Dollo 3D Printable 3D Printer

[spegelius]

Prusadollo progress: as I don't have the original aluminum bed carriage, I designed printable version based on the original. Of course it should be quite stiff so I printed it with ColorFabb XT-CF20 carbon fibre filament. Result is quite stiff, especially since I used 0.8mm nozzle, 3 perimeters (2.4mm), 1.5mm top and bottom surfaces and 25% infill. XT is bit finicky to print since it likes to stick to the nozzle and leaves blobs in random positions, so I had to babysit the print somewhat. Also had a fail when trying to print with 0.15mm layer, the material just wouldn't stick to the previous layer. Ended up printing with 0.25mm layers (KISSlicer synamic layers 0.25-0.3) and the result is more than good enough.
XT's glass transition temp is ~80C so I printed the corners with ABS, so that the heat from the bed should be ok even when heating the bed up to 100C. I do have insulation under the bed PCB, so XT should survive fine.

[spegelius]

Received some stuff I ordered from Aliexpress (stepper motors) and decided to do some assembly on PrusaDollo. Got the Y-axis, X-axis, Z-axis and LCD mount mostly done. Bed carriage moves properly on Igus dryling bearings. The smooth rods could be bit tighter, but shouldn't be a problem. Belt installed and tensioner solution seems to work nicely. Also the frame seems to handle the belt tension without problems. The knob for the LCD needs bit of redesign, it seems that this particular LCD assembly has a bit longer shaft on the encoder so the knob is quite high.
Z-axis needed some ad hoc remodeling, couldn't move the X-carriage all the way up and had to saw some material off of the top z mounts.
For X-axis and lower z mount parts I needed to use some painters tape to make the bearings and rods fit tightly. It seems that the Prusa parts are meant to be printed with Slic3r PE; KISSlicer does more accurate parts IMO, so couple of things are bit too loose.

The Z screws have 7cm of unused length. Might buy longer smooth rods and print more extensions to get full use of them.

[Roberts_Clif]

I too made a Prusa Dollo would like to see your STL enhancements are you going to release to the forum.

IMG_1811.jpg

IMG_1812.jpg

[spegelius]

Nice, looking at your build makes me think I could have gone a bit different way like the corners and rod holders. Have you got it running yet?

All files are here: https://github.com/spegelius/Origina...tree/MK2_Dollo. I'll be doing some fine tuning and PSU cover and RAMPS enclosure need to be done too.

Still missing few pieces: GT2 pulley for the X-axis and some M2x12 bolts for the endstops (probably could get something hacked together, but I'll do this properly). Also RAMPS fw needs to be modded from Prusa sources, which should mean basically just adding RAMPS pin assignments and tuning the steps. Found this: https://www.thingiverse.com/thing:2607463, there's the pins_ramps.h file I need.

[spegelius]

Got the PSU cover redesigned and printed for bigger PSU (not wattage, but physical size). Mount and wiring also done. PSU seems to work, adjusted it to 14V for little extra performance.




Modified the Prusa firmware for RAMPS and so far so good: display works, temperatures seem ok and the fan extender I bought a year ago also works after configuring the correct pins for it (pins 6 and 11, normally meant for servos). So now I can have 3 fans that are controlled by the FW: hot end, part cooling and electrinics cooling.
Took the RAMPS from my first Dollo build, going to put Megatronics in it someday.



Also modified the RAMBo cover for RAMPS. As my other printers are occupied currently, decided to try printing the parts with the new Dollo. And the results are good enough, even though they aren't top quality. What's important is that they are dimensionally correct.



Currently assembling the extruder and hot end. Getting close to actually testing this thing. Wiring won't be finished yet, nee to design some clips for wire routing.

[spegelius]

Got things wired up for some testing. Everything seems to work after fixing few things in FW, like steps for 0.9 degree motors, z-height and display button rotation direction.
Went through the Prusa printer selftest, which was pass. One thing the Prusa selftest is missing, though, is a smoke detectio; heatbed mosfet shorted and smoke ensued. My bad, didn't check the MK42 bed resistance which apparently is low enough to cause too much current to be pulled through RAMPS mosfet. Luckily I have spares so new one is soldered in and I'll just limit the max power in FW so everything should be good... hopefully.

