迪威網(wǎng)友a(bǔ)7f8ac
發(fā)布時(shí)間：2021/2/17 1:20:00
12/22/2017
This is another attempt to make PRUSA's MMU2 work reliably.
The nice little thing of this mod is that you need barely nothing to build it. Parts list is below.
PERFORMANCE SUMMARY
The thing works, and has made me complete my fit multimaterial prints. It could work better though. Results depend also on how careful I am with my tips. The cleaning trap has proven a great addition. It makes ipection when misloads really easy. The TPU seath for the bearing provides a better grip between the filament and the bearing and avoids lots of mis-readings.
I have discontinued this design (that works) and I'm looking out of the box trying to think of something different that solves finally the MMU2 problems. Waiting for parts from China now (yes, I had to abandon the NO COST approach of this design, but spending 10 € is not going to break my finances)
1/21/19 YESSSSS!! Got it!! It deserves its name: The Frog. No hai!! It's like 200 times better than this old MK5 design that you are looking now. I know that there are like 2000 extruder desig for free in Thingivee, and the Frog is not in Thingivee and it's not for free. Why? Because it's different and it makes the MMU2 work, and because I am pissed enough with PRUSA and their poor engineering to leave them this awesome design with a free licee . How different is The Frog? If you got this far reading you deserve to have a look at it:
You'll have to donate $5.5 to GreenPeace, UNICEF, MSF or whoever you want if you don't want to pay me, but I promise you it's worth it.
You ,might wonder why don't I keep on improving this design myself, but you should coider some hard facts about the MMU2:
This tip shaping, ramming, etc. is a complete sh*t. If it doesn't affect the final product it is only an inconvenience and just shouldn't be a factor
Wasting 10 % of the spool and countless hou just to find the right settings, sucks.
Some materials like PETG, my favourite, seem to be very difficult to work with, because of the strings. I haven't even tried. This is wrong, and I am not including TPU, I undetand FLEX is different and has its own problems.
Coming back to this design, here you have the STP files to play with them, but if I wanted to polish this design, and make it work better, I think I would;
-Give more elbow room to the bearing - when I designed it I did not coider the TPU seath. That was a later addition, so space is pretty tight in there. You'll have to print a good TPU part with no seams to make the device work well as it is designed now, either that or just...
Reduce the bearing size and modify the slider accordingly. The bigger the bearings are, the lazier they become, and more inertia they do have, but there has to be some iron in place to make the magnet work, so if you use a really small bearing, make sure there's iron (not stainless steel!) in the bolt you use in the bearing shaft. I finally broke my daughter's spinner and did this same thing myself with my MK3 in order to be able to work reliably until I get my other design up and running
Make the casing for the bearing smaller (shorten in the back the slider range, so the bearing bounces back to the filament). Also, it really doesn't need to be so long
Add a little slope (maybe 5%) towards the filament to the slider ramps to help the magnet
Lubricate the bearing
Lubricate the bearing again
And lubricate the bearing again until it rolls really smooth
All these measures will improve the device's performance.
12/15/2018
PARTS LIST:
1- One of the bearings that PRUSA includes with the MMU2 spool holde (aprox ?e 22 ?i 8 L 7.5 mm)
2- A magnet- The magnet housing in the plastic part is diameter 11mm x 5 mm length. I used two Neodymium magnet diam 10 x 1.5 bought in Aliexpress- Cost barely nothing. STP files are provided so you can modify the magnet case dimeio to your convenience
3- Three M3 x 20 bolts, like the ones that PRUSA includes with the MK3 as spares.
4- One M3 self locking nut
5- One bolt M3 x 5 to hold the seor in place.I used a Nylon6 slotted bolt because it was easier to cut to lenght
4- .One normal M5 nut to secure the Festo coupling. I use the hot tip of a soldering iron to fit it in place nicely
FIRMWARE - The idea is to put the thing to work with some firmware that takes into account the optical filament seor, like the one made by TheZeroBeast
Because the distance from the optical seor to the bondtech gea in my design (30 mm) is larger than in prusa's stock (16 mm), supposedly there are a couple of changes that had to be performed in the firmware. However, the thing works well with TheZeroBeast .hex files as they are.
