Improved bite depth and reliability by considering idler arm flexibility.

[mggevaer]

The geometry of Protoxtruder is spot on, where if the idler sits flush against the extruder body, the bite depth should be sufficient for any use case.
However, due to the nature of plastic parts, the idler arm itself flexes a bit when filament is inserted.
This means that even when filament is inserted, the top of the idler arm can be pushed flush against the extruder body at the top of the extruder, while the bite depth is still insufficient. The idler simply "flexes" out of the way.
Therefore the bite depth ends up being regulated by the force required to deform the idler arm, not the force created by the spring.
This means the protoxtruder is essentially running a fixed idler, the spring never compresses.
The immediate downside of this is that this limits the max amount of bite force.

The arrows in the image below show the forces at work.

solution 1

The first solution I came up with, is to modify the extruder so that idler arm does not touch the extruder when filament is inserted and the screw is tensioned to your liking. This will also ensure, that over time, if the idler arm starts to bend a bit, there's still enough travel on the arm to ensure proper bite force.

I have already partially achieved this by removing some material from the idler arm as shown in the image below. Allowing about 1mm more travel on the top of the arm. The idler arm still sits flush with the extruder at the top, but now grip is much better.
While before occasional filament slippage would occur, now the stepper motor skips instead, showing that the grip on the filament is no longer the limiting factor.

Downsides

This does make the extruder a bit less fool proof. It's now easier for the idler to be over-tensioned and for bite depth to be too deep.

Solution 2

Simply getting rid of the flex in the idler arm. Beefing up the idler arm will reduce flex, while still not allowing bite depths that are too deep. Although there will still always be some flex.
Ideally material could be added to the left of the idler arm, but that's more design work since it requires changing the extruder body.

Downsides

Can't think of any

Conclusion

The least amount of work might be a combination of solution 1 and 2, removing some material on the idler allowing for more travel, and adding some material on the outside to reduce flex.

Let me know if there's anything more I can do!

[nhchiu]

Hello! First of all, thanks for your finding and investigation.

I hope this question doesn't come as offensive. Have you seen this issue happening?
I'm asking because I haven't heard feedback with this issue, not trying to deny it.

Regarding your solution 1, the design already kind of have this solution built in. (Admittedly it might not be enough).
This is what I expect to happen. The gap between the idler and main body is larger than the clearance needed for printed parts. When no filament is inserted, the idler arm will "eat up" the gap at the top. Now the two drive gears are closer to each other and the gap between the grooves should be narrower than the filament diameter.
When the filament is inserted, it pushes the idler arm outwards and now the idler arm should be back to the rest position in CAD, or even slightly more "open".

What I'm trying to show here is that there is some room for the idler arm to bend before it cannot apply biting force on the filament.
But depending on the size and strength of the printed parts, the actual diameter of the filament, and even the softness of the material, the required gap in design can vary, so it might need more tweaking to work in most cases.

The reason why I don't just leave a much larger gap in there is that people often over-tension the idler, and the excessive force can cause other issues. For example, squeezing the filament too hard and deform the filament (especially when printing TPU), twisting the dual drive gears and causing bad gear contact, or causing permanent bending on the idler arm.

I can try to increase the space that the idler arm could bend and also try to beef up the printed parts if you are having this issue.
Thank you!

[mggevaer]

Of course your question isn't offensive! I get you need to check against common user error and such.

Yes, I was seeing this issue. I was getting insufficient filament bite, what I was seeing is that the filament slipped, instead of the stepper motor skipping in scenario's where flow rate limits where being hit.
This same stepper motors does skip when combined with RIDGA bmg gears instead of the gears slipping. This indicates that filament grip was lower than achieved with BMG gears (sherpa mini), despite the drive wheels being larger.
Running at these very high flow rate speeds doesn't make all too much sense, since filament ends up being not molten, but it becomes relevant again when dealing with large spools (5kg) or when that little bit more force can push through a filament clog on it's own and prevent a failed print.

