Stiffer Bowden tubes

[Ryan Carlyle]

One of the big disadvantages to Bowden drives is the flexibility of the tube, right? It contributes to extruder flow lag, forces larger and faster retractions, and so forth. And for high-force extruders, there's also the tendency to strip the PTC fittings off the end of the tubing. 

I'm thinking we should try to find a type of tube that has similar lubricity to the typical 4x2mm PTFE/PFA but is a lot stiffer/harder. That should decrease tube stretch and let the fittings grip better.

Some candidates that come to mind...

• Fiberglass-filled PTFE/PFA tubing
• Thicker wall PTFE tubing, eg 6x2mm
• High-hardness nylon (maybe something like this http://www.mcmaster.com/#9685t1 or http://www.mcmaster.com/#5173k86)
• Maybe PVDF? (http://www.mcmaster.com/#52395k21)
• Maybe FEP? (http://www.mcmaster.com/#5557k32)
• Maybe MFA? (http://www.mcmaster.com/#2133t9)
• PTFE-filled acetal?
• FEP-lined PE tubing?

[Scott Booker]

To be honest with you, IMHO the biggest problem with a Bowden is the clearance between the filament and the ID, in conjunction with any bend radius in the Bowden tube.  That generates positional errors and backlash.  There is a demonstration that they use when teaching how to run control cables (throttle, shift and steering) in a boat... they put one 360 loop in the cable and have you check the backlash.  Then they put in another 360 loop and you check it again.  Understandably, the slop and backlash gets worse and worse.  And of course, with a moving carriage and a changing bend radius, this becomes a variable.  It's an inherent "flaw" that you're just not going to get around by changing materials... and it's hard to eliminate the gap.

I proposed and designed a preload system for Bowden drives that I believe reduces the positional error... perhaps dramatically.  Since I don't have a Bowden drive printer, I never got around to building one or testing it.  But I still think the concept is viable and somewhat elegant (and passive and cheap!).

[whosawhatsis]

Is the flexibility really that much of a factor? My experience suggests that a more flexible tube (as long as it doesn't stretch), would have less backlash because it wouldn't have to be as long.

Incidentally, have you seen the tubes on the Cube 3? Those have bowden extruders built into the cartridges, with the drive mechanism in the middle of the spool and the nozzle built into the end of the tube. These use a thin PTFE sleeve with a (nylon?) skeleton around it that prevents it from stretching or from flexing in the wrong direction, but makes it much more flexible than the PTFE tubes we use in the desired direction. Seems to work really well, and I've been looking for some way for us to make an equivalent (and even considering designing an extruder to be built from a disassembled cartridge).

--
You received this message because you are subscribed to the Google Groups "3DP Ideas" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.
To post to this group, send email to 3dp-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3dp-ideas/2e567944-e3d4-4c48-ad77-aa86a3e614c6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Brandon Pomeroy]

Cube 3 has the advantage of running their own filament... you can't run normal PLA through it. Their blend of PLA is much less brittle than normal... it behaves very similarly to ABS (You can bend it back on itself without it snapping). We ran standard filament in it, and the filament snapped and actually punctured the wall of the tube.

[Ryan Carlyle]

For a solid wall tube, flexibility and stretchability are basically the same thing. Bending flex occurs via stretch of the outside wall and compression of the inside wall. All flexible solid tubes will stretch by some decent amount.

I see the ~400mm 4x2 PTFE tube on my Delta growing by a few millimeters when I accidentally run the extruder with the nozzle cold. Over 5mm of filament can "inject" without losing steps or stripping the pinch gear, some of which gets taken up by PTC fitting bite, and some in filament compression. Some of that filament travel visibly stretches the tube.

Wrapping a helical woven tension member around the tube is a pretty standard way of reinforcing pneumatic/hydraulic hoses against internal loads. The counter-helical wrap lay adds both axial strength and hoop stress strength, without significantly adding undesired bending stiffness. Problem is, you get some "Chinese finger trap" type effects. Axial tension causes diameter contraction. This can be greatly reduced by using multiple helical wrap layers with different lay pitch, but that's probably getting excessive for this kind of application.

