[Fermé] I challenge Prusa to use MMU2S  

RE: I challenge Prusa to use MMU2S

You know, I really cannot claim my unit works as intended. I have one filament type which works as intended (with up to 3 colors), which is, ironically, PETG. All PLA attempts have crashed and burned.

But at least you can see the concept is viable and necessary parts are there. A lot of the printed parts need work and the buffer is a joke (why not just give people a full spool and print files for one of the autorewind spoolholders). The software needs a lot more work. However if I compare this with my experience with the MMU1, at least I am confident it will work at some point without throwing more money at them.
I would like to see how they made all these factory test prints during their internal testing, though. It just seems impossible with a stock unit on stock settings, especially if you consider the firmware back then.

RE: I challenge Prusa to use MMU2S

Hi,

I totally agree with nealc, see my posts in the topic "Extremely disappointed with MMU

Dick V

RE: I challenge Prusa to use MMU2S

Agree totally with OP. This seems to have made my printer less reliable/not work at all. 🙁 I have spent hours upon hours. My issue? It doesn't seem to make it to the bondtechs. It's a snag right around the chimney. Take the bowden tube off top, and feed it yourself? Works fine. Let it feed as designed, NOPE.

RE: I challenge Prusa to use MMU2S

My loading issues mostly went away when I swapped out all the bowden tubes (except the short one in the hotend) with cheap chinese (I use Fystec 2mm ID). 1.9mm is a bit optimistic over that length, a small kink in the filament and the friction is through the roof.

If that doesn't solve it, reprint the idler door at 80% speed and a bit more infill. Mine was just terrible print quality and didn't properly trigger.

RE: I challenge Prusa to use MMU2S

I think it was bad design to use a small ID PTFE tube for the MMU2S for filament feeding. You want a large "funnel" per se to capture the filament and then guide it in to the single tube to the extruder. I would use a 4mm ID instead of 2mm ID. 

I'm also concerned with the angle of the MMU2S and the trash it's going to collect over time, you have to blow it out often. 

RE: I challenge Prusa to use MMU2S

Well it seems to me we (almost) all agree of one of the major issues: FRICTION!

I did some tests with various setups and buffers;

Unplugged the PTFE on the extruder and just pushed the filament forward and backward by hand.

There is a lot of friction even if you have brand new out of the box PTFE tubes without any kink or additional friction due to screws holding the tubes in place or PneuFit push connectors. 

I know the Bondtech gears are quite capable doing their job but you can also push them over the limit!

FYI: I am using the MK3(s) with all materials without any problems using only 2 PEI sheet build plates:

#1: Cleaned with Isopropyl Alcohol, used for PLA/PETG  and first layer calibration.

#2: Using the glue stick method for warping sensitive filaments and filament with a strong bed adhesion. (ABS, Polycarbonate and Flex)

The only thing I do after using Flex filaments is cleaning the nozzle afterwards with PETG or Cleaning (Nylon) filament.

Reason for this is that I noticed that leaving Flex filaments in the extruder after printing they tend to behave like

2K glues after multiple heating end cooling cycles.

My advice, if you want to spent the time setting up the MMU2s:

1) Check your filament flow for the lowest friction of the entire filament chain, by hand!!

Forget all the nice enclosures, buffers etc. for the moment,  just pay attention to the correct smooth filament flow!

If you need enclosures eg. for moisture influence on your filament, just print more or store your filament after printing with a bag of silica gel. 

2) Check the correct working of the filament sensors.

3) I use only factory settings and only adjust the nozzle temperature depending on the specific filament. Don't start tweaking all kind of parameters unless you know exactly what you are doing.

RE: I challenge Prusa to use MMU2S

yup, its all about friction.  BUT when loading odd filaments, you can "adjust" the pressure on the mmu abit.  it's not much but seems to help with flexi filaments or soft filaments.

I am still about 99% stock hardware and software. I only added a fan for the extruder motor, and that I kept off after the "S" install.  but soon after I put it back on.

all I can say is. take one issue at a time.  get one filament path working. print abit (enjoy the fast printing speeds) then add a 2nd filament path. figure out the issues. then enjoy the "not as fast but still fast print"  than by the time you have all 5 going.  enjoy waiting 2 days for a print. remember around 90 seconds per color per layer.... adds up fast.

RE: I challenge Prusa to use MMU2S

Well toaf,

What do you think of one of my master peaces regarding 3D FDM printing?

Entirely printed (> 400 hours) with transparant PETG! It is a future building block for houses which takes the solar energy during day time and will release the energy during Night, to warm houses......

And well believe me a whole lot of issues during printing........

Dick

RE: I challenge Prusa to use MMU2S

single color and it had issues?  looks neat. you should put a few more up. imagine that 5 colors each layer..... 😀 would take forever and a day

RE: I challenge Prusa to use MMU2S

Work in progress toaf!! 🙂

RE: I challenge Prusa to use MMU2S

On special request of toaf, just to put a few more up!

Here is one of my other projects..... The Netherlands is sinking below sea level, causing a lot of troubles due to our gas addiction.

Pumping up gas for centuries in the north of our country..... result: earthquakes!   Damaging a lot of our cultural heritage!

What should we do, beside of putting our fingers into the dykes..:) ??

Scan, preserve and master your 3D skills!

Attached an example of a church in the north of our country with high value paintings of Albrecht Durer:

RE: I challenge Prusa to use MMU2S

first get even off those islands. or sacrifice around a mile of shore line to build it up (and still be usable)

living that close to the ocean just seems silly. sure it looks nice, but we all know it gets mad sometimes.

