3D Printer problems - Monoprice/Wanhao D6

[Steamchick]

I like the idea of using 3D modelling for making casting patterns.....
And a guy made all sorts of jigs and fixtures for the assembly line at work...
And rapid prototyping speeded-up new model development dramatically.
But I understand that the polymers available are not all as good or durable as the "original" designers select? (Especially when compared to fibre filled materials, ABS, thermosets, etc.).
But I haven't bit the bullet yet and bought one....
Any suggestions... ? Maybe a "toy" for starters? And what size PC is really needed for CAD? (Mine is from 2008).
K2

 

[Shopgeezer]

I am in my third try learning Fusion 360. I have used Sketchup for years so I am
familiar with the concepts. But the poor training videos for rank beginners turned me off and I found Fusion too confusing and complicated to understand. Fortunately I found the series of Fusion videos on You Tube by Paul McWhorter. He pretty much starts with how to move the mouse so really gets back into what a beginner needs to know. The videos are great and I am getting to know Fusion better. Still an awful lot to learn. I highly recommend the McWhorter videos.

DonM

 

[awake]

To create the 3d designs, I use two different programs on a regular basis: OpenSCAD and FreeCAD. They are drastically different! In FreeCAD (analogous to what you would do in Fusion 360), you "draw" the model, step by step, seeing the results as you go. In OpenSCAD, you tell it what you want it to do, then click a button to ask it to render the results.

Let's say you want a 20x20x20mm cube with a 16mm hole through it. In FreeCAD, you could do this in a couple of different ways, but typically you would sketch out one face of the cube, and then extrude it to form the solid. Then you would sketch the hole on one face of the resulting cube, and extrude that to make the hole. Here is a rough 2-minute video illustrating the process:

With OpenSCAD, you write a "program" - just a text file with the extension set to .scad - to put together the various pieces of what you want. For example, to make the same cube with a hole through it, you would write the following in the OpenSCAD editor:

difference () {
cube([20,20,20], center = true);
cylinder(r=8, h=30, center = true);
}

Then you click the icon or press the F5 key to tell it to render what you just asked for, and you will get the following:

It is indicative of the differences in the approach that I could illustrate the OpenSCAD process with a single screen shot, while the FreeCAD process required a video. I find that each approach has its advantages for different 3d printing projects, and I instinctively turn to one vs. the other ... but I have a hard time articulating exactly what it is about any given project that makes it a better fit for one approach over the other. Over all, though, I find myself using OpenSCAD far more often for 3d printing projects, but not always. (Note that for anything other than 3d printing, at least for me, OpenSCAD is never the right choice. I do all of my design and modeling for woodworking, engine, and tool projects in FreeCAD.)

 

[Randoo]

I am in my third try learning Fusion 360. I have used Sketchup for years so I am
familiar with the concepts. But the poor training videos for rank beginners turned me off and I found Fusion too confusing and complicated to understand. Fortunately I found the series of Fusion videos on You Tube by Paul McWhorter. He pretty much starts with how to move the mouse so really gets back into what a beginner needs to know. The videos are great and I am getting to know Fusion better. Still an awful lot to learn. I highly recommend the McWhorter videos.

DonM

I've been trying to learn Fusion 360 also. One of the issues I've been running into is that the controls seem to change from when older tutorials were made. When shown how to do something I have to search Google to find where the control is now or what it's now called. Then the worst part, I have to remember! Slow and steady I guess.

Randy

 

[ddmckee54]

Randy, that's one of the things that I hate about Autodesk, they keep moving the commands/icons around. Somebody had a better Idea?? I'm not so sure.

I started out using ACAD10 I think, it might have been 8 or 9. Lately it seems like every time they come out with a new version everything gets moved around. Autodesk is the parent to Fusion 360 and AutoCAD. Frankly with the rates that Autodesk charges, I'm surpirsed that there's as MUCH of Fusion 360 still available at no charge as there is. I think that it will probably go to 100% subscription in the future, just like Dassault did with DraftSight.

I'm not QUITE ready to start compaigning to return to clay tablets and sharp sticks - but I'm gettting close.

