Been there done that. It comes with some cool advantages and two big disadvantages. It works and it is light weight so that’s cool. But Bowden tubes cause more oozing. I don’t like using Bowden setups as much. They suffer from problems that linear advance try’s to solve. But even with linear advance they don’t print as cleanly or as reliably as direct drive. In my experience. With that said. If I can’t get the tool changer to work I will switch to a dual head that has a direct drive and a Bowden.
To recap. I have decided for myself that I would rather have a slower direct drive than a Bowden. But my Prusa mini and Ender 3 both have Bowden and they work pretty well. They string and ooze more than my other printers but still look good.
I have considered for a while trying to design a shaft based remote direct drive extruder, or a solenoid based gear switch so 1 motor can drive 2 (or maybe more) extruders. If i make any progress i will poste it on here… maybe it will be the best of both worlds 
I do look forward to seeing what you do with a tool changer though. Having one opens up a bunch of options!
Again new at milling so please explain why this wouldn’t work; What if we did like @turbomacncheese said where we used 2 pieces bolted together and milled like shown below?
That might work… the only problem I see (I have done a ton of 3D printing but i’m a noob at milling) is that the top piece would be. Difficult to bolt down. Not impossible but difficult. We would need low profile blots.
We could split it at the bottom of the groove rather than the top, not sure if that gives us enough clearance for bolts or not though.
@jeffeb3 Gates has an app to calculate belt tension. Tell it what belt you’re using and the length, it gives you the htz.
Soooo. The first Timber bot 3D printer is doing well. I have moved onto the Laser Engraver version. I want to get it done soon so I can make some Christmas decorations. Here is a little video showing my progress so far. It is all just virtual right now but I hope to cut wood this weekend.
One thing I forgot to mention in the video. I intentionally built the laser and mechanics to be separated from the lower box. I did this so I can put a piece of glass under the laser. This will prevent soot from soiling the laser and it will create a thermal barrier. The disadvantage is I won’t have air assist. So the glass will be removable with this build. But I designed it this way because I may eventually try to build an SLS 3D printer from this design. SLS printers need to keep the powder warmed in a heated chamber. That way the laser doesn’t have to work as hard to get the powder to fuse. Well that much heat is bad for a laser diode so they need to be separated from the heat with glass.
Onto the next topic. I mentioned that when I rebuild my TAZ 5 (300x300 print bed) as a Timber Bot that I might try to build a tool changer. Well here are my current thoughts on the subject. Too many words to type so I explained it in a video.
Here is the link to the SMuFF I showed. he has video demos. SMuFF - Smart Multi Filament Feeder with #Bondtech gears by technik_gegg - Thingiverse
I think the issue with converting that to purely bowden would come down to the amount of slack you have in that assembly. If you have another motor on the hot end with the same gear ratio, you could run them from the same driver.
Your dog comes into the garage to grab something from the freezer? Clever dog! 
What do you think about smoke from the laser - do you think fans would be efficient enough to make a clean burn? It seems (from this forum) that people are having great results with compressed air pointing at the laser.
That is a clever idea. But my experience with the Prusa MK3s with the MMU2s has taught me that there is a LOT that can go wrong with a filament feeder and a direct drive extruder. I won’t list all the issues I have run into. I have tested about seven different configurations of a direct drive extruder. I finally settled on one that eliminated most of the problems. But I still get filament load failures about 1 out of 20 attempts. With softer filaments like TPU the failure rate is closer to 19 out of 20.
That is why I am willing to remove the direct drive variable all together. I have already exhausted most of my ideas for making that more reliable.
Now with that said. I have done some more research on the SMuFF over my lunch break. In the FAQ they claim that is does seem to work best when used in a bowden configuration. So that’s promising.
When you say slack are you referring to the length of the bowden tube? That is a concern for me. Bowden tubes work better the shorter they are. I don’t have any great ideas. I just know that I plan to keep the tube as short as I reasonably can.
Apologies in advance for the long reply. I am very detailed and this happens.
LOL. Yeah I heard that comment about the dogs. I really should make a script so I don’t make a fool of myself. But at least its entertaining. Yeah the dogs have access through the garage to their dog run. One is already half blind. I don’t want to cause any eye damage. The kids are the ones I send out to the fridge.
Most of the time I use the current laser engraver I stay in the area. I have already prevented 2 fires that could have gotten out of hand.
Fans/Blowers vs Compressed/pumped air (Here is my knowledge on the subject flawed as it may be. )
First let me define some of my vocabulary on the subject.
Engraving vs Laser Cutting
Engraving is just a shallow burn on the surface vs cutting is intended to go deep into the material.
Clearing smoke vs Air Assist.
In my mind they are different. Smoke can block the beam causing inconsistent or splotchy etching. Air assist can actually help remove material and debris from (potentially deep in) the burn area. This is helpful when you are trying to laser cut rather than engrave.
Smoke vs Soot
Smoke can block a laser beam causing less of the beam power to hit it’s target. Smoke moves with the force of heat propelling up and OUT in all directions as the burning material acts like a small explosion. Soot is heavier and potentially very sticky. It can get stuck inside of the material during a cut or it can explode out and destroy a laser lens. Soot can be bad when there is a lot of oils or sap in the material.
