Welcome to the SeeMeCNC Community

I've noticed that most "low" or "high" spots on the bed are actually produced when a tower (or a combination of towers) changes from moving down to moving up or vice versa.

Here is a a series of pictures that illustrates the results. Look at the brim, how on one part it's too squished into the bed, and how it's exactly related to the direction a tower is moving. Is this a belt tension thing? It's pretty consistent to each tower.

Then you have something "loose" or binding somewhere. It's what backlash looks like on a delta.
I would start by checking the UJoints on the effector all move freely.

The u joints all move easily. They are the acetal variety. I also have tension on the arms. I'm starting to suspect my belt tension is too loose as I previously had thought it was too tight and loosened them all off. I wanted to get some more thoughts before I adjusted anything. I agree that there is "backlash" somewhere, but I'm not sure if it's in the u-joints.

You can usually see backlash in the carriage motion, write some GCODE to just move the effector up and down, watch the carriage and belts, put you finger on the drive pulley and watch them again.

Hmm. Thanks I'll try that. I'll do a full tune up check tonight.

If you have the motors locked in position (just home the machine) you can wiggle things to see how/if they move. Try wiggling the hot-end, try moving each tower up and down, see if one of your arms moves more than the others, etc.

Check the set screws on the steppers pulleys also.

Watch the belts through the slots in the columns, if you see any slapping they are slack.

Well the hot end feels very very firm in place when the steppers are engaged. When I try to move the cheapskates up and down there is a TINY amount of movement, with the wheels turning a tiny tiny bit as a put a reasonable amount of pressure. Should the belts be tight enough that I can't move the carriage vertically at all?

Also, when I try and twist the cheapskate, the wheels are firmly in place, but the arms do deflect a little bit as if they are rotating a bit somehow...

I had a similar issues. My top pulleys are overhung, i supported all the three top belt pulleys with bearing and the problem resolved.. Its a mechanical issue...

What do you mean overhung? My top belt pulleys have always been a concern, re:level and tension. The belts also wear on the extrusion a bit but I can't seem to fix that.

bot wrote:What do you mean overhung? My top belt pulleys have always been a concern, re:level and tension. The belts also wear on the extrusion a bit but I can't seem to fix that.

Our friend does not have a true rostock, delta yes, but not a rostock.

Ahh hmm I see. I wonder what he meant, though. I might still be able to learn from his experience.

BTW, David, can you cast this in aluminum please and thanks: http://www.thingiverse.com/thing:330127

If I can get a decent print of it I think it will cast ok. Loks like a bugger to print though..
Did you check those set screws on the pulleys?? could be rocking on the flat of the shaft. My timing belts almost "drum" when plucked, but i still get some movement...

I have not checked the screws from the motors. I will do that soon. I made sure they were very tight and applied loctite so I didn't really suspect them but I should check.

if you follow the print back to the tower (Z axis??) your problem will be there. kinda like following the yellow printed road LOL If you can feel a difference in that tower you know you got it!

DavidF wrote:if you follow the print back to the tower (Z axis??) your problem will be there. kinda like following the yellow printed road LOL If you can feel a difference in that tower you know you got it!

An excellent point. Often people ask how should this feel or how should that feel... Luckily we have 3 of just about everything, and if one of them is different than the other 2, you've found your culprit

Yeah, this happens going in all directions... it's not specifically towards the Z tower. I'm accustomed to finely tuning machines, so each tower is very consistent... but they may all be consistently out of proper spec, which is what I'm trying to determine. I think I might print some belt tension adjusters...

So I checked my stepper pulleys. Rock solid. I removed my arms to adjust the cheapskates...

Hahahaha... oops. Again. It's always oops. They were rattling around they were so loose. (didn't seem so loose with the tension of all the arms on them).

I'm sure this will be it. I'm also tightening the belts a touch.

Thanks for the pointers... should have just checked myself. I was having too much fun printing (it was still doing pretty well).

bot wrote:Yeah, this happens going in all directions... it's not specifically towards the Z tower. I'm accustomed to finely tuning machines, so each tower is very consistent... but they may all be consistently out of proper spec, which is what I'm trying to determine. I think I might print some belt tension adjusters...

I've been contemplating a belt tension indicator, If for anything else, to have all three belts tensioned the same.