Don

 

[Steamchick]

Hi Don,
I have yet to make the move from "pencils to pixels" - but did some training courses (and daily work) on CAD... - not designing just sectioning others' designs and dimensioning for technical documents and pictures. But every couple of years we needed new training courses as the CAD changed (completely!) and we had to start learning again. I reckon that I spent more time "re-training" than actually producing drawing output. But the full time users always thought the "improved tools" were "Great!"
But my modelling design work is still done by pencil... It is what I enjoy and stress free... Surely that is the nature of hobby work? - If stressful and not enjoyable it is not the right hobby.
To that end, I am using a 12 year old PC to write on this thread - all be it with Windows 10... but I prefer it to an "Android". (My mobile phone doesn't even have a camera, but allows me to talk to people... or find out they are not available for a chat!).
I'll be dead before CAD gets easy to learn. Maybe I'll have "Pre-CAD Engineer" engraved on my headstone?
I used computers - when you needed a clever young Engineer to write the software - to do simulations and design work of pressure rises in the various chambers of high speed pneumatic valves and cylinders - but then they invented PCs and I went back to calculators and brain-power (PCs and I were simply innadequate for the first decade or 2!), as the generation of engineering software writers gave up Engineering for the huge salaries in "IT programming" for banks and people like that!
Of course, the advancement of technology is now enabling Boeing to design the modern Concorde - same size and similar range as the one from 50 years ago - in roughly the same timescale and with a similar number of engineers - and using computers! - Is this the technological revolution I expected in my lifetime? Is it just possible that we are spending more time "worshipping to god (= computer) than tilling the fields for sustainance"?
But this PC does let me "communicate with" (= "preach to" ??) all you interesting guys, so not all bad!

K2

 

[RM-MN]

I'll be dead before CAD gets easy to learn.

I doubt that CAD will ever be easy to learn but it has so many advantages over pencil drawing for larger projects that people will work to learn it. Along with that comes CAM for the CNC machines. I found CAM to be just as frustrating as CAD but since it can be done once for a part or assembly and the results fed to the CNC it will be the norm for the future. Making a single part takes more time to draw in CAD and program it for CNC in the CAM than just pencil and manual machining it but when you need several thousand copies of that part, manual machining is gone.

 

[Steamchick]

Having worked in industries that made one-off bespoke installations (for Power stations and factories), low volume products (high voltage switchgear), and high volume (Automotive) manufacture I agree whole heartedly that "computers" win in the professional world. From 1-off to high volume. But I doubt it will ever enter my home workshop. My basis is 1-off models. And even chalk on slate is useful for the simpler things I draw freehand before making them.
I left the drawing board behind when I movedcto automotive high volume... and didn't need to join the CAD -CAM revolution as it developed. Dedicated new young engineers did that, and I only use my "Old" skills on the drawing board for my own needs for models. I enjoy the application of the human skill of making parts that fit - maybe not perfectly, but as well as the capability of my brain, the machine and my hands. I suspect there are 2 schools. Those who want their models to be "as perfect as possible" for whom CAD and CAM are suitable, and those who focus on their own skills, where simple tools like pencils, rulers, manual calculations, scribers, squares, files and saws, as well as hand-set machines, suit the task. My 1978 Texas calculator is probably the most complex computing tool I use. But I may explore CAD when I need a new challenge.
Enjoy whatever you do and how you do it.
K2

 

[RM-MN]

You left out my category, those who are doing only 1-offs but have poor fine motor skills so that making parts that fit together manually doesn't really work well and the ones who do additive machining (3D printing) as well as subtractive machining (mill and lathe). Learning some CAD drawing has made my 3D printer more fun as I make items for the home and farm that I cannot buy locally and has made my milling much closer to what I intend.

 

[kf2qd]

You all fail to see what CAD is, it is just another form of you paper napkin or chalkboard. I do most of my design on CAD, things are more legible, I can modify the design to see how the mod works out and go back the the previous better one easily. I can work in 2D and 3D and they are very simple.

What I think your problem is - you have not used it enough that is becomes normal. You have not learned the language of your particular cad program (different, but simple). I fought with doing simple things in AutoCad until I learned about a few additional keystrokes that solved 90% of my problems. CAD makes a very nice way to share your ideas. Someone else can see it with the same precision that you do. Hand drawn might be crystal clear to you, and clear as mud to the next guy. And you can backup you design in various places and on various media. Stash them on a CD, on the cloud, on an SD card.

And you can try things in CAD that are very difficult on paper. Was recently working out the dimensions of a Daisy Wheel and Tri on a clock design. Could copy and move the TRI through the equivelent of 12 hours of motion and leave a copy of the TRI in each position which then let me see the clearance between the Tri as it indexed around for 1 hour. Also let me see the effect a different offset eccentric would have on the operation.

If you get into CAD either find a Book (can still be difficult because you still don't know the lingo), Find someone that will put up with a few dumb questions and ask them how to do a specific thing. (I can think of learning about the R(relative) keystroke when rotation a part. Saved tons of grief. Once you have a few basics of the language it really gets quite easy. By the way - your training sessions probably taught you a bunch of stuff you will never need, and jumped over the basic stuff making it appear unimportant. 99% of CAD is just the basics.

