Solved *Read pg 8 if you use a Geckotek bed!* Changed power supply, now y axis waves

I decided to try to swap X and Y stepper motors, but after I took them both off I realized that since I can't get the pulleys off, it wouldn't be possible. Shafts are different lengths anyway.

So, since I had the top off (so there is minimal support) I took a rigid tube and drove Z axis all the way up. It did not wobble at all. It did move at a slight, constant angle relative to the base, but that doesn't surprise me since the stepper motor hangs off center in the direction of the lean.

The way I see things, it HAS to be in the Y axis system, right?

 

Exactly. That's a perfect binary-search troubleshooting technique. You seem to have ruled out the part causing this in that step since the flaw follows the printer in this case (and you've printed more than one part).

 

I figured last test I did (raising Z to maximum height with the cover off) with none of the wobble eliminated Z wobble as a possibility. No?

 

Since I've changed RAMPS board to a generic and now only have 1 Z stepper driver, where else in Z could the wobble come from? The guide rods are straight according to a machinist square. Its almost like an acceleration type issue but I turned acceleration and jerk down to a crawl with zero difference.

 

So I printed two more towers in vase mode. My thought was to minimize running time (heat build up) and mass.

To my surprise, there is perhaps the most minuscule amount of wave action, but very acceptable. Now, I did replace the Y stepper driver and tune it. I am running full models now.

Now, I was taking a CLOSE look at the Z axis. I did notice that on the (my) left side of the machine, the lead screws had sort of shiny spots where oil wasn't building up. I ordered a new pair of threaded stepper motors, be here Monday. I did my best to use a square and it doesn't look like Z axis waves as it travels up/down, but I had no way of clamping the dial on to measure for sure.

 

A lonnnnnng time ago when dot matrix printers were around, Epson came up with a very innovative gear set which involved a pair of driving gears, as mated up with the driven gear. It was absolutely backlash-free and I was amazed at their cleverness.

Imagine now that the pair of gears include an internal spring to bias their teeth to separate as much as possible. This is what forces a rock-solid coupling with the driven gear.

http://www.designworldonline.com/new-gear-design-solves-automation-snafu/#_
http://www.automotioncomponents.co.uk/img/products/large/anti-backlash-pinions.r2080r.jpg

Ultimately, we could develop a split-gear approach to replace the stock driving gear (in each case) so that positioning could tighten up past what it is now. /daydreaming

 

Checked, all good.

 

Well, still no luck. The waves are still there. Interesting I printed two towers, the one toward the rear of the machine is much worse.

 

Just got the new stepper/belt/pulley in for Y axis. Starting a print now, but wanted to say I finally found a good hot end cooling fan! I've tried them ALL on Amazon, but I ordered some from here and that thing MOVES SOME AIR!!!!!! They say it has a "sleeve bearing". But it is light-years beyond anything I have bought to this point (which all seem to fail within a few months).

 

:fingers-crossed: I have a good feeling about this.

 

No luck...

Printing generic lead screw flange-nut adapter things for the new Z lead screw steppers next.

Going to have a brand new machine when I'm done (and spare parts enough to build a second!)

 

This is starting to seriously annoy. New Z axis lead screws/motors, STILL no better.

What is left?

Recap:
flashed firmware
rebuilt model
confirmed it wasn't model with known good model
new Y axis stepper motor/pulley/belt
confirmed Y axis belt bearings tight
Z guide rods tight, straight.
New RAMPS, generic
New Y axis stepper driver, calibrated.
Back to OEM PSU
X carriage seems tight
Still only starts becoming noticeable at about 30mm
New high output hot end fan


????????????????????????????????????????????????????????????????

 

Any chance that this is related to the entire platform wobbling/rocking?

 

It all seems solid?

For shits and giggles, I put a new Geckotek plate on. It did the same thing, although the model further to the rear is significantly worse than the one toward the front.

 

There were times back in the manufacturing plant where something big, industrial and motory would start going squirrely and you might simply put some rope on a cinder block and wrap it around something. The extra mass would be enough to slow down the oscillations. It's like sitting on your clothes dryer during the spin cycle when something inside is unbalanced.

My suggestion earlier was to apply some pressure on the top of the gantry during the printout when it starts to wave like that. If the extra stability/weight of your hand pushing down makes the problem go away, then you've troubleshot this down to some sort of oscillation or wobble in the entire platform itself. It would be one of many things you'd want to rule out, ticking off a long list of things you've already looked at.

(I used to repair washers/dryers as an amateur, btw. If the four feet aren't aligned in a plane on the floor, it will easily wobble during the spin cycle like this.)

 

So... what if the center of gravity of the entire rig—when the printhead is in the back-50% of the print volume (especially high up)—is now a leaning tower of Pisa, rocking backwards?

https://www.thingiverse.com/thing:217346

Maybe?

 

Its all pretty level and stable, but who knows. I moved my test print as far forward on the plate as I can, see if that makes a difference for some reason.

 

No difference moving the print to the front of the bed and using clips to extra secure the build plate to the heater plate.