MechMate CNC Router Forum

Go Back   MechMate CNC Router Forum > Common Assemblies & Parts > Motors & their mountings
Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
  #211  
Old Mon 31 May 2010, 11:25
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
The motor is rated at about 450 oz*in when wired half-coil. You would multiply that by the gear ratio of the belt drive, which would probably be at least 3:1 and possibly 4:1. At 3:1, you would have 1,350 oz*in, which is about 84 lb*in. That's the same torque that I have with the Oriental Motor Alpha 7.2:1 geared motors on my Shopbot. If you use Geckodrive stepper drivers that have 2,000 steps per revolution, a 3:1 belt-drive would give you a per-step resolution of 0.000654 inches if you used a 1.25-inch pitch diameter spur gear. Even at 25,000 steps per second, your jog speed would be faster than I would jog my machine.

A 35VDC to 45VDC power supply at 12A would give you plenty of power to drive four motors.

On paper, everything looks good to me.
Reply With Quote
  #212  
Old Mon 31 May 2010, 17:04
southernduckie
Just call me: Andrew (Duckie)
 
new south wales
Australia
more on steps

Hi mike,

sounds good then i am thinking 3:1 if i could, how many (true holding) steps per inch could i expect? im not sure on full steps/half steps can u hold on a half step? or is it just smother cut?

Im looking for acurcy of 0.001mm or aproxmatly 0.00004" would that mean that i need more like 7.2:1? that would reduce my ipm to very low levels though?


Regards, Andrew

Last edited by southernduckie; Mon 31 May 2010 at 17:12..
Reply With Quote
  #213  
Old Mon 31 May 2010, 19:33
David Bryant
Just call me: David #99
 
Western Australia
Australia
Hi all

I am about to order my motors and would like to know if anyone knows a reason why this should not work well on my MM.
Motion King 34HS9801 motors with 1/5 gear box reductions built on to the motor.
Motion King 2LA860 Drives.

Cheers
David
Reply With Quote
  #214  
Old Tue 01 June 2010, 01:27
shaper
Just call me: Jed
 
Perth, WA
Australia
Hi David

do you have any specs on the driver? as I couldn't find any for it on their site or is this why you are asking, they can be a little cagey with specs from my experience.

Jed

Last edited by shaper; Tue 01 June 2010 at 01:29..
Reply With Quote
  #215  
Old Tue 01 June 2010, 01:37
David Bryant
Just call me: David #99
 
Western Australia
Australia
Hi Jed
http://motionking.com/download/2LA86...tion_Rev.E.pdf

I am only unsure that the driver can be used in full step mode.

I do like the Gecko 203V spec, much nicer for the overloads but the Motion king are half the cost. I have a couple of Gecko's on order for work where I can do some testing
Cheers
David
Reply With Quote
  #216  
Old Tue 01 June 2010, 02:56
rotorzoomer
Just call me: Account - DISABLED
 
Account - DISABLED
Australia
Opinion

Their is no buyers remorse with the Gecko 203v and PMDX-125 combination...
Reply With Quote
  #217  
Old Tue 01 June 2010, 04:06
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
That "Motionking" drive is just a re-labelled Leadshine drive - the word Leadshine still appears in 2 places in their text. And Leadshine are quite acceptable drives.

(Mark, I have some buyer's remorse over the G203V drives. They are not killproof and we have a couple as evidence of that. The hype about their fantastic support doesn't help when you are far away from them. The G201X would have served us as well as the G203V, at a much better price.)
Reply With Quote
  #218  
Old Tue 01 June 2010, 06:00
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Andrew,

A resolution of 0.00004-inch per step would require about 50:1 reduction. Did you mean 0.0004-inch (0.0003937)?

With a 4:1 belt-drive and a 1-inch diameter spur gear, you would get about 0.000397-inches per step.

The bigger question is whether you should expect to have repeatable performance of 0.0004-inches per step. You're looking at moves that are about 1/10th the thickness of a piece of common copy paper. I think that you're expecting too much from a mechanical machine that uses a rack/pinion system and V-rollers.

To get that kind of resolution, your rails would have to be at least that accurate. Your mounting of the rails would have to perfect with no variation at any point. Your V-rollers would have to ride those rails perfectly and the V-rollers would have to be perfectly round and mounted perfectly centered. You could not have any debris on the rails that was thicker than 0.0004". The tension on the rack/pinion would have to be perfect with no slop or backlash. Temperature would have to be controlled so that the metal did not expand or contract during the cut. The material that you cut would have to be 100% stable, without any tendency to shrink or expand because of temperature or humidity.