 

[Steamchick]

Fair comment RM-MN, but there will always be lots of little alternative "groups" to the main ones I suggested. I didn't mean to disregard anyone, just explain that without machines capable of CAM, and the need for CAD, there are still some of us making perfectly good models, using more "traditional" methods. Perhaps we are the obsolete breed? But our forbears sent men to the moon and back, flew faster than twice the speed of sound, found oil in vast quantities and made plastics, that many rely upon and are now becoming a pollution problem, made bombs that could destroy the whole planet in minutes, etc., most of which was done without the aid of a computer. Incidentally, we are communicating by a faster version of the Morse code telegraph - that we call a digital computer. And we can now send pictures too!
By Population, the "Modern post-transistor Mechanised" world is probably still in a minority on the planet - I have never thought about it until now. The fact that a good proportion of the planet relies on diesel fuel or electricity for power does not mean it is available to everyone, nor needed by everyone (Consider farming the wheat fields of Canada versus the paddy fields of China, and the sizes of populations depending on the 2 forms of planting). The same applies to "electric thinking machines" - and the "Home-machinist". Of 4 feeds on my lathe, only one can be powered, and none on my mill. Yet I get "adequate" material removal rates, precision of fit and surface finish from "hand-feeding". I am sure CAD-CAM would produce parts hundreds of times faster, but for my "one-offs", the simplicity of my process keeps me occupied. That is reward in itself.
Enjoy "making", however you do it! I do...
K2

 

[ddmckee54]

I can relate to both sides of the CAD war. To some people CAD is the greatest thing since sliced bread and NOTHING is done without the aid of the computer and the latest software. Then there's the other end of the spectrum where CAD is Cardboard Aided Design.

I can relate to RM-MN's comment about the lack of fine-motor skills, BOY can I relate to that. A lot of my stuff now is CNC, or at least headed that way. CNC allows those of us with limited dexterity to enjoy this hobby.

I used to be a fair to middlin draftsman, and I also used to be able to read my own writing - not so much anymore. Anymore if I want to be able to comminicate an idea to someone else I have to use CAD, just so it's legible.

But enough of this nonsense, back to the salt mines... Ummmm, I think I meant 3D printers. Over the weekend I decided that no glorified hot glue gun that's controlled by angry pixies was gonna get the best of me! So I started from scratch. Alright, alright, SOMEHOW I managed to delete my printer profiles. No problem says I, I'll just reload my factory file in S3D and start from where I last saved it. I'm not sure IF I saved my factory file, and if I did I SURE don't know where it got sent - as I could never locate it. So I had to rebuild my D6 printer profile, and build my MP10 printer profile. By Sunday noon I had both printers calibrated to within a few hundredths of a mm on my 20mm test cube, so I started printing some REAL parts.

I turned the MP10 loose on the parts for the RC Benchy that I found on Thingiverse, and I turned the D6 loose on some dust collector parts - both printers were happily pooping out plastic parts. Only problem, when I tried the parts for the DC they wouldn't fit - WTF!!!! I had calibrated my 20mm cubes using PLA and I had gottten 20mm, why am I off when trying to print a 108mm adapter ring? Turns out I had used 106% as my scaling factor during the calibration of the printers, allowing for shrinkage dontchaknow. Much to my surprise, when I Googled PLA's shrinkage factor it's more like 102%-103% - OOOOPPSIE. That means both the D6 and the MP10 have incorrect steps/mm values and I'll need to fix that problem, just not right now.

By this time I'd already printed part of the RC Benchy and the MP10 was in the middle of a 14 hour print for another part. I'll print the rest of the RC Benchy parts, then fix the issue.

The D6 I just said screw it, I tweaked the CAD until I got the printed part I wanted. When I'm done printing these adapters I'll actually FIX this problem, not just put a band-aid on it.

Don

 

[Steamchick]

Hi Don. I do understand why you fight the gremlins to make plastic bits that way. But having made them 4% too big, can you not machine or fettle to size? After all, it is just a cast part? If you have a CNC miller, it should cope with the same CAD input?
But I am just a cardboard cut-out anyway, so no nothing of the complexity of your tools.
K2

 

[RM-MN]

Hi Don. I do understand why you fight the gremlins to make plastic bits that way. But having made them 4% too big, can you not machine or fettle to size? After all, it is just a cast part? If you have a CNC miller, it should cope with the same CAD input?
But I am just a cardboard cut-out anyway, so no nothing of the complexity of your tools.
K2

Some times the wall thickness on a 3D printed part is too thin to be able to machine it to size without exposing the honeycomb interior. Along with that is the problem of clamping them tight enough to keep them from moving without cracking them.

 

[Steamchick]

understand

 

[awake]

Some times the wall thickness on a 3D printed part is too thin to be able to machine it to size without exposing the honeycomb interior. Along with that is the problem of clamping them tight enough to keep them from moving without cracking them.

Of course, you can control the number of perimeters to give a thicker wall ...

