Hi!    Late to the party but here's my take on it. 

My background is polymers,  I actually went to college for plastics.    Just like the movie.    Think about it,  there is a great future in plastics.   

Anyway over 25 years of experience I have learned that there is no typical response curve for a material or family of materials.   That's why we re--re--embraced Scientific Molding and pushed it forward since the 1990s.    The key is to decouple the process from the material,  so you will get consistent results.   It's an approach that just works.    Molding shops that realize this excel at making product.    Those who fail,  well...

So what you want to do is look at the transition temps for the materials in the family and average them out.     Most filament I create is based off of MC6800 which is modified styrene and I also have three unique grades of Cycolac I extrude or have extruded for me.

If I wanted to eject at a low temp I would generally drop my temps to 200 degrees and let the machine handle the retract.     Any lower and ABS generally doesn't play nice.    I would probably end up at 210-215 for most of my materials.

The question is why.   I only do this for certain thermally sensitive materials like Surlyn or certain flex materials.     It is essential to get those materials out as quick as possible because they tend to break down at heat.   Typically I have it retract,  then kill the heat.  I'll be standing right there or nearby to catch the filament and pull it out the rest of the way.

ABS shouldn't have this problem.  If you are getting clogs,  you need to address other issues.   Either you have poor nozzles,  a problem in the hot-end itself,  or you are not getting proper temps.   

A potential solution would be to pull the material out hot and run a stick of Dynapurge through the nozzle then cool the machine down.   In fact I would strongly suggest doing that every so often anyway just to float any debris out.

When it comes down to it,  you are trying to solve a problem that a vanishingly small number of people have,  and that tells me you need to look at your setup.    Very few people in this hobby see a need to eject material at the end of a run,  and those of us who do,  do so because we run strange materials for unique projects. 

Lets address the real problem before we worry about thermal curves and material properties in a hot-end under specific conditions.

Very helpful, thank you.
I have not had any issues since stopping using the filament that I believed to be clogging, the material flecks in the filament I only noticed because of the way daylight interacted with the luminous green filament and I could see about every 9/10 meters a distinct mark and I believe that it was from manufacturing a filament in a dirty filament extruder or some fibres that really were not meant to be there were in the filament machine. The 3Dpintz filament and the other filament I have is still pushing though just fine.

As for the reason why, the filament is hygroscopic

Despite the place where I have my printer is the driest and warmest, the moisture in the air is still 28% when its 38% humidity. The thermistors both agree with the digital thermometer I have in there with it, its baseline temperature is 30.8 degrees C and can get up to 38.5 degrees C when operating.

So I wish to keep the filaments in their respective storage boxes in an air tight container the plenty of silca gel bags to ensure that there is little to no moisture in there.

My background is polymers,  I actually went to college for plastics.    Just like the movie.    Think about it,  there is a great future in plastics.   

Anyway over 25 years of experience I have learned that there is no typical response curve for a material or family of materials.   That's why we re--re--embraced Scientific Molding and pushed it forward since the 1990s.    The key is to decouple the process from the material,  so you will get consistent results.   It's an approach that just works.    Molding shops that realize this excel at making product.    Those who fail,  well...

So what you want to do is look at the transition temps for the materials in the family and average them out.     Most filament I create is based off of MC6800 which is modified styrene and I also have three unique grades of Cycolac I extrude or have extruded for me.

If I wanted to eject at a low temp I would generally drop my temps to 200 degrees and let the machine handle the retract.     Any lower and ABS generally doesn't play nice.    I would probably end up at 210-215 for most of my materials.

The question is why.   I only do this for certain thermally sensitive materials like Surlyn or certain flex materials.     It is essential to get those materials out as quick as possible because they tend to break down at heat.   Typically I have it retract,  then kill the heat.  I'll be standing right there or nearby to catch the filament and pull it out the rest of the way.

ABS shouldn't have this problem.  If you are getting clogs,  you need to address other issues.   Either you have poor nozzles,  a problem in the hot-end itself,  or you are not getting proper temps.   