Ill throw in my 2 cents on a subject I really know nothing about, but I will read up on the issue.  im guessing allot of people are working on it.

RE: I challenge Prusa to use MMU2S

@ toaf,

Please, could we go back to the purpose of this forum? PR 3D printing??

You asked me a question regarding 3D printing, right?

I just gave you some examples, no more no less.

RE: I challenge Prusa to use MMU2S

Posted by: NealC

I challenge Prusa to put MMU2S on all of their production printers. Change the i3 MK3S to be at least a 2 color print. See how many printers are actually produced in one month using the MMU2S.

Why on earth would they put an MMU2 on all their printers and print in two colors when they only need to print everything in one color?  Only an idiot would use an MMU2 on a system they know will always only print one color.  Why do you challenge them to be idiots?  

RE: I challenge Prusa to use MMU2S

Hi cybrsage,

It's very simple, US STYLE MARKETING!!

They released some stupid "invention" called the "MMU2 etc." and like teasing us all with some "far" future thoughts.......

Well, just disconnect your MMUxxxx and go back to single color filament printing....

Multi-color printing with the I3 MK etc. etc, is just a bridge too far, they are pushing their luck too far, just as simple as that!!

RE: I challenge Prusa to use MMU2S

Posted by: dick.v

They released some stupid "invention" called the "MMU2 etc." and like teasing us all with some "far" future thoughts.......

A one which is being sold very successfully and found many happy customers. There is still nothing comparable on the market at this price point.

RE: I challenge Prusa to use MMU2S

From all my reading, the MMU is at the same stage of development as the i3 Mk1 was when it started selling like hot cakes.  But then anyone purchasing a KIT must expect some tinkering to get it to work properly and even more to tune it to some level of satisfaction.

Another way to look at it is to see the project as a puzzle with no certain solution.  Or, as we all know with a Prusa order vs delivery, a gamble that has daily changing odds.

With that - we know even the MK3 printer is still in development ... but is somewhat reliable ... I'm waiting for the MMU3.

RE: I challenge Prusa to use MMU2S

@ tim-M30,

I totally agree with you, as long as you see the MK3(s)/MMUx as a research project, nothing wrong with it!

Worth every penny! But don't sell it as a product with pictures (like presented with the introduction of the MMU2) as a ready tested product.

Like I said in an other post, it's all about managing expectations!

RE: I challenge Prusa to use MMU2S

I have solved several issues on mine MMU2S.

1. Removed the Buffer (I use an high bar to hole the spools, for the moment until I will install auto rewinding spools)
2. Removed the long PTFE buffer tubes, and replaced by 5 cm short pieces out of the MMU.
3. I discover that one of the buffer PTFE tubes had a wider tolerance inside diameter of about 2 mm. I cut a piece of it and put inside the aluminum cooler, removing the old 1.78 diameter original PTFE. I did not had any filament blocked inside the cooler due to excessive retractions. (till now... 😎 )
4. I bough a 4mm ext. diam. PTFE with 3mm inside diameter on Amazon, and replace the tube from MMU to Extruder.

With all those changes, I have eliminated almost all friction issues.

TO DO:

• Change Prusaslicer settings to force a filament CUT after a color change. I noticed that is very difficult to have the "Perfect filament head shape" after a retraction. It depends on the temperature, material, manufacturer. Too many variants. It is never OK. It is better to cut it, to avoid a failure on the next loading.
• Improve the sensors and MMU firmware.
  I noticed that after some retraction / load, the firmware lost the idea of how much filament is out of the MMU. I see it push out around 2cm of filament with the extruder gear on, while the filament is really at 3 cm from the beginning of the MMU-Extruder bowden (too far from the  Extruder). Then it stops the gear, and push other 2cm and activate the gear and wait. And so on until the filament finally reaches the gear and it is loaded. This process take too much time.
  Why it does not push all the filament in one single long movement until it feels the IR sensor triggers ????
  To be checked.
• Improve the MMU blade section. The current blade is not stable, and could get out easily,and then is a nightmare to reinstall it.

I think that one those points are solved the MMU can be a reasonable and reliable device.

On the marked there is another MMU, that in order to solve some of this loading/unloading issues, decided to take a different path. NEVER unload the filament. The filament is cut and join/melted together during the print on the MMU, so no loading/unloading are required from the extruder, since the filament has a continuous flow into the extruder.
However also this uses a buffers like prusa MMU, even if it is inside the MMU itself (that is bigger).
But also this solution has potential issues: failure on the melting/joining process, possible dis alignment from the Gcode printed and the corresponding color really extruded. The need to use a "supervisor" that analyze gcode to be printed and in real time prepare the next centimeters of filament of the proper color.

At the end neither this solution is simple, and most important it cost almost more than the double of the Prusa solution, so it is not for all.

So I think the Prusa MMU can be a winning solution for its market position, it only needs more time. We should remember that Prusa products are Open Source and improved by the whole Community.

The fact we find the kits or the assembled unit in a web shop is a plus. May be this is misleading on the expectations. Everyone is free to download schemes and firmware, buy the pieces everywhere, and build the units. As opensource product, you can have access to any beta, and and even to experimental versions. However the build is in your hands, and this can add thousands issues. If someone wants something perfect, reliable, first shot hit, than it will need to buy something professional that could cost even tents of thousand time the Prusa MMU unit. 😍