Checking y-axis backlash/freeplay
Moderators: al wolford, sbk, Bob, Kayvon
Re: Checking y-axis backlash/freeplay
Now, THATS a coupler. Short of a loose screw to start with and one that loosened up later, I haven't had any problems but I think I'm gonna go ahead and get a set of these and the first sign of a problem, the solution will already be in stock. Thanks for sharing that.
CNC Shark HD ~ Control Panel 2.0 ~ Windows 7 & XP
Located in West Tennessee near the Tennessee River
http://www.eaglecarver4.com
Located in West Tennessee near the Tennessee River
http://www.eaglecarver4.com
Re: Checking y-axis backlash/freeplay
Just a quick note on couplers, the "spiral" type couplers are made to have some "give" in them to accommodate axial misalignment.
While that is a good thing in some cases where misalignment may be an issue, they can introduce some axial error (under load) to the cutting path if they stretch and compress.
The shark relies on the stepper motor bearings (and the couplers) to take all the axial cutting loads, so as you are cutting, the lead screw, coupler, and motor bearings are being pounded back and forth trying to move the carriage. The faster you go, the more inertia makes the forces higher. So the poor coupler has to handle all the rotational work, while it is getting hammered on axially. So they have a tendency to vibrate loose. The Y axis has much more weight to move, so it is usually pounded harder than the X.
More costly designs fasten a bearing to the lead screw, and attach that bearing to the machine frame to absorb the axial forces before they reach the couplers. So the couplers only have to worry about the rotational job and the motor bearings are not pounded axially.
I have seen heavy split couplers that are not spiral design, and they may be a better alternative to the ones supplied with the shark. Mcmaster or some of the other usual suspects may carry them.
Just changing the set screws to cap screws does make a big difference though. You just cant get enough tightening force with a teeny allen wrench.
Bob
While that is a good thing in some cases where misalignment may be an issue, they can introduce some axial error (under load) to the cutting path if they stretch and compress.
The shark relies on the stepper motor bearings (and the couplers) to take all the axial cutting loads, so as you are cutting, the lead screw, coupler, and motor bearings are being pounded back and forth trying to move the carriage. The faster you go, the more inertia makes the forces higher. So the poor coupler has to handle all the rotational work, while it is getting hammered on axially. So they have a tendency to vibrate loose. The Y axis has much more weight to move, so it is usually pounded harder than the X.
More costly designs fasten a bearing to the lead screw, and attach that bearing to the machine frame to absorb the axial forces before they reach the couplers. So the couplers only have to worry about the rotational job and the motor bearings are not pounded axially.
I have seen heavy split couplers that are not spiral design, and they may be a better alternative to the ones supplied with the shark. Mcmaster or some of the other usual suspects may carry them.
Just changing the set screws to cap screws does make a big difference though. You just cant get enough tightening force with a teeny allen wrench.
Bob
Re: Checking y-axis backlash/freeplay
These may take a very small misalignment but that is not the purpose of the spiral. The aluminum designed coupler will flex a bit more but the intent is to protect the drive and the driver (screw and stepper) so as not to twist off the machined ends in high torque applications. The coupler is more sacrificial. Now a Lovejoy coupler is better for alignment and it also compresses more thanks to the rubber cog. All parts of the Lovejoy are physically separated.rungemach wrote:Just a quick note on couplers, the "spiral" type couplers are made to have some "give" in them to accommodate axial misalignment.
While that is a good thing in some cases where misalignment may be an issue, they can introduce some axial error (under load) to the cutting path if they stretch and compress.
Bob
As you pointed out, there are straight split-case couplers without the machined spiral that will couple the stepper and screw better than the set screw coupler arrangement.
Bobby
Re: Checking y-axis backlash/freeplay
There is another issue that may be worth mentioning. I had posted about it in a different thread, but it may be more appropriate here.
Unwanted movement can also come from the stepper motor shafts moving axially when pushed in. On many steppers this happens because the main bearing is pushed into the flange from the back during manufacture, and held in place by the motor shaft and a rear bearing. The rear bearing may be spring loaded with a wave spring to apply preload to the bearings. These motors were not really intended to take axial forces, and after a while their main bearing is not overly tight in its bore anymore. What happens then is when you make a cut in +y direction, it pushes the lead-screw in the -y direction and into the motor. The motor shaft may push in to the motor housing and quickly spring back. The only way I noticed this happening on my setup was because I mounted light plastic knobs on the motor back-shafts to allow manual rotation. When the shaft was pushed axially, the gap between the knob and the motor widened for a moment, then snapped back when the force was removed. It is very easy to miss, and impossible to see if you watch just the shaft itself. This movement can get to be almost 1/16th inch.
Unwanted movement can also come from the stepper motor shafts moving axially when pushed in. On many steppers this happens because the main bearing is pushed into the flange from the back during manufacture, and held in place by the motor shaft and a rear bearing. The rear bearing may be spring loaded with a wave spring to apply preload to the bearings. These motors were not really intended to take axial forces, and after a while their main bearing is not overly tight in its bore anymore. What happens then is when you make a cut in +y direction, it pushes the lead-screw in the -y direction and into the motor. The motor shaft may push in to the motor housing and quickly spring back. The only way I noticed this happening on my setup was because I mounted light plastic knobs on the motor back-shafts to allow manual rotation. When the shaft was pushed axially, the gap between the knob and the motor widened for a moment, then snapped back when the force was removed. It is very easy to miss, and impossible to see if you watch just the shaft itself. This movement can get to be almost 1/16th inch.
Last edited by rungemach on Wed Oct 23, 2013 8:59 am, edited 2 times in total.
Re: Checking y-axis backlash/freeplay
BIG Thanks to all have given this information...... I have the Shark Pro for several years now and I went so far as to drill a hole in my table to gain access to the screw. I could never keep tight, I was at the point of retightening after every use. Very clever and solves a big issue.
Thanks very much.
Dan
Thanks very much.
Dan