A potential solution would be to pull the material out hot and run a stick of Dynapurge through the nozzle then cool the machine down.   In fact I would strongly suggest doing that every so often anyway just to float any debris out.

When it comes down to it,  you are trying to solve a problem that a vanishingly small number of people have,  and that tells me you need to look at your setup.    Very few people in this hobby see a need to eject material at the end of a run,  and those of us who do,  do so because we run strange materials for unique projects. 

Lets address the real problem before we worry about thermal curves and material properties in a hot-end under specific conditions.

1. I never operate or leave the printer running unattended.
2. The printer is in a cupboard that is sealed with a fire-proof door
3. Because of the location being some 30 ft from my desk and has to be controlled remotely I don't get to hear when the print has finished.

SO...

A simple retract feature to stop the hot end plugging up so I can remove the filament when I do discover that the printer has finished, it is hard to tell how long a print takes because NONE and I really do mean that NONE of the slicers used to date has given a real or close time to finish, I started a print one night at 8pm and it didn't finish until 4am when the slicer reported it to be a 3hr 57min print.

So the reality is this, the filament NEEDS to be retracted WHEN the temerature of the hot end has been cooled enough to stop the filament deforming or getting stuck when its pulled out the hot end, thats the problem here, the last several manual retactions has ended up with the filament getting jammed in the motor despite releasing it properly.

So this is a very much needed feature.

Nope, sorry - the actual reality is that your hotend is the problem.

In a properly assembled hotend, there is a straight path from the inlet to the nozzle, there is nowhere for the filament to go other than down and out of the nozzle, or be pulled back out the top

But an improperly assembled nozzle potentially has gaps between the parts, which would allow soft/melted plastic to expand into them, thus causing the problem you are experiencing - the filament getting stuck.

So again.. there is NO need to do what you are trying to do.
The problem is in the hotend itself, and the RIGHT way to fix it, is either disassemble and reassemble correctly, or replace it with one that IS assembled correctly.

As for slicer times being incorrect, yeah, that is a common issue since every printer has different print speeds, acceleration speeds, jerk settings, etc.. slicing programs can not typically take every possibility into account when calculating print times - the times are estimates based on ideal conditions for the print speed chosen.

However - with that said - I, personally, use Repetier Host to run my printers, and Slic3r for slicing, and I have successfully adjusted my time estimates to be within a couple of minutes either way on 98% of my prints.

the other 2% are typically within 10-15 minutes of the estimated time (very long prints tend to be further off)

The only time that we remove the filament from the hotend is if we are changing colors or material. Again if your filament is getting stuck or your hotend is not turning off then your machine or endcode has an issue that you need to fix. Your machine uses the same firmware that we all use and none of us have the issue you are having or the need that you seem to have.

Also to avoid confusion, Marlin is not gcode. It is firmware just like Repetier that runs the board in the printer. gcode is what the slicer converts your STL to so that the printer host can properly handle it and send the data to the board in the printer. gcode was and is what is used to run CNC based machines.

In the early 3D printer days the hotend was still referred to as tool head, not hotend. Also because of this many get confuse on Z axis moves as it is in reference to the hotend/tool head, not the platform. So when you move your Z axis you should be visualizing the head moving even though some designs move the bed.

Before the printer was fullt built, there was a glitch that allowed me to move the filament, it does not damage the hot end if a command is issued to move the extruder motor, I know, I have done it and all it does is chew in to the filament and clog the drive so that the teeth have no grip. If the command was issued to extrude several feet and no protection was in place, then yes, I can imagine that the stepper wouldn't be too happy and it is possible that the driver may burn out. NOTHING serious can happen in that regard and I wonder if this is yet another urban myth like I see on the sites I moderate for programmers, favourite of the moment is SEO and boy, have people produces some myths about what you have to do to get your site ranked number one in Google...

