Ender 5 Plus Dual Extruder (Part 2 — Planning the upgrade)
So, in the first part, I explained that this printer is truly crap. Unlike the Ender 5 Pro which is one of my favorite printers (with the addition of bed supports), this printer is a starting point for something glorious.
So, yeh, I have 7 working 3D printers and have been screwed repetitively by some of them and several others I’ve owned.
I decided it was time to get a decent FDM printer which met all my requirements. Unfortunately, no one makes one.
My requirements
- Mass economy
- Large
- Y-axis moves the head, not the bed
- Dual extruder
- Marlin firmware
- Auto-leveling sensor
- Quiet
- Enclosed chamber (for later)
- Serviceable
So, mass economy is pretty important. I’ve owned printers from XYZ and I’ve also purchased a $3000 FlashForge Creator 3 which is a total piece of shit that I’ll burn as soon as this mod is finished. I want a printer which pretty much everyone has or knows or something. I want the printer to be something where if a part breaks, I can either call the vendor or even make the part myself if needed. Printers break… simple as that… a warranty from a Chinese company who have service technicians who don’t respond for a month because it’s a holiday is a nightmare. I want to just be able to search online and find someone who has already done what I’m want to do and do it myself.
Never buy a printer which doesn’t have a Git repo and if it does have a Git repo, it doesn’t get updated for months. The exception is if you’re going to replace the motherboard and do it yourself. Which is what we’ll do here.
The plan
Start with an Ender 5 Plus, gut it, and fix it.
This sounds easy enough. All we need is the parts… which when all added up and including import tariffs and shipping… costs as much as the printer. But who cares? I have the printer and I don’t want to buy yet another one. So, we’ll gut it and build it up again.
Dual extrusion will be the goal. This is really important. If you’ve never had a proper dual extrusion printer before, then you really don’t know what you’re missing. Maybe you’ve had one and it never really worked. Ok, that’s fine. Let’s face it, mixing two different colors of PLA is pretty boring, but it’s something to do.
The real goal of dual extrusion is to provide supports. The reason is simple, you want every part of the surface of what you’re printing to look like those perfect finishing layers. You may be ok with sanding and post processing and maybe using something like PolyMaker PolySmooth to make your prints pretty, but I’ll be honest, I love my FormLabs Form 2, but I can’t even begin to imagine the discussion with my wife if I tried to spend over $10,000 on a Form 3L. And I want my FDM prints to start looking a lot better… I want all surfaces to be nice and clean.
Water soluble supports
Unlike normal supports which leave terrible artifacts on your prints, water soluble supports are truly specially because you provide a full removable surface on your prints. You effectively print a perfect mold to provide support.
If you think of a normal print as someone spending the day in the sun wearing a tight lace shirt, by the end of the day, the tan lines left behind look somewhat silly… generally uncomfortable and not overall nice.
On the other hand, if that same person would have work a t-shirt which covered their body somewhat evenly, there are no weird artifacts left behind.
This is what water soluble support accomplish for your prints.
Water soluble supports are not easy
I’ve experimented for a long time with printing them and the experience has been very mixed. At one point, I was sure the reason that Ultimaker 3 printers did water soluble so amazingly perfect was simply because of the dual nozzle mechanism. I’ve found out from multiple attempts so far that the real reason is not just the dual nozzle, but also that Ultimaker perfectly matches their materials.
An Ultimaker 5… which is probably the closest thing there is to a perfect FDM printer to date gets it right every time. Prints from that machine are nothing short of glorious. But again, I’d have to explain to my wife that I spent $12,000-$20,000 on a printer. And then she’d ask me what it is that I print that actually needs a $20,000 printer…and I’d tell her… parts for making printers. And she’d leave me.
Ultimaker combines 5 aspects which make them the perfect multi-material printer.
- Dual nozzles
Never ever try to print support materials or for that fact two non-similar materials through the same nozzle. It just really doesn’t work. Sure you can make a purge block, it’ll likely weigh twice as much as your print. It’s just a bad idea.
- Raised and lowered nozzles
The Ultimaker raises and lowers one of it’s nozzles when changing materials. This is the unique feature of that printer which makes it mind blowing amazing. When changing materials, the head travels to the side of the printer and flips a mechanical switch which lifts or releases a spring loaded nozzle. You have to calibrate the z-offset of both nozzles for this to work properly, but the software takes care of it and frankly works like a charm. At some point, I intend to design and 3D print my own version of this. For now, I’ll just have to skip this feature.
- Perfectly matched materials
Ultimaker filaments are all tested with their support materials and when you buy filament which claims to be compatible with Ultimaker, then they have tested as well. Ultimaker only makes 2.85mm filament, so unless you manage to find an alternative for dual extrusion on a Creality printer that can handle 2.85mm, that’s a no-no.
The solution however is to find materials that work together. I’m tired of playing the game of matching them myself. So I’ve already purchased matched materials. My choices are as follows :
- PolyMaker CoPa Polymide and Polymaker PolySupport these two are pretty much guaranteed to work together. The only issue I’m expecting is that Polymide is a Nylon which is extremely hygrophobic which means that it sucks up water like crazy. So when removing the PolySupport in water, it could damage the print. I’ll have to find out from experience.
