Hi,

I have an idea in my head for using dual extruders to print faster and cheaper but maintain quality.

Basically using a primary extruder with a 0.4mm nozzle for printing perimiters and using a larger nozzle for printing support and infill which can be printed every x number of layers at greater layer height.

I ran a quick test in simplify 3D as follows:

Using a 3D Benchy at 200% scale, 0.1mm layer height, 20% infill 50mm/sec the resulting build statistics are print time 8hrs 13min and material weight 69.75g.

If the material is £40/kg the material cost is £2.79.

I then re-run the exact same model, except put the infill on a secondary extruder, 1mm nozzle, sparse infill every 6 layers.

Build stats are print time 5hrs 22min, outer material weight 31.54g, infill weight 42.2g

If i use the same material as above to print perimiters but a cheaper material at £15/kg for the infill this makes material cost for perimiters £1.26, infill £0.63, total £1.89.

This basically gives a 35% saving on time and 32% saving on material cost, but the perimiters are still printed at 100 micron layers. And this secondary extruder approach can also be used reduce the cost and time for support material too, rather than simply throwing away the expensive material.

I know this is just what simplify3d build stats say, and real world maybe different, but I have usually found simplify3d to be farily accurate.

Has anybody out there tried this approach, had any success or failures with it?

Hi @MattG it’s an interesting idea, I’m just not sure it’s a complexity that I want in my printing life :slight_smile: I’d be concerned about running cheap filament through my hot ends/nozzles. I use high quality filament not only to pass that on to the customer but because it makes my life easier in terms of reduced jams/clogs, etc. You’d only need a couple of clogs/jams from the poor quality filament to wipe out the savings in time you’ve made. I’d also have to be swapping nozzles all the time; I use my dual heads primarily for printing support material which is usually either HIPs or, increasingly, specialist break-away material; this stuff isn’t cheap and works best at high precision so I wouldn’t want to be running it through a 1mm nozzle.

What’s your objective here? To get more profit per order or to be a better Hub for your clients?

I’ve used loads of cheap filament for my personal projects, designs, testing etc never had a clogged nozzle so maybe your unlucky there. This obviously won’t work where you want to use dissolvable filament and the objective is reduce time and cost.

PS when I say cheap filament I’m specifically talking about esun which runs fine, is just as strong as others and looks “OK”, I have had issues with completely “unbranded” before, but saying that I had a reel of colorfabb which was tangled to hell in the past!

The question then has to be, if you’ve never had a problem with cheap filament, why are you using the more expensive filament for the perimeters? :slight_smile:

I’m not really convinced the savings would be that dramatic. For a start, if you’re going to use two reels for every print that means you need to double the stock of your current filament type/colour range, which could be a substantial capital investment and increase in storage requirements. You’d also have to switch two reels for each print, rather than just one, which likely means you’ll have to switch more often. For example, I tend to have Black ABS permanently on one nozzle of one machine because I print it so often, but if I wanted to print white ABS, I’d have to switch it out to use the cheaper infill material in white and that means cleaning the nozzle both ways.

I’m not totally dissing the idea, it’s an interesting one, it just strikes me that the savings might be quite narrow once you add up the “peripheral” costs of implementing it. It’d be interesting to hear about your experiences if you go ahead with the idea for actual customer orders over time; heck, you may find it’s a revolution!

Esun works abosultely fine for me, but I use colorfabb and faberdashery, pretty much mainly for aesthetics, they always look better I can’t see any difference in printability or strengh. You wouldnt need double stock, you have one cheaper filament natural colour for the rapid extruder which stays there all the time, and load your other filament for the fine extruder. It doesn’t matter if the internals are the same colour as the external. I’m a little way off going ahead with it yet, it just popped into my head!

I think this can be a great idea, but the added complexity may not be worth it in the end. Adding a second printer to your repertoire will cut down print times more than fiddling with dual extrusion. If you have no space for more machines though, it is certainly worth it.

What is interesting, is that you don’t have to skimp on quality / strength as your second extruder can be set to run the PLA at a higher temperature (more flow, more melt, more strength), and the primary extruder is obviously doing its thing with your expensive material at a low layer height.

Things to consider: Nozzle cooling/heating times. If your printer allows it, you’ll want to allow your inactive nozzle to cool and heat up. Stops oozing and saves PVA clogs. Also, you will have a prime tower that you wouldn’t ordinarily have on a single nozzle set up. These things all affect time and material used.

On dual extusion there are a lot of rubbish designs out there causing terrible oozing problems and even more ridiculous work arounds. I think the best implementation i’ve seen is the bcn3d sigma, but to further improve on this there should be a blocking mechanism on the nozzle rather than a catch cup, I think that would work really well, and would save on priming towers, heating and cooling the nozzles etc, just retract, park, and restart! Wouldn’t work well for PVA because it would be sat there at high temps, but this whole idea doesn’t really apply to soluble supports unfortunately.

Also reading the blurb on e3d’s volcano, if used for the “rapid extruder” I think parts could actually be stronger whilst saving time and expensive filament.

Starting to quite like this idea. Wish I had the time to test it out…

I think you could essentially do this with one nozzle. I frequently use a 0.16 layer height and a 3.2 infill layer height printing it every other layer.

So maybe a 0.1 layer with a 0.3 infill layer every third layer.

Now I know it doesn’t take into account the cheaper filament etc. but may be a good compromise.

I do this at the momnet, I have a 0.4mm nozzle, 0.1mm layer height and sparse infill every 2 layers, going to 0.3 is starting to push the amount I can extrude out of that nozzle and would really need different temps or reducing the speed which is counter productive. It’s doing this that got me thinking about the idea. Plus the fact I printed a part a couple of weeks back that required A LOT of support, and it just didn’t seem right throwing all that good material away!

Actually, I love this idea, and am kind of upset that I didn’t think of it myself!

It is interesting for sure.

Isn’t this what the Robox does?

Is it? I haven’t looked at that printer, I will do now, thanks for that.

I think it can still be used for PVA. This is actually one of my gripes of S3D is that it claims ‘smart support’, but to me the perfect support structures would be able to use 2 tools to make the majority of support out of cheap plastic and finish it off with a thick layer of PVA. Cheap supports!! Using this method, you may never run out of a 1kg spool of pva…

Thats a good idea! That should be put as a feature request to s3d