In any case, the changes to be performed are:
1- File MM-control-01/MM-control-01/, line 21
Itead of "/2/ { 300, 330, 300},"
It should say "/2/ { 563, 619, 563},"
2- File MM-control-01/ , line 519
Itead of "moveSmooth(AX_PUL, STEPS_MK3FSeor_To_Bondtech, 385,false, false);"
should say "moveSmooth(AX_PUL, STEPS_MK3FSeor_To_Bondtech, 722,false, false);"
You can find the hex files already compiled with these changes here
NOT YET- NOT YET -NOT YET- Sorry but I am code blind, or github discapacited, whatever you want to call it
.
The thing is assembled and working. All files uploaded have been revised and are correct. I have included the two veio for the ge (longer and shorter belt) and the two veio for the seor cover (standard brass iert and M5 nut)
Just started testing, and the thing works as well as expected. No skipped laye for the moment, If the machine cannot make a load, itead of skipping a layer, it pauses and will do nothing until it's serviced and a button is pressed,
Service mea some cleaning through the trap with tweeze, checking that the bearing is rolling OK and that it can move freely close to the trap. Check also if the filament has some grinding, it usually does the trick, and there we go again. The trap is really useful to clean hai and globs.
Trap bolt cannot be too loose or the filament might find the wrong way out, but all loads are coming alright, even repeated loads printing multimaterial do work alright.
And what follows is the little story of the contraption and how the design got to be what it's today.
12/27/2017
OK, I've changed the bearing with a smaller one. It was too tight with the TPU seath and it colided with the walls producing seor misreadings. It would have been cleaner to print a slightly bigger housing in the extruder body, but this was way faster in terms of printing, disassembling and reassembling. STP files of the new parts are provided to adapt it to whatever bearing you have handy . New parts are only 1/2 hour print.
12/21/2018
Changed the slide to make them round because they do not slide so well after all, so if they don't slide, let them roll, baby!
Changed the filament cover for better printability
13/12/2018
Realiced that the pusher cannot be assembled because of interference with the z carriage. Erased it from design. Iddler's mod is unnecessary also.
Another drawback. This time assembling. I noticed the hard way that there are two veio of the x-carriage part. Of coue I had the wrong one and could not fit the belt. The wot thing about it is that I had disassembled the extruder. So I have to assemble it again, perform the calibration, print the part and repeat disassembling and assembling again... The kinf of thing I feared...
12/11/2018
Printing the thing right now. I did find the brass ierts in aliexpress. Cost barely nothing but I had to buy 100 units and wait three weeks to get them, so I have used just a brass M5 x 0,8 nut. It does the task just as well, but without the three week delay. I modified the seor cover to host the nut and set it in place nicely with a soldering Iron, I will include the M5 nut variant in the files.
Also I used longer bolts for the cover (25 mm itead of 20 mm - just because I had them handy. 20 mm is better and I was even forced to modify the seor cover's top, raising the bolt's head's beds, but that's better than cutting bolts with the saw...)
An inexpected drawback with the pusher, as tolerances seem to work tighter with PETG than with PLA, or perhaps it's the bridging that sags more. Fortunately, it's only a 15 min print to redo.
12/8/2018
Well, I think I am done with extruder designing, I mean, I think that without a complete rework, the PRUSA MK3 will only be just as good as it gets.
In any case, I am including all STP files just if anybody wants to go on and rework my design. There's a file called PRUSA_i3_MK3_V2.stp that contai the entire assembly.
The funny thing is that I still haven't put the extruder to work, but it will be fine if it works just as well as the filament seor housing for the MK3 (link below).
I think I have solved the two problems that the housing had to work with the MMU2:
1- The filament tip crossing through so many parts, Now it is direct.
2- And having to pause the printing and disassemble half of the machine to perform a little bit of cleaning up the debris, strings and hai...I have found a smart solution for that.
