This is what happens when you have an extra 0 on the wrong number...

This is what happens when you have an extra 0 on the wrong number… I was jogging the x carriage (is that the right name?) up to get access to the screws, had the jog distance set to 100mm instead of just 10mm… It tried to send the carriage through the top of the frame! I think it only went a little higher than “limits”, either the software detected being at size limit or firmware detected failure/resistance on motors. Makes me wonder if octoprint bothers limiting you to whatever print size you have set when manually controlling it… I killed the power right away but it was too late, the z motor mounts on both sides cracked, that was the saddest creeeak sound I’ve ever heard :frowning:

Hopefully I can get ahold of acrylic replacement parts from @Diego_Porqueras_Deez , I wouldn’t trust those parts in ABS or PLA… Hmm, maybe T-Glase would be strong enough? Printer might fail before then though as I haven’t gotten the settings dialed in all the way for t-glase, was still experiencing jamming problems. It still prints for now but I have a feeling if I push it too hard I’ll hear a loud CONK as the z-motors fall off…

Which printer model?

BukoBot 8 v2 duo, circa late 2013

Oh and if you can’t see it, zoom in on the pics, look for the line going through the clear plastic parallel to the Y axis (perpendicular to the frame).

On my printer the limits are set in the firmware making this sort of thing impossible. I thought that was the norm?

@Daniel_Bull I thought so too! If it does, perhaps the amount the print bed is raised up is eating into the build volume by a few mm giving it enough to do some damage…

I had to adjust my firmware and recompile to match the printer, maybe your firmware and hardware have different ranges?

First I would try is use this acrylic specific glue you can find locally. It’s very strong and should prevent from you replacing anything. If not, we can send you a new set…

Sorry about the powerful Z motors… With power, comes responsibility… ; )

@Diego_Porqueras_Deez Thanks I’ll give the glue a try!

Acrylic is one of the worst materials to build printers from, you would be better off with a 3D printed part or wooden parts.
The only good thing about acrylic is it’s stiffness and the looks. (The stiffness is why it cracks so easy)

@Daid_Braam I disagree with you. If the design is engineered well (like our Bukobots), it’s a great material to work with, you just need good thick stuff. Wood is a horrible material to work with, it’s very inconsistent and warps with heat and moisture (3D printer enemies). As for 3D printed parts, they might be good in some cases, but the if you have a Reprap style machine that can’t be squared tp precision very easily, it’s doesn’t work well either (especially if you’re stuck with low temperature PLA). Plus, the time for laser cutting to a production part ratio is a lot better than 3D printing a part…and you get a solid piece.

@Diego_Porqueras_Deez Then why did this printer fracture? Then why did every single feeder on acrylic Ultimakers fracture? (And had to be beefed up a lot to stay in 1 piece)
I’m not saying 3D printed parts would be good for larger scale production, but it can fix his printer right now, with the tools he has.

And plywood does not warp. It only expands and contracts a bit, but no warping. We tested this by repeatedly drowning and drying an Ultimaker. It’s also as quick, or quicker to cut as acrylic. And it’s consistent if you go to quality wood suppliers. And it does not fracture on stress or impact like acrylic does. (And it smells like a campfire after it’s cut. Which is great)
However, Derlin might be an option, if you are more into the laser-cut plastic then the wood options.

But, all discussion is useless as soon as you said “the design is well engineered like our Bukobots”. As you’re clearly not open for discussion and improvements. (I for one, see a lot of design problems in the Ultimaker. And think Cura, my own software package and baby, is crap)

@Daid_Braam Most machines are designed to a withstand certain loads and pressures (or at least should be). Unfortunately, sometimes this they are taking beyond the range because of other factors (like lack of software safety checks?), but that doesn’t meant it’s not tough enough for the job it was designed to do. In this case, the Z travel was taken beyond what it was designed for and with extra powerful motors and good torque on the Z screws, something has to give. If you want an unbreakable printer, would may need to make every piece out of steel. And yes, we use delrin in some areas where needed…

We are always open for discussion (which allow for disagreements too) and we thrive on improving things, but in this case, problem is not the acrylic, it’s was the software and controller that didn’t tell it to stop before the limit…or maybe we should just cut down the Z screws a bit more to not go beyond and lose a little height…(By the way @Jonathan_Foote if you loosen the top bar and remove it or slide it up a bit, you can have an extra inch of build height…)

I hate to tell you, but no design is perfect, nor will ever be. We all know this, but we aim for some goals we think are important and we try to reach them the best way possible. I can’t say our machines are perfect, but we do try hard to make them the best possible while keeping them robust, affordable, and open.

But despite this little discussion, we do love your Cura slicing engine and the accuracy it has, we definitely don’t think it’s crap. Although not perfect (which no slicer is anyways), we do think it’s very high quality as far as the math, options and the intelligent algorithms. Either way, thanks for making a great piece of open source software!

We are also looking forward to your new Ninja Kittens project for the CNC world (which badly needs a simple but good G-code generator that hopefully can work on Mac)…

But just like your motivation with your Ninja Kittens, when we experience something important that we think should be better, we try to make it better.

On a somewhat related note, if you don’t know about him, check out Panucatt devices (Panucatt.com) He’s making some great CNC controller boards that you may want to work with (I’m sure he may consider some freebies for you).

I don’t think that any material is “good” or “bad” necessarily. Good design is ultimately the prize to be won, and the practical limits of each material is a consideration. Wood scores well with strength and flexibility, but relatively low on the dimensional tolerance due to expansion and contraction with temperature and humidity changes, etc. With the addition of “binders” such as in plywood, this may change depending on what binder is used.

As a tip… try adding some sort of flexible gasket material or rubber rings between binding points like screws to avoid straining sensitive and breakable acrylic parts. Flexibility may alter overall accuracy, but it does keep your 3d printer from breaking as easily. Also remember that every moving assembly will break at its weakest point. It’s best to make that point the cheapest and easiest to replace.

An emergency endstop could be handy… Only requires 1 micro switch for each axis and they are all connected to a relay cutting power to the printer

I find it “odd” that firmware can auto-bed-level, filament change and have other cool features but does not do a simple check of endstop status before issuing steps to the motor. Surely that would prevent so many crashes like this?

@Diego_Porqueras_Deez Ninjakittens is a side project. I’m currently collecting SVG and DXF files which do not load correctly yet. Which, sadly, I have a lot of right now.
But it is producing GCode already, and should work on the Mac (but no release version yet, due to general incompleteness)
I’m using it a lot to generate laser files for our Trotec laser, which has one of the most horrible pieces of software ever to generate those normally.

@Neil_Darlow The firmware has that feature, but might not be properly configured.

@Daid_Braam Exactly. It seems to me the assumption made by firmware authors is that printers will be attended. In these days of Raspberry Pi and OctoPrint the opposite is often the case. Features such as endstop checking and e.g. Pronterface’s “Clamp Manual moves” should perhaps be enabled by default.