In general though, putting a tensile strength manner around a PTFE liner seems like a good approach.

[whosawhatsis]

Yes, bending a tube is compression of one side and expansion of the other, but if the distance between those sides is smaller, the difference in change of length will be smaller, which is why a tube that is not radially stiff will kink shut. So, a tube with the same stretchiness will be able to bend more if the OD is smaller. The Cube 3's tube uses solid reinforcement along the center line of the tube to achieve this. Maybe we can  get the best of both worlds by using a PTFE liner in some kind of tube that is round on the inside, but wide and flat on the outside? Might even be able to line up multiple tubes in it for a dual extruder.

--

You received this message because you are subscribed to the Google Groups "3DP Ideas" group.

To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.

To post to this group, send email to 3dp-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3dp-ideas/c22e705c-1b72-467d-bc4c-1c0c772c2f0a%40googlegroups.com.

[whosawhatsis]

Oh, and you could even put steel cables through parallel channels to keep it under tension.

[whosawhatsis]

Like this, but with a mix of hollow PTFE-lined tubes and steel cables for tension: http://www.google.com/imgres?imgurl=http://ecele.net/img/2014/01/33ljsjtab4z.jpg&imgrefurl=http://ecele.net/electrical_electronics/2014/07/2590966.html&h=533&w=800&tbnid=3Ml43BAH13BiYM:&zoom=1&docid=tVk0pslIeT0wiM&itg=1&ei=nYBBVfSLBcHCsAXB2YGACA&tbm=isch&ved=0CCwQMygQMBA

I bet you could get a company that makes custom cables to make that, but you'd have to be doing a big production run for it to make sense.

[Ryan Carlyle]

Yeah, something like that would work. Seems like overkill though. I just want a 4x2 tube with higher modulus than PTFE or PFA :-)  I'm not super worried about the increase in bending stiffness -- my direct drive bots use 6x4 PTFE feed tubes, and it really doesn't cause any issues. 

To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+unsubscribe@googlegroups.com.

[Ryan Carlyle]

If the modulus is sufficiently higher, a thinner-wall tube like 3x2 could potentially be used to offset the increases bending stiffness. But I'm not sure how feasible that is from a materials standpoint.

[tray]

Some 2x3 ptfe tubing stuffed into an AN3 brake hose might suit you. That runs 3.3-3.5mm I.D, depending on the brand. Unless you find just the right ptfe tubing, that'd leave a gap. However, spiral wrapping the tubing in some line would not only fix that, it'd reduce the friction of the pull while inserting the tube in the hose. (Still might be a bear to do.)

To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.

[Ryan Carlyle]

I still haven't given up on this. 4x2mm PTFE tubing is just too flimsy for the kinds of extruder force I want to push.

Something like this might work: https://www.mcmaster.com/#5201K71

It's a 3,000 psi rated hydraulic hose with 1/8" ID (3.175mm for 3mm filament). McMaster sells them in 6" increments up to 3 ft.These are build quite strong to handle pressure and shouldn't elongate under the kinds of loads we're dealing with. Nylon core with crimped metal end fittings. It MIGHT work -- I'm a little worried about the crimps having an internal shoulder that makes it hard to load filament. 

[dan...@puptv.com]

Would a double hose would work well enough?  A high pressure outer hose for the high forces, and a inner ptfe hose for the low friction.

Daniel - http://www.TriDPrinting.com/

[Ryan Carlyle]

It would work as long as 1) you can get the liner tube inserted and 2) the fit is tight enough for the load to actually transfer effectively. I'm not sure those two are entirely compatible. Might require a fairly specific ID/OD tolerance to work well.

What I've been considering is using a hydraulic hose (for the high-strength and nice end fittings) to load up an inserted PTFE liner in compression. For example, a 1/8" ID hose with a 2x3mm PTFE tube inside. That way, the PTFE tube has no load placed on it, so creep and tube fitting biting is no longer an issue. 

What you have to watch out for with a loose hose/liner fit is the inner liner helically buckling, which would greatly increasing filament drag. 

