Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: thedude on February 18, 2016, 06:16:21 PM
-
Hello all,
I wasn't sure where to post project questions. If this is the wrong place, please move to the appropriate section.
I have been working with Mach3 and the Hobby CNC driver for many years with great success, but I have a new project that I need a bit of help with.
I need someone to point me in the right direction for controlling two stepper motors on one axis and keeping them in sync.
Project.
I might be using the Geckodriver G540 for this, but am open to suggestions of coarse. Maybe the tried and true hobby CNC,,??
I am building 2 XY tables that will be separated and I need them to work in sync.
Any advice on what the best way to approach this would be very helpful.
Thank you in advance!
Steve
-
You should be able to slave them.
But if each table will have a spindle, then you'd need 6 axis, so a G540 might not be the best option.
-
Thanks for the reply,, sorry for my delay...
No,, only positioning, X and Y. No spindle in this setup.
So by "Slaving" the two.. Do you mean simply wiring the two motors into the X axis output, and the two motors into Y axis output?
If so, Will that not split the voltage supply to each? Or cause problems with current draw?
Is there another approach, options?
Thanks again for your reply.!
Any and all help is much appreciated here.
Steve.
-
Go to Config menu, select slave axis option and select slave A axis with The Y axis, you will be asking to re-start Mach3.
You can't slave X axis with Y axis, only X with A,B or C - Y with A, B or C and Z with A, B or C
Jeff
-
So by "Slaving" the two.. Do you mean simply wiring the two motors into the X axis output, and the two motors into Y axis output?
If so, Will that not split the voltage supply to each? Or cause problems with current draw?
It is not a good idea to just connect two motors together - each motor should have it's own driver (set-up as an axis) and then be 'slaved' as Jeff has described.
Tweakie.
-
That's great news!! And yes, didn't think wiring them on the same axis was the right thing to do.. That's why I ask you experts!
Ok, so that means that my x and y will consume all 4 axis. I still need a Z axis.
So, any advice on how to daisy chain two 4 axis driver boards through one mach3? (I say 4 axis as I might as well keep in redundant. I would only need 1 or 2 of the 4.)
Thanks for all your help so far folks!
-
Add a second parallel port.
-
And Mach software would accommodate?
That also brings up another topic that has a lot of folks talking. With new computer motherboards not having a parallel port, I have read that a parallel PS2 card would work. Any opinions on this?
-
If you use a 32 bit version of Windows 7 or older, you can use a PCI parallel port card.
If using a more modern or 64 bit OS, you can use a motion controller that offers multiple ports, like an ESS or UC300.
-
Opps, I meant PCI,, good lord.. haha
You mean something like this?
http://www.pmdx.com/2PARPCI
-
Yes
-
I see,,
http://cncdrive.com/UC300.html
This looks to be the best option.
Thanks and learning a lot about a good launching point for this project. I might have some connectivity questions, but I'll do some research first.
-
Oh,, wait,,, this board would be another option and would take the place of a gecko G540 or hobby CNC board... Skimming the manual, it seems like a nice option, and gives me the number of axis I need. Interesting.
BTW, for those who might be interested.
This project is to build a CNC controlled magnetic field mapping probe. It will map the magnetic field of superconducting magnets used in the FRIB beam line upgrade at MSU in Lansing Mi.
http://www.frib.msu.edu
-
No, the UC300 replaces the parallel port. You plug the G540 (or two) into the UC300.
-
AAAHH,,, I see. Well that sounds like a champ! I'll do some more research. I did find a review topic on here for the UC300, and folks have good things to say.
I will look into it further and I'm sure I'll have more questions. Thanks so much!
-
Hey Dude,
This board Davek0974 mentioned earlier in another thread will run dual axis drivers from the same output. X and Y.
http://www.machsupport.com/forum/index.php/topic,32020.msg222885.html#msg222885
-
Although the CWD556 does indeed allow two motors to be driven from one output you must bear in mind that it does not allow for synchronising the two motor’s exact position as, for example, in gantry squaring.
Tweakie.
-
Hi Tweakie,
I'd guess any drivers would work, I was referring to the CPO-10V BoB. (maybe a typo ?)
