Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: rabphxaz on January 03, 2009, 05:38:24 PM
-
The latest puzzle....my y and z stepper motors seem to work ok when they are following Gcode but I can't jog with them -- they just give off a loud buzzing noise. Anyone know what the cause might be?
Thanks, Bob
-
Think you have them tuned too fast, sounds like you are stalling them.
Hood
-
Hi Hood,
They were working fine and sounding good at about 40 inches per minute w/ an acceleration of about 30/sec/sec, but I just reset them all the way down to 10 in/min and 6/sec/sec, then all the way up to 60/40, and several stops in between. The result was various tones of horrible.
-
Can you command a G0 move to them from MDI and they will work?
Hood
-
I don't know how to do that, but if you'll give me the steps I will try it.
Bob
-
Go to the MDI page then click on the box beside where it says Input and then type in a move, obviously make sure you can make this move without hitting anything as a G0 move will be at full speed that you have set in motor tuning. What you want to enter is something like G0Y2 then press the enter key, your axis should now move to Y2 at full rapid.
Hood
-
They moved but they sounded terrible.
-
ok so its either overtuned motors or maybe your drives or the power supply. Try commanding a G1 move now and see how fast you can manage to get them going, for example try a G1Y2F10 and see if they move. If you have a voltmeter check out the power supply to your drives while you have a motor moving to see if it drops away significantly.
One other thing to try is see if you have enhanced pulsing enabled, its on General Config page from Config menu.
Hood
-
No volt meter, but checked Config and the Enhanced Pulsing is enabled. The G1X2F10 moved very slowly and made a clicking sound.
-
Are you set up in inches or mm? If you are in inches it should have moved at 10 inches per min, if in mm it should have been 10mm per min.
Can you copy and rename your xml and attach it so I can have a look.
Hood
-
I'm set up in inches, and I would guess that it moved about 10/min. Re: the xml, is it the backupxml in the mach folder?
-
not the backup, its the one you are using now I need. If you are using the standard Mill profile it will be called Mach3Mill.xml You may not see the xml extension if you have windows default which hides them so look in C:\Mach3 for a file called Mach3Mill and the icon of it looks like a piece of paper with the globe on it.
Hood
-
here, I think.
-
Do you have the steps per unit right for the Y and Z, they are different than your X but that may be correct.
Hood
-
I'm pretty sure I do -- they have a 3/8th screw and the x has a 1/2. And they all worked perfectly this morning.
-
ok thats fine. I have looked through your xml and so far dont see anything major wrong. If you have seperate drives might be an idea to swap them around and see if the problem follows the drive as I am thinking it sounds like a hradware issue.
Hood
-
Can I do that by changing the ports and pins settings? It seems strange that 2 of the 3 drives would partially fail at the same moment.
-
You could change the pins, it would show whether its a software issue if you did that, if the problem followed the pins then its software, if not then hardware. I am off for some sleep (12:30 here ) so hopefully you will get some help from someone else if you cant find the issue.
Hood.
-
Thanks, Hood, I will try to figure it out.
All the best, Bob
-
Did you try changeing the pins? If so, how did it go?
Brett
-
Hi Brett,
Thanks for your assistance. I changed the x and y pot/dir pins and as a result the keyboard up/down & left/right arrows changed their control to the other axis, but the original axes that are having the problems still have them. In other words, it looks like the inputs to the motors are OK, and the problem(s) are somewhere else. Any ideas?
Thanks again, Bob
-
Bob, could be anything from breakout board to power supply to drivers to motors or even axis binding. If you dont have a scope then it will just be a case of swapping things about physically until you can eliminate the problem. If you are using a breakout board then swap the step/Dir wires for X and Y and see if the problem follows, if it doesnt then move onto the drives etc etc.
Hood
-
ranphxaz,
Two things you said caught my ear.
-Buzzing sound - and they stall or don't move - reduce velocity and accel, which you did
thus you kind of found a max and min range for you system - for now just set so they run.
10 - 40 then use 15 or 20 for velocity, you need to just get them moving and don't worry about
"how fast the table will go"
-clicking sound - click-click-click ......or tap-tap-tap like someones taping with a pencil on a board
may indicate poor pulses, ( could be other things) maybe ringing on the pulse and in the last year there have been say 3 people wiith that problem.