The reason I haven't still italled the extruder is because, oddly enough, I thought it would be nice to print a couple of parts in the PRUSA orange PETG, but it is unavailable (looks like forever), and today, I have finally given up, and I have ordered just black PETG.
Also, designing a extruder is a bit messy unless you have two printe. If you do not get the design right at the fit time, you might go crazy assembling and disassembling the machine, so well, yes, just to be sure I printed a mockup in PLA to check everything is as it should be. Those are the photos included. BTW, the slider is not in the photos just because I forgot about it, but this part of the design has not changed since the fit veion of the housing.
The changes I have done in this final veion are:
I have included a little trap for cleaning purposes (hai, strings and globs do end there). Let's call it "the filament cover". It is interesting that the moment you open the trap the seor gets disengaged. It is also nice that when you close the trap there's a satisfying metalic noise as the bearing moves
Also, I have included a button to loose the iddler door in order to be able to manually free the filament. It's not that I think that I am going to use it a lot, but, you know, just sometimes I wished it was there...
The casing for the magnet is a bit bigger (diam 11 x 5 mm), and I have introduced a couple of funnel shape zones in the filament's path to facilitate the loads
Eliminated embossments
What is left to do:
The brass part of the seor cover. PRUSA does not sell it, so it is going to be either finding another nut where I can bolt the festo coupling and ierting it in my design or sacrifycing the part that came with the MMU2
Perhaps some PTFE washe for the filament cover?. For the moment I am using a M3 self locking nut for the hinge bolt in the seor cover, and it looks like it's good enough so. I just haven't allowed place for the washe in my design,
1/12/2018
Updated the X-Carriage part, modified from an STP file by jzkmath
Modified the extruder body and the seor cover to include square nuts to hold the part (itead of using plastic screws), to include a couple of fillets to smooth the filament's path, and to change embossment from "DON'T PANIC" to "KEEP CALM & PRINT"
11/28/2018
When I uploaded yesterday the PRUSA MK3 FILAMENT SEOR HOUSING ( I did not expect that it would have a rate of one download per hour, so it looks like I am not alone fighting the skipped laye of the MMU2.
The filament seor housing is surely good enough for the PRUSA MK3, but it probably falls short for the MMU2 requirements. With the filament going up and down so many times, introducing new possible jam points does not seem a good idea.
Therefore, here it is the redesigned PRUSA extruder body that includes the filament seor housing of my design. The elements (bearing, magnet) are the same as in the previous design, except for the bolts.
I still haven't printed the mod myself (I am waiting for some orange and black PTEG, and I still have to order the brass coupling, though I might canibalize the PRUSA part), but I am pretty confident that my design is correct. If you are adventurous enough to go for it, upload a photo and please let me know how it went
To build the PRUSA MK3 MMU2 EXTRUDER BODY W/ FILAMENT SEOR INDIRECT READING MOD you will need:
1- One of the bearings that PRUSA includes with the MMU2 spool holde (aprox ?e 22 ?i 8 L 7.5 mm)
2- A magnet- The magnet housing in the plastic part is diameter 11mm x 2 mm length. I used a Neodymium magnet diam 10 x 1 bought in Aliexpress- Costs barely nothing.
3- Three M3 x 20 bolts, like the ones that PRUSA includes with the MK3 as spares
4- One bolt M3 x 5 to hold the seor in place.I use a Nylon6 slotted bolt because it is easier to cut to lenght.
... and if you happen to start getting successful prints with the MMU2, you might even coider tipping me!! (just think how much money you threw in wasted filament trying to make the MMU2 work)
***** If you are anyone of the fit four peo to have downloaded this thing, (that would be around 11/29/2018 00:00 GMT ), perhaps you might have to redo both ends of the slider. Just make them round or a bit shorter and it'll be OK. I am sorry, but I fit uploaded a wrong veion... I noticed it quite fast, so if you were the fifth, then you are safe.
***** In this design I've only worked the seor cover and the top parts of the extruder body and the X carriage. For the rest of the design I just used the STL and STP files provided by PRUSA in their webpage. Credit of that design to whoever deserves it
This is another attempt to make PRUSA's MMU2 work reliably.
