Critique my BLV cube toolchanger draft!
-
Hmmm, that CAN bus sure is cool...
Hmmm, I do do a lot of color changes...
Hmmm, I know I haven't actually finished the current upgrade to the printer yet, but I was going to design a new hotend anyway...
...and here I am. I don't know the first thing about toolchangers, so would appreciate any insight from you all! At a minimum, I figure this could be a reasonably nifty quick-change manual toolhead, which I would also really appreciate given the frequency with which I like to fiddle with things.
The mount system was inspired by this delightful servo-less project, whose creator claims has gone through thousands of changes for her without a hitch. I am an enormous fan of magnets but loathe adhesives, so I adapted the clamped v-groove magnet mount from my kinematic bed into a Maxwell kinematic mount. I'll be testing 1/8", 1/16", and 1/32" thickness N42 magnets to try and find a secure fit without skipping steps. (In principle, the 1/8" magnets would have as much as 30 lbs of pull force!)
(I accidentally forgot the belt attachments on one of the plates, need to add those back in.)
It also occurred to me that the 2 magnets + the ball bearing on the toolhead form a switch, so I added channels in the backing plate for copper tape. I figure this can give me a way to sense a positive lock between the carriage and the toolhead.
The toolhead mounts to the carriage with 3/8" steel bearings, and has a lug to accept 2x 5mm dowel pins for the stationary holder.
The toolboard is mounted in a little cover that hinges open to access the bits.
Here's a little animation in the context of the printer. I figure I lose about 80mm of print area in Y (black are on bed is lost print area), which seems a fair trade. I might also put kinematic mount on both sides of the carriage so I can also use the printer in non-toolchanger configuration for larger prints.
Any thoughts or suggestions?
-
@whopping-pochard are you thinking the magnet to lock the tool to the carriage?
If so I would try and think of a better way of doing that last thing you want is it giving way mid move.
-
Yep, the idea is for the magnets to provide the clamping force. It's not a positive mechanical lock, but I'm hoping I can get away without one. I'm basing my optimism off a similar design that is reported to be pretty solid in practice.
At the least, I can detect the instant any of the six points of contact are broken and issue a pause command.
-
After using magnets for delta ball joints for years, I must advise you to never use magnets like that! The magnets will go wrong. There is no if.
*not actually a robot
-
@bot said in Critique my BLV cube toolchanger draft!:
After using magnets for delta ball joints for years, I must advise you to never use magnets like that! The magnets will go wrong. There is no if.
I agree not worth the risk of a Thermal Runaway if the thermistor comes adrift or anything nothing is ever really foolproof
-
@bot I'm curious to hear more about your experience! What does 'go wrong' mean?
-
They will decouple when you don't want them to. Or, they will attract something you don't want them to -- this is the one that is most annoying but often least critical. Every time you put a steel screwdriver or other tool near them, THWACK it will accidentally smack the magnet hard. AFAIK this is not exactly good for the magnets, but it's not going to ruin anything it's just annoying.
They can also interfere with electronics: preventing fans from spinning or other things.
If the printer is a fun project that is not going to be used as a workhorse, but to print things once in a while then they will be great and the flexibility provided by them may be worth it. If, however, you envision depending on this to print ultra-reliably or while left alone, then it's best to avoid magnets.
[edit: Another thing to think about is assembly. Are the magnets epoxied in place? If there is only a single magnet per tool, it may be manageable, but when multiple magnets need to be attached nearby each other, the magnets all want to attract to each other, and move. You have to hold them in place very firmly while the adhesive sets. Designs which do not use adhesive to hold the magnets are much, much better.]
-
@bot Thanks! Those are all really good points! My other two printers are currently my workhorses, so this is definitely in the realm of 'fun testbed' territory. For a certain definition of fun, anyway. Based on your input, I'm thinking it would be a good idea to add a postive-lock screw in the center so I can fix it securely to the carriage when I don't want to mess with toolchanging. Then if the magnets end up being too much of an interference, I can replace the magnets with some steel or aluminum stock and still take advantage of the kinematic mount for servicing or swapping the hotend.
And I definitely hear you on the assembly! I only use mechanical restraint for magnets. They press securely into the front plate, and then there is a thin backing plate that compresses them against their forward stops.
-
@whopping-pochard I'm digging that design with the swingout toolboard case and shared cooling fan. Are you sharing the design?
-
@I_Jedi Thanks!
And yeah of course, all open source! It will be on https://github.com/tanaes/BLV_MGN_cube-remix as I go. It’s a bit of a mess but if you want any help finding things, or a Fusion or STEP export, just ping me.
-
@whopping-pochard Outstanding, thank you! I am close to bringing my 're-mixed' BLV MGN online. (remixed as in borrowed ideas from all over the smashed them together in 3Dbuilder since I have no CAD skills
-
@I_Jedi There are a lot of modified parts in there that are basically just the original, but edited a bit for better printability / assembly. Most of the holes in the original STLs aren't the right tolerance, and there are a lot of unsupported overhanging holes. Anything in the "jon's mods" folder should have correct hole tolerances and single-layer sacrificial bridges to support overhanging holes. I also remade the front corner belt tensioners so they go on a little easier, are identical (just rotate, and screw on the cover 180°), and can be printed easily without support.
Happy printing!
-
Oh i'm blushing so hard to see my design here!
the idea of using block magnets like that on the other side too (instead of balls) is honestly clever too no idea why i didn't think of that 
it will save space and have more "suction". tho I worry it'll cause some alignment issues :?btw for a direct drive tool I suggest a middle ring magnet.
the ball/block design I had did handle the weight but it'd fly off at high speeds. but with an additional magnet it'd be very very very secure! but if something hits the tool there's always the possibility it'd get lose, for that I can suggest you to add a mechanical switch to detect an attached tool and stop and warn you!
-
Oh and don't be discouraged by others xD I had really no issues with mine. yes in theory a lot can go wrong but in practice it works very well for common use!
also if you want you can use the same motor for many tools but changing one of the couplings for a gear the extruder will go into or just using magnets to spin the other side (kinda hard tho. needs a special magnet too unless you make your own out of 2)
-
@amythebun Oh yay, you're on the forum here!! Thanks so much for the inspiration! I saw your printer on Hackaday and thought it was super clever! I'm very excited to see if I can get it to work.
I got the prototype assembled and it looks like it it can take a load of about 750 g at the farthest corner before it torques out, so I figure if I set accelerations above about 9000 mm/s^2 I should be able to send it flying!
-
on my side i will do the same system as the Paw printer because i really like it
https://www.youtube.com/watch?v=0ARCip5CrUk&feature=emb_title
https://forum.duet3d.com/topic/9392/latching-magnetic-tool-changer/31 -
@amythebun really clever about the tool engaged switch. thanks for the idea.
-
@amythebun said in Critique my BLV cube toolchanger draft!:
also if you want you can use the same motor for many tools but changing one of the couplings for a gear the extruder will go into
I just read this line and was wondering, why no one responded to it? it's pretty clever IMHO.
The downside is, you have to align both sides of the clutch eventually.
Maybe this can be done with the magic stall detection feature? While moving the motor to the extruder, you spin it with low current. When it stalls, the connection is OK. A manual switch might work too, but is lame
-
@o_lampe Wow yeah a good use for stall detection on the E motor! Very interesting.
-
@psychotik2k3 Yeah that is so awesome! If mine doesn't work out I might try to do that as well.
I'm finally back in town so hope to get back to it soon.