 

[ddmckee54]

Steamchick:

For right now on the MP10 that's kind of what I'm doing. Since I've already got many print hours invested in the Benchy using the incorrect scaling value, I'll continue using those settings until this project is complete. I may regret this decision as there are a number of commercially available products that will need to be installed and I'm not sure these will fit. Awwww Crap!

I think I just talked myself into fixing the step/mm on the MP10 first and then reprinting the parts I've already done. Redrilling/shimming the printed rudder tube and prop shaft holes might just be a hastle I want to avoid. The servo brackets probably wouldn't fit either, and who knows what else. YUP, it's ALWAYS the little things that bite you in the butt.

It's kind of the same thing with the adapter rings I'm printing with the D6. In the NEAR future I will get the D6 steps/mm set so that when I tell it print an object I will get an object of the correct size. That way in the future if I have some other brain fart brilliant idea and print a different 108mm object, they will fit together with a minimum amount of swearing at the STUIPID designer. I hate having to post-process parts, de-burring is expected, but having to file/sand parts to fit should not be required.

I'm also thinking that I need to print a larger object as a sanity check, maybe a 40x40x40mm hollow cube with thin-ish walls and no top. That would still print fairly quickly and give me a sanity check on size. I could even add lightening holes to the sides and bottom to speed up the print further, no need to waste TOO much plastic. I'll have to use my digital guesstimator to measure it though, my mike only goes to 25mm. To paraphrase Bullwinkle - No doubt about it, I need another mike.

Don

 

[Steamchick]

Hi Don,
Gee, you are not having fun just now. Have a day-off and relax then come back fresh and start again.
Do you have a vernier caliper? - Or even hand held external calipers, and a suitable gauge? (Maybe a 40mm diameter ball bearing, or some other ground tool or component of known size? - A light bulb can be used, if you know the size!). Or you can use a gauge made from 2 pieces of 20mm square ground bar... or 2 ball bearings smaller than 1" - but that add to the size of calibration cube you want to make?
Remember, Toolmakers use slips of various sizes and add them together to make gauges for comparing to the tools the are sizing. We don't need "dial gauges" for everything.
Stay cool mate, and enjoy the work - It's a hobby!
K2

 

[ddmckee54]

Steamchick:

On the contrary, getting these printers calibrated IS having fun. I went out to Horrible Fright last night, and they had a micrometer set : 0"-1", 1"-2", and 2"-3" for $40. So I figured I'd give it a shot, I wouldn't have wasted too much if they were crap. So far from what I can tell they aren't too bad, they even include 1" and 2" gage blocks. Every thing that I've got that can measure the 1" gage block agrees that it is indeed 1", even my digital guesstimator can't make up its' mind whether the gage block is 0.9995" or 1.0000", I'll accept it as 1". The 1"-2" mike says the 2" gage block is 2", and my digital guestimator says it's 2", but the 2"-3" mike says it's 2.005", I'm guessing that whoever zeroed the 2"-3" mike was having a bad day that day. They included the spanner to zero the mike barrels with the set, so I'll just fix that little issue. I always consider anything coming from Harbor Freight to be more of a tool kit, rather than a ready to use tool.

Anyway, I worked on the MP10 calibration last night, I got the 20mm cube to within a thou or two, which is about as good as I think you get with an FDM printer due to the way they build an object. As a sanity check I designed a 40x40x40mm cube and printed it. It's off by about 3-4 thou, so I'll do another tweak to the MP10's steps/mm tonight.

I'm not going to go too far down this calibration rabbit hole, if I can get the 40x40x40mm cube error to under a thou I think I'll call it "crose enuf". I am NOT gonna print an 80x80x80mm cube just for funsies, I wouldn't be able to mike it anyway since that's over 3". I might try a 60x60xsomething mm plate, THAT I could measure.

Once I get done printing 4" spacers on the D6, I'll go through the same procedure with that printer. The spacers take a little over an hour to print, and if I decide to make new blast gates for the dust collector - probably gonna happen - I'll need 6 more 4" spacers.

Don

 

[willray]

I'm not going to go too far down this calibration rabbit hole

You might already be further down the rabbit hole than you want?

If you already know this and I'm just insulting your intelligence and understanding, please feel free to ignore me, but:

Keep in mind that a couple thou difference comes from the extruded ribbon not having a nice accurate rectangular profile of the intended dimensions, so you've got a (pseudo) fixed extrusion-squish error that's added to your axis-calibration error.

Try to calibrate that fixed additive error out with the scales-with-size axis adjustment using small-size prints, and you'll get increasing errors at larger print sizes.

If you've got a mill on which you trust the dials, make yourself a set of go/no-go gauges of the largest possible sizes for your X/Y/Z axes, and just print strip prints along each dimension, then check them to the go/no-go gauges with gage blocks. That way the fixed squish error is as diluted as possible in the axis calibration.