Questions: No. of Extruders, MMU, Einsy upgrades  

Re: Questions: No. of Extruders, MMU, Einsy upgrades

4) The firmware is actually pushing the existing 8-bit architecture quite a bit. At least that is the current theory, that the problems with printing over USB are because the firmware doesn't have time to both read from the serial connection, and do linear advance, and do filament runout protection, and everything else at the same time. The reason they didn't go 32-bit for the MK3 is that they didn't want to introduce too many changes to the design at the same time. source

Re: Questions: No. of Extruders, MMU, Einsy upgrades

1) How many extruders can the new Einsy board support?
I've read, and find it hard to accept, that it can only control ONE extruder?
If so, why?
(I know, no one here is officially from prusa, but many here have enough experience to wager an educated guess, which is more than I have now.)
This is something that makes me question if I really want to buy a Mk3 - the MMU doesn't look finished enough for me, but I know I want to print with multiple materials.

To my knowledge and based on the market so far, there are two completely divergent and mutually exclusive paths when designing a multi-material capable printer: A) multiple extruders, or B) a MMU type setup like Prusa's. Because Prusa chose to go down the MMU path, there was no reason to choose a board or develop firmware or hardware that would support a second extruder.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

4) The firmware is actually pushing the existing 8-bit architecture quite a bit. At least that is the current theory, that the problems with printing over USB are because the firmware doesn't have time to both read from the serial connection, and do linear advance, and do filament runout protection, and everything else at the same time. The reason they didn't go 32-bit for the MK3 is that they didn't want to introduce too many changes to the design at the same time. source

Hm. Makes sense somehow - I didn't think of all the little things that have to be done at (almost) the same time. The 8Bit chips are probably indeed simply an old design, so nothing is optimized for concurrent processing (assuming there is a way to optimize this, I'm no chip designer either).
Didn't think of that, thanks.

To my knowledge and based on the market so far, there are two completely divergent and mutually exclusive paths when designing a multi-material capable printer: A) multiple extruders, or B) a MMU type setup like Prusa's. Because Prusa chose to go down the MMU path, there was no reason to choose a board or develop firmware or hardware that would support a second extruder.

Hm.
That is a good explanation. So they chose a conservative approach of changing as little as necessary (see also Zez's answer to point 4).

Thank you both.

Now I have a new question 🙂

5) Would it be possible to expand the Einsy, using this Stepper Expander? From what I've read, it can be used with a RAMPS, and the RAMBO is a variation of RAMPS?
(No need for details, a simple "No, they're too different." or, preferably, "Yes, no problem" is enough. Or perhaps a "Maybe, but that idea is useless anyway".
I'm merely daydreaming here. I want multicolor, but I'm less than convinced about the MMU. Not sure if something like the Diamond hotend really would be better, though.)

Re: Questions: No. of Extruders, MMU, Einsy upgrades

4) The firmware is actually pushing the existing 8-bit architecture quite a bit. At least that is the current theory, that the problems with printing over USB are because the firmware doesn't have time to both read from the serial connection, and do linear advance, and do filament runout protection, and everything else at the same time. The reason they didn't go 32-bit for the MK3 is that they didn't want to introduce too many changes to the design at the same time. source

Disclaimer: I really like the Prusa i3 MK3 and when I get a 3D printer this would be the one.

OK, this argument is just unreasonable in my opinion. Prusa keeps adding more and more cool stuff to the printer, which requires more processing power. On top of that, I've been reading people's opinions that the 8bit 16MHz (yes, "M" as in "Mega") have been too slow for some time now. I just don't understand the "too many changes" argument when it comes to the processor.

The main issue that I have with the argument is that they skipped a core component. Now, with issues related to processing power are starting to crop up, there is no way to address them until the MK4. It's not like they can just swap the core processor on the MK3.

Also, isn't the board developed by the Ultimaker people, not Prusa?

Re: Questions: No. of Extruders, MMU, Einsy upgrades

4) The firmware is actually pushing the existing 8-bit architecture quite a bit. At least that is the current theory, that the problems with printing over USB are because the firmware doesn't have time to both read from the serial connection, and do linear advance, and do filament runout protection, and everything else at the same time. The reason they didn't go 32-bit for the MK3 is that they didn't want to introduce too many changes to the design at the same time. source

Disclaimer: I really like the Prusa i3 MK3 and when I get a 3D printer this would be the one.

OK, this argument is just unreasonable in my opinion. Prusa keeps adding more and more cool stuff to the printer, which requires more processing power. On top of that, I've been reading people's opinions that the 8bit 16MHz (yes, "M" as in "Mega") have been too slow for some time now. I just don't understand the "too many changes" argument when it comes to the processor.

