HBot using fishing line

Hello,

For info, I will post many more information on another thread if somebody is interested (x-y axe file are made with solidworks, Z axe with Inventor, and some with freecad... do not asking me why, it's just because it's depend what software is installed on the computer who are in the front of me :-D ... )
It's not at all screw effective (but screw price can't be less cheap for me so I don't care so much). Screw size are M2, M3 and M4 with many differents lenght (and 4xM2.5 for the nozzle support in polyimide )

So the very first test on video:
[www.youtube.com]

And some picture of the current progress of the Hbot with fishing line.





The only speed test I have done, moving until 400mm/min without problems but I have not really tested repetability, precision etc... and I don't have made any good calibration of motor current, tensionning of the fishing line, distance of rod, so all sound good for me to go more fast :-D )

@+
Zelogik

I printed the Make Magazine torture test STL on my h-bot. I am pretty happy with the result. This was printed with support, which broke away very cleanly.

[i.imgur.com]

[EDIT: removed image tags because the formatting is bad.]

Some of the spots/imperfections in the image aren't there in real life, I think the camera flash is picking up the speckles in the silver PLA.

Edited 2 time(s). Last edit at 03/07/2013 11:31AM by crispy1.

That looks really nice.

robogrow Wrote:
-------------------------------------------------------
> I am try to come up with a 100' by 10' to 25'
> gantry system for an open source urban farming
> project.
>
> Anyone know of a drive pulley that will get decent
> friction on a branded wire cable?
>
> I expect a payload from 10 to 50 lbs. Ideally I
> would like to run several cross rails at once to
> carry cooperating robots (ie. one to harvest/plant
> and one to hold and transport).
>
> I thought a cable system might be simplest. I
> tried a double rail along the long axis and it
> wanted to skew and bind. Now I am trying a single
> drive rail down the middle with cable guides on
> each side. By having the robot centered when
> traveling lenghtwise I hope it will work.
>
> I am hoping to put the motors and micro
> controllers at the pay load and not the ends of
> the gantry, so as to reuce the number of
> microcontrolers batteries and wireless modules.
>
> Any design sujestion would be apprieciated. we are
> hoping to use cheap power wheel type DC motor on
> the long run with limit switchwitch that will
> detect the hydroponic gulleys and either a stepper
> motor or a DC moter and some simple image edge
> filtering to detect plant positions in the
> gulleys.
>
> Desinging a cheap viable and durable gantry system
> is harder then I first thougth.

You should email me directly. I spent 18 years in the commercial greenhouse industry designing and building just what you are looking for. I have designs to travel over 200 feet, and even a machine that can travel down 20 different bays each 200 feet long. I have two patents on devices related to this, all have expired so are public domain now.

gary.lucas@verizon.net

hi guys, if u need more friction on fishing line, i can suggest one good tweak

get some wax(i think its pure bee wax) and put it some to string, and some to "pulley", the friction will increase alot


i know about this trick because my mom used to work at shue factory, and they have strong string cooked in wax, so when u hand making shoes, and tighten stitch, the wax string stay tight even when no force is applyed

Just wanted to put a small update in this thread.

I designed a bowden extruder to replace my PG35L-based direct drive extruder. I got it printing successfully today, and I am very pleased with the initial results. Tuning in the bowden setup is ongoing, so I expect print quality to improve further in the future.

Here's a short video of the machine printing the twisted bottle in clear PLA:
[vimeo.com]

And here's a picture of the result:

Just looking at it in the video , did you change to a coreXY configuration over a pure Hbot?
Just wondering.

Quote

Just looking at it in the video , did you change to a coreXY configuration over a pure Hbot?
Just wondering.

Good catch. Yes, I did.

Edited 2 time(s). Last edit at 04/03/2013 10:42AM by crispy1.

I printed the Make torture test and it came out pretty well. There is no stringing between the cylinders, and the start/stop points are clean but not quite perfect. On the box section you can see some excess material buildup on the corner where it started each layer, so I have some more tuning to do there. There is no corner ringing evident, which makes me a happy camper. This was printed at 0.3mm layer height at about 60 mm/s perimeter speed.

[imgur.com]

Any further results/thoughts on the hbot using fishing line?

