:)Apparently, the oriental motor guys thought gears was a good way to go...since they are selling them, and I for one will be purchasing four of them...

One concern I do seem to recall in a post several months ago is that the angle of the motor (or gears or belts, I suppose) needs to be somewhat perpendicular in relation to the racks... but I could be wrong... not the first time this year either... and there's still one more day left for many more mistakes...
Bill

Larry,

Maybe I'm misunderstanding your question. Why would you build a gear reduction using gears when you can buy a stepper motor with a gear box made specifically for it from Oriental Motors? I guess you might save a little bit of money if you were lucky but I don't believe your creation would be nearly as reliable or as accurate as the factory product.

The 7.2 and 3.6 from oriental are getting reports of looseing steps due to the gear box binding from workload. I just bought four 7.2's and I want torque, so we shall see. Mike is saying dont ask to much of them because the gearbox strains/binds under load. Plus they are the smaller motor. Mike is useing 600oz*in motor with a 3to1 belt that transfers full torque. 600X3=1800oz*in
If the OM 7.2 gearbox can only handle 640oz*in of torque you dont get your full 300X7.2=2160oz*in transfered to the pinion. In fact you only beat Mike's direct larger stepper by 40oz*in. That also means that you can only use 89oz*in of the motors 300 rating. It just doesnt sound right. Like I said I own four of them(7.2's), building the 4'X8' Mamba and want torque so we shall see whats up with em. Plus I am fortunate enough to live in the same town as Mike. So with his trusted word and my sacrificial motors and full Mamba setup I trust we will get to the bottom of it.

Does all that sound about right Mike? Or at least close :)

Edit: As Bill said maybe its the angle of the mount that is giving some people problems. Or maybe other parts of the gantry are binding due the new found torque with3.6's or 7.2's on old shopbots? At this point I am still only guessing.

Bill & Doug,
My first post here was in Selecting motors thread and noticed the gear motors had less torque than the un geared motors.Orential motors are very expensive compared to others.When a geared version is only $57 more I ask what kind of gear box are you getting.Other gear heads are $600+from other sources.
From reading Mike's post and Charlie's post,I thought the general consensus or at least in my mind it is preferable to direct drive with a larger motor and or gearbox.
This means for some real performance I am back to belt drive.How does one get 7 or 10 to one without a bulky package or two stages?3to1 is max for two pulleys.I thought actually using the pinion gears as gears it may be possible to get higher ratios in a smaller package.
Happy New Year
Larry

I hope that I haven't started a panic about motors and gearboxes. Let's go over the things that we can gleen from the data sheets.

- The PK296A2A-SG3.6 motor has about 320 oz*in permissible torque.

- The PK296A2A-SG7.2 motor has about 720 oz*in permissible torque.

- The PK299-01A motor (no gearbox) has about 800 oz*in holding torque when wired bipolar series and about 550 oz*in holding torque when wired half-coil.

Those are the basic numbers that Oriental Motor publishes. You'll note that the PK299 motor without a gearbox and the PK296A2A-SG7.2 both put out similar torque, depending on how the PK299 motor is wired. So, the speeds that Gerald is getting with his PK299 equipped MechMate would be very similar to the speeds that you should expect with the PK296A2A-SG7.2 equipped MechMate. The main difference is that the 7.2:1 gearbox would give you 7.2X better resolution, but would also require a pulse train 7.2X faster than the non-gearbox motor.

All of the e-mails that I've received where people are having problems with the PK296 sized motors are NON-MechMate machines. As we all know, Gerald has done an excellent job of designing the MechMate, including the use of larger V-rollers for smoother operation and the use of welds and bracing to reduce or eliminate machine flexing. So, the problems that I've been reading about MAY never be seen with the MechMate.