The main issue that I have with the argument is that they skipped a core component. Now, with issues related to processing power are starting to crop up, there is no way to address them until the MK4. It's not like they can just swap the core processor on the MK3.

Also, isn't the board developed by the Ultimaker people, not Prusa?

I was going to defend their decision - in a way I still do, changing architecture like that isn't just slapping on a new chip, it's rewriting everything - but now I wonder: Changing to 32Bit would be a gargantuan task, no doubt (start from a fresh firmware branch, re-introduce everything, testing etc...), but then, why was it necessary to introduce the Mk3 NOW?
Why not introduce a Mk2S+ (or whatever) that has most of the new features (better stepper drivers, better heatbed, fans, maybe sensors?) and work on a completely new base - a Prusa i4 with 32Bit or whatever.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

Why not introduce a Mk2S+ (or whatever) that has most of the new features (better stepper drivers, better heatbed, fans, maybe sensors?) and work on a completely new base - a Prusa i4 with 32Bit or whatever.

That's what they did. They're not calling it Mk2S+ though. They're calling it MK3.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

Why not introduce a Mk2S+ (or whatever) that has most of the new features (better stepper drivers, better heatbed, fans, maybe sensors?) and work on a completely new base - a Prusa i4 with 32Bit or whatever.

That's what they did. They're not calling it Mk2S+ though. They're calling it MK3.

Bad naming then, IMO.
The naming MK3 sounds more grand than the actual changes, good as they may be.
I also don't understand why they redesigned the mainboard - that too seems to indicate a fundamental change, not just a few new support-chips.

In case anyone read my first reply - I've edited it, to sound less aggressive. Hopefully this is better.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

...changing architecture like that isn't just slapping on a new chip, it's rewriting everything - but now I wonder: Changing to 32Bit would be a gargantuan task, no doubt...

I am not sure that it's as bit of a task as you think. Going to a bigger CPU with the same instruction set is not that big of a task. After all, 8bit math still works in 32bit registers. If the instructions set has been kept compatible, it should more-or-less work. That is why 32bit software still works on 64bit machines of the same architecture without the need for a re-write. Testing it would be the bigger task but they had to do that anyway since they went to a completely new motherboard.

Again, I am in no way trying to bash Prusa. After all, this is the only manufacturer that I keep reading good things about and I really like the direction that they are taking. However, it seems that they are already on the edge of what the CPU can do (and a bit beyond it from some posts). So, it seemed natural to see a newer, faster, better CPU.

On the other hand, I am not sure that they had that choice. As I said, I don't think that they are the ones making the board. I think they are just sourcing it.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

I don't understand why they didn't go with the Duet Wifi for the board.

It would have solved some of the issues people are having. And you would have got instant web interface for updating / controlling the printer. No messing about with Pi zero or any extras. Quiet drivers. Plenty of extras - especially IDEX if / when it ever comes.

MMU is just not a good solution for multi-material. Too wasteful and they can't do flexibles plus blockages are still a big headache. Multi-colour, fine, but then how much use is it? Great for signs, sure. But you would have all that with IDEX too, plus proper MM capability, i.e. soft hinges.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

MMU is just not a good solution for multi-material.

I've been following all the drama about the MK2 MMU, and also drama from other multi-extruder printers all the way back to the Replicator. And I've come to the general conclusion that there are no good solutions for multi-material, at least not at the hobbyist level. There's some not-terrible solutions, and each not-terrible solution is less-bad in its own way.

The downsides to multiple extruders (including Independent Dual Extrusion) are that the Z-height of each extruder needs to be managed very carefully or else the nozzles can drag around already-printed material, and each extruder adds more weight (and resulting mechanical issues) to the print head. Going to larger number of materials (4 or more) doesn't scale well at all.

The MMU approach has got its own problems of course, and it's a relatively immature design. But with time and effort I can see this being a very fruitful solution. In theory there's nothing preventing an MMU-style design from allowing 8, 16, or more materials in the same print if one wanted to add enough extruder motors and build a big enough multiplexer.

So for what it's worth, I give props to Prusa for at least trying to push the envelope on a new way to do multiple materials, even if it's not really ready for prime time yet.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

Unfortunately MMU was the big reason I went with Prusa, Before learning about Prusa my list was Taz6 and Makergear M3 ID. Part of me still regrets not getting the Makergear on black Friday but I figure worst case I can use this one to learn from.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

On the other hand, I am not sure that they had that choice. As I said, I don't think that they are the ones making the board. I think they are just sourcing it.
My impressiong (based on nothing, actually) is that they design the board and ultimaker produces them. *shrug* Doesn't matter too much in the end - sourcing or designing it, it was their decision to go with it. I doubt anyone forced them.