[spegelius]

Finally got the damn thing printing. Took some effort, electronics was really the thing that caused some problems:
- as mentioned previously, fet was blown due to bed drawing too much juice. The bed resistance is 1.1 ohms, so @14V that is ~180W, with is bit too much for the fets that are rated for 10amps I think? Replaced the fet and tuned the bed max power from full power to ~115W (duty cycle from 255 to 160). I think there's bit room to go higer, but I'll see how this works. Bed seems to heats up reasonably fast.
- the electronics cover I printed was too small. The display connector cables were too high, so had to redesign the model 10cm taller and reprint it. Dollo3D.2 to the rescue again, new cover done and works fine
- while assembling the RAMPS to the new cover, some of the pins didn't go the the Arduino connectors and obviously caused problems with endstops and display. Common problem with RAMPS, you need to align the boards correctly and check that the pins aren't bent.
- while debugging this problem, I accidentally nudged stepper driver heatsink off and it caused a short on top of the driver. Result: z-axis is dead. Depression ensued. But resolved this problem by switching the E1 and Z driver pins in Marlin and now Z is driven by E1 socket driver. No big loss since won't be using E1 on this printer.

After I got all things sorted out, ran the selftests, XYZ, Z and first layer calibrations. Everything checks out, my frame is 100% correct if Prusa's algorithms are to bebelieved. Awesome.

Still, couple of things to sort out:
- Trinamics and the 0.9 degree motors doen't seem to like each other. Ugly coil whining, even though everything moves properly. Probably switch A4988 drivers in for now. Sad, was really hoping to have the Trinamics interpolation.
- X-endstop mounting screw head hits the left z-coupler at minimum z-height just enough to cause skipped steps. Need to file some stuff off of the screw head or the coupler.
- top frame isn't stable. Not a surprise since it's printed. Need to design clamps to tighten it. Wont affect print quality with small z-heights, but taller prints will suffer.

Classy Marvin scaled to 50%. Some PLA I purchased from Amazon. Seems to need bit more retraction and maybe cooling. Also first layer is a mess due to the z-skipping, lucky that print even stuck to the bed. I'll try Benchy tomorrow.

[Roberts_Clif]

As mentioned previously, Mosfet was blown due to bed drawing too much juice. The bed resistance is 1.1 ohms, so @14V that is ~180W, with is bit too much for the fets that are rated for 10amps

- while debugging this problem, I accidentally nudged stepper driver heat-sink off and it caused a short on top of the driver.

No, I had other problems with my secound 3D Printer, after fixing never got back to My 3D Printed System...

On my Controller I made heat sinks with an Aluminum Bar for a heat sink. As they were overheating did not need a large heat sink, no space for a good Heat sink.
With an adequate Cooling fan the Straight Bar Aluminum provides sufficient heat dissipation.

Controller 2.jpg

As mentioned previously, Mosfet was blown due to bed drawing too much juice. The bed resistance is 1.1 ohms, so @14V that is ~180W, with is bit too much for the fets that are rated for 10amps

Question is a separate supply for the bed that you are using or is this power controlling everything.

If you have a single power supply powering everything you may have many problems.
By using your supply powered at 12 volts power P in watts (W) is equal to the squared voltage V in volts (V) divided by the resistance R in ohms (Ω)

Equals 130.9090 Watts

I am not sure how it came to be that we started adjusting our 12 volt power supplies to 14 volts.
This must be from the movie "Tool Time" an "Tim the Tool man Taylor", if you remember the movie everything Tim made had to have more power.

PROBLEM : The Power supply was designed for 12V and a specific current draw. When we exceed this rating the supplies cause problems.
will take my Power supply because I know it.

My Supply 12volts at 30 amps = 360 watts
If I was to turn the voltage up to 14 volts
Then 14volts at 30 amps = 420 watts this supply was designed for 360 watts.
And what you do not see the the Head voltage this is the voltage above your 12 volts to keep electrical noise down.
You have essentially nulled the supplies ability to clean the noise so with your power supply operating in the RED you will see many noise related problems.

If you are using a 12 V Power supply I would lower the voltage back to 12volts. you will have a better operating 3D Printer.

[spegelius]

I adjusted the PSU to 14V because that's what Prusa did with MK2 when it was launched and it was suffering from temp faults when using the part cooling fan. Their fix was to tune the PSU to 14V. I haven't checked the PSU on my original MK2 (now MK2.5), but I assume it's running at ~14V.

You might be correct about lowering the voltage back to 12V, though. It's obvious that 14V is overkill for the MK42 bed, especially with RAMPS. Also I'm using a silicone sock on the hotend so lower voltage shouldn't cause problems with it. Also as I bought the E3Dv6 directly from E3D, it might be that it has a more powerful heater cartridge than what MK2 was shipped with.

Still, the printer has been working fine so far, no blown fets since I restricted the bed power in firmware.

[Roberts_Clif]

They Did What !!!!!

Then do a happy medium 13v.

OR

Add a Smaller dedicated supply for the Controller. Then using external high power mosfets drivers with the Larger 14 V Power Supply. thereby protecting the Controller from over current and noise.

https://www.ebay.com/itm/2pcs-ANET-A...LG5Q:rk:2:pf:0