I have some parts here for a 4th prototype, but I'm working on a couple of other projects at the moment.
I would suggest going straight to CoreXY unless you have someway to make a very stiff XY gantry where the racking forces won't be significant. The larger the XY area gets the harder to achieve this is.
The line is very slick and really doesn't like holding knots and tying a knot in it with tension is close to impossible, however tension is really important, so you need something to apply sufficient tension in the design. Also note that if your design allows the line to walk it changes the overall tension in the system, so you need some mechanism to counteract this.
If the filament can walk in your design then make sure it has enough space to walk because if it crosses itself it will cause issues.
Any out of round parts like pulleys translate very obviously into movement errors, so where possible use something you can guarantee is round as a pulley.

---

___________________________________________________________________________

My blog [3dprinterhell.blogspot.com]

Some work towards prototype number 4.




Just a prototype for the new filament drive, more details on my blog.

---

___________________________________________________________________________

My blog [3dprinterhell.blogspot.com]

I tried something like that, with printed parts. I found the friction to be very high, and not in a good way. Your implementation looks better than mine as far as using proper bearings and machined parts with no runout. I also still had slip somehow, and I had the equivalent of 3 full wraps...

Thee friction seems to be adequate, I'm currently basing that on how much force is needed to make the filament slip. The 60 degree machined groove might also be helping with the friction.
I think you have to minimally support the idler on bearings because of the tension on the line, without them I think it will just bind, the pulleys ought to be printable though, none of the dimensions are particularly critical. I machined them because I have a lathe and it was an easy way to get something I knew would be usable.
I'll probably use a smaller spacing between the grooves on the final part, but the 1.5mm I'm using in the image seems to work OK.

---

___________________________________________________________________________

My blog [3dprinterhell.blogspot.com]

Polygonhell Wrote:
-------------------------------------------------------
> Just a prototype for the new filament drive, more
> details on my blog.

Beautiful. I can't wait to see it in action.

Quote

the pulleys ought to be printable though, none of the dimensions are particularly critical.

Runout (out of roundness) is critical. I do not know of a machine that can print a perfectly circular object with runout < 0.0005" which IMO is what is required for precision linear movement.

You might be right, obviously any runout will result in changing tension as the idler and drive roller rotate.
If I get a chance I'll print some plastic parts this week and test them.

---

___________________________________________________________________________

My blog [3dprinterhell.blogspot.com]

For a while I was using out of round parts and could see a direct impact in the printed objects - ripples in the walls and stuff like that. So for me the issue wasn't so much changes in tension as impacting the positioning accuracy.

@Polygonhell: Is your idler slightly tilted like garyhlucas has described on page 2?
I am experimenting with a very similar setup myself, but ran into trouble:
After about 1/8th of a revolution, the motor axle won't move anymore (I am moving everything by hand right now) and the dyneema line slips. Might have something to do with the fact that I use loose distance washers to prevent the crossing over of the line and that the idler is parallel to the motor axle.
If the line wanders, it might move the washers, essentially increasing friction and clamping itself...



By the way, here is a good formula to calculate max. forces before the dyneema slips:

fm = fp * e^(µH*a)

fm = max. force
fp = preload force
e = Euler's number
µH = static coefficient of friction (dyneema == UHMWPE should be somewhere between 0.15µ and 0.3µ depending on source of information)
a = angle of winding (in radians)

so let's say 2N of preload and 3 windings; worst case scenario of 0.15µ static coefficient of friction:

2N * e^(0.15*6*pi) = 33.8 N

Mines entirely vertical offset by 0.5 of the groove spacing. Though the exact offset doesn't seem to matter as long as it's close, I let the idler "float" vertically but I'm using an M4 screw as the shaft with two 684 bearings in the center and it seems to stay pretty much stay where I put it (On a smooth shaft I'd expect it to self center).

As to force before it slips, anecdotally the version with 4 drive grooves and 3 idler grooves (shown in the images) I could physically make it slip by holding the line, thought there was IMO plenty of force before it did. I machined a version last night with slightly shallower (0.5mm) grooves with closer spacing (1mm) and 6 drive grooves and 5 idler grooves, on that version I could not get a good enough grip with my fingers to make it slip.

I'm cutting the grooves with a 60 degree threading tool and the odds are at the base of the cut it will be narrower than the line itself, effectively the walls then grip the line, I'm using 50lb spectra line.

