Mini Plus Review and fixes from engineer's POV
Bought it as a second printer for small parts with small details. Decided to pay premium to just print and avoid unnecessary hassle with tuning. Bought assembled version (non-kit).
I have printed for a few days with PETG and PLA, checking if everything is OK. I have other printer (slightly modified Artillery X1 v4) so it was easy to compare prints and make some conclusions.
Problems from the factory with solution:
- PINDA probe was set too low. Had to move PINDA probe 1mm higher, as during the first layer calibration having Z at -2.000 was still not enough to get a good layer thickness.
- Extruder gears had a loud click every time the gears changed direction due to gears backlash. Had to disassemble extruder, loosen two motor bolts, then push motor up closer to the gear and while holding it like that tighten the motor bolts. Now gears latch much better and are way quieter.
- Heatbreak was not set deep enough in heatsink (than allows for ptfe tube to move inside and increases the clog possibility), had to unscrew Bowden tube and then Bowden tube adapter from heatsink, loosen 3 grub screws and push hotend up (about 1mm) and while holding it tighten 3 grub screws back and then screw the Bowden tube adapter and tube back, then adjust PINDA probe and calibrate the first layer again.
- Noted that print quality was terrible (inconsistent layers, sometimes under extrusion) for small parts with small details (lot of retractions) due to Prusa slicer printing profiles retraction and de-retraction speeds (changing tension on extruder didn’t help). Had to decrease retraction and deretraction speeds to 20mm/s. That has improved quality of smaller details a lot, but still not as good as for big parts (all tested with Prusament PLA).
- The whole construction assembly took way more time compared to other printers of the same or lower price category.
- Cable management is not very pleasant and could be improved.
- Cable holder on the top of hot-end interferes with Bowden tube adapter screwed into heatsink and when unscrewing it catches a bit on adapter.
- Heat block should be enclosed in a silicon sock, so heat from the block is not affecting small PLA prints and cooling fan is not blowing at heat block.
- Cooling of part is not enough (should be from both sides) for PLA prints.
- Firmware will not allow you to store setting M500 into printer. You will have to use gcode.
Critical design flaws (after week of printing and dealing with low quality prints):
- Bowden type extruder. This is the cancer from the past which now have infected Prusa. Majority of the problems reported could be avoided with direct drive extruder. The only place where the Bowden type is feasible is for very high speed and high flow (big nozzles and big volume of filament) applications in CoreXY printers. Let me explain. Small printer – for small prints, here flow control is very important as 1mm of filament extrusion error will cause a big change in extruded length through 0.4 or 0.25 nozzle (you will get a long noodle from nozzle from 1mm of filament). In high flow application (nozzle 0.8+ ) 1mm of filament makes way smaller changes in extrusion length (noodle will be short), so slack in the Bowden tube has way less influence on the quality of the print. For example by extruding 1mm of 1.75mm filament through 0.4mm nozzle we get about 19.1mm extrusion (noodle length from the nozzle) and when extruding the same volume through 0.8mm nozzle we get about 4.8mm extrusion. So under/over extrusion will cause 4 times more problems on small nozzle (read on Prusa Mini). So for me it is unclear why would something like Bowden happened on Prusa Mini?
- Another problem is construction rigidity and squareness. For single sided Z for a printer of this size you should use 20x40 aluminum profiles with direct attachment instead of 3d printed PETG parts. And those designs already exist.
I don’t actually understand the placement of this printer. It makes a bad name for Prusa. There are a lot of much cheaper and sturdier printers with Bowden type extruders if you are willing to mod more than print. There are also cheaper printers with direct drive extruders.
You may think that with Prusa Mini Plus you are getting support, and software adapted to printer but… support and software cannot fix the design flaws.
Why Prusa have deviated from true and tested design? I don’t know.
I do not recommend Prusa Mini Plus, unless you wish to deal with a lot of problems instead of printing. You are better off with Prusa MK3S (at least it has direct drive extruder and a more rigid construction).
