Purge Bin vs. Purge Tower

Purge Bin vs. Purge Tower (some facts)

I've said it a couple of times now: I don't like purge towers! Simply, because they're a waste of precious material.

I've recently checked this statement and the surprising answer is: It depends!

First of all: Why do we need to waste material on a multimaterial 3D print at all? Let's step back for a moment and think about, what you do as you swap filaments manually on your printer:

• You pull out the old filament

• You put in a new filament

• You extrude the new filament until it doesn't show signs of the old material/color anymore

In some cases, when the previous filament was a totally different type (i.e. PETG => PLA), we even clean out the nozzle using cleaning filament. The reason why we do this is due to a small amount of the old filament that still remains in the nozzle, which we want to squeeze out before we start a new print. 
This procedure is exactly the same on an automated multimaterial print. As soon as the filament (color) changes, the printer needs to get rid of the old one before it carries on printing. Otherwise we'll experience "color bleeding", which is just another term for "mixing up different colors to get some nice brownish color".

Hence, the first fact is: We can't (FDM) print multimaterial models without purging. We've got to do it the one way or the other.

So, what's the benefit of purging into some kind of purge bin?

The big advantage of this method is, that you only have to purge after a tool / filament change has taken place. Depending on the model you're about to print, it may happen that a tool change occurs only a couple of times. For example:

Hence, fact #2: It all depends on the model. We have to decide which model deserves which method of purging.

So then, what's the disadvantage of purging into a purge bin?

The main disadvantage is: It's a messy thing to do. Purged threads are literally everywhere and you need some tool that removes the purged filament from the nozzle after purging (i.e. my Wiper design) otherwise it may bond with your model and ruin the print.

And even then, you'll need to set up an ooze shield (or draft shield) , which also will use up some additional filament, in order to:

• a.) prime the nozzle after each tool change 

• b.) protect your print from oozed out filament threads

Ain't there a better solution?

Well.. kind of. You can do what I did in the beginning: I mounted a vacuum nozzle right under the printers nozzle and used a mechanical endstop to turn on the vac when the nozzle was in position using a Solid State Relay. The the machine purged directly into the vacuum. Yay!

But you probably meant "better" and not "a lot more complicated". I have to confess, it's a bit of an overkill and pretty noisy too (which can drive you crazy when you're trying to get some prints done overnight).

So, no, there are no better solutions yet... except:

Don't purge at all

I've written a console app (SMuFF-PP - which you'll find on my Github repository) which tries to avoid purging in general by moving the tool change command in front of the original location. Say, you need to purge out a 100 mm of material to "clean" the nozzle. You run the app and tell it you'll need at least 100 mm in the configuration file. The app will parse the GCode given by the command line and move the tool change command more or less exactly 100 mm in front, if any possible. In this case the print will run continuously and there's no need to purge. 

The major drawback of this method is, it works out only on bigger models with less tool changes. The "Traffic Cone" would be such a model. As soon as you try this app on, say, the 3D Benchy which has lots of small extrusions (some of them  even less than 1 mm), the app won't be able to relocate the tool change and hence insert the "purge to bin" GCode instead (yep, the remains have to go somewhere!). 

However, on the right model it works great and since it's a console app, it can be used from a post processor directive within your slicer, in order to get the whole process automated.

Although there's no real solution in sight yet, there are some tips and tricks for multimaterial printing.

First of all: Look out for better tools. 

The slicer I've mentioned and have used myself so far, Simplify3D, is a great tool but it's heavily outdated (yes, we're all patiently waiting for version 5 to hit the market). 

Unsurprisingly, there is (at least) one better tool for slicing more than two materials: That's PrusaSlicer from version 2.3 on.

In case I intrigued you now and you'd like to give it a shot, here's a collection of screenshots for the slicer settings you'll need with the SMuFF. 

Second: Think about print resolution.

Do you really need that model printed with a fine resolution? The very same "Traffic Cone" model printed with 0.1 mm layer height will use about 13580  mm³ for the purge tower. That's roughly twice as much as with 0.2 mm layers, whereas the same model printed with 0.25 mm layer height will only waste 6901 mm³.

Third: Print more models.

Instead of printing only one  model, print copies of the same model in one go. The amount wasted for the purge tower will still be the same as for one model but the ratio between filament used for model and for purge tower will differ (from a mathematical perspective). If you don't want to print multiple copies of the same model, try finding different models which use the same colors.

Fourth: Check other options.

Prusa Slicer allows you to use objects to be printed as for wiping as well. This will also reduce the amount of wasted material, though this will only apply to models which do have quite some infill to print. The "Traffic Cone" will not meet this criteria for example.

However, there might be different solutions in the future, so keep looking for software updates for the slicer of your choice.

Conclusion

You've now read about the pro's and con's with each purge option you have. Whatever method you pick as your preferred one, keep in mind the SMuFF can handle both, it's just a matter of configuration. Eventually, that's what the S in SMuFF stands for.