yeah. I’m not a fan of rack & pinion though. a whole new set of things to keep in mind with tensioning and such
2 Likes
Not really. Once you set the spring tension for the drive gear and for anti-backlash. That’s pretty much it just check them when you do your yearly maintenance check. The tension for belts are variable like over time as they age.
For belts yes. Rack & Pinion no.
I have a screw drive machine (Gatton CNC not XC) and a few months ago I had a wire come off the z motor during a carve… so it would go down but not up… was cutting a sign… it did a 3/4" plunge (with a 3/8" straight bit) and continued on its way like it was nothing… I just sat there and watched it out of amazement for about 30 seconds (project was ruined anyways) before stopping it. I have beast motors too (420oz.)
At the level of going to nice screws or rack and pinion, you’d want to ditch GRBL, go with a 4/5 stepper driver and use like MACH3. If you go that route, you can actually home both sides of the gantry independently to square it up. That’s one of the cool things I learned during my research into CNCRP’s machines.
2 Likes
(this isn’t a direct reply to your post Phil, I wanted to add based on one of your earlier posts hence the connection)
Belts are faster indeed, resolution is also inherently less and so it available torque.
For media like plastics, foam, wood belts are easy, cheap and reliable. Simple to make big for very low cost.
With some attention to detail and machine capacity it will also be quite capable with metals (for a hobby machine)
For harder materials like aluminium screws are better suited in the long run. Lower rapids, increased resolution and increased torque.
Combined with a rigid platform you achieve very low flex => one can operate closer to the bit chip thickness value. This in effect allow one to go harder, gaining MRR (Material Removal Rate).
ACME is the low cost screw option, the SFU1605/1610 next in line.
Belts (stock) is 25mm/turn
ACME is 8mm/turn (available up to 1000mm)
SFU1610 is 10mm/turn (available longer than 1000mm)
SFU1605 is 5mm/turn (available longer than 1000mm)
Rack and Pinion is great for large machines where one want “screw” performance at a lower cost.
My current CNC is a “hybrid” with both ACME and SFU1605 screws.
Rapids are lowish, but capacity (MRR) is great 
3 Likes
Rest of your settings?
Absolutely doable, your Xcarve is a demon and you know how to operate it too 
Charley also carve aluminium extensively with his.
For reference, my CNC with screws (ACME at the time) did 2mm depth, full width using a 6mm 1F in aluminium (6082)
Approx 0.09" per pass. (18k RPM and 900mm/min)
(This is with a DIY CNC using MDF, not an Xcarve)
1 Like
Can you send me the file. I wanna try it on my system. I have not powered on my machine in a while. I’m just so swamped with orders. 
I was looking around the forum to see if anyone did the linear bearing upgrade on the Y and X there was one who started doing it and I have tried contacting him, but got no reply yet, I am still looking around at what is available and found these, I am in the UK, I was wondering if the rails need 2 sliders or just 1, the rails would fit where the belts were, and could be fixed with T-nuts, as I have a screw-drive now instead of belts
https://uk.rs-online.com/web/p/products/4885091/
this is a link to the other thread
Replacing V Wheels with Linear Guide bearings
Thank you Bill, yes I would be happier with 2 carriages, fixing to the Y axis is easy enough but I am still looking around for a way to fix the rails to the X carriage. I have a one piece X
I’ve done all the upgrades you mentioned so I figure I’ll share my anecdotes.
FYI, I cut mostly plywood and MDF with 1/4” ,1/8” and 1/16” bits and seldom do 3d detail jobs, so in my application DOC and cutting speed are much more critical than jogging speed.
After going through 20meters or so of 6mm belt and several ruined jobs I was frustrated and ready to throw money at this machine to beef it up. I upgraded to 9mm belts (gt2) got an early version of the cnc4newbie z axis and upgraded to the 269oz.in steppers. I actually was able to use the existing 6mm belt clips. I didn’t put a lot of analysis into the motors, I just assumed bigger was better. I still don’t know how it would have performed with the stock motors. Maybe just the same… This made a huge improvement but because I did that all at once, I don’t really know whether this was mostly due to belts or steppers. I didn’t break a single 9mm belt for several hundred hours. Well worth the $50 or so for new pulleys, idlers, and belts.
After this upgrade I would reliably run at 30 or 40 ipm with .1 " DOC through birch ply. I began to skip a step here or there when bits began getting dull or I increased speeds and you could see the lip between passes.
I did the phantom screw upgrade a month ago and have since been able to cut at 50-60 ipm at least .1”doc without losing any steps. Repeatability has been very impressive, but jogging is a bit slower, I use 3000mm/min.
This has improved my overall job time by roughly 30% and increased accuracy. For me this has been worth it, but it might not make sense for everyone.
I suggest start with 9mm belts since it’s the easiest and cheapest of the upgrades you mentioned. Even if you decide to go with screw drive some day, belts are a relatively small investment for some definite improvement.
8 Likes
I am wondering if the original X-carve can run the 269, 270oz stepper motors, or if these require the X-controller?
The drivers can drive them but you won’t be able to make them utilise their potensial.
Torque is a factor of power, and the older drivers can not deliver the power the bigger steppers will accept.
Power is voltage multiplied with amp (limit) of the steppers.
To extract the potensial of these bigger steppers a controller capable of this is required.
You want even more power? Use a larger PSU, 2x input voltage = 2x power capacity.
Note that steppers drivers have a input voltage maximum, do not exceed them so one need to match the whole system.