So you have to excuse me if I am a bit sceptical and keep poking around for information, its hard enough when you have several sites on the same subject all contradicting each other. Like a puzzle it will be solved.

as for contempt claim... Sorry, not me, it feels like there is reluctance to part with information to assis in this, I will look in to flashing the board with my mods in as I have to add mesh levelling to the printer.

Then I plan to take the printer apart and put the good bits in to a frame I designed and eventually I plan on installing my own interpreter, been fleshing one out as an idea the past couple of days. A more logical approach I believe, looking at Marlin GCode it can be improved on.

Oh and I found a standard for AMF additive processes AKA FFF as in Fused Filament Fabrication, so I wonder who coined FDM...  Anyway, there are standards and I will find them.

SO...

A simple retract feature to stop the hot end plugging up so I can remove the filament when I do discover that the printer has finished, it is hard to tell how long a print takes because NONE and I really do mean that NONE of the slicers used to date has given a real or close time to finish, I started a print one night at 8pm and it didn't finish until 4am when the slicer reported it to be a 3hr 57min print.

So the reality is this, the filament NEEDS to be retracted WHEN the temerature of the hot end has been cooled enough to stop the filament deforming or getting stuck when its pulled out the hot end, thats the problem here, the last several manual retactions has ended up with the filament getting jammed in the motor despite releasing it properly.

So this is a very much needed feature.

What I want to do is to have the hotend cool enough so that the filament can be wound back a few mm to be able to properly extract it without having problems that I have found, the filament when too hot is deforming on extraction so that it gets stuck. I have identified a problem and that is that the system is not well thought out in that regards, with ANY system the person running the machine HAS TO BE IN CONTROL that is ESSENTIAL to safe operation and you can't destroy a hot end, you can only damage the filament and clog the feed teeth up which means you have to then clean the part. You can not force cold filament that is 1.75mm through a hotend thats cold and a 0.4mm diameter nozzle.

Nannying people is NEVER a good idea, it breeds contempt. However, adding a 0.001p set of resistors to add over-current protection, it is something that could easily be implemented by these vendors and do it for free because these resistors that are used are not costly and compared to burning out the coils of a stepper, these vendors IMHO are nothing but greedy, they could quite easily add these parts at no extra cost.

So really this whole firmware issue is one of poor programming coupled with unwanted nannying.

There is absolutely ZERO reason to perform the "retraction" you are suggesting. Either remove the filament completely while the hotend is still at temperature, or leave it completely in. There is NO reason to pull it halfway out and leave it there, which is what you are suggesting doing. There is nothing wrong with having safeguards in place in the firmware. Those safeguards have saved people from catastrophe in the past.

And, have you even considered the possibility that there is an issue with your hotend assembly.. it should not be getting "stuck" as you call it, when pulling it out at temp. If it does, this suggests that there is an issue internally with the hotend itself.
But then again, this does not surprise me in the least with a POS GEEETECH machine - and yes, they are garbage - cheap, Chinese, clone, garbage. That particular "brand" is one of the worst. So - you get what you pay for.  Buy cheap - get cheap.

So, may I politely suggest that you take your contempt and move on?

IronMan wrote:

mark.giblin wrote:

I have found that retraction of the filament when at full temperature, the filament gets stuck easily because its soft. This blocks the extrude, by adding a 5mm retraction when the filament has had time to cool a bit would aid retraction when cold and not leave filament cooking in the hotend.

I am really surprised by this and that people are putting up with this being imposed on them.

I don't know anything about your hotend, but with my E3D's I simply release the tension lever on my extruder and pull out the filament manually either to change colors or just reload a spool. I have done it at full heat every time and have no issues pulling out or reinserting...and to Heartless' point above, I never had the need to pull the filament every time a print is finished; just let it all cool down and next time it fires up and extrudes just fine.

But keep trying different things until you are happy with it...that is part of the fun...AND BE SAFE!