The current design does prevent overtensioning, but unfortunately does not account for the flex in the extruder arm.

The no-downsides improvement would indeed be to increase the idler arm stiffness. It should be possible, I do suspect doubling the thickness of the idler arm would not be overkill. Functionally not much of an issue I don't think, it might look a bit funny though.
This could solve the idler arm flex issue, without allowing the possibility of over-tensioning.
That being said, there's very little room for flex, or idler deformation over time in the current design, even if the arm flexes only a little bit, bit force will quickly be limited.
Considering the drive wheels are very large, they are able to push back a lot on the drive wheels, so a lot of force is needed to get bite depth.

Alternatively, one could also only make room where I marked green on the image below. That way the bite depth is still limited, but the idler arm flex is accounted for.

[nhchiu]

Of course your question isn't offensive! I get you need to check against common user error and such.

Yes, I was seeing this issue. I was getting insufficient filament bite, what I was seeing is that the filament slipped, instead of the stepper motor skipping in scenario's where flow rate limits where being hit. This same stepper motors does skip when combined with RIDGA bmg gears instead of the gears slipping. This indicates that filament grip was lower than achieved with BMG gears (sherpa mini), despite the drive wheels being larger. Running at these very high flow rate speeds doesn't make all too much sense, since filament ends up being not molten, but it becomes relevant again when dealing with large spools (5kg) or when that little bit more force can push through a filament clog on it's own and prevent a failed print.

The current design does prevent overtensioning, but unfortunately does not account for the flex in the extruder arm.

The no-downsides improvement would indeed be to increase the idler arm stiffness. It should be possible, I do suspect doubling the thickness of the idler arm would not be overkill. Functionally not much of an issue I don't think, it might look a bit funny though. This could solve the idler arm flex issue, without allowing the possibility of over-tensioning. That being said, there's very little room for flex, or idler deformation over time in the current design, even if the arm flexes only a little bit, bit force will quickly be limited. Considering the drive wheels are very large, they are able to push back a lot on the drive wheels, so a lot of force is needed to get bite depth.

Alternatively, one could also only make room where I marked green on the image below. That way the bite depth is still limited, but the idler arm flex is accounted for.

Sorry for the late reply.

I guess it's caused by the update made in Aug.

I'll make a quick hotfix before I can make a overall fix for the flex in the idler arm.

Thank you for the feedback and suggestion.

[mggevaer]

I did further notice the issue doesn't occur with PETG (softer) but it does occur with ABS/ASA (harder) which I suspect need more force for good bite depth. This slightly further confirmes the idler arm flex theory... I think.... your idea of beefing up the idler arm seems the best initial fix to try.

[armandvanderwalt]

Is there any modifications I can do to the parts to increase bite on the extuder? I am having issues with ABS and there is virtually no grip on TPU, I printed it and then decided to have it printed by someone with a fully tuned printer also just to see if it was my tolerences or something, but both have the same issues. On ABS when I extrude I can see it slipping, sometimes I can print a whole bunch with no issues, but haven't been able to complete a print with this extruder yet, always stops extruding, have narrowed it down to slipping, so my printer is out of action while I try solve this, any advise? I will be adding a orbitor as a second toolhead hopefully next week so if there is different parts that I will need to print hopefully the orbitor will treat me well for that.

[nhchiu]

@armandvanderwalt As a temporary fix, you can sand down the faces marked in red here. If you are OK with CAD, you can move the holes for the pins on the idler arm closer to the mating face.

I suspect there might be some difference in the gears from different batches/vendors, from the experience of @mggevaer. I don't have any issues on the several gear kits I've got so it's a bit difficult to track down the cause of the sliding problem on my side.

[armandvanderwalt]

Makes sense, I will try sanding it, just a quick question, how much side to side play should the two feeding gears have? Now that I have it off my toolhead in hand I have a suspicion that they might just be going in and out of alignment with the filament path while extruding

[nhchiu]

@armandvanderwalt It's hard to describe. There is some play when not tensioned, but should be very minimal with the spring on.