As far as squaring, I'd seen it mentioned where after homing the (stepper) axis, one can usually turn one motor one click at a time by hand to true it up.
Or, square it up before power up and it is usually fine enough.
The OP Steve mentioned 2 separate X,Y tables in sync., so squaring might not be an issue.
-
Just thought up another squaring option that might work well.
If one was to put a simple SPST toggle switch on the PULSE line to one of the paired drives, isolate the signal to that drive while squaring with the other, then close the switch for normal operation.
A viable work around ?
Just thinkin",
Russ
-
Yes, sorry Russ - I was referring to the Bob.
Tweakie.
-
Why not use current sinking then any BOB should do. Just go from the UC300 to one BOB and then from each BOB output to each pair of drivers.
As far as squaring is concerned - you could just drive the axes against the end stops for a moment.
-
Yes, squaring will be needed. The two xy tables will be carrying a single bar of some sort for the hall probe along a Z axis.
Picture a 12 in DIA tube, 6 ft. long with an X-Y table at each end that will position a bar in the tube. A probe runs along the bar (Z axis)
to scan the inside of the tube, thus mapping the magnetic field. Positioning on X, Y, and Z.
-
Just thought up another squaring option that might work well.
If one was to put a simple SPST toggle switch on the PULSE line to one of the paired drives, isolate the signal to that drive while squaring with the other, then close the switch for normal operation.
A viable work around ?
Just thinkin",
Russ
This is an interesting idea.. I was thinking I would do this mechanically, but isolating them temporarily to adjust with Mach seems like a good idea. As long as they would not jump when reenergized.
-
Picture a 12 in DIA tube, 6 ft. long with an X-Y table at each end that will position a bar in the tube. A probe runs along the bar (Z axis)
to scan the inside of the tube, thus mapping the magnetic field. Positioning on X, Y, and Z.
Why do you need TWO XY tables for this?
-
Positioning accuracy needs to be .002-.005in.
I cannot have sag along the Z axis, so I thought the best approach is to tie the Z axis bar on both ends for proper support.
I don't have much room for a driveshaft along the tube axis to drive the other end.
This is not the typical use of CNC for metal or wood working, but I proposed this idea as an acceptable positioning device to map the magnet bore tube since I have some experience with a HobbyCNC controller with my lathe in my home shop.
Any other ideas?
-
I don't subscribe to an online pic posting site so I can't provide a sketch of the basic layout. Does this site not let you post a pic from your local PC??
-
Yes, click the Preview button and you'll then see options for attaching pics.
-
Ok, giving it a shot. A crude sketch of what I'm going to do.
-
It worked,, !! thanks ger21 !!!
-
Your labeling is probably guided by your familiarity with lathes. But if you swap the labeling for X and Z you have a standard XYZ gantry layout. Two slaved motors for Y and single X and Z. Sorry - I don't see a problem here. All pretty bog standard stuff. What am I missing?
-
I should have labeled both ends of the tube with X and Y, not just one end, sorry for the confusion. Z would be the probe positioning along the bore of the tube that is mechanically tied to both XY tables.
I need support for the Z axis on both ends given the length of the tube.
-
I'll try once more - a STANDARD XYZ gantry machine supports AND drives the gantry at BOTH ends. Try googling "gantry router pictures".
-
OK - I think I see where the confusion started here.
In your original post you started off saying:
I need someone to point me in the right direction for controlling two stepper motors on one axis and keeping them in sync.
So far so good - this is the way you need to go (as I've been saying) - you've been given the answer already - slaving.
But then for some reason which I don't understand - you said:
I am building 2 XY tables that will be separated and I need them to work in sync.
-
I think I can see the target here - you want to build a pipe bore scanner that has a travelling head on a support that runs up and down the centre of the pipe?
You want to control the vertical and horizontal position of the scanner within the bore?
If so then its an interesting project ;)
In my mind, X axis is the scanner head, Y axis would be towards/away from centre of pipe, Z axis is up/down from centre of pipe. This would be with my view having the pipe running left to right.
I would have my X on a removable support track, probably belt driven or nylon rack & pinion for weight saving, this would simply sit in two supports, one at each end and have both ends running slaved Y & Z motors so they track each other and keep the X axis / scanner support true to the bore of the pipe.