Unfortunately MMU was the big reason I went with Prusa, Before learning about Prusa my list was Taz6 and Makergear M3 ID. Part of me still regrets not getting the Makergear on black Friday but I figure worst case I can use this one to learn from.

The Prusa Mk3 seems to be great, but I want a printer that is capable of multi material print. If not immediately, then at least as an upgrade option. But since even the Mk3 can only address one extruder, I'm not sure what to think of it anymore.
It's a bit off-topic, but apart from those two, do you recall any other multi material printers you thought about?

Also this talk about the electronics has brought up a new question from me:

6) Would it be possible, and not too stupid, to replace the electronics?
I've seen a thread talking about Smoothieboard, that was about the Mk2, so I guess for the Mk3 it's probably even less of a good idea, or maybe they're better by now? No idea, the thread was inconclusive.

I realize this would make the printer a sort of Mk3 Zombie, but while I like the body of it, I'm not sure what I think of the brain. Before I spend a decent sum of money, I'm going to explore all angles I can think of.
My thoughts, regarding a brain-transplant, is that at least it'll still be a Prusa body, i.e. an open source RepRap, which I would prefer to some closed source machine where everything is custom-made just for them.

Or, alternatively

7) Would it be possible to change the MMU to a different kind of dual (or quad) extruder?
I'm still not entirely clear on how this thing works - the electronics part, I mean. But there are 4 motors that can be accessed, so it shouldn't be a problem to use a different extruder?
I suppose there is a good reason not to do this, or I'd have read about it here?
(I mean, here are people who literally take the thing apart not only to see what makes it tick, but also rebuild it so it has a better tock. So I guess I'm missing something obvious which will make me feel like a fool afterwards.)

Re: Questions: No. of Extruders, MMU, Einsy upgrades

I think it is quite possible to replace the electronics or the extruder. The i3 mk3 is open source, you can upgrade and design your own parts.

I generally plan to upgrade the control board eventually. I'm hoping it is in 3-9 months when a 32 bit Marlin Ultimachine/Prusa board is available, which would likely have Prusa support and firmware. If Prusa can get the current Einsy working with the advertised features and external USB Octopi, I'll likely wait that long.

If it turns out the Einsy board won't cut it, it'll be a tougher call. If you switch to another board, you are switching to another firmware and likely compiling it from source yourself. Doable, but not for the faint of heart, and you would lose any Prusa specific optimizations in their firmware.

As for upgrading the extruder to a multi-extruder version, that should be doable. You could drop in a Kraken or something. Just design the hot end carriage and mount your extruders somewhere. There would be lots of trade-offs (electronics, leveling nozzles, more mass on the hot-end, slower, etc).

I've got an mk3 MMU upgrade coming. I'm undecided as to whether I'll install it immediately or not.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

@rejutka I really wasn't looking at much else, the Prusa was at the low end of my budget though. I'm new to 3D printing so I'm about as far from being a good source of information on 3D printers as you can get. As @Paul.m27 the M3 is open source and that also was a big deal for me. I figure if the unit is widely adopted the community may be able to tweak and improve the unit over time. My thought is that prices will decrease as the technology matures so I won't be out as much when I feel the need to upgrade and Prusa units seem to maintain decent resale value. My ideal printer would be SLS but I don't see one of those hitting my budget anytime soon.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

6) Would it be possible, and not too stupid, to replace the electronics?
I've seen a thread talking about Smoothieboard, that was about the Mk2, so I guess for the Mk3 it's probably even less of a good idea, or maybe they're better by now? No idea, the thread was inconclusive.

I realize this would make the printer a sort of Mk3 Zombie, but while I like the body of it, I'm not sure what I think of the brain. Before I spend a decent sum of money, I'm going to explore all angles I can think of.
My thoughts, regarding a brain-transplant, is that at least it'll still be a Prusa body, i.e. an open source RepRap, which I would prefer to some closed source machine where everything is custom-made just for them.

Dealing with firmware is not the only side effect of replacing the brain. Since the main control board has the motor drivers built into it, you'be loosing all the advantages of the new stepper drivers. You'll have to not only compile but also maintain your own firmware. You'd have take the firmware provided by the new board manufacturer and add/remove/change anything related to the Prusa machine. Not impossible but definitely a lot of work, given you have to do it for every release.

