9mm Belts vs Screw Drive

I’m looking to upgrade away from the stock belts and was curious if anyone has used both of the 9mm GT3 belt upgrade and @Phantomm’s screw drive?

I’m wondering about price/performance. I’ve been reading everything I can about both and it seems @PhilJohnson is getting great performance with the belts and it’s a substantially cheaper route but as more people are figuring out what the screw drive can do I wonder if I should wait and save up.

Also, if I go with the 9mm belts, what benefit would I see from upgrading to the 269oz/in steppers vs keeping the 140’s? Would it be worth doing as an incremental upgrade or just do it all at once?

Any thoughts are appreciated.

it’s a win/win either way.

just depends how much time you want to spend on the machine and how much you drop $.

what’s the speed limit on 9mm GT3 belts? what becomes the issue at high speed? machine flex?
I seen Phantomm is looking to push the screw speeds to see what a reliable max is.

I’d take the speed of belts right now over screw drive. With the 9mm belts and the 270oz motors I haven’t really had an issue with any carve yet and I have been pushing things pretty fast without any step loss. The only lost off steps I have encountered with 9mm belts and larger steppers is when I forgot tabs on a small piece and my vac sucked it up and got stuck.

With that though, I brush off the rail and wheels with a brush and I check the vwheels to see if I can spin with one finger before every carve. (takes a whole 60 seconds). I check belt tension weekly. I’m sure I could make it skip pretty easy in some materials but slowing it down is probably still faster than screw.

If I jumped heavy into harder materials I’d definitely consider screw drive though.

Still, you could get away with the 9mm belt upgrade and the 269 or 270 oz steppers for like $160-$200.

1 Like

Where do people recommend getting 269 or 270oz steppers from? I did a little searching and didn’t find an obvious source.

Here are the 270oz that I use.

Here are the 269oz that a lot of others use.


Here’s the difference between the two.

Phase Resistance: 1.13ohms
Inductance: 5.4mH ± 20%(1KHz)

270: (I chose these for lower Inductance but the dual shaft is about 1mm from my Xcontroller)
Phase Resistance: 1.13ohms
Inductance: 3.6mH ± 20%(1KHz)
Dual Shaft

1 Like

The 270 oz-in are a better match to the 24V of the Xcontroller.


what’s the max motor you can power with the X controller?

It’s limited by the current that the driver can provide. For the X-controller it’s 4 amps per phase.


Is there a reliable way to measure how many amps the x-controller is putting out?

Yes and no.

You can directly measure the Reference voltage going into each driver chip and calculate the output current.

Current = Vref * 2.13 (The datahseet formula is (Vref/3) / RNF, the Xcontroller RNF = 6.383 ohms)

This would be per axis because Vref changes based on the pot settings. For a 2.8A output, you want a 1.315V Vref.

TP1 on the board will give you X axis Vref, TP2 gives Y axis Vref, and TP3 gives Z axis Vref.


I have 9mm GT3 but would like to build a screw drive machine one day.

Okay belts are faster in a no load situation, but thats not really a helpful spec.
You can push a screw driven machine much harder when it comes to feedrates and DOC so under load screw drive when done well will outperform belts any day.
On the other hand, especially when still learning, I prefer to have a weak link because when things go wrong something’s gotta give and if there’s no ‘weak’ belt in the system things might make a turn for the worst.

That is a seriously useful piece of info. Sometimes it’s hard to translate specs and numbers into application.

1 Like

As a screw drive owner I will confirm that the rapids achieved are nowhere near the belts. On mine with the Xcontroller I can get 100ipm on my 270oz motors. However, I can also take cuts on my Dewalt 611 that actually bog down the router. I’ve done .2in doc at 80ipm with my 1.25in surfacing bit that noticeably slowed the rpm, the setting on the speed dial made no difference. There are times (3D) where I wish I had the travel speed back. However, the chips you can create with the increased doc are substantial. I have also not lost a step since the upgrade.


some woods cut better going fast. I think I have about the same setup as you Phil except I have the bolted double X where you have the single fat extrusion. I once cut purpleheart (which is pretty dense) way too fast with an 1/4" by accident but somehow it did wonderfully well. But I’m not sure I would consistently use those settings on 5-hour carves with a 100 dollar piece of wood :sweat_smile:

You are right about the rapids though. Didn’t think of that. Although smart toolpath sorting will get you a long way with that.

Are there even pro machines that use belts? If so not many I’d guess. Maybe the xcarve is a bit too flexing to really benefit from rack/pinion of screw drive. If I ever do convert to screw drive, it’s not going to be a SO2 upgrade, but a new much stiffer build.

1 Like

Has anyone tried linear bearings with the screw drive,I don’t like V wheels, they need constant attention

1 Like

that would be a sensible thing to consider. But again, all these upgrades may not be very cost effective considering you’re beefing up an inherently flex-sensitive machine

1 Like

I have already done Phantoms screw drive upgrade,which is working great and just wondered if the linear bearings was a worthwhile upgrade, or is it too expensive, I have a linear Z drive and that is so much better than the Vwheels

Linear bearings would not be worth it in the Xcarve, at least in my opinion.

1 Like

Changing from 6mm to 9mm GT2 belt changed the accuracy of my machine a lot.

1 Like