Am I following?
How big is the pipe?
It just sounds like two vertical X/Y tables facing each other and running slaved motors on both axes, apart from the mechanics, probably quite easy to do?
Obviously all we need is a travelling linear device - could be a trapezoidal screw in the vertical plane running on a linear rail, doesn't sound like there is a need for any serious rigidity or speed here. No idea how fast the scanner runs though??
Just thinking out loud ;)
-
Picture a 12 in DIA tube, 6 ft. long with an X-Y table at each end that will position a bar in the tube. A probe runs along the bar (Z axis)
to scan the inside of the tube, thus mapping the magnetic field. Positioning on X, Y, and Z.
Why do you need TWO XY tables for this?
An XY table cannot support a 6' long bar and hold much accuracy given the sag of the bar.....
-
Your labeling is probably guided by your familiarity with lathes. But if you swap the labeling for X and Z you have a standard XYZ gantry layout. Two slaved motors for Y and single X and Z. Sorry - I don't see a problem here. All pretty bog standard stuff. What am I missing?
Nope,, my labeling comes from the physicists that calculate the effect on the beam line from the magnetic fields. They want the thousands of data points to be X and Y perpendicular to the field, (Along the radius) and for Z (along the axial bore) to be the field axially along the bore tube.
Oh I'm sure this is standard stuff to most folks on here. Which is why I posted my questions on here. I have always received great answers and alternative options to my projects from all the helpful folks on this forum.
So Glad to hear that you already have the knowledge to build a project like this, but since I have questions to learn and gain the same knowledge, I appreciate everyone's help thus far.
I could ask you,,, Why would you not label the bore of a magnet "Z"? But I won't ask you that.
-
I'll try once more - a STANDARD XYZ gantry machine supports AND drives the gantry at BOTH ends. Try googling "gantry router pictures".
I'm not building a simple gantry whereas the Z... "Head (spindle, plasma, router, laser) has a travel of a few inches.... read previous posts.
Sorry that you keep trying, but so far, everyone else gets it as I've got some great advice so far from them.
-
OK - I think I see where the confusion started here.
In your original post you started off saying:
I need someone to point me in the right direction for controlling two stepper motors on one axis and keeping them in sync.
So far so good - this is the way you need to go (as I've been saying) - you've been given the answer already - slaving.
But then for some reason which I don't understand - you said:
I am building 2 XY tables that will be separated and I need them to work in sync.
Whew.... ok.. again,, In order to carry a 6' long bar inside a 6' long tube,, I'm pretty sure I would need to support it on both ends, and that would mean I would need two XY tables working in sink as slaves to carry this bar as ostie answered for me in reply #3 and I thanked him in reply #5. (thanks again ostie)...... then other posters offered alternative suggestions which I've given some thought,,, replied questions, got some feedback, ya know like how forums do, overall, great dialog. So far I have been helped greatly and I'm moving forward with my project because of it!!! I love this forum!
So if your only going to offer options in the form of mild insults with no substance sorry, I'll focus on the others mentioned before.
-
Whew,, that was a chore.... Ok,, now back to the learning and productivity...
Thanks again,, and again all, (most),, And I'm moving forward learning the way to expand beyond 4 axis. I'm going to build a general schematic from what I've learned so far from you guys to look at and welcome any comments,, like to get your thoughts on IF I'm on the right track so far.
Thanks again all.
TheDude.
-
I think I can see the target here - you want to build a pipe bore scanner that has a travelling head on a support that runs up and down the centre of the pipe?
You want to control the vertical and horizontal position of the scanner within the bore?
If so then its an interesting project ;)
In my mind, X axis is the scanner head, Y axis would be towards/away from centre of pipe, Z axis is up/down from centre of pipe. This would be with my view having the pipe running left to right.
I would have my X on a removable support track, probably belt driven or nylon rack & pinion for weight saving, this would simply sit in two supports, one at each end and have both ends running slaved Y & Z motors so they track each other and keep the X axis / scanner support true to the bore of the pipe.
Am I following?
How big is the pipe?
It just sounds like two vertical X/Y tables facing each other and running slaved motors on both axes, apart from the mechanics, probably quite easy to do?