IF you're building a machine that MUST run as fast as possible, you'll PROBABLY want to look at large motors with a belt-drive transmission. A single-stage belt-drive transmission will have a maximum of 3.6:1 reduction. That is the result of the fact that the smallest pulley that can be bored out to accept a 1/2-inch motor shaft is the 20-tooth XL pulley. The largest pulley readily available is a 72-tooth pulley. The math shows that 72 / 20 = 3.6. I can't imagine a 3.6:1 belt-drive transmission being a problem. With a 30-tooth spur gear, the resolution would be 30 / 20 X 3.14159 / 3.6 / 2000 = 0.000654 inches per step. That is 5X finer than the thickness of a piece of copy paper. Believe me when I tell you that you will never get resolution anywhere near 0.0006 inches on any CNC router when you're cutting wood. Wood changes size from minute to minute in my unheated shop - especially as the humidity shifts from the deseret-like dryness to near tropical monsoon-like wetness in August. With a 3.6:1 belt-drive transmission and a PK299-F4.5 motor wired parallel, you'll get more than 2,500 oz*in of torque. The XL belt will probably be the limiting factor with that motor (although I haven't had any problems with XL belts on my 3:1 belt-drive setup).

If you already have 7.2:1 motors, use them. Cut thousands and thousands and thousands of parts with them. By the time that you've worn them out, you'll have both the experience and the money to buy whatever type of motor and gearbox that best suits your needs.

DIY Pulley calculation to fabricate T5 pulleys.
 

I recently joined this forum as I am contemplating building a router / Plasma cutter. My previous experience was converting a Bridgeport size import to 4 axis control using servos, driven by Geckos and Mach2.

Being a Brit and growing in an environment where hobby activities were possible only by doing as much DIY work as possible, I soon found that in 2001 there was comparatively little advice available on motor sizes and gearing ratios and that larger size pulleys were often very difficult to obtain and fairly pricey. I investigated and measuring a variety of pulleys found a couple of constants that enabled any size pulley to be fabricated.

(I just noticed in reviewing the word version of the AutoCad drawing that I just gave the diameter. It is calculated as PD - 0.839= 94.653, Published data = 94.65.)

Currently having drilled the center hole I mount the blank on an arbor and then drill the peripheral holes using a rotary table. However it is also possible to mount the blank and use a standard G-code program to drill the holes, this ability will be dependent on the accuracy of your system.

I have found that there is no real standard for either the pulleys or the belts for T5, some pulleys are obviously machined with a hob as per normal gears but there are others that appear to been drilled and after cutting to the Outer Diameter both sides of the cut holes are edge releaved. It is a similar story with the belts, some have chamfered teeth and slight gearing depth while others are very deep with virtually parallel teeth blending to a radius. Both combinations appear to work equally well together and in the intervening years have not had a broken or damaged belt.

I prepared an AutoCad drawing of a T5-60 tooth pulley. The reason for this size was because it was the largest that I could obtain Manufacturers data, for those interested it will be seen that the nominal Pitch Diameter is easily caclulated, though it is only used to establish the remaining critical dimensions.

On a different subject, I have noted that several people have discussed jerky movement and less than optimum performance. For those users that are interested I submitted some information that may be of interested in looking at ArtSoft's work in this direction, where they have been delving into these problems, the subject heading is: Quantum, long term software solution?

Hopefully some of the data that I have shown may be of some use to you.

When I was considering making a belt reduction I was going to use the parts from a craftsman 3 X 21 belt sander. They were very cheap and only needed slight mods. If anyone is interested I think I could find the model # and post it.

Bonding pulleys with Loctite
 

In reading through the various posts I have noted a couple of times that mention has been made of being limited to a minimum of 20 teeth for a 14mm shaft and I would comment as follows:

It is increasingly common that pulleys and pinions are bonded to shafts without any form of mechanical fastening. Loctite 609 is a classified as a general purpose retaining compound and is good to retain from press fit to up to 5 thou clearance. When the parts are properly degreased and Primed with 7649 strengths over 2,200 psi are attained.

I regularly use this product if I need to do turning operations on a previously completed part and then release the parts by applying a small propane torch to raise the local temperature above 170c.

Examining the xl profile the inner diameter of the teeth is 0.125" less than the outer diameter. Subtracting the 14mm shaft = 0.5512" diameter and using a 16 tooth pulley with a published OD of 0.999" and an inner tooth diameter of 0.874 - .5512 = 0.3028", i.e. a wall thickness of 0.15" which is more than adequate.
I used this method when working on the gearing of the knee (Z) axis of my mill.

