[neurofen]

.. im building my first cnc and i think the easiest(cheapest) way of mounting the x axis stepper motor is to put it on the top of the gantry and run a timing pulley to the ballscrew which is on the front face.

am i likely to be loosing accuracy and torque if a decide to use a timing belt and pulleys?

I'm pretty sure the best way to go is to attach the motor directly to the ballscrew via a coupling which i have.. its just that its likely to cost a lot more to manufacture a motor mount that will fit to do it this way, and my budget is now almost zero.

i found these

http://www.diycnc.co.uk/html/mechanical.html

the 350mm 5mm pitch 15mm wide high torque timing belt and 2 of the htdw24 pulleys.

and on rs online i found
product codes 474-6047 (timing belt) and 745-696 (pulleys)

i found a load of discarded L shape alumium angles which would make an ideal bracket for the motor on the top, all i have to do is cut them back slightly and drill a couple of holes.


are there better, more acurate types of timing belts? the ones on rs online are Polyurethane whilst the ones from diycnc are a different colour

[mactec54]

neurofen

The best timing belts, & the most accurate, are the GT2 profile

[neurofen]

thanks dude.just after posting i spoke with a mate whos made a handfull of cnc's and by all accounts i can kiss goodbye to the accuracy i gain from having tight pitch ballscrews if i go with pulleys and belts so i'm having to rethink my design again without them, back to the couplings.

[mactec54]

neurofen

You will not lose accuracy that you can notice,or measure by using the timing belts, Timing belts have been used on machining centre's for years, your friend must have the wrong timing belt set up, Ideal direct is better, The coupling that you have in the photo, would be less accurate than having a timing belt drive

[RICHARD ZASTROW]

Ditto what Mactec54 said. In addition, look at the G2 carbon belts, practically zero stretch after break-in adjustment. A bit pricier but you can use narrower belts for the same strength.

Dick Z

[hanermo]

There is absolutely no reason not to use exactly these HTD 5/15 mm timing belts and pulleys.

You lose nothing.
The proposed solution, on my 12x lathe with very high resolution servos (0.3 microns) delivers consistently better than 1 micron resolution in the real world.

Using GT2 etc. belts will be a little bit less noisy, and a lot more expensive.
Save the money, imo.
I am sure the GT2 may be better, but the HTD is a lot better than any machine, retrofit or coupling you are likely to buy. HTD alone needs very good mounts, if you are going for 1 micron or better resolution.

I did, and spent over 80 hours just fabricating the x-axis ballscrew kit for a 12 lathe.
And about 100 hours in finishing it (just finalising pics after doing finish work).
Polishing, bluing, lacquering and making a 3D model of the components took about 3 weeks.
Of course, I built mine for 1 micron resolution, and do get it.

I initially built the lathe with 1:3 at 10.000 count encoders, ending with 30.000 counts/rev. With a 3 mm screw, thats about 0.1 microns per count, and +/-2 counts/positioning about 0.4 microns real world.
In real use, I always got better than 1 micron resolution.
I cannot measure to better than that, yet. Probably will learn.

[mactec54]

hanermo

That is correct the HTD is ok to use but the cost is a little more for the HTD & it is not as accurate as the GT2, The pulleys for the GT2 that I checked were $1.20 cheaper then the HTD, The belts run around the same price

The GT2 is also a stronger system

[neurofen]

thanks guys, sorry, i posted then forgot to check back.. so you reccon pulleys and belts are good to go?it'd save me quite a bit in mounting the motor to the axis... have any of you got them driving 16mm , 5mm pitched ballscrews/ or something close to that with 2000uSteps per rev? max rapid feedrate will be about 4m/min(i think, not all that fast as they go ), but more than likely it'll be working a lot more slowly than this? i may well be using my machine for a lot of cross hatched engravings, so its important it can zip left and right at fairly quick rates over v small areas.