What I want to do is to have the hotend cool enough so that the filament can be wound back a few mm to be able to properly extract it without having problems that I have found, the filament when too hot is deforming on extraction so that it gets stuck. I have identified a problem and that is that the system is not well thought out in that regards, with ANY system the person running the machine HAS TO BE IN CONTROL that is ESSENTIAL to safe operation and you can't destroy a hot end, you can only damage the filament and clog the feed teeth up which means you have to then clean the part. You can not force cold filament that is 1.75mm through a hotend thats cold and a 0.4mm diameter nozzle.

Nannying people is NEVER a good idea, it breeds contempt. However, adding a 0.001p set of resistors to add over-current protection, it is something that could easily be implemented by these vendors and do it for free because these resistors that are used are not costly and compared to burning out the coils of a stepper, these vendors IMHO are nothing but greedy, they could quite easily add these parts at no extra cost.

So really this whole firmware issue is one of poor programming coupled with unwanted nannying.

Sounds to me your "PROBLEM" is self imposed. For one you are allowing your printer to run while you sleep or are away. That is a big nono. Your printer should never run unattended. Another issue is that you are wanting to change a standard. Many of us in fact most of us have all had to tear down a hotend to soak or burn out over heated filament many time till we learned how to prevent it or find the cause of it happening. This IS a standard practice. Again you seem to think this is a consumer market and as such you are imposing consumer level rules and expectations.

This is a HOBBY market. We make our printers the way we want. That being said, it is YOUR printer so edit the firmware, mod it, to do what YOU want and become a potential safety hazard as well. But don't expect others to do it for you or to take on your modifications that suit YOU. What YOU want may not fit or be right for OTHERS.

What you want violates a safety feature that was built into the firmware to safeguard those who might forget to heat the hotend and then try to advance or retract the filament.

The nannying as you call it is only unwanted by you. It has been this way since the firmware was ported to 3d printers from CNC and no one till you have voiced an issue. If they had an issue then they quietly modded their firmware to perform the way they wanted and life went one. Most of us only print when we can keep an eye on the printer if by some chance we do have to print a long print that will require running while sleeping or being away then we build or buy a fire proof box to house the printer in the event that things go wrong. We do not leave our machines running unwatched in a living area unless we have taken safeguards in the event of a failure that could happen at some point.

If your logic is crashing or your endcode is set up wrong and the hotend is not turning off then rather than band-aid the problem by modding firmware, fix the actual problem. Find out why your logic is crashing. Could be a bad power supply or faulty logic board. That is not normal. Find out why your hotend is not shutting off when the print ends, that is not normal.

Maybe be creative and add a motion sensor to the system that will shut power to the hotend off if it sees no motion within the print area say for 30 minutes. The function you want to add is not there for a reason because no one till YOU has found issue with it. Again instead of complaining about a function you want, just create it and move on. If tearing down a hotend in a HOBBY machine is too much trouble for you then make the changes for YOU that suit YOU or consider another HOBBY. This is NOT A CONSUMER market, these are NOT CONSUMER MACHINES, and YOU are expected to make the changes needed to make the machine perform as YOU want it.

mark.giblin wrote:

I have found that retraction of the filament when at full temperature, the filament gets stuck easily because its soft. This blocks the extrude, by adding a 5mm retraction when the filament has had time to cool a bit would aid retraction when cold and not leave filament cooking in the hotend.

I am really surprised by this and that people are putting up with this being imposed on them.

I don't know anything about your hotend, but with my E3D's I simply release the tension lever on my extruder and pull out the filament manually either to change colors or just reload a spool. I have done it at full heat every time and have no issues pulling out or reinserting...and to Heartless' point above, I never had the need to pull the filament every time a print is finished; just let it all cool down and next time it fires up and extrudes just fine.

But keep trying different things until you are happy with it...that is part of the fun...AND BE SAFE!

What I want to do is to have the hotend cool enough so that the filament can be wound back a few mm to be able to properly extract it without having problems that I have found, the filament when too hot is deforming on extraction so that it gets stuck. I have identified a problem and that is that the system is not well thought out in that regards, with ANY system the person running the machine HAS TO BE IN CONTROL that is ESSENTIAL to safe operation and you can't destroy a hot end, you can only damage the filament and clog the feed teeth up which means you have to then clean the part. You can not force cold filament that is 1.75mm through a hotend thats cold and a 0.4mm diameter nozzle.

