As of this writing, I'm still making a few changes to the design before it gets released to the official repository, so this update might have a few things you didn't see here in the video! It'll be merged in soon! :D
while the bolt might dig in a bit this will not be a big problem. I would say using a lock nut on the bolt, or a drop of loctite on the bolt would be a better improvement. It is amazing what vibration do!
You should use a flanged nut, or at least a washer, for the tensioner, to increase surface area and prevent the plastic to flow or crack under the belt tension with such a small area.
I am super happy you made this decision. I personally won't even use the conveyor belt at all, but i do want to build the index so having the basic core aspects of the PnP functionality perfected before moving on extra additions that probably a lot of the people won't need such as the conveyor belt is great, since now us who just want the basic PnP will be able to start building much sooner.
if someone hasnt already said something you may want to edit the x and y tensioners a bit to add some metal or a cap to the end of the adjustment screw, otherwise the plastic it is in contact will will deform/break over time, good solution though
if you _really_ want constant tension, considering adding complaint springs with a defined strain (preload). Directly tensioning the belts without such a spring makes it susceptible to creep in the belt, and thermal effects (mismatched tempco of the belt vs. the aluminium-frame)
I want you to know that I really appreciate the work you are doing. I will be making a pnp machine for work next year, I will surely hop on the Index design. Thank you very much for the project you are sharing
Theres just something really satisfying to it to watch you making progress with the Index. Keep it up! Edit: It would be cool if you would replace the innerducts with some drag chains, that would really help the cable management.
Congrats! The whole thing looks better and better and the tensioning system is really clever. But you might want to support the loose end of the motor shaft with a bearing. I did destroy 2 motors on a corexy printer by trying to achieve the perfect tensioning (over-tensioning) of the belts. The motors got louder and louder until one started to skip.
Brilliant! Really excited to hear about your switch in focus. Lots of us are super keen to build an Index that works, even if it doesn't have all the cool R&D features yet :D :D
The adjustment screw will push into the plastic, which should be ok but will wear. A little blind hole could be added, and people that worry can press in a little bearing ball for the screw to press up against.
You should race to an MVP as soon as possible. It will get things going faster both technically and financially. I would guess the marlin mod should be on the top of the list since you already have a board. Since auto feeders are optional, you should be able to get a minimum kit going pretty soon. I am anxious for it.
I think 0402 is a more honest minimum goal designating whether people would choose to move to a pick and place, rather than 0603. Both can be hand soldered but a very tight design with 0402 gives you MUCH more in terms of scaling down the size, and is way harder to hand solder once it gets really tight.
the very best channel really. I have a PNP machine that I made after watching your videos. I agree with these changes. Cool. I am mostly interested in paste dispensing (automation) and testing after oven. I hope the index repitabilaty testing will be successfull. very inspiring videos good luck
Maybe somebody already said this, but if you want other people printing these parts on there printers it might be a good idea to make so kind of test/calibration part to print. With some known measurements on there to check and maybe the different size holes for the push in nuts. So people can validate that the parts are the correct size before printing al of them.
Nice job on the tensioners! Lots of comments about the screw tip contact on the plastic. A small piece of steel glued to the plastic could be an easy fix, or a pair of jam nuts tightened against each other with a touch of thread past them and a washer against the plastic are easy fixes. I'm more concerned with the mechanical limit switches... they look out of place (IMO) and require unnecessary PCB's (IMO), but more importantly, are not very repeatable. Use U-shaped optical make/break sensors and a flag... very repeatable, solid state, and can be direct mounted without PCBs.
Not to be a party pooper here, although you might consider some kind of plate for the Y belt tentioner (the rend thing with the bolt that tentions the belt). I noticed that when you start tentioning with the bolt being screwed in, this one starts to eat into the plastic part, will eventually make a hole into the part.
You need to make your MOBO available and show how you are getting parts to drop/place....I had to get an air release valve to get mine to work. So an updated copy of your Marlin code would definitely be good!
Awesome project Shephen ! Why not use linear guide rails on X/Y ? It would simplify the gantry and make the joints more compact. I would use a Voron 2.4 like X/Y gantry system that is very accurate and handle high accelerations. Also, you should have rubber feets ! ;)
and are there any other quantitative methods for someone new to building an Index to know they have a good belt tension? maybe pinging a belt and measuring the sound frequency with their phone?