Unfortunately, getting a newer microcontroller is not the same as swapping the CPU on a desktop, where you can keep everything else.. It's more like swapping the motherboard on a laptop - every peripheral goes with it.

Basically, my assumption is that unless you are very knowledgeable in both HW and SW design and development, we'll have to wait for Prusa to move to a newer microcontroller. Hopefully, that won't be too far into the future given they are already pushing the limits of the current one (and just barely came out).

Re: Questions: No. of Extruders, MMU, Einsy upgrades

I generally plan to upgrade the control board eventually. I'm hoping it is in 3-9 months when a 32 bit Marlin Ultimachine/Prusa board is available, which would likely have Prusa support and firmware. ...
I've got an mk3 MMU upgrade coming. I'm undecided as to whether I'll install it immediately or not.

I've been lurking and waiting for the system I want to appear on the market. I am tempted by the mk3, but it just doesn't do that much more than my existing printer, and most importantly, even with MMU, it'll never do what I really want.

The printer I'd like would be Duet Wifi based - silent stepper drivers, 24V or 12V, wifi enabled interface for updating firmware / tweaking.

It would be a rail based system. Smoother, quieter, less chance of layer shifting, able to carry more weight for direct drive systems.

IDEX, because MMU isn't what it claims, it should be called MCU. But the use case for colour printing is weak, unless it's full mixed colours. 2 or 3 colours have about the same number of useful applications. Signs, logos, etc. After that, it's mostly cosmetic interest.
IDEX gives the chance to do real multi-material, especially stuff like flexible hinges / fasteners, which is a killer use case and there's no crazy wasteful purges.

But mk* is unlikely to ever go the IDEX route. And upgrading to something like a Duet Wifi wouldn't be worth it - you might as well start from scratch and design your own.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

The stepper drivers are the least problem, I think - I've seen boards with removable steppers, so you could drop those in you like.

Firmware is a bigger problem, but I'm not sure if it is that much of one - from what little I've seen, firmware can be configured easily (I'm thinking of the repetier configurator).
Of course, that still requires the firmware to KNOW about the features of the printer...

My main problem is, like with biscuitlad, that the Mk3 doesn't really do THAT much more than my current printer. Quantity, not quality!
I'm sure it'll be better in every respect, but in the end, it's a single extruder system with no way of upgrading to multi-extruder.
The other Mk3 features are nice, but not enough to make me want to lock in into it for years. And I don't intend to buy a new printer every other year either...

Unless the MMU does its job. Of which I'm not convinced, even if I take support-forum-bias into account, the deafening silence from Prusa on this OFFICIAL forum (check the logo on the left, and the domain!) is quite disheartening and feels as if they don't care at all. Either for their customers/fans or their product (i.e. either look around for ideas, or at least tell those looking for solutions if you've tried and discarded a potential solution).

Bottom line: this isn't fun at all.
I see three choices for me:

a) lock into Mk3 single extruder
b) MK3 zombie
c) something else

Part of me wants to go A, because I'm convinced it's a solid printer.
But I'm also convinced that with multi extruder there's lots of cool stuff to do (of which I have idea yet, since I usually only know it once I've handled it).

Meh, for now the goal is to actually reactivate the Wanhao (ironically, also needing a brain transplant. Melzi power connectors melted). I was intending to wait for a few months anyway, to see how the Mk3 develops and matures.
Maybe the MMU will get a magical upgrade that'll make everything better 😀

Re: Questions: No. of Extruders, MMU, Einsy upgrades

But I'm also convinced that with multi extruder there's lots of cool stuff to do (of which I have idea yet, since I usually only know it once I've handled it).

If what you really want is to experiment with multi-extruders (and not just MMU), why not just buy a Flashforge Creator Pro? It's a mature design, and $900 from Amazon fully assembled.

Re: Questions: No. of Extruders, MMU, Einsy upgrades

Unless the MMU does its job. Of which I'm not convinced, even if I take support-forum-bias into account, the deafening silence from Prusa on this OFFICIAL forum (check the logo on the left, and the domain!) is quite disheartening and feels as if they don't care at all. Either for their customers/fans or their product (i.e. either look around for ideas, or at least tell those looking for solutions if you've tried and discarded a potential solution).

Personally I'm glad they are not spending their time replying to all the questions in this forum...I'd much rather them spend all their available time fixing the firmware, getting the MK3 MMU ready to ship, and providing direct technical support. If you want answers to your questions directly from PRUSA, I'd email or chat with them directly.