RICH
-
Hi Hood,
I will try to bypass the BOB and see what happens. Part of the problem is that the guy who sold me the BOB won't respond to my questions so I don't know if there is a simple way to check that out. (Arturo Duncan @ cnc4pc.com, by the way).
Rich, I never really cared how fast the table ran, I just set things up so that the motors sounded good -- in the middle of their range. Since the problem I have tried speeding them up and slowing them down, but the sound (imagine a blender filled with gravel) gets worse in both directions. I have also tried adjusting the pulse debounce in both directions with no luck. The fact that it is happening with 2 of the 3 motors ought to be a clue, but I don't know how to read it. And it was working perfectly during its first real project, which it finished with no problem. Then when I tried to jog the Z higher to remove a clamp, it started. Weird.
Thanks again, Bob
-
There have been a lot of reported issues with the CNC4PC BOBs in the last while and a lot of these seem to be related to the optos. I believe they are socketed so can be replaced. Now I have never owned a CNC4PC BOB so dont really know how they are set up but if you could swap two of the optos around that may show if the problem lies with them.
Hood
-
I bypassed the BOB and it made no difference. My conclusion, therefore, is that it has to be the drive box, which is what I call the thing that Xylotex sent along with the stepper motors. It, I think, generates the pulses. I sent them an email yesterday and they haven't responded, but it's the weekend, so that may be the reason. Do yu think I'm right in thinking that it's the drive box?
Thanks, Bob
-
Certainly sounds like it is, what you could do is swap the wires around for the motors and see if it follows them if it doesnt then it has to be the drives on the xylotex.
I think the xylotex are 3 drives on one board which at first seems a good idea but the problem being if one drive packs up then the whole board is basically scrap, much better to have three seperate drives in my opinion. I suppose there may also be a chance that if one drive packs up it could cause damage to another one as they are all on the same board. It may not be the case as they may not share any common circuitry but I would guess otherwise.
Hood
-
I changed the ports/pins and that changed which motors were acting up. Wouldn't that tell me the same thing?
Bob
-
It could be the bob but it may be your PC. So if you have a scope and look at the signal
along the way you could possibly find the source.
BLENDER FILLED WITH GRAVEL SOUND
The sound of the stepper is a good tool. If a stepper gets hot ( maybe because of a current
setting ) it will sound and behave funny and hopefully noticed before that burnt smell is sensed.
Mrs stepper is getting a mr nice pulse and walking like a champ,
but some Kremlin dosn't care for mr nice pulse and also throws some pulses at mrs stepper, some hard, some soft, sporaticaly.
Now mrs stepper is still walking along and the mr nice pulses are stronger and keep mrs stepper walking
along.
The Kremlin pulses are hitting mrs stepper somehwere during the walk and all they can do is
bonk against the stide.
The bad pulses give mrs stepper a ringing in her heart and can cause walking problems and sometimes even making her stop and sit down.
You really don't want the technical version! ;)
RICH
-
OK, if I can get someone over here with a scope to look at the signal, where would I look and what would I expect to see?
The steppers didn't get too hot, and I also tried them after sitting for a few hours -- no difference.
Bob
-
Bob, this may have been asked before, but how is the controller hooked to Mach? Parallel port? If so, do you have a straight through cable or a printer cable?
Brett
-
Bob,
1. The pulse signal out of the PP
2. The signal at the BOB input termnal
3. The signal at BOB ouput termnal
4. The signal at the drive input terminal
5. The signal at the drive output treminal """"caution do not do unless you really
know what your doing as you can short and blow a drive """""""" but before you get
to this one, i think, if there is a problem, you will have found it and thus no need
Scope displays can vary depending on adjustments, make sure the probes are calibrated / adjusted.
If they have a scope they will know what to do and how to interpret.
See attach:
RICH
-
I changed the ports/pins and that changed which motors were acting up. Wouldn't that tell me the same thing?
Bob
Ok lets start over as I get easily confused :)
X axis asigned Pin 2 and 5
Y axis asigned Pin3 and 6
So you jog X and X moves, jog Y and Y moves
Now you swap pin asignments in Mach
X is pin 3 and 6
Y is pin 2 and 5
So you jog X and Y moves and jog Y and X moves
if its a software problem the the bad pulses would still be coming out the Y axis which now physically moves the X axis and it would now become bad
Is that what happened?
Hood
-
Brett -- the controller, which I have been calling a drive box, is hooked up to the pc with a printer cable through the printer port. It was going from the pc to an opto BOB, then to the controller, but I bypassed the BOB.
