The v3 extruder stopped working

[wcogen]

The v3 extruder stopped working. I'm not sure exactly why. I feel it isn't a hard failure.

I was printing and thought I was over extruding. Since most prints looked good, I hadn't calibrated the extruder. I never checked to see if when asked to extrude 100mm of filament if in fact it did extrude 100mm of filament.

I've been using a RaspPi 3 to control the v3. I've used this setup for over 2 weeks and have primed the extruder using this very setup when I change filament. I disconnected the PTFE tube from the extruder so I could measure the amount of filament moved. I've done this several times when first testing the v3. I know, I know, why didn't I calibrate it then? I used the home command and then tried to extrude 50mm of filament. No movement. In the past I've seen this after sending several extrude commands in a row. I'm careful to not send a command until the previous command finishes.

I reset everything and tried again. No good. I disconnected the RaspPi.

I connected Repetier-Host via USB. I've used R-H several times with the v2 and v3 with complete success. Again, I couldn't get the extruder moving.

I noticed while moving around the LCD menus a Position menu with the Extruder shown. I tried using this as well. No good.

I then unplugged the Y axis motor from the Rambo board and plugged the extruder motor into the Y axis socket.
Results: The extruder motor moved with either X or Y commands and the X and Y motors moved as well. Didn't totally understand that.

I then plugged the Y axis motor into E0 socket and left the Extruder motor wire unplugged.
Results: The extruder motor did not move. X and Y motors moved.

My guess would be that the E0 driver is dead.

However, inspecting the top of the Rambo and sniffing I detect no hard failure.

I looked at switching to the E1 driver but I don't know enough to feel confident in doing this without some guidance.

Finally, there are three other things I've noticed recently.

1. There is a AutoLevel Active (1/0) switch in the EEPROM configuration shown in R-H. Mine is set to zero(0). I've had inconsistent results with the autoleveling feature. It seemed to work only sometimes. I calibrated the horizontal radius the way I do for my MAX v2. So, what does that switch do? Should it be set to one(1)? Seems to me it should. Perhaps I inadvertently switched it off?

2. I noticed that the three carriages don't rest in the same position after 'G28' Tower Homing. See attached photos.
[img]https://drive.google.com/open?id=0B0TTH ... UxobXgyS0k[/img]
[img]https://drive.google.com/open?id=0B0TTH ... lM4SFRFZEE[/img]
[img]https://drive.google.com/open?id=0B0TTH ... 2VCUUw1cjA[/img]

3. I've been trying to balance the tightness of the bed clamps between secure tight and secure loose. Then I started using the thin end of the clamp to hold the build surface in place. Mostly it holds the plate securely in the X-Y plane and the little thin fingers put mild pressure down on the edges. Mostly I did this to put even pressure holding the blocks that support the hot bed plate. I found I was fighting myself with leveling the bed and uneven tightness. Make sense?

Any help would be appreciated.

Best regards,
Bill

[Xenocrates]

Two things to watch for.

A) the extruder will not function while the hotend is cold

B) if you did not turn the printer off while playing with the stepper leads, the drivers may have died due to the disconnect

[wcogen]

I did not know that the hot end must be hot to run the extruder. I'll try that. Sounds like that is my problem with the extruder. Explains why there were times it worked, loading filament, and times it didn't. How hot does it have to be? Any minimum?
And, yes, I powered down, counted to five, switched the steppers, powered on.
Thanks,
Bill

[mhackney]

Yes there is a parameter in firmware for that:

#define MIN_EXTRUDER_TEMP 150

Set at 150°C for the V3

[wcogen]

Happily the extruder works just fine. Minimum 150, check. Thank you both.

Any wisdom for the other things I see?

1. There is a AutoLevel Active (1/0) switch in the EEPROM configuration shown in R-H. Mine is set to zero(0). I've had inconsistent results with the autoleveling feature. It seemed to work only sometimes. I calibrated the horizontal radius the way I do for my MAX v2. So, what does that switch do? Should it be set to one(1)? Seems to me it should. Perhaps I inadvertently switched it off?