The nice little thing of this mod is that you need barely nothing to build it. Parts list is below.
PERFORMANCE SUMMARY
The thing works, and has made me complete my fit multimaterial prints. It could work better though. Results depend also on how careful I am with my tips. The cleaning trap has proven a great addition. It makes ipection when misloads really easy. The TPU seath for the bearing provides a better grip between the filament and the bearing and avoids lots of mis-readings.
I have discontinued this design (that works) and I'm looking out of the box trying to think of something different that solves finally the MMU2 problems. Waiting for parts from China now (yes, I had to abandon the NO COST approach of this design, but spending 10 € is not going to break my finances)
1/21/19 YESSSSS!! Got it!! It deserves its name: The Frog. No hai!! It's like 200 times better than this old MK5 design that you are looking now. I know that there are like 2000 extruder desig for free in Thingivee, and the Frog is not in Thingivee and it's not for free. Why? Because it's different and it makes the MMU2 work, and because I am pissed enough with PRUSA and their poor engineering to leave them this awesome design with a free licee . How different is The Frog? If you got this far reading you deserve to have a look at it:
You'll have to donate $5.5 to GreenPeace, UNICEF, MSF or whoever you want if you don't want to pay me, but I promise you it's worth it.
You ,might wonder why don't I keep on improving this design myself, but you should coider some hard facts about the MMU2:
This tip shaping, ramming, etc. is a complete sh*t. If it doesn't affect the final product it is only an inconvenience and just shouldn't be a factor
Wasting 10 % of the spool and countless hou just to find the right settings, sucks.
Some materials like PETG, my favourite, seem to be very difficult to work with, because of the strings. I haven't even tried. This is wrong, and I am not including TPU, I undetand FLEX is different and has its own problems.
Coming back to this design, here you have the STP files to play with them, but if I wanted to polish this design, and make it work better, I think I would;
-Give more elbow room to the bearing - when I designed it I did not coider the TPU seath. That was a later addition, so space is pretty tight in there. You'll have to print a good TPU part with no seams to make the device work well as it is designed now, either that or just...
Reduce the bearing size and modify the slider accordingly. The bigger the bearings are, the lazier they become, and more inertia they do have, but there has to be some iron in place to make the magnet work, so if you use a really small bearing, make sure there's iron (not stainless steel!) in the bolt you use in the bearing shaft. I finally broke my daughter's spinner and did this same thing myself with my MK3 in order to be able to work reliably until I get my other design up and running
Make the casing for the bearing smaller (shorten in the back the slider range, so the bearing bounces back to the filament). Also, it really doesn't need to be so long
Add a little slope (maybe 5%) towards the filament to the slider ramps to help the magnet
Lubricate the bearing
Lubricate the bearing again
And lubricate the bearing again until it rolls really smooth
All these measures will improve the device's performance.
12/15/2018
PARTS LIST:
1- One of the bearings that PRUSA includes with the MMU2 spool holde (aprox ?e 22 ?i 8 L 7.5 mm)
2- A magnet- The magnet housing in the plastic part is diameter 11mm x 5 mm length. I used two Neodymium magnet diam 10 x 1.5 bought in Aliexpress- Cost barely nothing. STP files are provided so you can modify the magnet case dimeio to your convenience
3- Three M3 x 20 bolts, like the ones that PRUSA includes with the MK3 as spares.
4- One M3 self locking nut
5- One bolt M3 x 5 to hold the seor in place.I used a Nylon6 slotted bolt because it was easier to cut to lenght
4- .One normal M5 nut to secure the Festo coupling. I use the hot tip of a soldering iron to fit it in place nicely
FIRMWARE - The idea is to put the thing to work with some firmware that takes into account the optical filament seor, like the one made by TheZeroBeast
Because the distance from the optical seor to the bondtech gea in my design (30 mm) is larger than in prusa's stock (16 mm), supposedly there are a couple of changes that had to be performed in the firmware. However, the thing works well with TheZeroBeast .hex files as they are.