Rich -- If I can get someone over here I will show them your instructions.
Hood -- Yes, when I switched the pins from the x to the y, the problem switched from the x to the y, and visa versa. I concluded therefore that I could eliminate the motors and the PC as the problems, and was left with the BOB and controller. When I bypassed the BOB that left the controller as the major suspect, although I guess the pc or software could somehow be corrupted. Is there a way to check that?
Bob
-
Its getting late here again and my brain is working at 1% of the normal 2% it is usually at but dont see how swapping pins in Mach could eliminate motors from being a problem but the drive still being a suspect. If you swap pin asignment in Mach all you are really doing is swapping which axis will jog when you jog with the keys. For example if you Jog X with right arrow and you now change the pin asignment so X connects to the pins that the Y axis is on and vice versa it would mean pressing the right arrow would now jog the Y axis, I think ;).
Hood
-
With electronics, I sometimes feel like a "pig staring at a wristwatch", as my grandpa used to say. But my thinking is this: if I switch a path that I know is working to a motor that has a problem, and that motor runs fine, then I figure that the problem can't be the motor. It doesn't tell me too much about what part of the bad paths are bad, but I think it tells me that the motors are good. Does that make sense? (By "path", I mean the electronic signal coming from the pc, through the controller, and arriving at a motor.)
-
Hood,
By the way, please don't stay up on my account. This will get solved in time, and you have been very generous with your time. And I am grateful. Go to bed.
Bob
-
No probs, hope you can get a scope as that will definitely be the way to get the answer. I couldnt do without my scope now, they are so easy to use and tell you so much I just dont know how I managed without one :)
Hood
-
Watch out, Hood, the next thing you know you'll be having to teach me how to use a scope, too.
-
Ha ha if I can teach myself in a few mins then I am sure you will manage :)
Hood
-
Hi, Bob
Sent you a PM Message, Looked at your xml, It may not be the one your using as the pin assignment seems different.
With BOB's & Axis Cont. Boards, There a couple more things that need to be checked, I'm thinking that Xoltex board's may need longer step pulses to see the pulses properly and are sensitive to leading/trailing edge step/dir. pulses.
In the pic's below, Set mach3 to sherline 1/2 step mode,Apply, OK and shut down and restart Mach3, This set's wider Pulses.
Set all 6 of your Dir/Step LowActives to ether X on Checked, All the same, If you need to change a axis direction do it in motor tuning page "Reversed" state to Check.
Give it a shot, let me know, It could be your velocity settings as Hood Sed or your default start-up feed settings also..
Chip
-
Ah think you have it there Chip, no experience with xylotex but certainly seems to ring a bell that they need the wider pulses :)
Hood
-
Hi, Hood
Your Head, Ring's all the time like mine, HU.!!
Leading and Trailing Edges, can get'ya also.
Chip
-
LOL yep :)
Hood
-
Heres a very basic guide to scopes, might be of some use to you, certainly helped me when I first got one.
Hood
-
I just rethought the whole good motor/bad path scenario and you're right -- it doesn't tell me anything. Which means that all I've eliminated is the BOB board. Maybe I need to go to bed.
Thanks again, Bob
-
I also just realized that I missed some messages -- I guess I skipped ahead a page. However, Chip, I just printed out your message and will try what you suggest.
Thanx, Bob
-
Hi, Bob
I posted it pretty quickly, So if you don't understand some of it let me know, Did you get the PM I sent. ?
Chip
-
Hi Chip,
OK, I found the info sheet that came with the Xylotex steppers and drive box. It says,"
The STEP impulse must be an ACTIVE HIGH signal. The DIR line must be stable 200 nanoseconds BEFORE AND AFTER the rising edge of the STEP line. The STEP signal should be high for a minimum of 2 nanoseconds." (capitals theirs). It also goes on to specify the step and dir pins, which I have correct.
So, with that in mind, what I did was enable the Sherline pulse, make sure I had the other stuff correct, and try the steppers. No change.
I don't know what the info sheet means about the dir line being stable 200 nanoseconds before and after, etc., and I don't know if there's a setting for that, but one thing to remember is that everything worked perfectly yesterday, and I made no changes to the pins/dirs/low actives, or anything else. It ran a program, finished, returned home when I clicked on Go To Z, and then started acting up when I tried to jog the z axis up.