The link to Loctite 609 data sheet is: https://tds.us.henkel.com//NA/UT/HNAUTTDS.nsf/web/3E906D6B842166B0882571870000D855/$File/609-EN.pdf
609-EN.pdf (application/pdf Object).

Depending on distributor a small bottle of 609 is in the region of $US12 and the Primer 7649 around $14.

Hope this info is of some use.

Gear info ?!
 

http://www.mechmate.com/forums/attac...1&d=1196783106

Hi Fabio,

May I be audacious and ask you to tell/share more on the how you did & conceive you motor gear belt reduction system ?
I’ve gone through your personal pages, read post 3, got the pics from #1 & #19 but for me, I’m still in the dark as to how you built them ! Never done any but would very much like to give it a try. I’m daring to ask “Better yet, any plans & parts list ? “ :eek:

From what I can understand there are some advantages to make your own “gear system” vs buying one that is part of the motor as many do to keep it simple The KISS approach !!!
I‘ve read many sources say it can be an advantage & flexible if ever needing for some better resolution or speed !!
But, with this in mind, I can’t seem to see how you kept it all in a close confined mater and with what type bearing/bushing on the shafts & how they stay/hold from the alu plate to shaft ?
Thanks for considering !
Robert

Hello Robert,

Well, the idea to build my own belt reduction was born when i tried to search already made reductions for stepper motors, or even motors with gear reduction like those from Vexta that are used by a lot of people here (in forum)...

As I told before, Brazil is not so good to find these kind of products, and importing taxes are very high...

I used as a starting point the original pieces (M5 10 312 and M5 10 322 motor swing plates) and imagined how to place one big aluminium 72 teeth XL belt gear as driven gear and a small steel 20 teeth XL belt gear as driver gear mounted on motor axis...
I also bougth a XL size belt (sorry don't remember correct length size), calculated the center distance between gears and started scatching some ideas.

The reduction is made basicalliy by two steel 10mm plates (but you can use aluminium) separated by 3 small collumns.

The final design is attached in DFX format - sorry if they are incomplet, but I asure you it's possible do build funtioning unit from them...

Each main axis, as I named the axis that the big 72 T gears is mounted is a 12mm diameter steel axis mounted on two 6001 ball bearigs.

Each bearing is press fit in a recess made in each plate as you can see in drawings, and there are two thin spacers between gear and plates (one each side) to maintain gear centered and not touching plates...

I then milled a flat recess in the shaft and installed a Allen screw to block the gear in the shaft.

The main shaft is protuding to one side to allow the module 1,5 pinion to be fixed with two allen screws (in milled recesses)...

In the plate where the motor is fixed, I milled a kind of recess to allow the motor slide in and out to adjust belt tension...

If you note in drawing, the width of the recess is smaller than the front of a NEMA 34 motor, and because this I milled two paralell flats in the motor front cover plate (without dismantling it! ) - (sorry I don't have photos).

The homemade reduction was them assembled over bench and then easily assembled in machine the same way the original one is mounted...

Attachment 1122
Attachment 1123

Fabio…the Fabiolous !!! < :0))

This is very well and superbly explained. It is very motivating & inspirering for me.
I’m no genius when it comes to metal working, I’m a meticulous custom & studio furniture maker.
Love Metal working, but do not practice nor do much, so… I’m not inclined to see those little things that may be obvious to others.
Thank for showing the way on this one to me… I owe you one ;)
Robert

Robert,

Please, feel free to ask me any other detail that interest you.

I'll be glad to help someone else...

Best regards,

Fabio
O.T:
I don't know if you know this, but i'm very impressed with the detail:
http://br.youtube.com/watch?v=kvHYaR...eature=related
http://br.youtube.com/watch?v=GK5_Bl...eature=related

Marc and I have designed this 3 to 1 motor reduction. If time permits I will install one on the Z tomorrow and report how it does.

Jr,
Those reduction units look slicker than snot......
patiently waiting for more info. I would like to build one of the MM's to run a Plasma cutter. Keep Up the good work.

Happy Hunting
archie =) =) =)

(newbe to this forum but not cnc)

Jeep 534,for more information on the trannys look in my personal thread. Later when Gerald returns, he will possible transfer info here?

Some help needed on for my future belt / trany system !
 