On my Shopbot, on a good day, when I'm really watching what I'm doing and I'm cutting slow (less than 60-inches per minute), I can sometimes hold resolution to 0.002-inches, which is 5X worse than you're hoping to get. That 0.002" is the difference in sizes when I cut parts for inlays. Sometimes the parts fit and sometimes I have to do a little sanding to get the inlay to slip into place.

I have a good friend who is a machinist who operates machines that cost well over $100,000. He described the difficulty and expense of holding 0.0005" tolerances in mild steel. His machine had a resolution of 0.00025", 2X better than the tolerance that he was expected to hold. He had to make several finishing passes, each one progressively finer. Even then, he said that most of the parts ended up in the trash. What upset him the most was that he knew than there was no reason for that specification on those parts. He said that the completed piece would have functioned flawlessly if it was specified to have a tolerance of 0.0015". The customer was the Government. The company was paid on a cost plus basis, meaning that they charged for the time and materials required to make the part, plus a percentage for overhead. There was no incentive for the company to talk to the engineer that specified the tolerances. The company was more than happy to see most of the parts end up in the trash; it meant more money in its pocket.

When I was much younger, I had a fascination with rifles. I found a gunsmith in Murray, Utah, who had been trained by the famous P.O. Ackley. That gunsmith prided himself on assembling rifles that had a tolerance of 0.002", and those rifles were expensive. He told me that even with that amount of looseness, his customers complained. At the range, as temperatures inside the rifle changed, the shooter had to use more and more pressure to close the bolt. They had to use cartridges that had been previously fired in that particular rifle and then reloaded with minimal neck sizing only during the reloading process. The barrel floated in the stock, with no point of contact with the wood. He said that their was no known way to keep a wooden stock from changing dimensions, so the only solution was to not let the wood touch the barrel.

What I'm really trying to say is that in the world of cutting wood on a CNC machine, there is a very practical limit on the repeatable resolution that you can expect from a machine AND from the material that you're cutting. A good operator who keeps his machine in good repair can expect to hold that practical tolerance, but I'm guessing that the real world resolution, even on a well built and well maintained MechMate is much, much coarser than 0.0004".
Reply With Quote
  #219  
Old Tue 01 June 2010, 08:03
southernduckie
Just call me: Andrew (Duckie)
 
new south wales
Australia
machine accuracy

Hello Mike,

Thanks for your reply

by coincidence some of the parts are for firearms (i shoot center fire bench rest 500-1000m), but a tollerance of 0.001mm is not that tight in an engineering perspective, i worked in a toolmaking shop for several years, 30 year old mills can do it (although the scrap bin is heavy). you are right very light cuts are required with razor sharp tools.
the minimum specifications i am looking for are,

resolution of at least 0.005mm to gain the following
repeatability to 0.01mm (allowing for hand lapping due to accuracy issues)
x,y 90' acuracy of +-0.01`
x,y/z 90' acuracy of +-0.05`

looks though i might need to revise designs/use

cheers Andrew
Reply With Quote
  #220  
Old Tue 01 June 2010, 13:25
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Andrew,

I don't want to discourage you because I've never built or used a MechMate. When my Shopbot bites the dust, I'll most likely get busy and build a MechMate.

You might ask Gerald about what to expect IF you build carefully and adjust everything to its optimum point.

The machine that I own is a PRT-Alpha 120 X 60 Shopbot, which has a lot of design compromises that Gerald addressed in the MechMate design. The 'ripples' on the rails on my machine are still a major factor in "chatter marks". The general flexibility of my machine contributes to unexplained errors.

However, even with Gerald's improved design, when I look at your requirements, I'm concerned.

My expertise, if I can claim any expertise, is in the electronics portion, particularly stepper motors. I've never had enough money to buy a gearbox that had zero backlash. The Alpha motors on my machine have a list price of $1,800 each (with stepper driver) and they have a perceptible amount of backlash, probably much more than 0.0004". A belt-drive is virtually backlash free, but, depending on the belt material, it may have a little elasticity. The belts that I used on my machine, before I bought the 7.2:1 geared motors, were tight, but I'm sure that they had some elasticity.

Maybe the main reason that I'm concerned is that I simply expected my Shopbot to give me cut quality equal to the cut quality of most other CNC machines. Boy, was I surprised! After I had the machine long enough to verify that the chatter marks were not operator error, that "surprise" turned to anger, particularly because the sales literature advertised +/- 0.003" accuracy. The "chatter" marks were easily 0.06", or about 10X worse than their advertised +/- accuracy.

