UM2 Dual extruder to single hotend?

[Crihz 1]

Hi everyone,

I have extensively searched the forums and googled but have been unable to find an answer to my question. If anyone can shed a light on this, please do:

What would it take to setup dual extrusion to single hotend on a UM2? So more or less; what would it take to make an e3d cyclops work on a UM2?

I have been thinking about a number of solutions but I am not sure which one would work or if these solutions are at all possible.

1. Tinker firmware: I know there is a dual extrusion version of Tinker firmware but does it allow to use just one heater and temp sensor? Would this mean compiling a custom version? (which I have no clue about) Or is this something you can select in Cura? How about the offset between the nozzles (which should be 0 for a single hotend)? Is that something that should be done in firmware or can this be done in Cura?

2. Hardware mods: use diodes to connect heater 1 and heater 2 outputs to trigger a mosfet that's connected to a single hotend heater cartridge. Use two temperature sensors in the hotend. Or will having two temperature sensors in one hotend trigger a safety when only one of the heater outputs is activated?

Best regards,

Chris

Edited February 23, 2017 by Guest

[krys 206]

Hi there,

Welcome to the forums!

I have not seen anyone talking on the forums about a UM2 single hot-end dual setup like what the Cyclops or what Mr. Prusa is doing. Maybe you are the first!

But there are a couple UM2 dual print projects going on right now using different techniques.

You might want to check out @foehnsturm's project here, for example.

(Disclaimer: I am part of the beta test for the above project.)

I am not sure about the other project (as it is not being coordinated in English so I cannot read it ), but I hear it is working too.

Or, maybe you want create the single nozzle version and then UM2 can have even more options! In which case, good luck!

Anyway, hope this helps.

[Crihz 1]

Thank you for your warm welcome!

Yes, that is exactly what I have in mind. Similar to "y-mixer" setup from Prusa (and many others out there).

I noticed the foehnsturm project, I wish I could be a beta tester. It looks very promising already.

Since you have been experimenting with dual extrusion yourself, would you be able to address the question about the thermistor? Will the firmware trigger a safety when for instance the second thermistor detects a rising temperature when the second heater cartridge is switched off?

I want to make sure what I have in mind will work before I start buying extra parts for this project.

What other project (the one that's not in English) are you referring to? Crema's water cooled setup?

Best regards,

Chris

Edited February 24, 2017 by Guest

[krys 206]

Yes, that is exactly what I have in mind. Similar to "y-mixer" setup from Prusa (and many others out there).

I noticed the foehnsturm project, I wish I could be a beta tester. It looks very promising already.

Since you have been experimenting with dual extrusion yourself, would you be able to address the question about the thermistor? Will the firmware trigger a safety when for instance the second thermistor detects a rising temperature when the second heater cartridge is switched off?

I want to make sure what I have in mind will work before I start buying extra parts for this project.

What other project (the one that's not in English) are you referring to? Crema's water cooled setup?

 

Well, I can say that @foehnsturm's project is working very well. It seems to me at least that we are just tweaking/improving on a working system at this point. But of course, I cannot speak to foehnsturm's release plan or anything. The design works very well though.

As for your idea, I am pretty sure you will need to customize the firmware. @TinkerGnome is the person to talk to about your idea in this regard. And probably for the hardware aspect too.

But it seems to me that if you are having to go the custom firmware route, then the extra electronics is probably not necessary. Then again, your hardware idea just might make custom firmware less/not necessary. I do not know.

I have tagged TinkerGnome (above), so hopefully he will see this post and respond.

Oh, and I might also be concerned with how retracting the filament affects the TFM coupler. You are going to have to be careful with temperatures, I suspect. But I am just guessing there.

As for the other project, yes it is @Crema's, here. I had to go find it again to remember

Anyway, it seems like you are taking on an ambitious project. Good luck!

[AreDigg 8]

Another interesting concept is this https://www.mosaicmanufacturing.com

I have ordered one of these but they only plan to make a version for 1,75 mm, not 2,85. So a conversion is needed for ultimaker.

[Crihz 1]

That is indeed what it looks like. If you only glance at Ultiarjan's picture post from feb 12 you know you are looking at something reliable. Even at this early stage of development it seems to be more than enough to satisfy most people looking into dual extrusion. Just think about what it will be able to do in the future.

