Flexible drive shaft as bowden alternative?

I was just contemplating trying to modify my Printrbot+'s extruder to be fed via bowden tube, and while reading through the pros and cons I wondered if anyone's tried to instead move the extruder motor off the print head and connect it to the gears via a flexible drive shaft (like the kind sold for use with drills and dremel tools). That would leave the weight of the gears and hobbed bolt in place, but might not have as many issues re: ooze, hysteresis, etc.

Would the torsional rigidity be too low, the increase in backlash too much to make this a viable option? Are there other problems or reasons people haven't been doing this? Is it worth a try?

I think it would have a lot more hysteresis rather than less.

---

[www.hydraraptor.blogspot.com]

Well, it might be a farfetched idea. I wonder what's easier -- compensating for hysteresis in filament delivered via bowden tube or compensating for the angular deflection of a flexible shaft? And would an encoder at the extruder end of the shaft be of help?

Reading of bowden tube bots with 20 and 30mm retraction settings to avoid ooze led me to think I should seek other ways to move weight off the print head. I've never looked into flexible driveshafts but figure with the relatively low force that needs be applied to the extruder gears (relative to the torque of drilling or milling) that perhaps angular deflection would be negligibly low.

There are alternative ways to move the motor off the print head. Gary Hodgson was working on one here. It may overcome the hysteresis problems you could get.

Off carriage extruder

There are several designs which look good on paper, and even tested in a non-printing mode, the question for me is how well do they perform when driving an extruder during printing. The force required to push the plastic is not insignificant, I suspect it will cause the designs to not work so well in practice, e.g. due to binding on the bearing.

Extra forces added to the X axis are likely to lead to reduced positional accuracy. Apart from some complex modelling, the only way to find out is to try printing, which obviously takes a bit of work to do.

It appears that someone has implemented an extruder driven by a (900mm!) flexible shaft. I wonder how they dealt with the hysteresis and X-axis forces:

[mutley3d.com]

According to this writeup it works well, or more specifically…
[numbersixreprap.blogspot.com]

Quote

And there was this curious configuration at the excellent Ooznet stand... looks like a Prusa i3, which it basically is, but fitted with an unusual extruder driving mechanism, called a Flex3Drive! It allowed the best of both worlds, the light x-carriage with the motor remotely, but still the benefits of direct drive extrusion (lower retract distances). It worked very well.

The creator of the Flex3Drive has posted about it to this thread:
[forums.reprap.org]

Quote
martinprice2004
There are alternative ways to move the motor off the print head. Gary Hodgson was working on one here. It may overcome the hysteresis problems you could get.

Off carriage extruder

Is that the same guy as The Shrub over in this thread: [forums.reprap.org] ? I was thinking that it was a cool concept, but it hasn't really had much action on it in about three months.