Jan 13, 2019, 03:31 AM
A very simple yet effective way to make anti backlash nuts.
I needed some anti backlash nuts for my CNC. Ballscrews were too expensive at the time, but I found a way to eliminate backlash with this method. It works by moulding the delrin nut around the lead screw. This creates a slight interference fit and there is just no space left for any backlash. These nuts have been on my CNC for 5 years now without noticeable wear. If that happens, I can just remelt them. A very cheap yet effective method. The delrin slides very well on steel. I lubricate with grease.
The Following 21 Users Say Thank You to Max Maker For This Useful Post:
baja (Jan 21, 2019),
boxofrogs (Jan 15, 2021),
Corm (Jan 14, 2019),
emu roo (Mar 25, 2022),
high-side (Mar 24, 2019),
JoeVanGeaux (Jan 15, 2021),
johnsmachines (Jan 14, 2019),
Jon (Jan 14, 2019),
LMMasterMariner (Jan 16, 2019),
Marine2171 (Jan 14, 2019),
mwmkravchenko (Jan 15, 2019),
nova_robotics (Jan 15, 2021),
oldpastit (Jan 15, 2019),
old_toolmaker (May 30, 2020),
PJs (Jan 15, 2019),
rgsparber (Jan 13, 2019),
rossbotics (Jan 13, 2019),
Saxon Violence (Mar 31, 2019),
Sleykin (Jan 20, 2019),
that_other_guy (Mar 22, 2019),
tooly (Apr 6, 2022)
Jan 13, 2019, 06:38 AM
These are also called EvaNuts and there is more to the story.
First of all, they will deform under pressure. So on the one hand you don't get backlash, but you do get an error that is a function of depth of cut.
The second issue has to do with leadscrew alignment with the ways. If a perfect leadscrew is perfectly aligned with the ways which are perfectly straight, you can achieve zero backlash. But as this alignment error increases, one of two things must happen. Either the leadscrew must bend so it always aligns with the zero backlash nut or the mechanism binds and you can't move the full length of the axis. This is most often seen at the ends of table travel because the leadscrew can bend in the middle but not at the bearings.
If the leadscrew binds, the side force will wear on the plastic nut and cause backlash over time.
Another problem has to do with the uniformity of the leadscrew diameter. The EvaNut is cast on the leadscrew and can be a tight fit if no packing is used. I found that a variation of a tenth was enough to bind the nut on the leadscrew. Add packing or cast the EvaNut at the point of maximum leadscrew diameter. Either way, you introduce backlash.
For more information on making these nuts, see https://rick.sparber.org/EN.pdf
For more information on how leadscrew alignment and backlash are related see https://rick.sparber.org/XAA.pdf
My goal here is to encourage others to prove me wrong and advance the "art". Just because I could not make this work does not mean others can't solve the problem.
The Following 5 Users Say Thank You to rgsparber For This Useful Post:
boxofrogs (Jan 15, 2021),
jimfols (Jan 20, 2019),
Jon (Jan 13, 2019),
PJs (Jan 15, 2019),
Sleykin (Jan 20, 2019)
Jan 13, 2019, 07:43 AM
What you wrote is the same that I found. Allignment is very critical. My leadscrew bearings are flexible in a spherical housing and I can move them up down and left right. So I move the gantry to one side with loose bearings, then tighten them, move to the other side and tighten them. I repeat this process a few times. On my next CNC I would mount the lead nuts floating to avoid that issue. But now I make more money and I could just buy a ball screw. However this method could be useful for other stuff like router lifts or tablesaw jigs.
Jan 13, 2019, 07:57 AM
Using EvaNuts on a CNC router has two big advantages. Your cutting forces are much lower than on a mill cutting metal so less distortion. The accuracy requirements are also much less.
I did play with EvanNuts that could slide perpendicular to the leadscrew's major axis. It didn't work very well because the force needed to push the nut sideways has to come from the bending of the leadscrew and that means excessive wear on the EvaNut. Furthermore, when under load, it takes a lot of force to slide sideways. So I found that the nut would jump back into alignment only when the cutting pass was over. This made for some very strange error behavior. Very dissipointing.
Jan 14, 2019, 04:56 PM
Jan 16, 2019, 05:51 AM
Thanks Max Maker! We've added your Anti Backlash Nuts to our CNC category,
as well as to your builder page: Max Maker's Homemade Tools. Your receipt:
Jan 16, 2019, 05:37 PM
Thanks, interesting now if I could find a loose leadscrew of the right pitch.
Mar 27, 2023, 06:56 PM
Mar 31, 2023, 03:32 PM
Do you have any suggestions for eliminating backlash from a rack and pinion arrangement? I've thought of shimming down the rack or putting the pinion gear on an eccentric so I can adjust it to push it more tightly into the rack. Any other suggestions are most welcome.
This is for a mini-lathe, if that helps. I have probably 15-20 degrees of slop in the main carriage handwheel.
Mar 31, 2023, 05:15 PM
I’ve done it by preloading with a gas piston (used on hatch backs).
Users Browsing this Thread
There are currently 1 users browsing this thread. (0 members and 1 guests)
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts