Gear Extruder - Anyone interested?

[mendells 52]

I'm trying to gauge interest in who would be interested in buying a gear extruder? I've been using the design for a little over a week now and it works like a champ. I am still refining the design but will be making another 2 sets at least for our other Ultimakers. If there is enough interest I could see about making enough to sell to those who are interested.

 

 

Features:

 

• More torque
  
• No grinding/jamming!
  
• Better accuracy (more steps/mm of filament)
  
• Easy pressure settings (no more worrying about too much, or too little pressure)
  

 

What would be included:

 

• Drive gears (For standard or flexible filaments)
  
• Bearings
  
• Hardened Shafts
  
• Screws
  
• NEMA 17 5:1 Geared Stepper Motor
  
• Extension Cable
  

 

What you would print:

 

• Gearbox
  
• Shaft coupling
  
• Motor Adapter
  
• Motor Mount
  

 

I don't have a price yet but my rough estimate is something in the $100 range with the motor, cable and hardware included. If you are interested please let me know. I will look into making machined aluminum components to replace the printed ones.

Edited October 1, 2015 by Guest

[Labern 774]

Does the stepper motor have to be mounted externally?

Is there a program change to use the geared stepper motor?

[mendells 52]

Unfortunately the geared stepper motor is too long to mount internally. You could potentially mount it internally if you removed the sheet metal shield that covers the stepper motors but I wouldn't recommend that. I am not 100% happy with the mounting location or method yet so it may still change in the future. I would like to mount it closer to the top of the machine and offset far enough so it doesn't stick out the side.

As for the program changes, the only thing that has to be changed is "E-Steps per 1mm filament" in the machine setting. The default is 0 (which I believe the default is something like 86) and the new value is 443.46.

[gr5 2,069]

Nice! I'd love to know how much weight it can lift. A UMO or UM2 can lift about 5kg or about 10 pounds:

https://ultimaker.com/de/community/view/5222-pulling-force-of-um-extruder?page=1

[mendells 52]

Nice!  I'd love to know how much weight it can lift.  A UMO or UM2 can lift about 5kg or about 10 pounds:

https://ultimaker.com/de/community/view/5222-pulling-force-of-um-extruder?page=1

 

I could rig something up to find out, I can guarantee its significantly more than the stock UMO or UM2 feeders. The stepper motor is larger and has a 5:1 planetary gearbox so it should have roughly 5x the pulling force.

Edited October 2, 2015 by Guest

[MrByte 4]

I'm certainly interested in this.

[MrByte 4]

Although, any particular reason for using a different stepper motor than the stock motor?

[mendells 52]

Higher torque and better resolution. The gears have a larger diameter than the stock knurled feeder, which negatively affects resolution (lowers steps per mm) so to compensate and add more torque (something the original drive motor is a bit lacking) I'm using a geared motor..... and I had one laying around

Having a more powerful extruder means you can print at lower temperatures and faster speeds.

Edited October 2, 2015 by Guest

[MrByte 4]

Thanks, that really helps explain things. I'm definitely interested in this though (I've been following your progress for a while now) and would love to give it a try. I really want to be able to effectively print XT CF-20.

[neotko 1,417]

Anyhow no matter with how much force you can push something it won't change the fact that the filament needs x time to reach the print temperature. There's physically a limit of how much filament can reach the temp per sec. The only way to reach much more speed/sec its to have a larger hotzone. Ofc more force will help but also there's a limit of preassure and heat/time.

[eldrick 64]

There is also "lash" to consider, i.e. imprecision introduced by having a couple of levels of gearing involved.

Edited October 3, 2015 by Guest

[mendells 52]

Anyhow no matter with how much force you can push something it won't change the fact that the filament needs x time to reach the print temperature. There's physically a limit of how much filament can reach the temp per sec. The only way to reach much more speed/sec its to have a larger hotzone. Ofc more force will help but also there's a limit of preassure and heat/time.

 

This is true but the stock extruder really lacks the torque required to extrude even at normal speeds. With our stock Ultimakers we need to run ColorFabb XT at 240C to print at 40mm/s without underextrusion. With the gear extruder we can at 60-70mm/s.

 

There is also "lash" to consider, i.e. imprecision introduced by having a couple of levels of gearing involved.

 

The backlash in the geared extruder seems to have little if any effect. I did have to increase my retraction a bit but I believe that is just due to the increased pressure on the filament.

I am running a series of test prints to effectively measure the impact of the gear extruder vs the stock one. I will post it when I have it done.

Edited October 4, 2015 by Guest

[mendells 52]

Did a quick speed test. The tower on the left uses the gear extruder the one on the right uses the stock extruder. The gear extruder looks good up to 120mm/s where the stock extruder starts showing signs of underextrusion around 70mm/s. Both printers were printing clear XT at 240C.