In the video I think I'm running at 12000 steps/second (200 step steppers, 16x uStepping) with no acceleration with a 14mm drive pulley my math says roughly (12000/3200)*43.9 or about 160mm/s jerk equivalen in marlin or repetier, now obviously there is no load on the mechanism, but that's pretty agressive acceleration with no slipping and I don't see any even when applying significant resistance with my fingers.

I still have to design the carriage and I need to order some raw material before I can compete the XY stage and do some real testing.

---

___________________________________________________________________________

My blog [3dprinterhell.blogspot.com]

@Polygonhell: hm, yeah, offsetting the grooves half a line width sounds like a simple to implement and logical approach.

Crazy how much force you can apply to this simple system without it slipping...

Ok, I have probably found the culprit in my system: Although I drilled the hole in enclosure plate that holds everything together larger than it need to actually be, the set screw of the shaft collar might have snagged on a bit, preventing further motion. After I have filed out the hole, everything is running well. Man, that was a major derp move not to consider possible snagging. lol. Just need to make sure that the edges of the distance washers are rounded, less they be cutting the dyneema.

Has anyone tried maybe using multiple loops of fishing line in lieu of wraps. make Identical lengths of fishing line with loops tied at the ends and possibly some sort of reinforcement ring on the inside of the loop. these could then be tightened to correct tension with zip ties or spring. I have little knowledge in 3d printing and am just now building my first. (a prusa i3/ mendel90 sturdy hybrid fondly called the "Big 12") So my two cents may not count for much in this older discussion between far more experienced builders. I think the multiple loops would increase friction (thereby reducing slippage and if on ribbed pulleys should stay stationary and not walk on each other. Tension should also be consistent on both sides of the pulleys.

And another update


[www.youtube.com]

This is pretty promising, there is an issue with the carriage geometry, the screws to attach the line too are too far apart, for what ever reason I located them such that the line would be correctly spaced when attached to the inner side of the screw, of course the line centered on the screw. The result of this is that as the carriage moves to the extremes and the angle of the line changes you get some unwanted motion orthogonal to the bar in the H, across the length of the bar it's surprisingly large, I measured it with a dial gauge at 0.015in which is about 0.4mm. So I'll have to remake the carriage.

---

___________________________________________________________________________

My blog [3dprinterhell.blogspot.com]

Last week, I took apart an old HP plotter at my hackerspace. It's drive system used two steel cords wrapped around a pulley somewhat similar to what you're using. I can take pictures of it if you're interested.

I'd be interested in seeing pictures as well.

K. I'll try to remember to bring my camera to the hackerspace on Tuesday. I'll shoot a bunch of pictures and put them in an imgur album.

Ok, I remembered to bring my camera to the space today. Here's the gantry of an old HP plotter that uses a double wrapped steel cable and a v-groove cart.

http://imgur.com/a/tsyAr#0

Edited 1 time(s). Last edit at 08/23/2013 02:49AM by iquizzle.

hi guys, i have one question about hbot:
i am making my own hbot, i have few more parts to print and waiting for gt2 belt.
the idea is to make very robust and heavy metal chassis so i have welded my chasis, and then cnc milled places where printed parts are seating. with very strong motors(3A @24V) and 10 mm smoothrods i think it will be very fast.


finally the question:

i see that most of printers have one or two very narrow placed linear bearings on "Y" axis, so i am wondering how those printers maintein Y/X axis in right angle? is the belt/fishing line or whatever held this two axis @90 degrees?

also why corexy or h-gantry? whats real difference IRL?

woo Wrote:
......
> finally the question:
>
> i see that most of printers have one or two very
> narrow placed linear bearings on "Y" axis, so i am
> wondering how those printers maintein Y/X axis in
> right angle? is the belt/fishing line or whatever
> held this two axis @90 degrees?
>
........

I think that the forces involved in plotters, 3D printers and other things which commonly use the HBot method are very low - it would not work nearly so well if it were a milling machine where the forces can be quite large. It is like being able to stand with your feet together normally, but having to keep your legs wide apart in a high wind.

Having said that, if you get slop in the linear bearings or an increase of friction on one side, this narrow stance in the bearings can lead to a jamming situation quite rapidly.

Mike