Have a problem as far as where to post my question, here or with the power supply section, so Gerald please see to this & feel free the reroute my question ?!

Initially I got soooo intrigued by Mike’s posts on his polycarbonate belt reduction but then I got itch & influence with Fabio’s version that it got me to the drawing board but finally got the kick and seduced by Marc & JR interpretation of their belt reduction trany system that I have no doubt anymore that I too will be making my own version but only after finishing & getting to run with success my version of the MechMate.

Question I would like someone’s help with is with the famous motor selection question.
Unbelievable the amount of time one like me who nose didly on motor / power supply section can spend on reading, sorting info & trying to learn to make the right final decision on those !!. :o

A note about my future set up is I’m designing my MM with an indexer. So, as far as this motor selection goes I’m still confused about the proper transformer/power supply.
should I consider getting all identical 5 motors to balance my future application or 4 identical for the MM and a different 5th for the indexer would not matter?
Would it cause some challenges when time comes to properly select the power supply ?

Example of what I came to select after some reading. No certain if those are ok but I’m aiming for 4 - PK299-F4.5 for the basic MM with 4:1 belt reduction and one PK2913-F4.0 for the indexer. I also intend to have belt reduction to the indexer motor set up in the range of….say +/- 15 to 25:1 via a double gear / belt set up ?

To some, maybe the possible torque obtained by this belt reduction with a PK299-F4.5 will be enough for the usage of the indexer, but on the other hand the extra torque available with the PK2913-F4.0 should be welcome if it as no penalty or negative effect in the electrical control box components selection & set up aside of it’s extra +/- 100$ for it ??

Amicalement, Robert ;)

Calculate the voltage of the power supply by taking:

32 times square root of milliHenry inductance

So for the PK299-F4.5 you get 50V and for the PK2913-F4.0 you get 65V. That means your power supply, if you use only one, must be 50V. That means your PK2913-F4.0 will produce less torque because it has a reduced voltage. About 23% less.

Thanks Gerald for your input.

Before going further Gerald, I thing this post should go to another thread. Please fell free to edit this one and the one before and redirect them !! ( or Maybe I should think starting my own personal page ?? )Anyway…..

Allow me to question on yours :eek:
But before, I should have been more specific about the how I intend to run the indexer motor at max torque possible since no speed is in need, therefore I’m assuming a Bipolar parallel motor mode is the right choice. If I’m perceiving this wrong, please redirect me, but bipolar parallel voltage needs to be set at 0.707 of the unipolar !
If I understand this math thing correctly, in the case of the PK2913-F4.0, inductance is 4.2mH unipolar or parallel, then the math goes:
S.root of 4.2 x 32 = 65V, but 65V x .707 = 46V
Due to my ignorance in this field, I’m not in a position to assume correctly, but couldn’t I use this motor set up ( Pk299-F4.5 need 50V, & PK2913-F4.0 @ .707 needs 46V ) with a +/- 50 to 56V or even higher ?
Some are using the 56V with the Pk296A2A-SG motors, rated at 1.5mH and reported good results, questionable about their findings but still !!! ( S_root 1.5 x 32 = 39V ??? )
Where I’m getting even more puzzled about this voltage math, is when I read out of the Oriental Motor spec sheet their testing the PK299-F4.5 in parallel at 60V ??? See page C-229

where I’m with another dilemma is all this complexity worth my trouble for some extra torque, especially if I intend to have a +/- 30:1 to 36:1 multi-belt / gear set up ! ( I revised my gearing calculation vs the 15 to 25:1 as previously written :o )

Pk299-F4.5 Bipolar Parallel = 880oz-in x 30:1 = 26 400oz-in = 137.5 ft-lb of inertia capability
PK2913-F4.0 Bipolar Parallel = 1320oz-in x 30 = 39 600oz-in = 206.25 ft-lb
That’s a 50% more inertia torque capability, but still only +/- 69lb !!!
Me to sort out, you to tell me I’m nuts or on track ??!!
Again, sorry to pull some of you guys hair, but haven’t you all gone through or close to this in the past :cool:!!

Thanks you very much, Robert ;)

MULTI national belt reduction
 

Sometimes its nice to see how the big companies put there product together.