Terrific! I imagine there isn't a quantifiable answer for this question, but what IS the right amount of tension for the belts? How will I know if they have the correct amount of tension on them?
@@KarlMiller If you model a belt as a string under tension, the tension observed from the audible tone of a fundamental resonance frequency resulting from a step change (plucking) should be as follows: T=4×[belt span in meters]²×[measured peak frequency in hertz]²×[belt linear density in kilograms per meter] The frequency can be determined cheaply with a free spectrum analyzer app for a phone (there are several). Target tension is determined by the belt manufacturer but tends to be "the minimum so that it doesn't slip over a tooth." The purpose of a sonic tuner is to get consistency on every assembly.
@@amoose136 Holy crap! That has to be the best answer I have ever received on TH-cam. Not only did it teach me how troubleshoot my own solution, gave a rule of thumb characterization that I got. This channel is so educational, and it has some great people contributing as well.
Amazing update as always! I'm curious if the gantry belt attachment points are in line with the pulley width - It's probably moot, but it looks like the gantries pull the belt a bit away from the track for the attachment point, which could cause some cosine error, or at very least some belt loosening between midpoint and extremes of the travel as the 'triangle' portion of the belt length changes.
Presentation well done, with explanation of goals. I believe you will find the interest in videos for improving the present design, testing, and validation will provide more views in total than the future R & D videos.
Its really cool but there is a even better way of doing it you can see it on the ender 3 v2 it does not use a hinge so the Roller is not moved of center so the Belt has no chance of rubbing against the frame and another Bonus is that it can have a waaay longer travel range so you can put on the belts completly lose and still have perfect tension it works by having the Roller floating and with 2 printed pieces and a screw you get linear motion
Perhaps put a short bolt/screw into the contact surface to address all the comments below. Or a smaller diameter captive nut. Ideally, whatever it is, it is already in the BOM.
I think you should put something where your tension is pushing on the plastic print. Maybe a screw with a head big enough for the tension screw it bear against so it doesn’t destroy the plastic.
Does it matter that the tension screws at the front are tension "against" plastic? maybe add a piece of metal there so the tension does not "eat" into the print
what is the current price you are estimating(without kitting just the parts you need) also I would slap stone plates under the legs to give it more mass on the stationary parts
If you want more TH-cam views, create a small intro that gives some sort of idea wtf an index is. TH-cam is pushing out random videos of your long vlogging series.
As of this writing, I'm still making a few changes to the design before it gets released to the official repository, so this update might have a few things you didn't see here in the video! It'll be merged in soon! :D
Where are the STL file
You should probably put something metal behind the new tensioning screw. Otherwise it will dig into the plastic.
while the bolt might dig in a bit this will not be a big problem. I would say using a lock nut on the bolt, or a drop of loctite on the bolt would be a better improvement. It is amazing what vibration do!
@@allenlorenz or use nuts with nylon inserts for those.
@@JonathanBuford That is a lock nut :-)
You should use a flanged nut, or at least a washer, for the tensioner, to increase surface area and prevent the plastic to flow or crack under the belt tension with such a small area.
I am super happy you made this decision. I personally won't even use the conveyor belt at all, but i do want to build the index so having the basic core aspects of the PnP functionality perfected before moving on extra additions that probably a lot of the people won't need such as the conveyor belt is great, since now us who just want the basic PnP will be able to start building much sooner.
if someone hasnt already said something you may want to edit the x and y tensioners a bit to add some metal or a cap to the end of the adjustment screw, otherwise the plastic it is in contact will will deform/break over time, good solution though
if you do a kit don't forget stickers...people love stickers
Yeeeeees
You're like a whole engineering team rolled up in an infectiously endearing overly excited young fella. Good work man
Wonder if you can dial the PNP in to pick that door up from your hallway and put it back on its hinge 😉 Great content as always!
if you _really_ want constant tension, considering adding complaint springs with a defined strain (preload). Directly tensioning the belts without such a spring makes it susceptible to creep in the belt, and thermal effects (mismatched tempco of the belt vs. the aluminium-frame)
It’s like Wintergatan doing the million marble validation. I’ve really learned a lot from his last few months of testing reliability.
