If you have not got more PTFE tubing to replace the length in the flexistruder when it wears out, I would recommend it soon.
Why? Does it wear out THAT fast?
Already looked but the comments give me pause as to where to get it:
Hardware HD-TB0006 .25”ODx.125”ID PTFE tube (~4”) - 50’ roll Zoro Tools G3251001 0.25 ft $0.70 $0.70 only comes in 50’ rolls ($105) This Zoro tools PTFE should be much more consistent – need to test before using in production – only comes in 50’ rolls (~150flexy’s)
Hardware HD-TB0006 .25”ODx.125”ID PTFE tube (~2.5”) McMaster-Carr Supply Company 8547K23 1 This mcmaster PTFE needs to be replaced due to inconsistent quality and diameter
First, I want decent quality. Second I don’t want to buy 50’.
Anyone know of a vendor of reasonable lengths?
Well I was told by a person that uses a lot of nijaflex that the less flexible filaments will damage it pretty fast. And the checks I did last year, it was crushing/deforming the walls of the tubing where the screw is so that it had feed/buckling issues. I have not since ran ANYTHING but flexible filament through the flexistruder I built for testing.
McMaster Carr has it in 1ft lengths for a reasonable price. Shipping is usually about $8.
This is great work everyone, I look forward to following along, and I if I get excited about going to a dualy flexi, i will find it even more useful!
I haven’t even tightened that screw 
Its very loose as I hadn’t needed to use it for rigid filament.
I am not too thrilled iwht the flexi design. Access to the hobbed bolt is, well, non-existant. I have been penciling down sketches for a converted wades that would incorporate the tube but still give access to the hobbed bolt for cleaning.
What gave me pause was this quirp at the bottom of that page:
OD is oversized and ID is undersized for finishing to the sizes listed. These tubes are FDA compliant and meet UL 94V0 for flame retardance and ASTM D1710.
So, is it really .125 or is it only .120. Or .110? What is it I am actually getting?
I also noted that its max temp is 500f, only 260c. I know it doesn’t touch the hot end, but is this a factor here?
In the end it doesn’t really matter. The printed part is drilled out (6.5mm) to accept the tube. I’ve now printed and assembled two of the flexystruder bodies and after drilling out the hole to 6.5mm (the size recommended by Aleph Objects), the tube fit snugly. On one I needed a small allen wrench to push the tube in. On the other I could push it in by hand. The inner diameter needs to be drilled out to 3.5mm.
I agree that there’s probably a modification to the normal extruder that should work with flexible material… Probably just a matter of extending the lower extruder neck, reducing the gap to the hobbed bolt. This should reduce any kinks to the filament and retain use of the idler.
The problem is the PTFE tube does this:
|
( - hole for the bolt
|
when you put the hard filament through. The amount of the tube that is left there after the drilling is
small and it stretches/thins it out as the hard filament is pushed past the hobbed bolt.
Thus the fit is now loose there for the flexible filament. You are not gripping it as well and can get feeding/retraction problems. Ooze!
This makes so much more sense! Thanks for the explaination. I kept wondering why ABS and PLA filament was so harsh on the PTFE tubes as they seem pretty sturdy.
Thanks this is great.
Ironically when I went from the single hex on my Taz 5 to the dual extruder, my PID’s for the extruders are perfect, only the bed PID settings weren’t working. LOL go figure. Strange.
PS I thought the flexistruder was strictly for flex type filament only and not ABS?
To understand why the flexistruder exists you have to consider why the standard extruder doesn’t work. When you feed filament into a “standard” extruder (is there a specific name for this?), the filament is squeezed between the hobbed bolt and a ball bearing. Although there is a hole in the extruder body to guide the filament, there is a small gap on either side of both the bolt and the bearing. With hard filaments like ABS or PLA this is no problem as the filament is rigid enough to span the gaps. With flexible material, the filament can bulge out (it is like pushing on rope) and jam.
The flexystruder solves this problem by keeping the filament in a tube the entire way and only a small notch is removed to allow the hobbed bolt to grip the filament and push it along. There is no gap for the filament to bend. In order to apply enough pressure on the hobbed bolt, a screw presses against the tube on the other side of the bolt.
There is nothing that prevents this from also working with harder filaments. In fact, the two extruders are working almost exactly the same way. What kmanley57 was describing was why the PTFE degrades so quickly when using the harder filament. Because it is so rigid, it doesn’t give when squeezed between the PTFE tube and the hobbed bolt. Since the filament is harder than the tube, it acts like sand paper grinding away at the tube anywhere it is rubbing. Eventually this can degrade the tube enough that gaps can start forming that would allow the soft filament to buckle and jam.
On other direct drive systems, the tension between the idler and hibbed bolt isn’t a problem. Any gap between the hobbed gear and extruder body is the biggest culprit. Printing an adapter for ALU direct drive mechanisms were a good fit.
I think with the printed extruders on these printers, it could be possible to modify the normal extruder design to get a better contour with the hobbed gear. Which should reduce the gap… similar to what the flexystruder accomplishes. Though I’m sure someone smarter than me has already tried, which is why we have the flexystruder design…
So long story short it’s a bad idea to use the flexistruder with anything but soft filament, which is what I’ve read elsewhere. You can run a gas engine with alcohol, but it’s still a bad idea if you want a working engine tomorrow.
My PID values are not sticking. If I restart the printer the bed goes right back to heating up to 2° below the set temp and stops. Yes I send the updated values, and yes I send the M500 command. The extruder values stick, but the bed values do not. If I resend the PID values for the bed, it functions correctly until I power off. Any ideas?
You can either make the changes in the LCD (these changes are stored in the EEPROM, I believe) or update the values in the firmware and reflash it.
I can’t locate the Bed PID’s in the firmware menus. I can only find the extruder values.
Are you saying I should reflash with a revised config.h? Why won’t the M500 get the values to stick?
Sorry, I have no idea. I don’t know much about the Marlin firmware or how/when it writes to the EEPROM. Sounds like your only option is to set the bed PID values in the configuration.h file in the marlin firmware and recompile/upload.