[Whosawhatsis]

For getting a PTFE liner into another tube, you might consider PTFE heat shrink. In theory, if you heat it and stretch it, you should be able to make the diameter narrower than its "shrunken" size. Slide this in and heat it again, and it will try to return to its proper shrunken size (which is larger than it's current size) and it will press against the inside of the outer tube.

Also, how much do you need to increase the stiffness? Maybe just go with a stiffer fluoropolymer like FEP?

--
You received this message because you are subscribed to the Google Groups "3DP Ideas" group.
To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.
To post to this group, send email to 3dp-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3dp-ideas/2bc0bd43-36df-47ca-a9ac-9742ccf01dbc%40googlegroups.com.

[Ryan Carlyle]

PFA is probably better than PTFE or FEP. It's as stiff as FEP and a little harder (which should help with pushfit grip). But all the low-friction fluoropolymers are similar enough that there isn't going to be much difference. What would be really nice and easy for comparison testing is something like 6x2mm tubing --  just to get more wall thickness. But that might be stiff enough in bending to start causing issues. 

Putting 4x2 inside 6x4 might work ok. It'll have less bending stiffness than 6x2 (good) but the 6x4 has 67% more cross-section than the 4x2 so it should be stronger than just 4x2. And you'd still be able to use 4x2-oriented designs like the Lite6. The main thing is that I think you might need to PULL the smaller tube through, not push it. Really depends on tolerances. 

On Thursday, April 13, 2017 at 2:39:49 PM UTC-5, Whosa whatsis wrote:

For getting a PTFE liner into another tube, you might consider PTFE heat shrink. In theory, if you heat it and stretch it, you should be able to make the diameter narrower than its "shrunken" size. Slide this in and heat it again, and it will try to return to its proper shrunken size (which is larger than it's current size) and it will press against the inside of the outer tube.

Also, how much do you need to increase the stiffness? Maybe just go with a stiffer fluoropolymer like FEP?

From: Ryan Carlyle <temp...@gmail.com>
Reply: Ryan Carlyle <temp...@gmail.com>
Date: April 13, 2017 at 09:17:26
To: 3DP Ideas <3dp-...@googlegroups.com>
Subject:  Re: [3dp-ideas] Stiffer Bowden tubes

It would work as long as 1) you can get the liner tube inserted and 2) the fit is tight enough for the load to actually transfer effectively. I'm not sure those two are entirely compatible. Might require a fairly specific ID/OD tolerance to work well.

What I've been considering is using a hydraulic hose (for the high-strength and nice end fittings) to load up an inserted PTFE liner in compression. For example, a 1/8" ID hose with a 2x3mm PTFE tube inside. That way, the PTFE tube has no load placed on it, so creep and tube fitting biting is no longer an issue. 

What you have to watch out for with a loose hose/liner fit is the inner liner helically buckling, which would greatly increasing filament drag. 

On Thursday, April 13, 2017 at 10:59:02 AM UTC-5, dan...@puptv.com wrote:

Would a double hose would work well enough?  A high pressure outer hose for the high forces, and a inner ptfe hose for the low friction.

Daniel - http://www.TriDPrinting.com/

--
You received this message because you are subscribed to the Google Groups "3DP Ideas" group.

To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+unsubscribe@googlegroups.com.

[Whosawhatsis]

For the record, I use PFA for all of my bowden tubes. The main benefit seems to be that it doesn't get eaten by the teeth of the pushfit connectors as easily as PTFE. That and it's clear, which is sometimes useful.

To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.

To post to this group, send email to 3dp-...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/3dp-ideas/a2611a93-f917-4bd1-973f-7ac87135e88b%40googlegroups.com.

[JasonB]

2AN automotive brake hoses are 1.98mm ID, PTFE lined and steel braided around the outside.  I wonder if they would work nicely.

To unsubscribe from this group and stop receiving emails from it, send an email to 3dp-ideas+...@googlegroups.com.
To post to this group, send email to 3dp-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/3dp-ideas/2bc0bd43-36df-47ca-a9ac-9742ccf01dbc%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ryan Carlyle]

The hose dimensions sound great, but the crimp ID is probably smaller. That's the big challenge with hydraulic components. Both compression fittings and hose crimps shrink the inside diameter.