1 of the manual the variation in pulse timing should be small. Your pic shows quite a high variation and that may well be the cause of your motor stalling.
As a guide the pulse timing line should look something like this;
(http://www.machsupport.com/forum/index.php?action=dlattach;topic=31366.0;attach=42177;image)
-
Is there a way to reduce it?
-
I think not, it is a function of the particular computer being used and just how that computer runs Windows. Laptops often produce bad results (as well as power saving problems) and Dell computers have been reported many time in the past as having poor results.
One way forward is to use an external motion controller (such as the ESS) - there are many to choose from but those that support both Mach3 and Mach4 will satisfy your current and perhaps future needs.
-
Hi,
sorry I agree with tweakie, that DriverTest.exe screenshot is rubbish. Its showing a jitter of 20us on a 40us signal, ie 50%. No wonder the steppers go cranky.
For reasonable service you want less than 4us jitter.
There are lots of little tricks that MIGHT, and I stress MIGHT improve the situation. A lot of those tricks belong to 'Mach Lore' and as the parallel port loses favour so does the
lore which generated the tricks.
About 18 months-2 years ago bought an Ethernet SmoothStepper and have never looked back, I recommend that course to you. I know a lot of dyed-in-the-wool parallel port affectionados
are going to jump up and down at my sacrilege but there you have it, ditch the parallel port and get yourself an external motion controller. If you buy Chinese rubbish you'll get what you deserve.
A lot of bleeding heart liberals are going to jump up and down at me castigating all Chinese made stuff as rubbish but I don't care. If you don't believe me just read the posts from the last month,
I will guarantee that 50 or more are along the lines...motors not working...supplier doesn't answer emails....cant read the Chinglish instructions. The list goes on and on.
Even if you decide to buy a good brand buy direct from the manufacturer or their nominated distributor...the Chinese have flooded the market with rip-off copy junk that doesn't work on places like Ebay and
Amazon. Before you part with any money get some advice or feedback from the forum about the device you are contemplating.
Craig
-
It was the crappy computer after all. Replaced it, and the machine is running much better now, with faster rapids.
However, the smooth stepper is something I will think about in the future, when the budget allows it.
Thank you for your help and insight