The Meticulous Z-Axis

Something like this one. It would be a good idea to add an external fuse for the previously discussed potential immolation issue

1 Like

The voice of experience here! Thanks for the pointer, I will make sure to get a PS that is overpowered for my needs. I think the link that @mooselake posted is a pretty good one. Itā€™s more expensive than what I budgeted for, but Iā€™m okay with spending a little more money on a more reliable machine (up to a point, I canā€™t break the bank in the process).

I think ā€œDo NOT be on fireā€ is a pretty good rule! :wink: Thatā€™s a good point with the fuse. I also want to add a couple of those auto-extinguishers in the top of the enclosure, just in case.

Itā€™s not entirely my time, Iā€™m able to go off and do other things while the printer works. Usually Iā€™m designing or playing video games while itā€™s working. For my own projects and a couple 1 offs, the 3D printerā€™s 15 hours arenā€™t that bad. If I was cranking out orders, then yes, a vacuum former would have much better cycle times. I was looking into building one a little while back, I think I still have the CAD files somewhere.

If youā€™re going to spring for a couple exploding anti-fire balls and an enclosure might as well go for a higher quality Meanwell like this sample. A 3D printer would benefit from the extra power, a Maslow could get by with something smaller, although Iā€™d still want about 1.5 to 2x what you think you need. Besides moving into a higher efficiency range and having some extra breathing room thereā€™s no kill like overkillā€¦

1 Like

Hey guys, Iā€™m new to Maslow only posted a couple times and still waiting on the next batch to ship out to get my machine. Iā€™m designing and building my own full size airplane so aerodynamics are an interest of mine and the ductwork for the dust collection could use a little tweaking. I couldnā€™t find a good image on the web so I drew one up quick and dirty. The optimum shape for a intake duct is similar to the end of a trumpet. This applies more I suspect to the plywood and sheet metal version of the dust collector than the 3d printed one but there is probably a little room for improvement there as well.

After changing the profile probably the best way to improve the efficiency of this simple duct is to add some clay into the corners especially around where the vacuum nozzle attaches to smooth the airflow into the hose a little better. Iā€™ve seen examples where epoxy putty was used to smooth out airflow in intake ports so simple modeling putty ought to work for us in this application.

I like this thread and think Iā€™ll be looking to upgrade my z axis when it arrives

4 Likes

Will, if no one has already done so, welcome to the Maslow forums! :smiley:

First off:

Thatā€™s AWESOME! As both someone interested in flight, and a maker, that is just so cool to me. Iā€™ve never been in a cockpit before, but I hope to fix that before too long!

Your knowledge of aerodynamics certainly helps here. Up to this point Iā€™ve only really been calculating flow based on the section of the chute. What youā€™re saying completely makes sense to me. That hump would create an eddy current.

Using clay to help smooth out the corners is a good suggestion. That would work better than the caulking
I was talking about before. Also, as a former art student, I have too much of it sitting around so it would be good to put it to use.

Iā€™d be curious to know what your analysis of the plastic dust chute would be. Itā€™s a consistent section for most of the body, but there is a transition to the hose adapter. This is the section along the center-line:

Thanks for the welcome. First off let me establish my bonafides or perhaps lack thereof. Iā€™m a finish carpenter by trade, although I spent 3 years in college never learned anything useful much less about aviation. What I know Iā€™ve taught myself just by studying as much as I can about airplanes although I do have my pilots license. I can only run basic math formulas. My plane is kind of a Frankenplane meaning I bought plans for different homebuilt aircraft and borrowed elements from them as needed and combined them into a unique design. The wings and tail came from one airplane, fuselage from another, landing gear from another etc (all with a few minor tweaks). I have been careful not to do anything out of the mainstream. Itā€™s a very good looking plane but nothing that others havenā€™t done before. Iā€™m not an engineer and donā€™t claim to be one so most of what Iā€™m telling you is mostly gut feeling with a basis in fact.

Having said all that here are my thoughts on your design and possible improvements that could be made:

The red lines are places of less than optimal airflow (again a only slightly educated opinion).

If it were mine I would raise the exhaust end a couple inches to allow a more gentle transition without any major direction change in the airflow. As you can see from the top sketch I think the way you take the air up and then back down a little (on the bottom side of the duct) is going to cause some loss of efficiency. Also I think that the big direction change to go up immediately at the intake opening is a problem.