In any case, the changes to be performed are:
1- File MM-control-01/MM-control-01/, line 21
Itead of "/2/ { 300, 330, 300},"
It should say "/2/ { 563, 619, 563},"
2- File MM-control-01/ , line 519
Itead of "moveSmooth(AX_PUL, STEPS_MK3FSeor_To_Bondtech, 385,false, false);"
should say "moveSmooth(AX_PUL, STEPS_MK3FSeor_To_Bondtech, 722,false, false);"
You can find the hex files already compiled with these changes here
NOT YET- NOT YET -NOT YET- Sorry but I am code blind, or github discapacited, whatever you want to call it
.
The thing is assembled and working. All files uploaded have been revised and are correct. I have included the two veio for the ge (longer and shorter belt) and the two veio for the seor cover (standard brass iert and M5 nut)
Just started testing, and the thing works as well as expected. No skipped laye for the moment, If the machine cannot make a load, itead of skipping a layer, it pauses and will do nothing until it's serviced and a button is pressed,
Service mea some cleaning through the trap with tweeze, checking that the bearing is rolling OK and that it can move freely close to the trap. Check also if the filament has some grinding, it usually does the trick, and there we go again. The trap is really useful to clean hai and globs.
Trap bolt cannot be too loose or the filament might find the wrong way out, but all loads are coming alright, even repeated loads printing multimaterial do work alright.
And what follows is the little story of the contraption and how the design got to be what it's today.
12/27/2017
OK, I've changed the bearing with a smaller one. It was too tight with the TPU seath and it colided with the walls producing seor misreadings. It would have been cleaner to print a slightly bigger housing in the extruder body, but this was way faster in terms of printing, disassembling and reassembling. STP files of the new parts are provided to adapt it to whatever bearing you have handy . New parts are only 1/2 hour print.
12/21/2018
Changed the slide to make them round because they do not slide so well after all, so if they don't slide, let them roll, baby!
Changed the filament cover for better printability
13/12/2018
Realiced that the pusher cannot be assembled because of interference with the z carriage. Erased it from design. Iddler's mod is unnecessary also.
Another drawback. This time assembling. I noticed the hard way that there are two veio of the x-carriage part. Of coue I had the wrong one and could not fit the belt. The wot thing about it is that I had disassembled the extruder. So I have to assemble it again, perform the calibration, print the part and repeat disassembling and assembling again... The kinf of thing I feared...
12/11/2018
Printing the thing right now. I did find the brass ierts in aliexpress. Cost barely nothing but I had to buy 100 units and wait three weeks to get them, so I have used just a brass M5 x 0,8 nut. It does the task just as well, but without the three week delay. I modified the seor cover to host the nut and set it in place nicely with a soldering Iron, I will include the M5 nut variant in the files.
Also I used longer bolts for the cover (25 mm itead of 20 mm - just because I had them handy. 20 mm is better and I was even forced to modify the seor cover's top, raising the bolt's head's beds, but that's better than cutting bolts with the saw...)
An inexpected drawback with the pusher, as tolerances seem to work tighter with PETG than with PLA, or perhaps it's the bridging that sags more. Fortunately, it's only a 15 min print to redo.
12/8/2018
Well, I think I am done with extruder designing, I mean, I think that without a complete rework, the PRUSA MK3 will only be just as good as it gets.
In any case, I am including all STP files just if anybody wants to go on and rework my design. There's a file called PRUSA_i3_MK3_V2.stp that contai the entire assembly.
The funny thing is that I still haven't put the extruder to work, but it will be fine if it works just as well as the filament seor housing for the MK3 (link below).
I think I have solved the two problems that the housing had to work with the MMU2:
1- The filament tip crossing through so many parts, Now it is direct.
2- And having to pause the printing and disassemble half of the machine to perform a little bit of cleaning up the debris, strings and hai...I have found a smart solution for that.
The reason I haven't still italled the extruder is because, oddly enough, I thought it would be nice to print a couple of parts in the PRUSA orange PETG, but it is unavailable (looks like forever), and today, I have finally given up, and I have ordered just black PETG.