I've learned to live with the "accuracy" that my machine produces. When a customer wants high precision, I usually run a test to show him what I can do. Sometimes that means that he takes his work to another shop.
Reply With Quote
  #221  
Old Tue 01 June 2010, 13:54
Gerald D
Just call me: Gerald (retired)
 
Cape Town
South Africa
Quite simply, the MechMate concept & design is meant for processing wood and holding tolerances measureable with the tools found in the typical woodworking workshop.....ie. tape measure, steel rule and sometimes, in our wildest dreams, a vernier caliper. Our inconsistencies we would blame on the wood.
Reply With Quote
  #222  
Old Tue 01 June 2010, 19:45
southernduckie
Just call me: Andrew (Duckie)
 
new south wales
Australia
thanks

Thank you for your time Mike and Gerald,
I can see where you are both comming from im going to continue to build a MM to learn if nothing else i will put the machine to very good use, It can build the next monster of accuracy which can be 1' x 2' this will be hopefully cheaper to get prefect results on.

Regards Andrew
Reply With Quote
  #223  
Old Sat 21 August 2010, 21:13
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
So what motor setups would people suggest today?

I am looking for suggestions for motors and drivers from people with what's available in todays market (looking at last post date being 2008 and the link for first choice is broken) thanks all.
Reply With Quote
  #224  
Old Sat 21 August 2010, 21:44
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Mojo,

If you read Gerald's first post (27 October 2006), you'll find that his information is just a good today as it was when he wrote it. (I tried most of the links in that post and they still worked for me.)
Reply With Quote
  #225  
Old Sat 21 August 2010, 21:56
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
Geralds post

I did read his post but the link for first motor doesn't bring me to that motor type and researching the part number comes up with a blank currently I am going to contact the company and try to find out what they suggest as a replacement motor or what the new part number may be thanks
Reply With Quote
  #226  
Old Sat 21 August 2010, 22:11
MetalHead
Just call me: Mike
 
Columbiana AL
United States of America
Mojo

OM Motor info is here.
http://catalog.orientalmotor.com/ite...=prod&filter=0

Let me know if you are wanting these motors.

I saw the links were broken a few says back and have not had a chance to fix the links yet.
Reply With Quote
  #227  
Old Sun 22 August 2010, 10:24
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
I plan on buying these motors in a week or so. thanks again for the quick response
Reply With Quote
  #228  
Old Sun 22 August 2010, 15:38
Richards
Just call me: Mike
 
South Jordan, UT
United States of America
Mojo,

Let's review the basic requirements. The PK296A2A-SG7.2 motor from Oriental Motor is limited to 3A by Oriental Motor. That means that the Geckodrive G201, G201x, G202, G203v, G250, G251 and G540 could all be used. The G250 would require a special header or a customer "mother board" to be of any practical use. The G540 at $250 or $300, depending on your supplier, is perfect IF you don't need more than two output lines and four input lines, otherwise you should consider a Break-Out-Board, such as the PMDX-122.

That motor has an inductance of 1.5mH when wired half-coil. Because that motor is actually under driven at 3A, you could probably use a power supply voltage up to 50V @ 3A per motor and still stay within the 80C temperature rating of the insulation on the motor. Personally, I use a 35VDC power supply with those motors. They still run at full usable speed and at full usable torque.

Although most of us use a linear power supply, i.e., a toroidal transformer, a bridge diode and one or more capacitors, you could use a switching power supply as long as you add a large capacitor (10,000 to 30,000 uF) on the output of the switching power supply. Some people also add a high wattage diode in series with the output of a switching power supply. I don't. To me, that diode just drops the voltage by about 0.7 volts without offering significant protection. I could possibly see the use of a reversed biased diode in parallel with the capacitor that would act as a steering diode to handle voltage spikes when the power was turned off, but even that shouldn't be necessary with a properly designed switching power supply.

I have tested those motors extensively using the Geckodrive G202 stepper drivers, a 35VDC power supply and a PMDX-122 BOB. I've also done extensive testing with G203v stepper drivers, a 35VDC power supply and a PMFX-122 BOB. My favorite is still the G540 with a 35VDC power supply. No BOB is necessary. All three test setups worked flawlessly. I would only use the G20x stepper drivers and the PMDX-122 IF I needed additional I/O.

Keep in mind that SOME SG geared motors have a little backlash. The motors that I own have no detectable backlash.

