HS6-GT – DIY H-Pattern Shifter

[Bitacaia]

I have made some fast investigation regarding the dimension for the Ball Bearing.
Note:
The engagement movement is 12mm in one direction. So the total length of the "Offset Lever" is 24mm

Hope this image makes sense... This is how i design the Slopes.

The conclusion.
-You can have maximum of 1.5mm of compression of the springs using the POM wheel with 625 Ball Bearings. And very shallow slopes...
-Using the actual components, and assuming a radius of 1mm for the edge of the slope (we don't want a sharp edge, because it will be prone to break, so bigger radius is better). The maximum compression is 2.3mm. And medium slopes.
-Using the M105zz without the POM Wheel, we can have more liberty to play with the slope angles and spring compression.

I really don't believe the big wheel will give a better feeling... As I said before, smaller, in this case, is better!

 

[pandax]

Beautiful picture.
Thanks for taking the time !
I don't expect much improvement, in any way, tbh ... I just want to try, for the sole purpose of science
I feel the fail coming from miles away (or disappointement should i say), but it's sooo cheap on aliexpress, meh, why bother ?
It was like 40 cts for the POM wheel alone, and i bought the 625 bearings on the side.

 

[Jbixby]

There you go, works great.

 

[pandax]

Looks great ! But it looks like it's too lose ?
The one from OP looks more sturdy, less wobbly !

 

[Fatapoc]

Looks great ! But it looks like it's too lose ?
The one from OP looks more sturdy, less wobbly !

I was thinking the same...

 

[Bitacaia]

There you go, works great.

View attachment 354145

That is a good point! Why you have so much play? I know you had a tolerance problem before. You changed something meanwhile?

 

[Jbixby]

I haven't changed anything yet. I'm going to get a few different springs and see if that tightens up the mechanism before I make any new parts.

 

[Bitacaia]

I haven't changed anything yet. I'm going to get a few different springs and see if that tightens up the mechanism before I make any new parts.

Question:

Did you oriented the "Spring Side Stopper" on the correct position?
I need to redesign this part to avoid wrong assembly...

The Big fillets should go to the bottom position. The distance from the contact surface to the axis center is different. If you rotate, the lever will be loose. Because the ball bearing will be further away from the "Offset Lever".

Can you check this? Tnx!!

Tiago Viana

 

[Jbixby]

Question:

Did you oriented the "Spring Side Stopper" on the correct position?
I need to redesign this part to avoid wrong assembly...

View attachment 354581

The Big fillets should go to the bottom position. The distance from the contact surface to the axis center is different. If you rotate, the lever will be loose. Because the ball bearing will be further away from the "Offset Lever".

Can you check this? Tnx!!

Tiago Viana

I'll check when the new springs come, but I'm pretty sure they are in the correct orientation.

 

[Fatapoc]

I tried the nylon bearings, but they are too thick, so they didn't fit . I might order different ones and let you guys know how that worked out.

 

[pandax]

I tried the nylon bearings, but they are too thick, so they didn't fit . I might order different ones and let you guys know how that worked out.

Allow me 48h, and I'll modify the parts for the nylon bearing.
I intend to use them too, so I might as well do the modifications now
Just let me some time, I won't be able to do it before tomorrow; but I'll gladly do it and share them.

Cheers

 

[Jbixby]

Alright, so not great news. I got some new springs to try, opened the case back up and found this. Probably only a few hours of use over a 2 or 3 days testing it out.

So I have a couple options now and one in the future. Right now, the first offest lever that's pretty well ruined I printed with a 0.4mm nozzle. I'm setup with a volcano and 0.8mm nozzle which might make the part a little stronger but I don't think will solve the problem if the part now starts to fall apart after 5-10 hours of use. I have petg which I guess is stronger than pla+ but I don't see that lasting all that much longer either.

The last idea I have is to cnc something out of aluminum, I'm just starting to setup the cnc machine I've been building so that's probably not something I'll get to any time soon.

 

[Jbixby]

I found these plastic bearings at McMaster Carr if you're in the US.

McMaster-Carr

McMaster-Carr is the complete source for your plant with over 595,000 products. 98% of products ordered ship from stock and deliver same or next day.

www.mcmaster.com

I'll probably order one to try out when I have some other stuff to buy.

 

[pandax]

Try printing the part on it's side !
A bit more work, but it will be much stronger and won't delaminate so badly.
And you could try petg, of course.

Good luck !

 

[Bitacaia]

Hi Jbixby.
That's some bad news... My best guess to solve this problem is to make the slopes bigger (bigger radius and more "meat"). And, as i have showed before, the only way to do that is to use small diameter ball bearing.
Your last post, with the Nylon ball bearing looks good, but it still has outer diameter of 16mm. So you will not be able to change much the slopes.