2. I noticed that the three carriages don't rest in the same position after 'G28' Tower Homing. See attached photos.

3. I've been trying to balance the tightness of the bed clamps between secure tight and secure loose. Then I started using the thin end of the clamp to hold the build surface in place. Mostly it holds the plate securely in the X-Y plane and the little thin fingers put mild pressure down on the edges. Mostly I did this to put even pressure holding the blocks that support the hot bed plate. I found I was fighting myself with leveling the bed and uneven tightness. Make sense?

Again, thank you.
Bill

These three photos show X Y Z endstop screws in relation to the endstop. The X endstop is about 2mm lower than the Y and Z after G28
I had trouble getting my phone to focus. The Y and Z endstops were identical.

[mhackney]

1) the special SeeMeCNC Repeater firmware has: #define FEATURE_AUTOLEVEL 1 so yes, it should be set to 1
2) this is likely because the length of X is 2mm shorter than the other 2. Is the screw screwed in further? It could also be that the switch is mounted lower (r the others higher) or the edge of the bed at X is lower. In this state, does the nozzle stay at the same high above the plate as you jog out to each of the towers near Z=0?
3) my recommendation on the bed clamps is to only use 3 of them and tighten them. This guarantees that the bed will not warp since 3 points define a plane.

[Polygonhell]

The heights might be correct, depends on your build.
The probing will measure the heights to the glass under the towers and if there is that much difference, that's what you'll see.
You can always check it manually after the auto level by moving the platform to the bed under each tower and checking what the height is, if they are all pretty close the leveling is working fine, The auto level in repetier is very basic, if one tower is miles off from the others the issue is with the probing, you have either a mechanical issue or an electrical issue that's causing the probe to trigger differently on that tower.

It's worth noting that you do want to be close to mechanically correct before you let any automatic level feature correct the error, if for example one towers had an endstop screw of significantly different length than the others it would affect it's ability to accurately calculate the delta radius, though even a few mm difference probably wouldn't make that much of a difference in the final calibration.

[wcogen]

After trying several things I reloaded the software and started over. I'll report the results tomorrow. I'm tired. Good night.

[moo77777]

When you home the platform, do your steppers stutter after all 3 hit their respective endstops before completely coming to a stop? This was the big problem I was having during my initial setup. This was causing any changes made to my endstop screws to be completely ignored after a home. I cant say what ended up fixing the problem but it managed to solve itself after some belt tightening and with making sure that the stepper wires were not near any of the high current carrying wires of the heated bed and the hot end.

[wcogen]

The support here is wonderful. Yes, there is some jitter/jutter/shutter. Different from my MAXv2.

We are expecting house guests in a couple of hours for the weekend. I won't be able to get back to this until Monday. Ugh.

So much to print, so little time.

[wcogen]

Figuring that the problem I'm seeing is self-inflicted, I've been troubleshooting on my own for a while. I'm stuck.

This is what I've done.
Based on what moo7777 described, I set about un-bundling and un-neatening all my wires. I've separated all endstop, stepper and power wires. I've rerouted to ensure no power wires are near signal or stepper wires. This happens with the bed and hot end off. I've tightened the belts very carefully. I reloaded the firmware. The stutter persists.

As an experiment, I changed the 'Tower X endstop offset' to 100 from 286. Y was 0 and Z was 59. This positioned the endstop screw at the same height in relation to the underside of the top section as Y and Z after sending a HOME command. I then ran the auto-leveling script. After it ran, the 'Tower XYZ endstop offsets were X=281, Y=0, Z=55.

The stutter is the problem and I've failed to find the solution. Please review the videos. One shows the HOMEing of my MAX v2. No stutter.
[youtube]https://youtu.be/igRX-fGAILg[/youtube]
[youtube]https://youtu.be/RxGOMFCudFM[/youtube]

Thanks for listening.

[Kevinvandeusen]

videos not working

[wcogen]

I made them public. Please try again.

[moo77777]

After doing significantly more experimenting, I have come to the conclusion that the end-stop screw problem you are seeing comes from a bad calibration run. Occasionally the calibration routine would do a bad tap but not bad enough to reject it entirely and that is when the problems occured. Make sure the calibration script is run cleanly and that should solve the jitter and weird offset you are seeing in the end-stop screws.

[wcogen]

Thanks for the guidance. I've run the probing gcode 5 times and the stutter persists. I'll put a video of my v3 running the script up tomorrow. The only other thing I can think of is the speed of the home after the first endstop touch. Seems faster than my v2. Hey, I'm almost at my wits end.