For those people who live in a part of the world where the Oriental Motor products are hard to get or are too expensive, the second option would be to buy a good quality 34-size motor that has about 300 oz*in to 450 oz*in of torque (similar to the PK296-02A, PK296-03A, PK296-F4.5A) and that has inductance of 3.5 mH or less. Using that size motor with a 3:1 to 4:1 belt-drive and a power supply sized to the inductance of the motor, would give excellent resolution and excellent torque. The belt drive is more bulky than the SG gearbox and it usually must be custom made, so it isn't the ideal solution for everyone. I've made several belt-drives using material ranging from polycarbonate to aluminum to Delrin. I prefer those made from 1/2-inch thick Delrin. They were light weight and easy to machine on my Shopbot.
Reply With Quote
  #229  
Old Fri 10 September 2010, 11:42
Mountaincraft
Just call me: Mark
 
Shingletown, Ca.
United States of America
I have heard someone say elsewhere, that when using 2 motors on the X axis, an encoder should be used on each motor for 'feedback'.. Is this true? What exactly does the encoder do, and do the gecko drivers listed work with encoders?...

Also, I'd like to learn why using two motors is preferable to one motor and a drive shaft to connect both sides of the X drive on the gantry?... (just curious on this as to the general design considerations of one style over the other)

I have also seen elsewhere that it is a good idea to use motors with the second shaft on the back of the motor, as then a 'damper' (ie harmonic balancer) can be attached.. Any comment on this as well?

Because of those two factors, I'm thinking it is a good idea to plan on the 'B' model motors, but I'm not sure if it is something that is necessary or justifies the extra few dollars per motor..

Anyways, as interesting as reading this whole thread has been, and somewhat overwhelming, I didn't really see anyone address these topics, so I thought I'd toss them out there for comment from the gurus of all things mechmate and CNC..

Thanks,
Mark

Last edited by Mountaincraft; Fri 10 September 2010 at 11:48..
Reply With Quote
  #230  
Old Fri 10 September 2010, 11:54
danilom
Just call me: Danilo #64
 
Novi Sad
Serbia
Encoder could be used with stepper motors but is primary used with servo motors, ,it tells relative or apsolute position of motors shaft while it rotates it, so a driver starts the motor and waits until it gets the needed position from encoder and then it stops it (its servo motor principle of work)
Stepper motor gets command to move one step and it moves, no need to monitor its shaft with encoder, unless you need a system that does not miss steps becouse of some other reasons (use more powerfull stepper or gear/belt reduction then) and then the driver would "notice" that the encoder did not emit enough pulses so he would order a steper motor to move one or the number of missed steps.
stepper drives from geckodrive are not designed for use with encoder. Only their servo drivers.

I dont know much about dampers but they might be used I read somewhere but have not seen a lot talk about them.

sorry for somewhat crude english
Reply With Quote
  #231  
Old Fri 10 September 2010, 12:30
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
Mike, thanks for the in-depth reply. I was thinking i would use the PK296A2A-SG7.2 motor from Oriental Motor (geared) with the G540 drivers. I plan on using 3 motors for X,Y,Z and 1 for A which will be for the indexer. any thoughts on this would be much appreciated. This part of the build will be the toughest for me i think as it is all new territory.

Keep in mind that SOME SG geared motors have a little backlash. The motors that I own have no detectable backlash. I am assuming you own the motors listed above?
Reply With Quote
  #232  
Old Fri 10 September 2010, 14:35
bradm
Just call me: Brad #10
 
Somerville(MA)
United States of America
Mark,

The problem is that real world shafts aren't as stiff as you might think. They twist.

Mike,

You need two motors for the X. If you're thinking shaft, you need to (A) figure out where it would go in the mechmate design, and (B) determine if the motor needs to be upgraded. I think you'll discover that a suitable shaft and associated hardware will cost out to be more than another motor, and that you can't drive a large enough motor with the G540.

So now you're probably wondering how to make an indexer work. Here's an idea: get a G251 as well. How to drive it? Depends on how comfortable you are with electronics. One way would be to tap into the X step and direction inside your G540 and bring them out. Another is a second bob on a second parallel port. Another is using signals from a serial port if you have one, possibly inserting your own optocouplers. Ask if you start thinking about going this route.

As Danilo says, a well designed stepper system does not need encoders.
Reply With Quote
  #233  
Old Fri 10 September 2010, 14:58
Mountaincraft
Just call me: Mark
 
Shingletown, Ca.
United States of America
I wasn't really considering using a shaft on the mechmate.. but was just curious as to why when it was being designed, the decision was made to use two motors vs one motor and a shaft..