For those interested in aerodynamics here is a video of a guy that went 213 mph on 65 hp in a plane he designed and for awhile was the record holder for speed in his category. https://youtu.be/rxvoDbZpoY8

He does a good job explaining in laymans terms about aerodynamics, although Iā€™m not sure any of it is directly applicable here I find it fascinating. Heā€™s one of my big inspirations for designing my own plane. If youā€™re interested in flying go to your local airport, most FBOā€™s (Fixed Base Operators, operations that sell fuel sometimes have maintenace facilites and rent aircraft and employ flight instructors) have a discovery flight program to get you your first flight for anywhere between $60-70. Itā€™s a lot of fun. Or look up your local EAA chapter someone there will probably be willing to take you for a ride.

If youā€™re really interested in optimizing the airflow you might consider printing your part out of clear plastic and running water through it with a streamer of food coloring, anywhere the food coloring starts to disperse you have flow separation. Thatā€™s probably a lot more work than itā€™s worth but if youā€™re interested in such things it would be fun to see. Also it will be interesting if you kept the different iterations and found some way of testing them to see if I really know what Iā€™m talking about or just blowing smoke. Maybe dump 1 cup of sawdust say 2 inches from the intake of each design, turning on the vacuum and seeing which one sucks it up the fastest.

2 Likes

Iā€™ll just throw in that the air speed through that duct is slow enough, that there may not be any separation issue at all in Maynardā€™s design. But thereā€™s no harm in avoiding the question using @Willja67ā€™s suggestions.

Flow separates when there is an adverse pressure gradient (meaning the pressure locally is increasing as the air goes downstream). The best way to avoid an adverse pressure gradient is to keep the local stream tube area constant or decreasing. A wall that is curving away from the flow produces these bad things while a wall that is curving toward the flow produces a favorable pressure gradient.

A sharp corner toward the flow (figure below) is okay, although the flow will form a small eddy in the corner, but that wonā€™t amount to much loss.
Screenshot%20from%202018-05-28%2012-00-57
Source: NASA TMā€”2010-216771

A sharp corner away from the flow will almost definitely cause a separation and be lossy.
image
Source: https://www.grc.nasa.gov/WWW/wind/valid/backstep/backstep.html

Incidentally, in the aerospace business we use computers to simulate flow around and through objects. Itā€™s called Computational Fluid Dynamics (CFD). When I started my career, CFD was brand new and got lots of things wrong. Now, 30+ years later, CFD does a much better job on most things, but it still has a hard time predicting separation.

2 Likes

@Willja67 This has been a learning experience for me for aerodynamics. I can see how the changes youā€™re proposing could have a positive impact on airflow. However, Iā€™m going to continue moving forward on the version I have already printed for testing. Iā€™ll be testing each of the dust chutes with an anemometer to quantitatively measure the performance of each system. When I get to designing a MK3 dust chute, I will take your suggestions into consideration. I would like to update the system for a 3" hose at some point and airflow optimizations such as what youā€™re suggesting could go a long way towards giving me the needed CFM.

I could be entirely wrong, but Iā€™m going to try to defend my design a little. I like @jwolterā€™s point of air speed affecting the amount of air separation, although that could just be because itā€™s less work for me. >.>

The section I used provides more flow than the hose that it is rated for. The section should provide somewhere around 154 CFM, whereas a 2 1/2" hose is rated to 140 CFM. If there is some efficiency loss due to air separation, it may not be drastic enough to slow the air more than the maximum airflow in the hose.

The last item I have in support of the MK2 dust chute is that it is designed with the Maslow in mind. Looking at just the dust chute may give some idea for air flow, but itā€™s not the entire picture. The chute is mounted on top of 3/4" plywood and a 1/4" pad of HDPE, which sets it 1" above the work surface. This means that the air doesnā€™t only come from the ā€œfrontā€ of the chute, but also below it. Also, the Maslow is set up at a 10*-15* angle, so the dust would have a tendency to ā€œfallā€ down to where the dust chute is. I didnā€™t go into as much detail with my airflow diagram, but this is how I visualize it:

There may still be some of the air separation you showed at the bottom and in the tight radius, but I donā€™t see it quite as extreme as what you had shown.

Any time I see NASA as a source on aerodynamics, I know itā€™s legit! xD

3 Likes

I just want to influence everyone to be pragmatic. The dust chute would likely be useful as-is. While optimizing the design based on CFD simulation may provide some benefit, the benefit may not be worth the effort. We should avoid rigor that does not produce substantial improvement.

4 Likes

I agree @Joshua. Iā€™ve been enjoying reading along and learning about air flow, but letā€™s keep in mind that the stock ridgid router dust collection runs through a 1 inch inlet which seems to work OK so weā€™re miles ahead of that

2 Likes

I agree completely with @Joshua and @Bar here. I am not changing the design of the 3D printed chute for now. I am going to make the recommended change to the wooden one because I feel there will be a bigger difference in the performance.