Investing millions in making and maintaining their reputation in the CNC market they pin it on belt drives, a rugged, simple way of gear reduction that looks to have the ability to have backlash adjusted out when ever needed.

I have seen this method used on their smaller budget machines and also on their higher end models with complete tool changing systems.

The gear reduction is 3.75:1 . Sorry I did not count the pitch on the drive pinion or the rack.

D

Interesting parallelogram they have there - seems like the pinion is already on the small side and the motor is in danger of fouling the housing at the top. Thanks for the pics :)

The yellow are the horizontal links of the parallelogram - they can swing up and down:

Hey thanks for that!

I had been scratching my head trying to figure out the two adjusters. At first I thought the slots you drew were at some kind of slight angle that allowed the adjusters to draw the gear up to an exact position , but that didn't explain the spring. Now it makes more sense.

I can only think of two benefits.
1) ergonomics, it makes for a very flat side as the motor is tucked in the cross beam for the Y axis.

2)Possibly the motor cannot drive the gear off the rack , it is constrained by the slots and linkage from moving too low to grind the teeth and stalls instead. But the slots are still big enough to allow the spring to pull the assembly upwards and keep the backlash very low.

Bottom line is that belt drives are a valid method for driving your CNC machine

I love my Alpha Step SB machine but I am not convinced that it was the best way to go. I now think gear reduction through belts with no feedback on the similar size motors would give better cut results for less cost.

D

The need for gearboxes
 

Busy making prototypes for a customer on a tight deadline who's normal supplier of this item has dropped them. Not the most technically challenging job but interesting none the less - apparently these are destined for Cape Town newest swishest Hotel in the Waterfront (refer to recent star studded opening of another hotel from the same group in Dubai)

Sorry for the delay but the MM webserver or ISP blocked my IP from connecting to the forum.
Well here goes, we completed the machine last week and have done some cutting. I now find the biggest problem is to obtain the tools we need to do the cutting jobs that we have. I have a 8mm and a 12mm collet with the bosch router but found out that those are not the popular sizes for tools. I also need longer tools to cut the 3d plugs from foam. I guess this will be an ongoing battle.
The dust collection system works very well, almost to well. It leaves nothing behind.
I opted for a 4:1 reduction on all motors. See pics.
Thanks Gerald your effort made it possible for us to have a good working machine in under two months.
Also thanks to Greg from CncDirect for all his help.

Marius, that's a beauty!

Even your dust hose is a matching colour! :D

Those belt drives are very interesting, it looks like you got standard flanged bearing unit somewhere . . . . . . (some car engine water pumps are similar . . . old merc. . . )

Geared motor details
 

Allan,
I attached the drawings. Take note that the Z axis has a problem. I had to lift the motor plate by 40mm on order for the plate not to snag on the spider plate bolts. I took one of the large spacers and welded a 40x10 flat bar with a hole onto it and therby lifting the motor plate. I also had to extend the rack in order to achieve the correct travel that I wanted.

Good Job Marius - especially the belt drives.

where's the video?

I meant to include these two pic to clarify the reduction assymbly. What is not shown on the drawings is that you must counter sink the bearing flange mounting holes and mount with m6 x 12 CSK cap screws from the front.
Take note of the hand of the motor plates. They only mount one way after you counter sink them. The Z axis plate only fits one way as well.

I haven't posted in a while so I thought I had better give an update.
I have not made anything spectacular just cut out a few MDF letters for a friends kids. Something for them to paint and put on the wall no pictures but you get the idea.

Now on to something I do have pictures for :)

I have started working on the belt reductions. They are based on the ones that J.R. made, thanks for sharing... I ordered the pulleys and belts from a local supplier who got them in promptly for me. Unfortunately they did not get exactly what I wanted and instead of a 3:1 reduction I ended up with what would have been 2.9333333333333333333333333333333:1. This was not going to be acceptable as far as I was concerned, so they returned the 44 tooth pulleys and got me some 48 tooth. This results in a ratio of 3.2:1. There was not as much room with the original belt as I had hoped but at least it should still work.

Attachment 3962

Due to the belt and pulleys not being exactly the same as J.R. used I had to change the design slightly.

