Backslash on Y-axis
I'm trying to square the Y and X axis but I've stopped my activity 'cause I've noticed a backslash on the Y-axis.
Here the problem: you can rotate the pulley in (1) of around 15 degree but the portal is not moving.
After dancing around the machine for more than 2 hours, the only way to eliminate the problem that I've found is to hard-tighten the screw in (2).
Unfortunately, when doing that, the pulley becomes very hard to rotate by hand. I think that the motor will be not so happy of my solution...
Quite curiosly, the pulley on the other side is perfect. You can rotate it by hands with minimal force. It runs smooth, precise and with zero-backslash even if the screw in (2) is not hardly tightened.
Maybe is better to leave things as they are? How much disalignement error (in mm) can 15° of backslash produce?
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
A further note:
the screw in (2) keeps the bronze threaded guide in position. This is the guide where the threaded axis is passing through.
Maybe the guide has a poor quality and I should substitute it with a new one? :dry:
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
Nobody has an idea? :unsure:
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
Here the answer from Stepcraft:
"I think, the backlash is based on wear oft he leadscrew nuts on Y axis.
You need two new nuts."
Mmmmh... maybe they are right. There is something else I could check before buying the new parts?
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
Hold ballscrew with pliers / vice grips on inside (close to orange plate)
see does the pulley move independent of the ballscrew.
this can be a common cause of backlash. Backlash can be defined as were any rotational movement of the motor does not translate into linear movement of the cutting tool.
this can be
ballscrew / leadscrew nut wear
play in the ballscrew direction due to thrust bearing play
pulley grub screw play etc.
there are lots of places where play and backlash can come into the equation.
Ok Rory.
Tomorrow or aftertomorrow I will do further tests. I will report my results here.
Anyway, pulleys seems well tightened and fixed. Zero play. I was strong holding the ballscrews with my hands while moving the pulley. Anyway I will try with pliers also just to be 100% sure.
I suspect that there is some movement of the leadscrew along his axis. But is not easy to distinguish it from a wearing of the leadscrew... mmmmh... I have to think about it...
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
Here my results:
1. pulley has no play
2. leadscrew has no backslash
3. the long ballscrew has play!
I can shake the ballscrew by hand, moving it forward and back along his longitudinal axis. I can clearly hear "tump, tump,...", so it is moving and stopping, moving and stopping...
As I can see, by using just my eyes, the play is around 0.3/0.4 mm.
I have strongly tightened the big screws to make the distance between the orange plates shorter. No results. (I have also tried to add a washer between the screw head and the orange plate, like somebody in this forum has done.)
There is no way to prevent the ballscrew from moving?
Maybe a washer between the pulley and the orange plate?
🙁
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
Hi Julius,
Is not the longitudinal free play on the leadscrew set by the position of the pulley (part 62) on the end of the leadscrew? I think if you slacken off the setscrew and retighten whilst pushing the leadscrew against the flanged ball bearing (part 21) and the pulley towards the bearing you will be able to reduce the free movement to virtually nothing ( <.05mm).
Don't overdo it though, the leadscrew still needs to be able to turn easily to avoid overloading the motor. Check both leadscrews - if one has worked loose the other might have the same problem.
Peter
SC 420/2, Industrial VFD spindle from StoneyCNC
UC100 + UCCNC
Cut2D, Autosketch10, Draftsight, Eagle 9.5.1
There is no problem, however simple, that cannot be made more complicated by thinking about it.
Boom!
Hi Peter, you're 100% right!
I have done what you said this afternoon before reading your post. I have unmounted the pulley and the bearing and then started watching at them and thinking... thinking... thinking...
I've tightened the setscrew while the leadscrew was pushed against the orange plate (see the image below) and.... boom! No more free play that I can notice.
SInce I have no three hands, to keep the leadscrew in a pushed-position, I have simply rotated the other pulley in a such way that the portal was moving against the orange plate (even if a little warped).
By the way, you're right. One should not tighten that screw too much or the leadscrew will become hard to rotate.
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
By the way!
For everyone:
don't use pliers to keep the leadscrew in position, the risk of damaging the thread is too high!!! (Even by interposing some plastic material.)
Rather, use a block of material between the portal and the orange plate with the belt. When you tighten the setscrew clockwise the portal will be pushed against the block. This will allow you to provide a fine regulation of the setscrew.
MORE GENERAL HINT:
many of you yet know: the stepcraft is a bizarre machine, if you strongly tight the big screws of the orange plates (the ones keeping the aluminum sides) you will go in troubles for sure. The portal will not travel all along the Y-axis and you will hear that noise: grrrrr... grrr... indicating that the belt is jumping and the friction is too high.
Because of what above, you will need to tighten those screw individually: left and right, top orange-plate and bottom orange plate. Is a repetitive tighten-and-try process.
When the process is ended you will need also (final step) to clamp the portal to the orange plate (bottom or top) provided that it is perfectly aligned with the plate itself. The solution is to interpose two blocks of solid material and same size between the portal and the plate.
Hier the hint:
if you don't have the american 1-2-3 blocks (they costs a lot) I suggest you to buy some chinese mini-clamps! The clamps can be unscrewed from their base and the base can be used as stop-block for fine regulation of your steppie!
In fact, quite surprisingly, the base are really well made. The dimension error that I've mesured is lower than 0.05 mm, this is good enough, isn't it?
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
Hier my chinese blocks that I've used intensively to align the portal:
I've ordered four of them and waiting around 24 days to get them.
I've made my order sul Aliexpress.
B)
P.S: blocks are 45x30x12mm (if I remember right).
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
BTW, chinese block are dimensionally correct but not square 🙁
SC300 + Spindle HF500 + Portalerhöhung + LinuxCNC + gsimple
- 44 Forums
- 7,394 Topics
- 63.3 K Posts
- 9 Online
- 26.5 K Members