E-Step, Random Length extruded with same E-Step Value

My printer (Prusa I3 Wood Frame) has been up and running for almost two years now. Recently (2 weeks ago) I upgraded my hotend to an E3D "clone". Once I got my cooling fan up and going on the new hotend I started printing a few calibration cubes. It was then that I quickly realized I needed to re-calibrate my E-Steps.

This is where it goes sideways. I start by heating the hotend up to temp, Prime the hotend, mark 40mm on the filament, then through Printerface I tell it to extrude 40mm. It extrudes, I measure the remaining distance. I get number "X". I do the math and input the new E-step value. I mark, extrude, measure again (M/E/M). This time I go WAY past by a factor of two or more. I reset to old E-step value. I then run three more M/E/M cycles to make sure everything is consistent. I get three different "X" VALUES!!!

So then I run 20 or so more M/E/M cycles. Everytime I get a different value.

So is this a software issue, a mechanical issue, or a "Idiot User" issue?

Just to be clear that I am doing the math correctly here are the numbers from my first M/E/M cycle.

E-Steps= 344.8

Marked at 40mm. Extruded 40mm. Measured 11.16 remaining. Ended up with a new E-step value of 475.824. That went WAY over.

Any help would be WONDERFUL!

Your math seems correct. Is it possible that your extruder is slipping, either physically or missing steps? Also you did not say if you are reloading firmware or saving to eprom. Is the filament easy to push thru the at temp hot end?

Saving to Eprom. Yes if I turn the wades gear by hand it runs smooth as silk. There is no visual sign of slipping or any sound that would indicate missing steps. If the Nema is missing steps, I would assume the would pause momentarily then begin moving again, right?

How about the teeth on the hobbed gear. I have a direct drive and the teeth got flattened after running pc-abs alloy. It didn't make noise, just didn't extrude properly.

I thought about that last night. The teeth totally clean. They are good shape. I tore the entire extruder down last night to make sure everything was in good condition, and working well. As far as I can tell there is nothing wrong with the extruder (Nema, hobbed bolt, printed parts, hotend, nozzle, fan.) It is just failing to provide reliable extrusion. I would think that if it where a software issue, there would still be a reliable and repeatable error. I.E. failure to extrude at a repeatable rate. This is so random. One M/E/M cycle it is short 4.65, the next it is 7.94, the next 1.34 and so on.

All I can think is there is a mechanical problem somewhere. I just cannot find it.

Yeah, if it's not the gear set then it would have to be either the motor, driver, signal wires, or firmware, would be my assumption. But, I don't have enough experience to know where to start.

Well after much searching there is nothing I can find. I am getting my friend to print me a new gear set for the extruder. Once I get that on I can see if it fixes itself. LOL

The only other issue I have had, which I assume is connected to this issue, is after a few layers the extruder fails to extrude correctly. Then if I just let it continue to print, it will start right back and be fine for the rest of a print. I would have to think that is a result of whatever this current issue is.

All of this started after a really bad clog with my original 2 year old J-head. It has not been right since.

You could try a new stepper driver in e1 if you are currently using e0 on ramps. This woul tell if ramps or the driver are the culprit.

Did you update the thermistor value to the new E3D thermistor ? If not, it could be your hotend isn't at the temp you think it is which might explain both issues ?

Paul.

Can you post photos of your extruder setup? That would make it easier to suggest possible causes.

I swapped out the Pololu with an extra that I had laying around. No change. I did not try swapping it to a the E1 location. I am currently installing new gears on the extruder as I found a hairline crack in the smaller gear. I will try swapping the Pololu over to E1 and changing Marlin to match if this does not work.

I used the existing heating element, and the existing thermistor as I am lazy and they where wired and where the same as what came with the kit. SO I did not update any settings for them.

Pics

Please forgive me for the Ghetto Rigged fan holder. The printer went down right as the new hotend came in and the fan on it was a joke.





The "hairline" crack that I saw this morning. WAY more than a hairline. I think this may have been the cause.



Edited 1 time(s). Last edit at 12/19/2016 07:27PM by windcalmer.

One other thing that can cause random extrusion - having an extruder max acceleration value set higher than the controller can reliably send pulses to the stepper motor.

Stick to 20000 mm/s or less. You only really 19200 (2G) on retracts at most.

That is the "Amax e" value right? It was at 4000. Changed it to 2000. If I need to go down from there I will.

Well, I got the gear set installed. Updated the Amax e setting to 2000, and what do you know. It works. I was able to get consistent results calibrating my esteps and a fully successful print!

THANK YOU EVERYONE!!!

Yep, Amax is one way that firmware describes it.

It is great that it works now.
Good collaborative effort. But what was the real fix? Was it the new gear set or the acceleration (amax) settings?

Definitely the broken gear was the main culprit, but the too high acceleration might have caused the gear to crack and split, over time. It certainly does not help.

Edited 2 time(s). Last edit at 12/20/2016 12:21PM by DragonFire.