Still baffled, Bob
-
Also, I didn't get your PM, that is, nothing came to my email address. Are they somewhere else?
Bob
-
Hi, Bob
At the top of this page on the left you'll find your name and Message to read, see pic, Once you click on it you should see the Msg., If you click your name there's an option to set Option's / Preferences also.
Thanks for pointing out it was working yesterday, I missed That, These setting are still worth Checking at this Point..
My earlier post second Pic, Set all 6 of your Dir/Step LowActives to ether "X'es or "Checks"'s, "It's the Proper way to set them up" and isn't really properly addressed in the manual. , All the same, One way or the other may make a Difference. On some controls it has no effect, Others the axises won't run at All or Poorly.
If you need to change a axis direction do it on Motor Tuning page "Reversed" state to Checked for the axises that are running backwards.
Getting late Hear, Chip
-
Morning Everybody,
Maybe today will be my lucky day. Re: Chip's last note, I rechecked all the step/dirs for active high pulse and axis direction , and they are set correctly. One thing I noticed, however, is that whenever the Reset button flashes, there is a message that says "No shuttle detected of type selected". Is that important?
Also, I'm hoping that Xylotex will respond this morning to the email that I sent them on Saturday...if they do, I will update everyone.
Thanks again, Bob
-
One thing I noticed, however, is that whenever the Reset button flashes, there is a message that says "No shuttle detected of type selected". Is that important?
I only get that when I have my Shuttle Pro disconnected and it's never a problem, I just assume it's letting me know the status.
Dave
-
I've got a real bad feeling re: Xylotex....After waiting all morning for them to answer my emails, I looked up their number and called. Their phone is disconnected. I guess I need to assume that they're out of business. I guess what I will do is take my office computer to the shop and copy the mach file to it, and see if that changes anything. If it doesn't, then I can eliminate the pc along with the bob, and I will buy a new drive box. Question: for a plug-and-play type unit, can I use the existing steppers or do they need to be factory tuned to it? Also, does anyone have any suggestions for a good unit, brand-wise?
Thanks again, Bob
-
What you are calling a drive box is really just a group of stepper drives on one pcb. Generally the make recognised as the best are the Geckos and there are a few different versions you could , the newest of which are the lower power G250, G251 and G540 which I would think would be plenty for your motors. The G540 is kind of a similar idea to your xylotex in that there are 4 drives connected in one unit but it also has a breakout board included, the good thing though is they are all seperate components so if one did happen to go down you just get a replacement drive and not the whole unit.
Any idea what kind of motors you have and their current ratings?
Hood
-
Hi, Bob
At this point I think we need to revisit your velocity and acceleration values in Motor Tuning.
I'm thinking you have "as many do" tuned your speeds with the default F6 feed rate that Mach3 default's to with new Installs.
Let's start from scratch, Go into Config, General Config screen and set the Initialization String to F100 and "Check" the Use Int. String on All Reset's Box, See Pic below.
Now go into Motor Tuning and set the Velocity and accelerations to Vel 10 Alel. to 1 for all 3 axies, Dont forget the Save axis settings button for each axis, then Load one of your files and do some test's and some Joging around.
Post the result's, Did you find the PM If not, It was, "If your in the US, PM your phone # and I'll give you a Call".
Edit: If it was working before we should be able to find the problem at some point.
Chip
-
News: I got an email from Xylotex, which of course said I should check the ports/pins/acceleartion, etc. I wrote back asking them to call me .
Hood - the motors are 425s with 1/8 microstepping.
Chip -- I will try what you outlined and post back.
-
Chip -- I went through the list and they sound the sam -- the y axis is fine, but the x and z still sound horrible. I started a file to see if by chance it was just happening in jog mode, but it's there too.
Bob
-
Bob, ok I am presuming xylotex list these motors so will have a look but the 1/8th microstepping is a drive feature and nothing to do with the motor. Hopefully you will find the problem is other than hardware but if it is then I would certainly consider going the Gecko route, not sure how much more expensive they are compared to xylotex but ask Jim Pinder if the Geckos are worth the money, I loaned him one on a semi permanent loan and even though his wallet has double locks on it he has now splashed out on geckos for his other axis ;D
Hood
-
Hood -- I will email Gecko right now and ask them what they suggest.
Bob
-
Bob, I have sent you a personal message.
Brett