I have tried adding Bondtech dual drive extruder (grips filament quite well), Bondtech heatbrake and Bondtech plated copper nozzle. Quality of prints have improved a bit, but not really worth the money. What this printer really needs is a direct drive extruder. I do hope someone will come out with an upgrade kit solution.
I've had my Mini+ Kit for a few weeks and after careful assembly, mine has been performing flawlessly right out of the box. No tweaks. No upgrades. It just prints with incredible accuracy.
Not sure what you were expecting. You probably spent weeks (if not months) tweaking and upgrading your Artillery X1 v4 yet the need for a couple of adjustments on your Mini seems to have you totally upset.
Maybe you should just return it and buy something else?
Not trying to be mean. Just don't understand why you posted this or what you were expecting.
@JordanK I am glad that your printer works well for you and for the things your print. By assembling it you have controlled every step and if something doesn't work you can blame yourself only. That is fully understandable.
The post above is more dedicated for Prusa engineers, product designers, marketing and quality control. I do get some logic behind their solution (let's make one handed printer as entry machine for this niche of the market, but construction not rigid enough so we can't put direct drive extruder as it would be too heavy and compete with MK3, let's go Bowden - here you are another Creality printer).
As I have already mentioned that the idea of the whole thing - that if you pay premium for something, you would expect it to work. When you ask mechanic to do work on your car and you get your car back with misaligned wheels, does it feels right? Same with the printer.
If it is interesting:
For my Artillery X1 I did these and you can guess it doesn't cost a lot. Also when buying this printer I was expecting it to require tweaking as it's price is less than Mini.
Replaced Z axis nuts with anti-backlash ones (I have noted that in majority of videos people install them incorrectly so springs work against gravity - and they should work with the gravity to have good effect. Nut should be on the bottom and anti backlash part on the top).
Replaced rails wheels and adjusted them.
Replaced PTFE tube in heatbreak to a bit longer one.
Put the latest Marling firmware.
Majority of the time I spent tweaking printing profiles (here you are right, this work took the longest)
Good post, always good to get a person's opinion who has knowledge of how things ought to work. Thanks for this post!
I agree, its always good to het others views and learn.
I too had the same issue with my mini+ where i had to raise the sPinda probe by 1mm because it was hitting -0.2000 too early. this has really helped the quaility of my prints, however my latest challenge is keeping the print from lifting off the steel bed shortly into the print.
....and no, i cant be washing the steel sheet with soapy water like you do with dishes ! as has been a suggested solution elsewhere. i have a dishwasher for that !
People assembling the Kit version of the Mini+ don't have issues with the Pinda height because you adjust this during installation. As far as washing the steel plate goes, I typically wash the plate every 4-5 prints and just do an IPA wipe in between. This cleaning doesn't really take any extra time.
The biggest complaint from the original post seems to revolve around Prusa's decision to go with a Bowden vs. Direct Drive extruder. Obviously they went with a Bowden because they wanted less weight on the carriage. Correct me if I am wrong but less weight typically means faster, quieter, and higher quality prints. I don't think this decision is a flaw but simply a design trade off for a less expensive printer.
As far as inaccuracies caused by "slack" in the Bowden tube, the tube on my Mini+ is extremely rigid and the Pursament filament I'm using fits the tube quite snugly (with no slack). I bet there is slack if a cheaper filament was used with inconsistent thickness. I'm just not seeing these inaccuracies with my printer. I will say that the extruder height/gap issue that is causing some people grief is definitely a valid complaint and something Prusa should fix before shipping any new Mini+ units.
All good discussion.
There are many factors which can influence and the most common already mentioned in the forum. So here I will mention other reasons. Some early versions of Mini had problems with the surface quality of metal sheets.
