Your thoughts concerning the standard «Groove Mount» for hotends

I've been experiencing and testing hot ends since the last few months and the groove mount is making me curious. I don't want to point out any specific hot end, just the groove mount in general.

After testing retraction, I've realized that many hot ends would slightly move axially, giving wrong layer height. If the groove mount isn't perfectly at the same thickness as the groove in the hot end, there will be a gap allowing it to move, no matters how tight I set the hot end in place.

I have an idea that could solve this problem, but changing a standard isn't very easy.
Do you think it could be a good idea to suggest an alternative design for the next hot end design?

I would be very interested to hear what you have in mind

I agree with your diagnosis of the problem with groove mount hot ends. The groove mount plate has to be very tight and its thickness has to be very accurate for it to work effectively. I am also in the process of designing an all-metal hot end and I just ended up using basic groove mounting partly out of convention but also because it is actually a universally applicable mounting system for almost all extruders. I am going to use a 4.5mm groove mount. I designed and laser cut the mounting plate out of 4.5mm acrylic and so that it fits very snugly. Acrylic is very rigid so it is an ideal alternative to having a machined aluminium groove mount. If you have a small laser cutter but you don't have a cnc mill, this also cuts down on the number of outsourced machined parts. I imagine you could easily get away with printing a groove mount using ABS so you could design it to fit your hot end perfectly.

Anyway, it would definitely be a good idea to suggest an alternative mounting scheme!

Eric

I have to agree, the groove mount is a weak spot. I've switched to mounting the hot end via extending the extruder mounting bolts, and even that simple work around is much better.

---

Yvan

Singularity Machine

What is the problem you are trying to fix in the parts you are producing?

The extruder gear is just pushing filament into hot end to pressurize the tube
in the hot end to push out plastic thru the nozzle.

It seems almost impossible to seal the output of the gear movement
to the input of the hotend.

The filament diameter also varies.
It is desirable to have a thermal break between hotend and gear mechanism.

Is the precision you are trying to achieve, worth it?

I was recently introduced to the groove mount, and was not impressed. This first kit I got was PLA, and the heat travels through the bolts and the extruder melts, the hot end falls out.

The second J-head I got does not even fit in the extruder.

Perhaps if makers were careful to meet tolerances, it would work, but of course people don't.

The Prusa nozzle comes with a mounting plate, so it simply bolts together and is quite secure. For j-head types, a two piece metal collar could be used to clamp the head and bolt to the extruder.

Sounds like a little job for the weekend

i was waiting for a thread like this, i noticed a while ago that the tiny amounts of wobble were having a huge affect in the printers ability to stack the layers on top of each other and did this [www.thingiverse.com] it's a clamp system instead of the double screw thing,
i'm going to modify the wades extruder for the same thing, but this is how it really should have been done from the start IMHO

---

-=( blog )=- -=( thingiverse )=- -=( 3Dindustries )=- -=( Aluhotend - mostly metal hotend)=--=( Facebook )=-

I made a clamp like that and it did not work well as it moved slightly when hot because it is only held at one side. It only has to move a minute amount at the top for the nozzle to move a lot.

Acrylic plates are no good because it has poor thickness tolerance as do most sheet materials.

The way I mount them is with three screws and washers, see [github.com] That holds it solidly and is also easy to disassemble in the event of a jam.

---

[www.hydraraptor.blogspot.com]

woodencase01 Wrote:
-------------------------------------------------------
> I have an idea that could solve this problem, but
> changing a standard isn't very easy.
> Do you think it could be a good idea to suggest an
> alternative design for the next hot end design?

I think this is a good opportunity to collect information from use in the field to determine what would be a good step forward.
I have not completed my printer yet so my knowledge is still theoretical but I have read up a lot.

I have a couple of suggestions for a new generation hot-end mounting.

Firstly It would be nice if it was compatible with at least some of the most popular existing systems.

Secondly making it rigid is the most important thing, The vertical play and also the angular play must be minimised as this error is magnified by long hot ends.

