3d Printing with Tinyg ( rotational axis A trouble ) page 2

Thanks cmcgarth5035 - The character of the extrusion (slow-fast-slow) is mechanical, and not a result of jerk settings. It is a peristaltic pump that squeezes a rubber tube to meter the ceramic slurry. Each cycle has beginning and end - one point of interest, with this mechanism, the amount of torque required of the stepper builds and then drops at the end of each cycle.

And thanks for your clarification on the board pots. I think I am missing some steps, so perhaps the power is a little low. It looks like a way forward would be to conduct some tuning/tests with the pot on the A axis. Still doesn't explain the repeatability of the erratic results, but it is a start.

thanks cmcgarth - That sounds like a good test. I think I can run it today. If you still think that the jerk may need to be addressed, would you suggest that I increase or decrease the value?

I ran a test yesterday - without slip - ran single curves from the bottom, middle, and top of the vases from the video I previously posted. I timed extruder rotations per 30 seconds and they were all identical - with the current feedrate, approximately 11 revolutions.

To further confuse, one thing I noticed was that even without a load there was buzzing from the extruder stepper - the pot was initially set at 12 o'clock, and an increase to 1 and then 2 increased the volume of the buzz. Turning the pot down to 11 and then 10 reduced the volume, but the buzz remained. The stepper did not buzz when it was new.

Amazon link to the stepper that is currently on the extruder:

https://www.amazon.com/gp/product/B00PNEQUZ2/ref=oh_aui_search_detailpage?ie=UTF8&psc=1

I switched to this larger stepper some time ago when the original stepper begain clicking. Thinking it may be underpowered, I swapped it for the larger one above. clicking still occurs after the machine has been running for an extended periods. Any thoughts on the cause, or if it might be related?

I'll be swapping back to the previous stepper today, and I think I can run the suggested test with both steppers over the next 2 days. Link to the original stepper which I'll be switching back to today. It is identical to those on all of the other axis:

https://www.inventables.com/technologies/stepper-motor-nema-17

I would go along with that suggestion... in the extreme make your pump speed constant and vary only the x,y, speed... to produce the irregular effect... otherwise you have about exceeded my ability to contribute other than to observe... it is an interesting project. I am curious about the pump stroke... does the stepper for A axis come to a stop in the cycle or just slow down. If so it seems the jerk could be a factor unless you've overridden it somehow. If not it would seem worth experimenting with... as if it works as described would help create the fat in center deposit without having to create it in the program.
You could also experiment with making A, a fucntion of distance traveled in X,Y. and introduce some jitter into that to avoid or reduce the condition of the droplets stacked exactly on top of one another.
I realize that's not addressing whats causing your issue... but any fix is a good fix ;)

I started the post above before I had seen whitecloud's most recent post... thought I could edit it but it appears not. Added this so it would be more clear that I was responding to cmcgrath5035 suggestions. I also wonder if you've had this issue from the beginning or it developed along the way.

Thanks at yawstick - The extrusion values are currently defined by distance traveled for each subsequent point (every 5mm along a curve or single layer of a vase). I like the idea of jitter in the extrusion value. I will try that at some point - after I get the darn thing behaving.

The A stepper only stops between curves. As I understand, the tinyG will evaluate it's path and the related speed/jerk/acceleration every time it is given a new point to travel to. The extruder stepper only 'stops' or pauses at the end of a completed curve before traveling to the next curve and starting again.

thanks again yawstick - trouble seems to have developed recently - in the past I had some irregularity that was easily managed with the onboard pot or addressing material consistency. My ceramic slurry is very consistant now.

I originally envisioned your extruder pump as a screw in a barrel type pump. I had to google yours to see what you meant. I would think that type of pump would naturally have a bit of a pulse to the output when running at a continuous speed. Do you have a link to it? How many steps required for a full revolution?

Is it possible to remap the motor outputs so that you could swap A & Z to different driver chips.

Have you tried just making just one vase form to see if it still happens in the same place in Z?

Curious if you came up with anything on this?

I have conducted a few tests - to no prevail:

3/1/18 printed a series of circles utilizing the same gcode but varying the feedrate (7 single layer circles that are approximately 6" in diameter). The results is that the extrusion rate remains consistent regardless of feedrate (100 - 1500) (250-1000 is my normal operating range).

3/3/18 - Reduced Micro stepping on the extruder while printing the same file as above. Tested at 4 and 1 microsteps. No visible change in extrusion rate - was able to reduce the power to the stepper with the POT and found that there was a great decrease in operating temperature of both the chip and the stepper (I've been monitoring this with a laser thermometer - never any higher than 120 f - much cooler with lower microstepping) - the trade off appears to be increase noise.

3/4/18 ran the tall vase gcode (the problematic one in the video) from my PC instead off the Raspberry Pi - very similar results (3 apparent extrusion rates at the bottom, middle, and top). seem to be at approximately the same points in the code.

3/5/18 ran tall vase gcode without microstepping on the extruder - again very similar results (3 apparent extrusion rates at the bottom, middle, and top)

3/4-3/5 - saved and reviewed response from tinyg - no errors detected

Still having some trouble. I've experimented a bit with adjusting the jerk settings on the A axis.

If I understand, reducing the jerk would create smoother acceleration

I have reduced the value by half and then quartered - it hasn't seemed to help. I have not tried increasing the jerk value. Any advice on tuning the A axis?

Also - could the erratic steps be heat related? Thinking about heatsinks and/or a better fan. I currently have a 12v 40mm fan blowing directly down on the board from just a few inches away - I could mount a small blower directly toward the chip on the A driver.