Nannying people is NEVER a good idea, it breeds contempt. However, adding a 0.001p set of resistors to add over-current protection, it is something that could easily be implemented by these vendors and do it for free because these resistors that are used are not costly and compared to burning out the coils of a stepper, these vendors IMHO are nothing but greedy, they could quite easily add these parts at no extra cost.

So really this whole firmware issue is one of poor programming coupled with unwanted nannying.

I have found that retraction of the filament when at full temperature, the filament gets stuck easily because its soft. This blocks the extrude, by adding a 5mm retraction when the filament has had time to cool a bit would aid retraction when cold and not leave filament cooking in the hotend.

I am really surprised by this and that people are putting up with this being imposed on them.

I don't know anything about your hotend, but with my E3D's I simply release the tension lever on my extruder and pull out the filament manually either to change colors or just reload a spool. I have done it at full heat every time and have no issues pulling out or reinserting...and to Heartless' point above, I never had the need to pull the filament every time a print is finished; just let it all cool down and next time it fires up and extrudes just fine.

But keep trying different things until you are happy with it...that is part of the fun...AND BE SAFE!

But by all means, go ahead and screw around with things and destroy your extruder motor, your hotend, and who knows what else.. you seem to think you know more than everyone that came before you, so go for it.

I just hope you don't burn your house/apartment down in the process...

*another OMG WTF has this existed for so long like this without being sorted out*

or

*a case of microsof-itus* which is where bugs exist for years but never get fixed until SHTF.

Really, Firmware over rides a direct command... How bloody stupid is that, the point of manual control is to ensure the operator is in control.

This is what I call "poor programming" and is a really OCD moment pet hate for me, the standard of programming has gone out the window, its why we don't have operating systems with software, we have a bloatware system running bloatware software and programs are Gigbytes when they really only need to be Megabytes.

If you have been a mod on a developer website for as long as I have, you get to see this slide in to bloatland with each iteration of programmer.

There is a learning cure, like with anything, which is why I am asking questions that to some may be obvious, I don't know, I have looked at that much I got in some cases conflicting information.

SO when I draw on my existing knowledge of making, it helped me get the printer built and with some time spent when I had the odd hour at the weekend, was able to in a few hours get the printer printing, well at a stage I was confident I was going to be able to print.

Now there are things I as the operator MUST be able to do, I have found that retraction of the filament when at full temperature, the filament gets stuck easily because its soft. This blocks the extrude, by adding a 5mm retraction when the filament has had time to cool a bit would aid retraction when cold and not leave filament cooking in the hotend.

I am really surprised by this and that people are putting up with this being imposed on them.

You don't go out and buy a Ferrari and have it speed restricted to 55MPH do you? This is pretty much the same as nobbled software 
where the user can't affect a change or setting that is needed.

I don't suppose theres ANY G CODE for this setting change..?

If you mean to change the cold extrusion prevention setting then that requires a firmware edit in Arduino then a recompile and reflash. The reason for the limit is for safety that you don't damage your extruder while trying to push or PULL cold filament.

As for your peev with programming, gcode is as old as the first CNC based machines as that is what it was written for. It was actually ported to 3d printers since they are essentially a CNC machine as well. So gcode programming is older than Windows.

You should have a copy of your firmware saved. If you don't then that is a bad on you as there is a chance it can get corrupted and you need a safe copy. IF you have then copy then all you need to do is open it in Arduino then open the configuration. h tab.

Once you get that tab open find the line that looks like this.

//this prevents dangerous Extruder moves, i.e. if the temperature is under the limit
//can be software-disabled for whatever purposes by
#define PREVENT_DANGEROUS_EXTRUDE
//if PREVENT_DANGEROUS_EXTRUDE is on, you can still disable (uncomment) very long bits of extrusion separately.
#define PREVENT_LENGTHY_EXTRUDE

#define EXTRUDE_MINTEMP 170
#define EXTRUDE_MAXLENGTH (X_MAX_LENGTH+Y_MAX_LENGTH) //prevent extrusion of very large distances.