Fans and blowers
Fans and blowers move a small amount of air and are pretty good at clearing smoke. To work best they should be high CFM meaning they move a lot of air very fast. Usually very loud. The air flow should be focused and aimed close to burn point. Otherwise the moving air will dissipate and the movement of the smoke will overpower the airflow from the fan. Fans work better for clearing smoke than they do for air assist because they can’t affect the burn zone with any real force. The heat from the laser beam usually over powers the air movement from the fans. Even with a fan soot can still reach the laser lens potentially causing damage.
Compressed/Pumped air
These usually have enough pressure focused right where it needs to be that they can be used as an air assist. They can over power the heat forces from the laser beam and cause smoke and soot to move out of the way. Problem is after the smoke and soot clear the path of the focused air blast they can still get in the way. Smoke can still block the beam though soot is less likely to reach the lens.
On my current engraver I use a high CFM 40 mm 12 DC fan with a shroud that focuses the air flow. It works well enough for clearing the smoke. But at one point I thought about adding an air pump so I would have the best of both worlds. But I don’t laser cut often enough yet to make it worth it.
When I build this new machine I will still have a fan to help clear the smoke. But I want to experiment with the piece of glass in place. That will prevent me from using anything like a fan or air assist. But I want to try a few things. Like trying to create a wind tunnel under the glass. Not focused but with all the air moving in the same direction. Also I want to try putting the glass directly on top of some of the material and see how well it works. like in this video.
Ultimately the piece of glass separating the core XY and laser module from the lower box will be used later on the SLS printer I hope to someday build.
Don’t worry about the slip of tongue, I enjoyed your vid alot! Down to earth and to the point without much fuzz. Worth the watch!
Thanks for your thorough elaboration. You have obviously put a lot of thought into these plans. I’m just asking out of curiosity, trying to understand the different variants. Looking forward to seeing the machine being built
No. I mean if the bowden extruder is advancing 1mm of filament, and the nozzle is providing resistance, how much goes into the nozzle? If there is a little wiggle room between the geared part and the part that holds the bowden tube, then that slop will mess you up
It sounds like they intended it to be bowden, so that should be taken into account. But if you were the first to try it, I would expect to have trouble if things didn’t hold together rigidly.
For your fans, try getting a high cfm revers incline fan. They are a bit more expensive, but they do not stall under pressure. So if you 3d print the proper duct with a reduction in volume you should be able to get somewhere between smoke removal and air assist. It will also be very helpful in creating your wind tunnel. To give you an idea, I have seen 50hp revers incline fan implode a factory (im not kidding that has literally happened) so even a relatively small one should be able to create all the force you need. And simplify your experiments by letting you control the pressure and speed by changing your duct volume and inlet/outlet sizes. They also will still react to pwm controll and give you some adjustability as well
another idea for your filament changer. have you considered making a modifaction to the design, where one “bay” is a remote drive extruder rather than an actual filament. i will try to explain;
you have say 4 bays, 3 with filament then on the 4th bay rather then clamping down on a filament it engages a flex shaft, the shaft would then drive a remote extruder like this one https://www.thingiverse.com/thing:4009262. this way your still only using 1 stepper to drive your filament, you get a mostly direct drive system, and you have a filament changer.
let me know what you think.
Remote direct drive… that is fascinating. I will need to look into this. It may not be feasible to make it work with a filament/tool changer but then again maybe it can work. Interesting idea.
worst comes to worse, you can always add a solenoid as a peripheral. let the solenoid engage a gear that connects the filament changer stepper, to the flex shaft. all the remote direct drive extruders need gearing anyway so it shouldn’t be to difficult to add it. then as long as the filament selector is not gripping on any of the filament choices and the solenoid is engaged you have direct drive. when the solenoid is disengaged and your gripping a filament in the changer you have a Bowden.
also if your doing research, you may want to start with the Zesty Nimble. I believe they where the first to invent the remote direct drive approach.
@mordiev, I miss seeing your progress with this project. I hope all is well and you can get back to it soon
Well… I have made some good progress on the laser engraver. But I need to pause all my pet projects for a few months. Without trying to get super political I will say this. Elections have consequences. Here in Colorado (and in my county and city) they passed three laws that will make it so they can raise taxes/levies on my house every year without voters like me voting on it anymore. Also they passed a new payroll tax that will take a large chunk of my paycheck starting in 2024. Sooo in the next few years it will be a lot harder for me to afford to live in the same house I have lived in the last nine years. In that time the property taxes have almost tripled. I am already financially stretched to make that payment. I expect the taxes to double again in the next few years.
I need to do something or I will lose my house. So I am looking for a new job and getting ready to sell. If I can earn more money and have a smaller house payment then I will be more secure for several years. Probably going out of state. I have heard Texas is favorable… but I have not decided where I am going yet.
So my free time has been spent job hunting, packing and finishing up home improvement projects that have been “half done” for several years.
I’ll take some pictures of the current engraver build and share them.
Eventually I’ll share my idea for the tool changer. It is a unique idea I think. I may even apply for a patient on it and then do the Tesla thing of making it open source.
On other news. Prusa announced in their latest blog post a small update on their Prusa XL. It will be a core XY design as well. They announced that they are planning to add a tool changer to it. So that is exciting to see when it eventually happens.
Ungh, That’s a bummer. Moving is no fun. Neither is job hunting. Hopefully you can get back to it soon, or find some time to play with these to give yourself a break from that chaos.
But also, yeah, please no politics.