It's going to be a while before I get to the 'dedicated' mechmate router table, but I will probably consider adding an indexer when I do... The plasma gets built first, and it will double as a part time router (lighter duty) until the router can be built..

I will be sticking pretty close to 'recommended' components, so I feel pretty confident that mine will be a 'well designed' stepper system.... I'm building a temp cheapo table to use while the mechmates get built, but I want to buy as much of the hardware and electronics as possible sized and purposed for eventual use in the mechmate tables...

Gracias,
Mark

Last edited by Mountaincraft; Fri 10 September 2010 at 15:05..
Reply With Quote
  #234  
Old Fri 10 September 2010, 15:15
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
Quote:
Originally Posted by Mojo View Post
Mike, thanks for the in-depth reply. I was thinking i would use the PK296A2A-SG7.2 motor from Oriental Motor (geared) with the G540 drivers. I plan on using 3 motors for X,Y,Z and 1 for A which will be for the indexer. any thoughts on this would be much appreciated. This part of the build will be the toughest for me i think as it is all new territory.

Keep in mind that SOME SG geared motors have a little backlash. The motors that I own have no detectable backlash. I am assuming you own the motors listed above?
Oops forgot one motor for the Y car so i will be using 5 motors total
Reply With Quote
  #235  
Old Fri 10 September 2010, 19:22
smreish
Just call me: Sean - #5, 28, 58 and others
 
Orlando, Florida
United States of America
Mojo.
Okay - (Brad you can chime in here if you can help - Theatrical Brains think alike)

I would most likely use the G540 for the x y and z format of the machine and then add a PMDX board and single G203v driver for the indexer. Thus, utilzing two (2) LPT ports for communication.

My thoughts on this is experience - when adding the indexer on MM #5, I wished like heck I had some extra I/O ports on the machine to do duty purpose index switches or edge finders for really complex part work. Or that single contact closure to turn on the webcam viewer in Mach when the G-code hit a fault - to record the event.

Nevertheless, that option of having a separate, HIGH Current driver will allow you have much more freedom on picking a stepper motor sized appropriately for your indexer axis.

...Makes me think of the PK 296-F4.5A (which draws 6.3 Amps bipolar) which I used on my indexer. Lot's of torque required to turn large diameter (heavy) projects. It got hot and would draw every bit of that current load on the 4' x 4' x 8' blocks of foam (even thru the 40:1 double reduction gearbox it was attached to)

My thoughts.

Last edited by smreish; Fri 10 September 2010 at 19:24..
Reply With Quote
  #236  
Old Fri 10 September 2010, 19:43
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
I would assume a separate power supply for this driver and stepper would be in order? thank again
Reply With Quote
  #237  
Old Fri 10 September 2010, 20:04
smreish
Just call me: Sean - #5, 28, 58 and others
 
Orlando, Florida
United States of America
Nope,
Depending on your choice (remember, the Mh determines the proper voltage) so, a choice would be find a Stepper motor with a similar inductance (to maintain similar voltage requirement) and high current. It's a math game. Lucky for us in the US, it's easy to get almost any flavor of stepper from Oriental Motor - same day service on the west coast.

I would plan for that extra driver in the total CURRENT of the transformer. Same voltage, just bigger capacity on the wattage (current size)

I'll do some math later if you want.....but there are far more experts here on the forum than me.

(Mike, Gerald, Brad, Mike.....thoughts on this?)

Sean

Last edited by smreish; Fri 10 September 2010 at 20:07..
Reply With Quote
  #238  
Old Fri 10 September 2010, 20:59
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
ok thanks very much will do more research on my end then.
Reply With Quote
  #239  
Old Fri 10 September 2010, 22:24
bradm
Just call me: Brad #10
 
Somerville(MA)
United States of America
As I kind of hinted at up in #232, I'm with Sean on this. As long as the indexer motor can be kept in close spec to the others, think in terms of adding one more driver, and a BOB. Since you gain from extra I/O pins anyway, it makes sense.
Reply With Quote
  #240  
Old Fri 10 September 2010, 22:44
Mojo
Just call me: SolarImages
 
Bisbee, AZ
United States of America
thanks Brad
Reply With Quote
Reply

Register Options Profile Last 1 | 3 | 7 Days Search Today's Posts Mark Forums Read

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -6. The time now is 10:07.


Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.