FineLine Saturn2 4' x 4' with ClearPath, UC300-ETH / UB1

[rcmcdee]

Auto Tune video 2 ->

[rcmcdee]

Auto Tune video 3 ->

[DDgitfiddle]

Robb,

I have some answers for your questions, but not alot of time...so quickly:

- Once you have the tuning for one of the gantry motors, you can load it into the other motor using the Clearpath software. You shouldn't need to adjust the tuning numbers. You just need to remember to reverse the direction of the motor. I forget if I did this in Clearpath or UCCNC. If possible, I recommend trying to startup UCCNC and run the motors together before re-attaching the gantry. These gantries are stiff and it is hard to tell if one motor is running the wrong direction.

- Once you have UCCNC/UB1 running, the UB1 sends a pulse train to the motors via the motor signal cable to command motor motion. The tuning parameters stay in the motor. You will not need the USB cables connected anymore unless you need to adjust the tuning for some reason. If you need to retune for some reason after the gantry is reattached and UCCNC is running, you can command UCCNC with G-code to step back and forth, then use the Clearpath software to adjust BOTH motors (i.e. 2 computers, 2 USB cables).

- I will send photos of how I routed my spindle cable. It is pretty much self-supporting (no cable chain), but you have to make sure that it doesn't rub on anything as it moves.

-Robert

[DDgitfiddle]

Here is a quick shot of how I ran my spindle cable:



It is tied to the Z-axis motor cable bundle and then just attached to the back of the motor. I think this was possible partially because I had to twist the cable in order to get it to line up with the motor connector. That tended to push the cable outwards. You can achieve that by rotating the spindle motor if you need to. They key is to make sure that the cable is as short as possible when at the top of the stroke, and to try to get it to lean away from the mechanism. You could also add some corrugated plastic tubing to the outside of the cable for extra protection.

I also wanted to clarify my second bullet point above. When I tuned my gantry motors, I used a back and forth motion that was pretty fast (something like 600 IPM with a very high acceleration). This was generated by the Clearpath software and I used it for tuning. I believe that the Clearpath software will also let you tune the motors while commanding the motor motion using the normal inputs. So you would program a back and forth cycle using G-code and command the system to run it with UCCNC. At the same time, Clearpath software would be connected via USB cable to allow tuning to be adjusted. If it is possible to have two separate computers connected to the motors, it will be easier to adjust the gains on both motors at the same time. I have not tried any of this, but it should work based on what I saw in the Clearpath software. I need to go back and try to re-tune my motors a little because I think I have some of the gains wrong and it makes the motors hunt a little when they are stopped.

-Robert

[rcmcdee]

Back in the shop working to get the machine to the next level. I had seen in another post that the motor carriages mount directly to the frame with no spacers. I had previously substituted a thin hex nut in place of the nylon shipping nut on the swivel shaft and that provided about 3/16 additional height for the X axis motor. When I removed that spacer nut and bolted down to the carriage, the motor gear needed to be lifted on the motor shaft to clear the block for the prox sensor. Once done the main (large) drive gear just kisses that block and grinds (see picture). I could easily mill a little clearance off that edge of the block for clearance. is there any reason not to do this?

I believe this would happen with all current shipping systems and could use a little attention in manufacturing.

I am also trying to get a good sense of how tight to set the backlash preventers. I am getting gear hop at the hard stop on one end of the X axis. Is there a good height measurement for that spring? Or a number of turns past contact?

TIA,
Robb

[rcmcdee]

Back in the shop yesterday trying to finish up the tuning. I am still having backlash issues on the X axis. I am running autotune in the middle of the axis at 100% torque and at about 35% through the program there is an attempted high speed run that causes the gears to hop and grind. If I tighten the backlash spring trying to eliminate this the motor gets pulled off vertical and the gear hop is worse. I lowered the torque to 50% but still get gear hop at all spring settings while hard stopped against one end of the axis. This does not happen at the other end, even at 100% torque.

Any thoughts on how to cure this? Why would this be an issue at one end and not the other?