Crema's approach also seems to work very well. But I wouldn't feel completely comfortable having water run through the hotend without taking extra precautions to keep the other electronics safe in case of a leak (which I have experienced on a watercooled PC in the past). I also wonder about vibrations from the pump traveling up to the hotend. The print results are amazing though and it's very reasonably priced, even if you include the cost for the pump and heatsink.

Palette isn't exactly what I am looking for but it undoubtly has it's place in multicolor printing. The unmistakeable steep price of such a complex system would also make me hold off on it.

My idea probably isn't the best one out there but it is very cheap and should be able to produce reasonable results, depending on what type of filament and extruders you will be using to accomplish it. I do realize I will probably have to switch to a different hotend, maybe a V6 lite will work best for this approach as it has the PTFE tube running all the way down to the nozzle. Some simple g-code scripts should make it all come together nicely. There will be a need for a purge on color change obviously but using a V6 lite will keep purge time to a minimum (but it will also limit print speed).

I hope @TinkerGnome will be able to chime in.

Chris

Edited February 25, 2017 by Guest

[tinkergnome 927]

Oh no, please do not have another project...

May i quote Mr. Průša?

 

The whole process is stupidly simple in theory, but devil is in the details and execution.

He says in the comments that all the hard work is done by a post processing script. I think, this is the correct approach to achieve the wanted result without a need for customized firmware and with minimal additional hardware.

Personally I would not connect two outputs to one heating element. There is only one hotend, i would stick with one heater and one temperature sensor.

The only problem is: the dual Marlin versions are working fine without a second heater, but not without the second temperature sensor (for safety reasons). But a PT100 can be easily faked by a simple resistor. If cold extrusions are allowed (M302), Marlin has no problem to move the second extruder motor (assuming that the faked sensor simulates room temperature).

As long as you don't want to mix both materials, some custom tool change scripts and/or a bit of post-processing should do the trick (and is needed anyways for the retraction/priming/wipe tower stuff).

[Crema 123]

Hi,

Do you want to produce something like this?

prometheus

cheers

Marin

[Crihz 1]

That was pretty much all I needed to know . Thank you so much @tinkergnome ! Any idea if a safety is triggered when using two pt100s in the heater block and only one heater cartridge? Because I'm thinking about slight temperature adjustments that might be needed for different colors of the same filament material.

Hi Marin, that is indeed what I am planning to do. Maybe not exactly the same but close to what prometheus does. I think it will be more than enough for my needs.

Time to order some parts...

Edited February 25, 2017 by Guest

[tinkergnome 927]

Any idea if a safety is triggered when using two pt100s in the heater block and only one heater cartridge? Because I'm thinking about slight temperature adjustments that might be needed for different colors of the same filament material.

 

To be honest: I assume i missed something... why do you need a second temperature sensor at the same hotend? It would always report the same temperature as the other one (more or less).

The second temperature sensor is not a problem, but a "heater error" is triggered, if you try to heat up a non-existing heater. That's why your gcode should never try to set a temperature for the second extruder. You can always set the target temperature for T0, and change it on every tool change (e.g. with a post-processing script). I bet that's one of the things that the script from Mr. Průša does.

It should also be possible to change all this behavior in the firmware. You can decide at some point whatever is easier for you in the end. @gr5 shared his notes about a proper build environment for Windows here.

[Crihz 1]

I believe I am the one who missed something . I haven't flashed the controller with the dual extrusion firmware yet and I haven't bought a second pt100 and extruder so far so I haven't really been able to test any settings in Cura or s3d yet or been able to see how the printer responds. I could do a simulation of the pt100 but then again if a second pt100 will work it might come in handy when diagnosing a problem...or it could cause one for that matter. I think the G code script will be fine for my use but I will read up on the link you posted for setting up a windows build environment. Thanks @tinkergnome!

Edited February 26, 2017 by Guest

[DeathMachine 23]

Very interested in this Idea. Thinking the prometheus system or the diamond hot end with only 2 in feeds due to room with the UM head.

Ed

[Crihz 1]

I have ordered some parts and will keep this thread updated as soon as there is any progress worth mentioning .

[Daid 306]

Note that I tried this with an UMO somewhere in 2013 I think. I didn't get reliable results. It kept jamming because of string in the Y piece. It never printed more then 20 layers without some kind of issue.

Times have changed, printers have improved, hotends got smaller. But I don't think this will ever achieve high reliability. Moving to 1.75mm might help a bit more.