Attachment 3963 Attachment 3964

The plates machined well and I managed to press the bearings in without much hastle. (made on my CNC mill)

Attachment 3965 Attachment 3966

My motors have 14mm shafts, but the only bearings I could get quickly and cheaply have 1/2 inch bores. This meant that I needed to make shafts with a small step in them, not a big deal to do just a little more time.

Attachment 3969

I also make a few spacers for the bolts between the plates

Attachment 3970

I was a little concerned as to how I was going to bore the small pulleys but I came up with a plan. I bored a piece of aluminium the same size as part of the pulley and then used loctite to "glue" the pulley into this while it was still in the lathe. I then drilled and bored the pulley. This kept it as close to centered as I could get it. I was very happy with the way this worked, and none on them slipped.

Attachment 3967 Attachment 3968

That’s about it for now. Hopefully I will get the pulleys drilled and tapped over the next few days so I can test them out.
There will be no working on the machine on Saturday though as I am off to an Air show with a friend but I am sure to be eager to get it finished once I get back.

http://www.airshow.net.au/avalon2009/index.html

Sorry all for not posting on the weekend, I was just too excited about getting the gearboxes finished. They went together perfectly and have made a difference of course. My maximum speed has been reduced from 27 meters per minute down to 15 meters per minute. I am still happy with 15m/min rapids, the table is not all that big so it gets from one end to the other quickly. The added safety of the slower speeds is also a bonus ;)
Some pictures...

Attachment 4019 Attachment 4020
One X axis gearbox and the y installed

I got a delivery of some 6mm acrylic sheet during the week that I had hoped to make an edge lit sign with. The sign is 230mm high by 530 long and is to go over a friends BBQ area. I was not sure how it would go so I was conservative with my machining feeds and speeds. The finished depth was 4.5mm and this was done in 4 passes at 1900mm/min using a 12mm 45 degree V cutter. It took around 45 minutes to machine. If I was to do it again it would probably be in 2 passes at the same speed.

Attachment 4021

The picture does not do the finished product any justice, it does look fairly good even if it’s me that says so :)

Attachment 4022

Here is a close-up of one of the letters just so you can see how badly the machine does curves...

I still need to mount some LEDs and a way to hang the sign but so far so good.

That’s it for now, off to dream up another small project before I start the next big one.

Jayson,

I assume you're being sarcastic about "how badly the machine does curves". This looks really good. Congratulations!

Jeff

Hi Jeff,

Yep I was being sarcastic. I feel very happy with the cut quality.

Regards,

Jayson.

P.S.
Attachment 4051
Sign Finished.

Here's a couple of pictures of the complete reduction units...all axis's now have a 3:1 reduction drive installed.

There is a noticable difference in the smoothiness of operation and cut. Definently a worthwhile upgrade.

I decide to try and gear my Mechmate, the arc cutting quality began to be a little irritating. So I went out and bought some parts and the last ones came today.

Looking forward to get it all mounted, hopeful this will work out better for me.

The gearing is 3.2 : 1

Mike,
Did you machine the plates for your reduction on the MM? Give us some details on your reduction build! :)

I had the plates machine in a machine shop just 2 min drive from my home, the also made the shaft and machine the drive wheels. I made the drawings after J.R.'s model with a small modification. Thanks J.R. for sharing with us.
I just finish mounting the gears and I must say that it is running much better with the reduction on, tomorrow I will test cut to see if my ARC is smoother.

My feed rate is now 12000 mm/min and I'm very happy with that.

A few more gearbox pictures.

Dennis

Belt reduction drive
 

I am planning on having a belt reduction drive. (3.2:1). The geared motors are too expensive for me an i think i remember reading the performance of the belt drive should be superior than the gearbox motors (esp. w.r.t backlash).

Here are some CAD pics

Attachment 4789
Attachment 4790
Attachment 4791

Matt,
Those pics look good. Are you planning on cutting them on the MM once it is done with the direct drive llke JR did?

I am in the process of seeing what they will cost to be made. I would try and cut them with direct drive but my motor brackets have a bore slightly smaller than my motors so they don't fit directly to the plates. So the Aluminium plates has been sized to fit my undersize motor plates and accept my motors. I will post pics when/if/how they are made.