I would very much like some thought to be put into having a system that can somehow handle switching from direct drive to Bowden hot ends

Lastly it would make sense if there was some thought put into making change over of heads easy/fast to allow for experimentation/development with new extruders or milling heads though this might be more relevant for the cold end mount unless the hotend mounting was the datum point and the cold end was mounted to that.

The JGR clamping looks positive and could hold a Bowden hot end as well..

What are your plans?

---

Kalle
--
Lahti, Finland
The only stable form of government is Open Source Government. - Kalle Pihlajasaari 2013

**Deleted**

Edited 1 time(s). Last edit at 08/22/2013 07:38PM by woodencase01.

I think you guys are making this more complicated than it needs to be.

Mounting the hotend with transverse bolts is a bad idea because you can't provide any upward clamping force. This causes the hotend wobble some people see.

The original groovemount mounting method used a plate that compressed the hotend into the extruder body. See the attached pic for an amateurish sketch of what it should look like. Notice that the groove in the hotend extends out past the lower edge of the extruder body when it is fully seated (exaggerated in the sketch). This allows the clamping plate (red) to provide upward clamping force.

The hotend should be a light to moderate press fit into the hole in the body. This aligns it with the filament path.

The slot in the clamp plate should, IMO, have a slightly loose fit on the hotend, so that it does not align the hotend, it only provides clamping force.

There's no need to involve bevels to align the assembly.

Hmm, strange how in some cases RepRap technique seems to have regressed. I did a google for "groove mount plate" and found several examples.

This thing is close to what I was thinking, [www.thingiverse.com] but seems to be what crispy1 describes.



This thing is very like what I was thinking, [www.thingiverse.com].

Cripsy, thank you for the clarification.
Your mounting drawing is the same as nophead has previously shown. I think too that it this the way to do it.

I still wonder why it isn't mounted as you suggested as a standard...

nophead Wrote:
-------------------------------------------------------
> I made a clamp like that and it did not work well
> as it moved slightly when hot because it is only
> held at one side. It only has to move a minute
> amount at the top for the nozzle to move a lot.
>
> Acrylic plates are no good because it has poor
> thickness tolerance as do most sheet materials.
>
> The way I mount them is with three screws and
> washers, see
> [github.com]
> bond/manual/extruder_hot_end_assembly.png That
> holds it solidly and is also easy to disassemble
> in the event of a jam.

before i got to the clamp i tried a single side quick release wedge thing, which worked quite well, it started to wobble the tinyiest little bit after a week of printing, the double screw clamp one i put on thingiverse already have about 3 weeks of solid printing on it and it's still rigid ,

i think we're getting to a point now with the kind of printing we are doing , tiny little amounts of movement and flexing anywhere in the printer are going to become much more obvious compared to when we were printing objects with a 0.5mm nozzle and a .35mm layer height, now i do everything with a 0.3mm nozzle and a 0.15mm layer height, and even now 0.15mm is starting to look "low res"

---

-=( blog )=- -=( thingiverse )=- -=( 3Dindustries )=- -=( Aluhotend - mostly metal hotend)=--=( Facebook )=-

Quote

I still wonder why it isn't mounted as you suggested as a standard...

Because someone who didn't fully understand the tolerance stack and theory behind why/how the groove mount plate was used saw an opportunity to replace a custom-machined metal (or wood) plate with 2 M3 bolts.

This simple bowden design for a delta end effector is working out pretty well for me. It's laser cut acrylic all bolted/clamped together with spacers to the end effector. The acrylic is beveled a bit with a file and when clamped together it is not going anywhere.

The simplest idea i came up with was to thread the external groove mount area on the hot end and then just screw on a thin panel nut. This would clamp the hot end to the plate. Quick and easy.

This is how I overcame the problem that you talk about.
The photos will show how it is done, but basically it is a mounting plate that also acts as a heatsink as well.
A M6 thread is used to mount the hotend to the mounting plate, the nut goes inside the recess that the groove mount usually goes in. Nothing to change on the extruder and the mounting plate can be made to suit any type of machine quiet easily.

Photo1
Photo2
Photo3
And here is a dual head Bowden version using the same method
Photo4