Obviously all we need is a travelling linear device - could be a trapezoidal screw in the vertical plane running on a linear rail, doesn't sound like there is a need for any serious rigidity or speed here. No idea how fast the scanner runs though??
Just thinking out loud ;)
Absolutely accurate!! Yes! Other than the axis labeling, which is not standard I know,, but doesn't matter in my mind since I made my own control screen with MachScreen.... Due to the physicists requirement for X, Y and Z,,
The "Pipe",, or "Bore" of the magnet(s) is DIA 12-16 inches, (Designing to map different magnets). Designing the XY tables to accommodate the variance in travel with a replaceable Z bar that carries the probe that will measure the field strength at each location along the length of the bore tube.... Thousands of them throughout the entire bore.
-
Ok, cool, then its 99% a mechanical i.e. build problem. Forget about axis labelling it's irrelevant at this stage as you will need slaving on two axes but this is really simple - my breakout board can run two drives from each output as standard, not many can so that would work there.
Picture a bog standard x/x/z plasma table, then remove the frame between the two sides and chuck the torch out. make the side frames self supporting, I don't know if there is room to join them, maybe you need a crane or forklift to load the pipe?
So now we have two carriages running along two side frames on linear rail probably. On those carriages build another linear motion unit - a trapezoidal screw lifter would be good as it holds its position and does not self-drive from the weight on it, probably decent belt reduction from the stepper motor as speed is unimportant.
Now we have two points (the nuts on the trapezoidal screws) that can move in horizontal as well as vertical planes, separated by a distance that could theoretically be many metres, both these points will move in sync as they are both slaved.
Now we just need to bridge those points with a support, maybe a box section with triangulated support etc to keep it rigid and light, then fit some possibly nylon rack along it, build a carriage to hold the scanner, the traverse motor and coffee machine ;) then some type of cable manager under the rail and it will run up and down all day happy in the knowledge that both ends are perfectly in sync.
Is this a field or lab machine, as if field, dust and crud will need to be factored in.
-
My apologies for winding you up. That was definitely not my intention. Believe it or not - I was trying to clarify something.
I've attached a (crude) schematic of one of my routers. To give an idea of scale/size, the gantry is approximately 4 feet long.
Now this MAY be an unsuitable design for your purposes for details yet to be disclosed - but what I was trying to clarify/get across is that this does not have two TABLES - it simply has a master and a slave axis. As you can see the gantry is fully supported and hence there is no sag.
Anyway - as sometimes happens when conversing with text - what should have been a small point to try to clarify, has taken on a life of it's own beyond its usefulness - so I'll back off and leave you to it.
Again - My apologies for winding you up.
-
My apologies to Sterling for following him around the forum but his diagram reminds me of Steel Nick’s excellent foam cutter design.
You just need to replace the hot-wire with a horizontal axis ??
Tweakie.
-
My apologies to Sterling for following him around the forum but his diagram reminds me of Steel Nick’s excellent foam cutter design.
You just need to replace the hot-wire with a horizontal axis ??
Tweakie.
Thats the one - exactly what i was trying to describe, just beef it up a bit and replace wire with beam/scanner :)
-
My apologies for winding you up. That was definitely not my intention. Believe it or not - I was trying to clarify something.
I've attached a (crude) schematic of one of my routers. To give an idea of scale/size, the gantry is approximately 4 feet long.
Now this MAY be an unsuitable design for your purposes for details yet to be disclosed - but what I was trying to clarify/get across is that this does not have two TABLES - it simply has a master and a slave axis. As you can see the gantry is fully supported and hence there is no sag.
Anyway - as sometimes happens when conversing with text - what should have been a small point to try to clarify, has taken on a life of it's own beyond its usefulness - so I'll back off and leave you to it.
Again - My apologies for winding you up.
All good... I have received the answer to my OP question, then some questions came along so I was answering them. Glad to hear folks are interested in this project as I'm sure I will have more questions along the way. As I said before, I have a good bit of experience with this but more than 4 axis was new to me. I hope to post a general schematic for review to make sure I'm on the right path. Also, I'll post some pics of our MSU magnet mapper when it is installed, maybe a vid or two of it in operation. !!