Regardless, any of the systems we have been looking at here would be better than the stock Rigid base.

I will eventually look into a system rated to a 3" hose, as that is typically the hose size of a commercially available machine. In that design, @Willja67ā€™s suggestions will help me tune a more effective system. But thatā€™s a ways down the road yet. Iā€™d imagine that will be the chute used in the aluminum sled, that I will probably be building 5 years from now (being pessimistic).

1 Like

I am almost done with this sled. I have a couple of questions:

Can I use the stock ring and rollers for the chain hookup or do I have to go with what the design shows?

What firnware updates need to be made and how/where do I make the changes?

Canā€™t wait to try this out. GREAT idea!!!

1 Like

I love so much that the ring is now considered ā€œstockā€! :smiley: I have not made any predrilled holes in the cut file for the ring, but the design should be able to accomodate it:

image

If you have any feedback for fitting the ring, let me know. I have not tried it myself. Also, if you have any feedback for building the sled, Iā€™d like to hear it as well.

The only change that I can think of that would need to change is your Z-Axis pitch. I measured a lead of 4.7625 mm from this kit, and I believe it is a 4 start screw. That would give you a pitch of 1.1906 mm. Also, make the pitch negative in Ground Control to reverse to motor direction, because of the gearing.

Also, not sure if the new dust chute model is anywhere useful to you guys. I thought I saw in the email from Bill that he was asking about the dust chute model. I need to take a minute to clean up the community garden page, so in the meantime, here is the STL file:
Dust Hood Chute V2.stl (593.2 KB)

Keep in mind the new chute is larger than the last, so you may need to cut away a bit of the spine to get it to fit.

1 Like

OKā€¦Would be glad to give feedback I need to mount the ring and make the changes to the firmware yet so hopefully try this out in the next couple of days.

2 Likes

You forgot one more setting. The Z axis mm per revolution. Right now, I set the Z axis to move 5mm. It moves 64mm. I have the lead screw setup with a smaller gear than the drive motor. Where do I go to set the correct mm per revolution so the z axis moves the correct distance?

Sorry, itā€™s been a little bit since Iā€™ve been in the settings page. I remembered the setting just being the pitch, but I was incorrect. With equal size pulleys, this setting should be -4.7625.

image

Thatā€™s not a small difference! The axis is moving 12.8 times more than it should! D:

By my math, with the setting at 1.1906 (which is what I previously stated), the motor would need to turn 4.2 times. With the correct setting, that should need to turn 1.05 times. That error alone would give you 20mm of travel (my bad).

Whatā€™s youā€™re gear ratio? You would need to divide that by -4.7625, the mm per rev of the screw.

Example:

Motor: 10 tooth
Lead screw: 30 tooth

Ratio: 0.333
screw mm/rev: -4.7625

Final mm/rev: -14.2875

I believe I got the 10 tooth and the 20 tooth gear. I did set the pitch to -1.1906 as you previous suggested. So, instead of the -1.1906, what should I set it to?

I hate math! :smile: but I got -9.525. Can you please check me?

1 Like

Honestly, based on your 5 mm input and 64mm output, try -15.24. that should get you in the right ballpark. Youā€™ll probably need to tweak that a little to get the accuracy dialed in.

1 Like

OKā€¦just to confirm. Your suggesting that the Z axis pitch be set to -15.24 to start and adjust further if needed?

1 Like

Thatā€™s what I get for your gear ratio as well.

I only think it would be somewhere around -15.24 because of the amount of travel you got last time.

If I were you, I would first try -9.525 and see how much closer that is. If itā€™s still not right, then try the -15.24 and see if that is better. It might be that one of the two is close, but not quite right. If that is the case, try changing the pitch in small increments both up and down from your current value until you have the desired accuracy.

For example, letā€™s say you found the -15.24 to be close. But it moves the axis 6mm when you want it to move 5mm. So youā€™d want to change it by .5mm and see how it affects your calibration. So first you try to increase your mm/rev to -15.74 and found that now you move 6.5mm. Next you would try -14.74 and you find that you now move 5.6mm. You could continue to iterate from there, reducing the mm/rev until it moves 5mm.

The difference between the numbers could mean that the lead that you have is different than the one I measured on mine. Are you using the 8mm lead screw from the 400mm travel kit that I linked in the BOM? Itā€™s also possible that my measurement was off by a little, which would translate to a lot of error across the gear ratio and several rotations.

2 Likes