Robb

[ger21]

Any thoughts on how to cure this? Why would this be an issue at one end and not the other?

When you say it does it at one end, and not the other, are you moving in the same direction?

If so, it's probably an alignment issue.

You may just have too much power for those drives. They probably aren't really a good choice for high power applications, as the required spring force at high power levels creates a lot of wear.

[rcmcdee]

When you say it does it at one end, and not the other, are you moving in the same direction?

When I jog down to one end of the axis and hit the hard stop the motor stalls and stops. When I jog to the other end and hit the stop the motor the motor does not stall and continues grinding gears. The gear hopping and grinding happens when under hard acceleration in the middle of the axis in either direction. It behaves the same when I reduce the torque to 50%. It could be that these have too much power as Robert had made a similar comment on his build.

[ger21]

When I jog down to one end of the axis and hit the hard stop the motor stalls and stops. When I jog to the other end and hit the stop the motor the motor does not stall and continues grinding gears

Probably because of the design, that in the one direction it's easier for the gear to climb out of the rack than the other direction.

[rcmcdee]

Probably because of the design, that in the one direction it's easier for the gear to climb out of the rack than the other direction.

Thanks for the feedback Gerry,

I understand this in the context of a router bit (i.e. climb cutting) with rake to the blade... but it does not makes sense to me with gears???

[ger21]

I think it may have to do with the pivot point of the drive assembly being on opposite sides of the direction of travel.

[mactec54]

Thanks for the feedback Gerry,

I understand this in the context of a router bit (i.e. climb cutting) with rake to the blade... but it does not makes sense to me with gears???

This is quite common with Rack and Pinion with this type of design not being quite correct with the design, with a helical Rack and pinion this happens in practice, in one direction it tries to climb out of engagement in the other direction it tries to pull in, this is normal, so you basically, you have the same affect with this design, there is no substitute to mount everything without the spring loading, this requires machining of all mounting surfaces for it to work, the Rack then must mount parallel to the linear rail

[klipsch]

Any updates on progress made with the motors and gear grinding/hopping?

[wmgeorge]

When I jog down to one end of the axis and hit the hard stop the motor stalls and stops. When I jog to the other end and hit the stop the motor the motor does not stall and continues grinding gears. The gear hopping and grinding happens when under hard acceleration in the middle of the axis in either direction. It behaves the same when I reduce the torque to 50%. It could be that these have too much power as Robert had made a similar comment on his build.

Your running the carriage into the rubber bumpers at the end, On purpose? Yes of course you can expect some grinding and jumping the gears. If you run normal operation with the Limits and Homing active is it ok then?

[DDgitfiddle]

Robb,

Per your request, I am trying to upload my UCCNC profile for my machine. I am having trouble getting it to upload, though. If it comes through, you need change the extension to ".PRO" from ".TXT" because I couldn't upload it with the original extension.

-Robert

[rcmcdee]

That uploaded fine. I have changed the extension and loaded the profile. This handles a lot of the questions I have. I still cannot get the machine to jog from UCCNC. Not sure where the issue is but will spend time in the manual and crosschecking my wring to see where the problem is.

Thanks again for your valuable input.

Robb

[DDgitfiddle]

That uploaded fine. I have changed the extension and loaded the profile. This handles a lot of the questions I have. I still cannot get the machine to jog from UCCNC. Not sure where the issue is but will spend time in the manual and crosschecking my wring to see where the problem is.

I've attached pics of my controller both on and off to hopefully help a little. Can't seem to attach pics or files in private messages...must be a limitation of the site or my browser.



-Robert

[rcmcdee]

This is very helpful Robert. I wonder what the LEDs are supposed to indicate. Why are some on with no wire plugged in? Why do the lights for the servos not synch?

Robb

[ger21]

Led's indicate the input or output is active.

[rcmcdee]

Thanks Gerry,

Here is a picture of my board powered on with the profile Robert provided loaded. There doesn't seem to be a lot of correspondence between lights and wires.