Whats the best kind of pulley to go for? i've seen loads with different tightening mechanisms and made from different metals. ideally i'd like something that tightens to the rotors without any additional machining, as simply as possible.

are there potential issues with using pulleys and belts i should be aware of?

someone said something about T5 belts? does this mean anything to anybody?

would it be neseccary to incldue a tensioner in the belt loop, i do have good alignment options on the motor mount due to the extrusion frame im using..ie the motor can be pushed back to the back of the x beam and moved up and down slots in the mount itsself.i can probably also twist it slightlt if a bore out the hole in the bottom of the L bracket to a slighly larger diameter.

[mactec54]

neurofen

Just look up GT2 5mm, Go to SPI-SI web & you will even be able to work out the centre distance/belt size /pulleys Etc

You can use the HTD for sizing the belt & pulleys, but use/buy the GT2 pulleys & belt

You want to key them to the shaft it you can, No tensioner needed, When you make your motor mount, slot the mounting holes, so you can ajust the belt to the correct tension

SDP/SI

[RICHARD ZASTROW]

There are tension techniques in Gates' eng. section of their cataloques. Do not apply too much tension to avoid radial overloading of the motors shaft end bearing

Dick Z

[neurofen]

thanks, im going to give it a go with a belt and 2 pulleys.

no gearing, 1:1.. is a larger diameter pulley with more teeth better for accuracy?i sit it all dependant on mass and speed etc

the gt2/gt5 belts are really difficult to find in the uk/ edit cant find anyone who sells them, i might have to go with the htd for the time being

[neurofen]

http://www.sdp-si.com/D265/PDF/D265T042.pdf

not sure about the copyright here, but its worth a read

[neurofen]

ok, so suddenly, as if from nowhere i found about a dozen uk based sites for timing belts etc.

so do yous till recommend gt2 over T5 belts.. found this

T5-340-16 Timing Belt :: BearingBoys.co.uk

[mactec54]

neurofen

As we have said in other posts, GT2 is the most accurate, If you can get the GT2, then that's what you need for your machine, The PDF will show you some more about the different belts profiles

If the HTD T5 is all you can find, then it will work for you, but won't be as accurate a drive system, As you will have a little more back lash than with the GT2 belt

[neurofen]

i'd read that , but a lot of the time on paper the newer products can SOUND a lot better than the ones that came before but in reality give no real noticable(maybe on paper) gains . Couldn't see a reference to the T5 type belts, just wandering what they were good for. I thought i'd do a benchmark and some tests, ordered one of the HTD belts and when i can find a supplier in the uk of the GT2 type will get one of those as well. I'll run them both through their paces, see what the real world difference really is.

I'd go for the GT2 first on your recomendation but still no-ones selling them here.

[mactec54]

neurofen

Both are made by the same company,(Gates) It is in the PDF I posted, you can see the difference in there illustrations

This is not a new product,It has been in use for many years

It's not just the Belt that is different the timing pulleys are different as well, As I said the HTD-T5 will work well for what you are doing,Just the GT2 is the best choice if you can find it

[neurofen]

thanks dude, i just cant find a supplier for the gt2 belts here which is a shame. the other tooth is going to have to do. like you said it probably doesnt matter since my machines hardly a leviathon.

[harryn]

I have done a fair amount of reading about belts, although I still am learning. It's interesting that different companies push different designs as "the most accurate". Brecoflex seems to be really sold on the AT style belt vs others, but in general, the T style is definitely not the way to go.

When you run the numbers, belt stretch seems to be a bigger design criteria requirement than tooth strength, at least for the belts I looked at, unless your application is just power transfer.

[RICHARD ZASTROW]

And the GT2 carbon fiber belts have practically NO stretch after initial run in. Also, as mentioned before, narrower belts can transfer the power than belts that are not carbon fiber.

Dick Z

[neurofen]

I have done a fair amount of reading about belts, although I still am learning. It's interesting that different companies push different designs as "the most accurate". Brecoflex seems to be really sold on the AT style belt vs others, but in general, the T style is definitely not the way to go.
.

They've all invested their money in their technologies, theyve got no other choice but to sell what they've got i suppose, anyway i went with the none T5 HTD belt in the end, which is good.