I strongly recommend the use of the MR105 bearings. And if you don't like the noise of metal hitting the the plastic, you could always make a printed ring to cover the bearings, and raising the outer diameter to around 12mm.

About printing on the side, like Pandax, suggested, could also solve your problem. Although you will need to add support and probably make the part hard to print...

For those that don't understand about 3D Printing, I must explain, 3D Printing isn't an exact science! There are so may variables that can change the result. In my case, i don't have this delamination problem, so far (it can happen...)
It can be a filament problem; printer setting problem; Printing temperature problem; Slicer parameter problem; or finally a Design problem.
So, don't be afraid or disappointed with this problem, for sure we can find a solution.

Tiago Viana

 

[Jbixby]

pandax, I'll try a few a different ways. Since I've been printing with a 0.8mm nozzle I've had a really hard time getting supports off my prints. Not sure what to do about that.

bitacaia, My thought with the plastic bearing was that it would wear the offset lever part less than a metal bearing, is that not correct? I actually like how the shifter sounds.

 

[Bitacaia]

pandax, I'll try a few a different ways. Since I've been printing with a 0.8mm nozzle I've had a really hard time getting supports off my prints. Not sure what to do about that.

bitacaia, My thought with the plastic bearing was that it would wear the offset lever part less than a metal bearing, is that not correct? I actually like how the shifter sounds.

Hi Jbixby.
I think it can be interesting to try to solve your problem in this thread. even it is related with 3d printer settings.
So, i would like to know the printer you have, the slicer you use and the PLA brand and settings you are using.

After looking at you first layer of the shifter case (i know you are printing with different nozzle and settings the "Offset Lever", but it is the only picture i have.) my conclusion, you need to tweak the printing settings. The filament is too spread. They are not touching, so your printed parts aren't working as solid, but as a bunch of strings...

Here you have a good first layer (right) and your first layer. It is easy to see your first layer isn't very good. The filament isn't fusing together because they are too spread out. (I know, i'm just seeing the first layer, the inside layers can be different, but i'm analyzing what I can see)
My guess:
1- Your nozzle is too far away from the printing bed. Because of that, the filament isn't being squeezed against the bed.
2- For a 0.4mm nozzle, the extrusion width normally is 0.45mm. But, when the lines are too spread out, i raise the flow up to 110% (Cura slicer: Flow; Prusaslicer: Extrusion multiplier)
Here a site about this thema:

Prusa Knowledgebase

All information you need to know about Original Prusa 3D printers. Assembly manuals, print quality troubleshooting, calibration, PrusaSlicer and much more.

help.prusa3d.com

3- Higher temperature and less part cooling can help to fuse the filament together. The problem, the part can look bad, but who cares... this is a functional part...

Hope this tips are correct, and if they are, I hope it helps. I'm no 3D Printer expert...


About the Ball Bearing question:
True. The Nylon will absorb some of the impact force. But nylon or POM is much stronger then PLA/PETG, so it will not absorb much... So, i don't think it will improve that much the problem. But is better than nothing. At least it will reduce the noise.
But if you like the noise. Just go straight to the MR105 Bearings.

 

[Jbixby]

I printed the case parts while experimenting with my new setup, ender 3 pro, hemera direct drive with 0.8mm volcano. I left the case as is because of the amount of pla it takes to print, my prints are fine now wrt first layer. I'm using Cura 4.4 and eSun PLA+ general settings are .48 height .9 width 190 hot end 50 bed.

I did print the offset level on its side last night, so I'l try that out when I have some time.

 

[Bitacaia]

I printed the case parts while experimenting with my new setup, ender 3 pro, hemera direct drive with 0.8mm volcano. I left the case as is because of the amount of pla it takes to print, my prints are fine now wrt first layer. I'm using Cura 4.4 and eSun PLA+ general settings are .48 height .9 width 190 hot end 50 bed.

I did print the offset level on its side last night, so I'l try that out when I have some time.

First, congratulations for the Hemera. If I hadn't bought a bondtech extruder clone, for a direct drive setup, I would buy one!
I believe you wrote the parameters you use for the case. Because 0.48 layer hight is really high for a 0.4mm nozzle. Is more suitable for the volcano 0.8mm setup.
If it is so, what was the parameter user for the "Offset Lever"?

 

[Jbixby]

Those are the settings I printed the new offset levers with. The first one was likely .28 height, .44 width. The prints with the 0.8mm nozzle are quite a bit stronger than the 0.4mm nozzle. Here's the part printed on its side.

It doesn't look nearly as good as printing the part on its flat side, but it should work.

Next step is seeing if I can make the part with the cnc machine I'm building right now.