The line a made bold is the one you will need to change to temp lower than what you will need. Or you can just comment it out by placing // in front of it instead of the #. Once you edit it, save it as a different copy from your safe backup. Then compile it and upload it to the printer.

Again I think you have this entire 3D printer market confused. 3D printing is a hobby and as such we are used to editing code and making changes to the machines to do what we want then to do. Heck some of us even write custom codes to make it do things others may not need.

The point is, it is not a consumer market like some advertise it as. It is not plug and play and it does require some work/skill/know how on the owner/operators part. You will need to change firmware, you will need to make mods to your machine, you will always be making updates and mods to make your machine better and to perform as you want. That is the fun part and what makes it a hobby. It will consume time and it take away part of your life. If you are not prepared for this then maybe look at another hobby.

or

*a case of microsof-itus* which is where bugs exist for years but never get fixed until SHTF.

Really, Firmware over rides a direct command... How bloody stupid is that, the point of manual control is to ensure the operator is in control.

This is what I call "poor programming" and is a really OCD moment pet hate for me, the standard of programming has gone out the window, its why we don't have operating systems with software, we have a bloatware system running bloatware software and programs are Gigbytes when they really only need to be Megabytes.

If you have been a mod on a developer website for as long as I have, you get to see this slide in to bloatland with each iteration of programmer.

There is a learning cure, like with anything, which is why I am asking questions that to some may be obvious, I don't know, I have looked at that much I got in some cases conflicting information.

SO when I draw on my existing knowledge of making, it helped me get the printer built and with some time spent when I had the odd hour at the weekend, was able to in a few hours get the printer printing, well at a stage I was confident I was going to be able to print.

Now there are things I as the operator MUST be able to do, I have found that retraction of the filament when at full temperature, the filament gets stuck easily because its soft. This blocks the extrude, by adding a 5mm retraction when the filament has had time to cool a bit would aid retraction when cold and not leave filament cooking in the hotend.

I am really surprised by this and that people are putting up with this being imposed on them.

You don't go out and buy a Ferrari and have it speed restricted to 55MPH do you? This is pretty much the same as nobbled software 
where the user can't affect a change or setting that is needed.

I don't suppose theres ANY G CODE for this setting change..?

Ok, lets bash this out in stages, see if it can be ironed out a bit...

G28 ; home all axis

Park the extruder

M140 S0 ; turn off hotbed

Turn off the hotbed, its not needed anymore

M106 S255 ; set fan to full

Switch fan on to full as we are cooling down to the temperature being set from 250degC

M109 S160 R170 ; set hotend temperature 160 - 170.
M116 ; wait for temperature

Set and wait for the cooler temperature

G1 E-5 F300 ; retract 5mm of filament
M107 ; turn fan off

retract filament 5mm and turn off the fan. Fan indicate job done

M84 ; after that turns motors off

to me, that makes a logical order to do things with, remember its END CODE here and cooling down to an optimal temperature (which no one has yet said what the temperature is and I can't find anything other than people quoting the ideals for extrusion)

Ideas, I am open to them...
M300 doesn't make a beep, so I haven't bothered and not that I want anything beeping at me in anycase...

The reason you find no temperature specced is that there is none. It varies by as much as 50 degrees +/- per roll/machine and it also depends on how much heat is in the core of your hotend versus the read temperature. Notice in hotend design the thermistor is not reading the exact core temp but a temp a few mm from the core and even a few more mm from the heater. That could be as much as 20 degrees different. This is why when people attempt the Atomic or cold pull it takes multiple attempts till they get it at the right temp.

In the world of 3d printing there are no exact or set temps as each machine varies due to their PID settings, hotend configuration, accuracy of the thermistor and it's location, quality of filament used, and many other variables that cannot be controlled. Even the temp that the printer is setting in will effect the actual temp of the hotend.