If you get your first layer torn off (dragged by nozzle) from the bed (assuming you do print on correct https://help.prusa3d.com/en/materials and clean surface with z level calibrated right), then I suggest decreasing printing speed of the first layer to 10mm/s, decrease first layer acceleration to 250mm/s and movement to 70mm/s. So the there is more time for filament to "unstick" from the nozzle. You can also adjust printing speed in your printer menu during printing. Also as I have noted in the first post, you may need to decrease retraction and deretraction to 10mm/s for fine details.
If you get your print torn from the surface after a few layers, please check if your room temperature is not too low, as warping occurs for any material. I do usually cover a printer with a cardboard box (for ASA/ABS all the time) for PETG only if room is cold or printed object is large. On one type of PETG filament I had to turn off cooling fan for the whole print.
There could be a lot of other things affecting your print (temperatures, flow rates, layer thickness…)
Update: Sometimes small parts warp and get hit by nozzle, knocking the print off. Here I don’t have any clear solution (enclosure?).
Yes, You are right. Mini frame is not meant to have direct drive extruder. The frame is really flimsy – especially between bed and Z axis stand (all screws tightened well). Even with Bowden type extruder it changes Z level a bit under weight of hotend as head moves on X axis. So fast printing is a question. Here I see one solution (a bit questionable) is to change nozzle to 0.6 so it is possible to print with a lower speed and keep volumetric flow the same. However this way the details are lost.
I ordered MINI basically that weekend it was announced out and had one of first units with few quirks unique to first batch (think cable management zipties so tight they broke cables inside the loom). Other than that, it's been OK. I learned to live with some of its quirks, but there are 2 you mentioned which I am critical about and hope for improvement. I fully agree with the need to make XZ connection to Y frame more rigid and not through plastic part; and hotend with PTFE melting at the bottom and shortening in length is another pet peeve.
As for refusing to wash steel sheets ... well, the surface simply requires de-greasing, and washing it with dish soap had been proven to work best so far. Use body wash if you don't have a single drop of dish soap in your household. But don't come back complaining about prints not sticking to surface when people keep telling you what to do but you don't want to do that for whatever reason...?
What are your thoughts on remote drive systems? Would that not give you your direct drive but at an acceptable weight for this type of printer? Or would you find there to be comparable slop to a bowden style?
I have no experience with remote drive systems (remote direct extruder). Logically thinking I would be more concerned about mechanical system slack or elasticity of connection; having in mind that you would need a flexible shaft between stepper motor and extruder (I know that flexible shafts are actually quite rigid to twisting). Thinking further, to minimize the effect of elastic connection and to minimize the torque twist (when shaft tries to rotate the print head), you would need to gear the transfer of torque, so shaft rotates much faster than your stepper motor and then gear the extruder so fast rotation of shaft is transferred back to slow rotation of hob wheel of extruder.
With lower moving mass than direct drive extruder and minimal filament slack they look quite promising for fast and accurate printing.
I see you may not be expiereced with them, but im glad to see how knowledgeable you are on the concept. I havent seen many printers useing remote drive systems. Ive never really know why this is, other than "its different". Or they dont work as well as intended.
I have seen them used on delta printers where weight is absolutely critical.
Heres one provider of a remote drive system. https://flex3drive.com/g5/
Do you think youll attempt installing such a modification onto your mini?
I am not planning to do any upgrades on my Mini any time soon (I am a bit tired of tweaks and this type of upgrade is not on the list).
I have checked the extruder (the link you gave me). It uses worm type reduction on extruder with 40:1 ratio. So mechanically that looks good.
The problem I see is that stepper motor will have to rotate very fast to keep the same retraction and de-retraction speeds. For example if geared bondtech extruder uses 3:1 gearing ratio and E415 steps per 1mm, then after replacement of one extruder with the other (40:1 gearing ratio), stepper will have to rotate (40 / 3 = 13.3 times) faster for the same retraction speeds. E-steps will increase the same number of times (from E415 to E5519). So retraction speeds in slicer should be lowered to limit stepper motor rpm (retraction distance could also be lowered as remote drive needs less retraction than Bowden).