[gr5 2,069]

 

Nice!  I'd love to know how much weight it can lift.  A UMO or UM2 can lift about 5kg or about 10 pounds:

https://ultimaker.com/de/community/view/5222-pulling-force-of-um-extruder?page=1

 

I could rig something up to find out, I can guarantee its significantly more than the stock UMO or UM2 feeders. The stepper motor is larger and has a 5:1 planetary gearbox so it should have roughly 5x the pulling force.

 

I'd really like to see the test - it's very easy if you have a scale and a bucket and something heavy to put in the bucket and a bungee. Oh and I guess you need pronterface so you can move the Z slowy and watch the scale carefully.

The feeder may be 5X but the grip strength is probably only 2X (two sides instead of one). I agree with your comment about retraction by the way.

I've heard that splined extruders (versus diamond knurled pattern) have the problem that they can get PLA dust in them that doesn't fall out as easily as the knurled pattern.

[gr5 2,069]

Pronterface is a fantastic tool if you don't have it - it lets you measure all the sensors like end stops easily and change acceleration parameters, max speed parameters and so much more but most importantly you can move Z slowly by any amount. Free download here:

http://koti.kapsi.fi/~kliment/printrun/

[mendells 52]

Pronterface is a fantastic tool if you don't have it - it lets you measure all the sensors like end stops easily and change acceleration parameters, max speed parameters and so much more but most importantly you can move Z slowly by any amount.  Free download here:

http://koti.kapsi.fi/~kliment/printrun/

 

Thanks for the link, I'll see what I can do about getting that test performed.

From some of the testing I have already performed the geared stepper motor will skip steps before the gears loose grip on the filament, so from that I believe I can safely say the motor will be the limiting factor not the grip. I haven't had any slipping/grinding so far so there isn't any dust in the gearbox so far.

[memory556 18]

Very interesting! Looks awesome!

[mendells 52]

Anyone know how to change the maximum retraction speed? I've had to increase the retraction distance to 10-12mm and I want to up the retraction speed to something like 55-60mm/s. I'm guessing its something I would have to change in the firmware but I'm not sure how to change it? Any help would be greatly appreciated.

[peetersm 124]

You can use pronterface to see the firmware current max retractions acceleration with the M501 command. Then you can issue M203 amd M201 commands to change it.

I tried asking about the default setting bing too low here

https://ultimaker.com/en/community/view/16201-is-the-umo-firmware-1502-maximum-feedrates-for-e-too-low

But never got any replies, but that post shows what you should see.

Don't forget to save your firmware settings after you change them you have to use a M500, otherwise reboot the printer and your adjustments will go away, sometimes this is what you want.

Edited October 13, 2015 by Guest

[mendells 52]

Updated version of the gear extruder:

 

• Moved mounting location.
  
• Updated tension screw with thumb screw (no tools needed).
  
• Re-Printed housing with XT-CF20
  
• Added SemiFlex cushion to dampen motor vibrations.
  

 

I have also updated my fan shroud design to accomodate the stock Ultimaker nozzles, improve airflow and include a template for the Kapton tape cutouts.

https://www.youmagine.com/designs/um2-olsson-block-fan-mount

Edited November 4, 2015 by Guest

[mendells 52]

Printed a few doorstops with the gear extruder and Olsson block. The 0.8mm nozzle and 0.4mm thick layers makes a sizable dent in the print times. The print time went from over 4 hours to just over 1 hour. The gear extruder could extrude faster but the heater isn't strong enough to keep the nozzle hot.

.4mm nozzle,

ColorFabb XT

0.12mm layers

70mm/s

Print time: 4h 34m

.8mm nozzle,

ColorFabb XT

0.4mm layers

50mm/s

Print time: 1h 4m

Nonlethal Door Stop

Edited November 4, 2015 by Guest

[trevorjtclarke 3]

@mendells I have also been following your progress for a while. Great work!

Count me in!! Name the price, and tell me when it will arrive

Whats your ideal launch date?

[skinny-kid 64]

Count me in as another person interested in your design. Excellent work you have done thus far!

[mendells 52]

There is a lot of work to do before production so I don't know for sure when that may be.

A question for those who may be interested. Would you rather?

A: Receive all hardware and print the gearbox, motor mount and motor adapter yourself ($100-150)?

B: Receive all hardware and 3D printed gearbox, motor mount and motor adapter ($130-180)

C: Receive all hardware and machined aluminum gearbox, motor mount and motor adapter ($200-250)

Prices are rough estimates, just trying to get an idea what people would want.

Edited November 5, 2015 by Guest

[drofnas 15]

Option A.

Since the original feeder would be un-harmed in this setup, I could always print myself replacement parts if something was to happen with the previously printed parts.