You can get a numerical value for belt tension by measuring the pitch when you pluck it like a guitar string.
I doubt this will work... it's not going to freedom to vibrate.
I want you to know that I really appreciate the work you are doing.
I will be making a pnp machine for work next year, I will surely hop on the Index design.
Thank you very much for the project you are sharing
Not gonna lie, seriously one of the best channels on youtube! 💖
Great tensioner idea! Love that hinge mechanism. My 3D printer Anet A8 got similar tensioner update that I print and boy, results are incredible.
Theres just something really satisfying to it to watch you making progress with the Index. Keep it up!
Edit: It would be cool if you would replace the innerducts with some drag chains, that would really help the cable management.
I really like that belt tensioning mechanism, very simple and effective; a much better design than what I came up with for my laser engraver!
Congrats! The whole thing looks better and better and the tensioning system is really clever. But you might want to support the loose end of the motor shaft with a bearing. I did destroy 2 motors on a corexy printer by trying to achieve the perfect tensioning (over-tensioning) of the belts. The motors got louder and louder until one started to skip.
Brilliant! Really excited to hear about your switch in focus. Lots of us are super keen to build an Index that works, even if it doesn't have all the cool R&D features yet :D :D
The adjustment screw will push into the plastic, which should be ok but will wear. A little blind hole could be added, and people that worry can press in a little bearing ball for the screw to press up against.
black and red, totally cool! GJ Stephen
You should race to an MVP as soon as possible. It will get things going faster both technically and financially. I would guess the marlin mod should be on the top of the list since you already have a board. Since auto feeders are optional, you should be able to get a minimum kit going pretty soon. I am anxious for it.
I think 0402 is a more honest minimum goal designating whether people would choose to move to a pick and place, rather than 0603.
Both can be hand soldered but a very tight design with 0402 gives you MUCH more in terms of scaling down the size, and is way harder to hand solder once it gets really tight.
I’m really looking forward to see the index doing its job!
Highly recommend you use threaded inserts instead of the captive nuts. Cost is fairly similar to the nuts and is much more reliable. Awesome work!
the very best channel really.
I have a PNP machine that I made after watching your videos.
I agree with these changes.
Cool.
I am mostly interested in paste dispensing (automation) and testing after oven.
I hope the index repitabilaty testing will be successfull.
very inspiring videos
good luck
Maybe somebody already said this, but if you want other people printing these parts on there printers it might be a good idea to make so kind of test/calibration part to print. With some known measurements on there to check and maybe the different size holes for the push in nuts. So people can validate that the parts are the correct size before printing al of them.
Nice job on the tensioners! Lots of comments about the screw tip contact on the plastic. A small piece of steel glued to the plastic could be an easy fix, or a pair of jam nuts tightened against each other with a touch of thread past them and a washer against the plastic are easy fixes. I'm more concerned with the mechanical limit switches... they look out of place (IMO) and require unnecessary PCB's (IMO), but more importantly, are not very repeatable. Use U-shaped optical make/break sensors and a flag... very repeatable, solid state, and can be direct mounted without PCBs.
I love the production quality on this video and all the content!
Nice work - impressed with how well you're pumping out the videos! Great update, keep at it :)
I like the color on you build. For testing i would use the Cams and as much speed as the motors can handle.
Not to be a party pooper here, although you might consider some kind of plate for the Y belt tentioner (the rend thing with the bolt that tentions the belt). I noticed that when you start tentioning with the bolt being screwed in, this one starts to eat into the plastic part, will eventually make a hole into the part.
You need to make your MOBO available and show how you are getting parts to drop/place....I had to get an air release valve to
get mine to work. So an updated copy of your Marlin code would definitely be good!
Awesome project Shephen ! Why not use linear guide rails on X/Y ? It would simplify the gantry and make the joints more compact. I would use a Voron 2.4 like X/Y gantry system that is very accurate and handle high accelerations. Also, you should have rubber feets ! ;)
Red looks cool
Does it make sense to give a recommended numeric torque range for the belt tension bolts? if users have a torque driver of course
and are there any other quantitative methods for someone new to building an Index to know they have a good belt tension? maybe pinging a belt and measuring the sound frequency with their phone?