Also, designing a extruder is a bit messy unless you have two printe. If you do not get the design right at the fit time, you might go crazy assembling and disassembling the machine, so well, yes, just to be sure I printed a mockup in PLA to check everything is as it should be. Those are the photos included. BTW, the slider is not in the photos just because I forgot about it, but this part of the design has not changed since the fit veion of the housing.
The changes I have done in this final veion are:
I have included a little trap for cleaning purposes (hai, strings and globs do end there). Let's call it "the filament cover". It is interesting that the moment you open the trap the seor gets disengaged. It is also nice that when you close the trap there's a satisfying metalic noise as the bearing moves
Also, I have included a button to loose the iddler door in order to be able to manually free the filament. It's not that I think that I am going to use it a lot, but, you know, just sometimes I wished it was there...
The casing for the magnet is a bit bigger (diam 11 x 5 mm), and I have introduced a couple of funnel shape zones in the filament's path to facilitate the loads
Eliminated embossments
What is left to do:
The brass part of the seor cover. PRUSA does not sell it, so it is going to be either finding another nut where I can bolt the festo coupling and ierting it in my design or sacrifycing the part that came with the MMU2
Perhaps some PTFE washe for the filament cover?. For the moment I am using a M3 self locking nut for the hinge bolt in the seor cover, and it looks like it's good enough so. I just haven't allowed place for the washe in my design,
1/12/2018
Updated the X-Carriage part, modified from an STP file by jzkmath
Modified the extruder body and the seor cover to include square nuts to hold the part (itead of using plastic screws), to include a couple of fillets to smooth the filament's path, and to change embossment from "DON'T PANIC" to "KEEP CALM & PRINT"
11/28/2018
When I uploaded yesterday the PRUSA MK3 FILAMENT SEOR HOUSING ( I did not expect that it would have a rate of one download per hour, so it looks like I am not alone fighting the skipped laye of the MMU2.
The filament seor housing is surely good enough for the PRUSA MK3, but it probably falls short for the MMU2 requirements. With the filament going up and down so many times, introducing new possible jam points does not seem a good idea.
Therefore, here it is the redesigned PRUSA extruder body that includes the filament seor housing of my design. The elements (bearing, magnet) are the same as in the previous design, except for the bolts.
I still haven't printed the mod myself (I am waiting for some orange and black PTEG, and I still have to order the brass coupling, though I might canibalize the PRUSA part), but I am pretty confident that my design is correct. If you are adventurous enough to go for it, upload a photo and please let me know how it went
To build the PRUSA MK3 MMU2 EXTRUDER BODY W/ FILAMENT SEOR INDIRECT READING MOD you will need:
1- One of the bearings that PRUSA includes with the MMU2 spool holde (aprox ?e 22 ?i 8 L 7.5 mm)
2- A magnet- The magnet housing in the plastic part is diameter 11mm x 2 mm length. I used a Neodymium magnet diam 10 x 1 bought in Aliexpress- Costs barely nothing.
3- Three M3 x 20 bolts, like the ones that PRUSA includes with the MK3 as spares
4- One bolt M3 x 5 to hold the seor in place.I use a Nylon6 slotted bolt because it is easier to cut to lenght.
... and if you happen to start getting successful prints with the MMU2, you might even coider tipping me!! (just think how much money you threw in wasted filament trying to make the MMU2 work)
***** If you are anyone of the fit four peo to have downloaded this thing, (that would be around 11/29/2018 00:00 GMT ), perhaps you might have to redo both ends of the slider. Just make them round or a bit shorter and it'll be OK. I am sorry, but I fit uploaded a wrong veion... I noticed it quite fast, so if you were the fifth, then you are safe.
***** In this design I've only worked the seor cover and the top parts of the extruder body and the X carriage. For the rest of the design I just used the STL and STP files provided by PRUSA in their webpage. Credit of that design to whoever deserves it
|
共個(gè)文件
壓縮包大?。?MB
|
||
|---|---|---|
微信小程序
微信掃碼打開此模型
迪威模型
微信掃碼支付