[youtube]https://youtu.be/-CKvpsth7is[/youtube]

One more thing. I disconnected the X stepper. Ran the home command. Y and Z reached home and stopped, waiting for X to home. When I manually moved X to trigger the X endstop, Y and Z homed quietly. No stutter. I have to re-run this test to be sure. When it happened I was sure it was something in the X tower. Haven't found it yet.

Good night.

[wcogen]

... long story short, I unscrewed the X tower endstop screw enough that the stutter is diminished to almost gone.
I ran auto-level several times. I printed a 100mm disk. See results below.

I've attached the log of the auto-level if that helps. I know something is eschew. The towers are all the same height. I'm tempted to try switching the X and Y endstop switches.

[Kevinvandeusen]

Yes, swap the stops. Also use only 3 clamps on the bed, and not tight. i fought the same battle on the second item

[wcogen]

I'm only using three clamps. And not too tight. I won't give up.

[wcogen]

OK all, this is where I am.
I changed the X tower endstop switch. This produced a small change. The stutter still exists. However, the X tower endstop screw stops in a very similar position to the Y and Z endstop screws. The stutter still exists.

I run the auto-leveling gcode and results are as before. See above photos.

I then checked the tower heights as I do manually on my MAXv2. The X tower height was about 0.2mm tight. I set it with the endstop screw and printed with good results. After making this change I don't run the auto-leveling gcode again.

It is the stutter. I don't understand it. I've had the stepper, endstop and power lines separated. Run straight. Kept in bundles. Twisted. Curled. No change. Please take a look at the latest videos. The stutter is clear when compared to the MAXv2 HOME.
[youtube]https://youtu.be/pNMh_3krOv8[/youtube]
[youtube]https://youtu.be/TQx9enTFk_U[/youtube]
[youtube]https://youtu.be/b5LPjmRc7gw[/youtube]
MAXv2 Homing
[youtube]https://youtu.be/igRX-fGAILg[/youtube]

I'm at a loss. I'm open to suggestions.

Thanks,
Bill

[mhackney]

Try swapping the X and Y stepper AND endstops on RAMBo. Make sure it is turned off! Then test. Does the stutter stay on X or move to Y?

[wcogen]

Seems to stay on X.

This is what I've been doing to test the levelness of the bed after running the v3probing script.

I set the debugging switch, Dry Run = On. I print a 200mm diameter ring. It is 10mm wide. As the hot end comes down to the bed I hold a piece of paper under the nozzle. As the printhead moves I move the paper with it. If the leveling worked, the paper slides between the nozzle and printbed. If it isn't level, the paper is pinched between the nozzle and printbed.

In this test, the results were the same. I was expecting the position on the printbed that grabbed the paper would have moved. It didn't.

This testing method seem OK with you?

Since changing the X tower endstop switch the relative positions of the endstop screws to the underside of the printer are very close. I couldn't determine if switching XY endstops and steppers.

100mm Ring_v03.gcode

(314.11 KiB) Downloaded 120 times

[aodhan]

Bill,

My v3 has done the little "stutter" since I first set it up. Some may argue that it may not be optimally aligned, but I have no problems with printing. I've made full bed grid prints in order to map the temperature of cold and hot spots on my bed and I'm able to use the entire bed with a consistent output from the nozzle. I'm just not convinced it's a problem, overall.

That being said, the assembly instructions didn't mention anything about how to properly calibrate the end stop screws and I concluded that it didn't matter so much as long as each one engaged and prevented the carriage body from slamming into the upper assembly.

I would certainly stop the endless swapping out of parts that you're doing. I think I would have gone insane by now. Good knowledge for us all, though. Thanks for sharing!

-Matt

[mhackney]

Keep in mind that RAMBo and Repetier do often give a little stutter when the printhead is lowered to print after homing. My V1 used to do it (now running Duet so it doesn't) and my V3 does it, even printing from OctoPrint (on raspberry pi). This behavior has always driven me batty. If this is what you are seeing, then it is "normal".

[wcogen]

I'm shifting this thread to "P003A HE280 [Solved?] Auto Leveling Issue/Bed Leveling Issue/Calibration". It started here and really belongs there. Thanks, Bill.