Terrific! I imagine there isn't a quantifiable answer for this question, but what IS the right amount of tension for the belts? How will I know if they have the correct amount of tension on them?
I’ve heard of people tuning belts by plucking them and adjusting until a particular tone plays. Just like a musical instrument.
@@amoose136 what is the note for the correct amount of tension?
@@KarlMiller If you model a belt as a string under tension, the tension observed from the audible tone of a fundamental resonance frequency resulting from a step change (plucking) should be as follows: T=4×[belt span in meters]²×[measured peak frequency in hertz]²×[belt linear density in kilograms per meter] The frequency can be determined cheaply with a free spectrum analyzer app for a phone (there are several). Target tension is determined by the belt manufacturer but tends to be "the minimum so that it doesn't slip over a tooth." The purpose of a sonic tuner is to get consistency on every assembly.
@@amoose136 Holy crap! That has to be the best answer I have ever received on TH-cam. Not only did it teach me how troubleshoot my own solution, gave a rule of thumb characterization that I got. This channel is so educational, and it has some great people contributing as well.
Nice work, but i think the M3 screws of the tensioners will dig into the plastic over time. Maybe some way to distribute the pressure might help.
Amazing update as always!
I'm curious if the gantry belt attachment points are in line with the pulley width - It's probably moot, but it looks like the gantries pull the belt a bit away from the track for the attachment point, which could cause some cosine error, or at very least some belt loosening between midpoint and extremes of the travel as the 'triangle' portion of the belt length changes.
Presentation well done, with explanation of goals. I believe you will find the interest in videos for improving the present design, testing, and validation will provide more views in total than the future R & D videos.
Ahah.. my current favourite TH-camr 😀 Always cheers me up AND great content!
Its really cool but there is a even better way of doing it you can see it on the ender 3 v2 it does not use a hinge so the Roller is not moved of center so the Belt has no chance of rubbing against the frame and another Bonus is that it can have a waaay longer travel range so you can put on the belts completly lose and still have perfect tension it works by having the Roller floating and with 2 printed pieces and a screw you get linear motion
Perhaps put a short bolt/screw into the contact surface to address all the comments below. Or a smaller diameter captive nut. Ideally, whatever it is, it is already in the BOM.
Totally stealing that belt tensioner idea for my own machine!
I think you should put something where your tension is pushing on the plastic print. Maybe a screw with a head big enough for the tension screw it bear against so it doesn’t destroy the plastic.
can you make a video on how you are making this a product that is scaled? amazing work!!
5:12 why does it look like a skull from Día de Muertos celebrations?
you should get a light/medium duty press
Index might not be the optimal name, I think there is a whole category of 3D Printers/Brand name called Index
That’s IDEX
Can I send you some braided split loom? I can never go back to the plastic stuff.
Looking fwd towards your validation exercises. What is your target for smallest component size? Are u aiming for 0402 or even smaller footprint?
Hi Stephen, don't you think " Belt tensioner spring" would be better and simple solution to tension the belt on Index?
Also I would beta test the hell outa this for material costs for SURE
Does it matter that the tension screws at the front are tension "against" plastic? maybe add a piece of metal there so the tension does not "eat" into the print
what is the current price you are estimating(without kitting just the parts you need)
also I would slap stone plates under the legs to give it more mass on the stationary parts
Keep plastic creep in mind - it can be significant over the longer term.
such a great video!
For pressing in nuts I would recommend a Knipex Zangenschlüssel, because of the parallel jaws.
th-cam.com/video/WS0oyQevFLs/w-d-xo.html
how about a video of your ender at work???
Good work man!
Where are the STL file
Please sell me a kit that I can afford for a pnp.? The cost is out of this world for someone like me on a very tight budget. I would love to have one.
STL ?
Index voron
Once again, I would like to point out that I was first 🥇🥇.
I know it's somewhat cringey, but it feels weirdly good lol.
If you want more TH-cam views, create a small intro that gives some sort of idea wtf an index is. TH-cam is pushing out random videos of your long vlogging series.
And I watched this whole video, because it had elements I liked. But still have No Idea what this gantry does
Acaket
废话太多