- BASF UltraFuse PAHT CF15 and BASF UltraFuse BVOH these are extremely expensive materials, and the PAHT should have the same problem as the Polymide in water, but I’ve smashed this material with a hammer before and nothing happened. It’s an amazing material. But at those prices, I have no intention of anything less than perfection. Hence I must have proper support structures. We warned about the PAHT CF15, it’s not a high resolution material, it requires a 0.6mm nozzle or it almost certainly will rip the nozzle apart or clog.
- SpoolWorks Scaffold and well SpoolWorks anything else.
- An excellent bowden configuration
- 75mm filament and Bowden is just a terrible idea. Don’t get me wrong, it works and with Capricorn tubing, it’s not at all bad. But it’s not good for many materials. Ultimaker doesn’t have these problems because they have 2.85mm thick filament which frankly is barely flexible. As such, Bowden extruders work very reliably with this configuration.
- Dry materials
The Ultimaker 5 added a filament enclosure which not only could automatically change materials (meaning a lot more than just two spools) but also included a dehumidifier. This is critical, not just for the water soluble support material, but for nylons and well… pretty much anything else.
Summarizing a great dual material printer and what we can do
We can’t currently do the raised and lowered nozzle configuration the Ultimaker 3 and 5. And as I said, I intend to conquer that problem another time. As I have a FormLabs Form 2, I can make some pretty impressive parts for printing. In fact, I’m almost tempted to make my own heat sink using FormLabs High Temperature resin.
Also, consumer printers are all 1.75mm filament and so are the accessories. So, we’ll have to deal with that problem by simply using dual direct drive.
Other than that, I think we’re good to go.
The start
I’ve always loved shopping. I have tons of printer parts laying around the house ready to build yet another printer… and yet every time I start a project like this, I realize I need to buy more.
At this point, I have everything… or close to it and just received an e-mail saying the last piece is in the mail.
The dual extruder assembly.
I initially considered using two Micro Swiss all metal hot ends. In fact, I am very tempted to mod my Ender 5 Pro with these. I have two of them on my desk in front of me. But I’ve decided against that. After all, I just feel like it will be a calibration nightmare.
Instead, I’ve opted for the E3D Chimera+ which I ordered as 24V and air cooled.
I had started the project with the intention of using a cheap Chinese clone I bought from AliExpress some time ago. This way, I wouldn’t have to spend the $100 or so on the Chimera+ and if I got it all working, then I’d of course consider getting it.
Well, that failed because… well, the Chimera+ and the clone are not quite the same.
When trying to figure out how to mount this beast, I stumbled across
This lovely BMG-X2 Chimera Creality Ender 5 Extruder Mount on Thingiverse. And I instantly fell in love. It looked like a near perfect configuration for the printer. I could start with this and mod it as I went along.
I have a few BMG clone extruders which I’ve bought from Banggood and they work well enough, but they’re very big and bulky and even using the matching left and right models, it would be very difficult to mount them. This one made use of something special.
It used the BMG-X2 Extruder which is designed specifically for working with the E3D Chimera! It was like it was made just for me!
So, of course I ordered one as soon as I saw it. It got here the next day too!
And then I tried to mount my cloned Chimera to the bracket… and… it simply didn’t fit. The clone uses external collets where the real thing uses embedded collets. So, I’ve ordered the real deal from E3D as I mentioned above.
Printing the mount
Ugh!!! So, the mount from Thingiverse is flimsy… that’s not fair, it’s made with thin parts on it. This would be fine, but I didn’t want to print a twice as much support material as fixture material and the printer I was using was at the office (the Ender 5 Plus in question) and it has PolyMaker PolySmooth installed. And it printed… and it failed.
Then there was the issue of the fan ducts. I mean, they’re benign enough, but I want them to look amazing… and to do that I would need to print them at 45 degrees to minimize the amount of internal supports… but it would be at the cost of a lot of external supports… and of course the first thought that came to mind would be :
If you’ve never printed in Proto-pasta Candy Apple Metallic Red HTPLA, you should start now. It’s the most beautiful filament I’ve ever seen. It’s absolutely lovely. But you need to run heat tests on it. On my Ender 5 Pro, I have to print at 212 degrees to get that beautiful appearance. Of course, if you do this, then printing at 45 degrees will likely not make it sparkle nearly as well as it should. So, it’s not the best option here.
The end result was to start printing on the Form2.
I printed all these parts directly on the plate which is generally a bad practice with SLA printing. It causes bubbles up from the plate, but I figured… why not? If the print fails completely, I’ll just have to try again.
As you can see, the pieces which came out were amazing. The only real problem was that the fans ducts have internal supports which are pretty unobtrusive, so I’m not worried. But I think I can jam something springy into the vents to break them loose.
The main bracket would have been finished by now, but sadly, I was stupid and let it print for 4 hours this morning before realizing that I didn’t install the wiper properly on the tray and as a result, the print just didn’t stick to the bed. I had to start over. It’ll be done in 4 more hours. And that’s fine since I don’t actually have the Chimera+ yet.
The Motherboard
Ok… here’s where I have a serious dilemma.
Since I have an X, Y and two Z motors as well as two extruders, the BigTreeTech SKR 1.3 (seen on the left) which only has 5 drivers can be a problem.
On the other hand, the PandaPi v1.1 seen on the right wasn’t even supported by the time I received it in the mail. But it has room for 6 drivers… but I only own 5xTMC2209 drivers at the moment.
I love both boards, and I’m tempted to order either a newer PandaPi or a bigger SKR Pro v1.2 which I know is great. Another option is hold out a little while and buy the BigTreeTech GTR V1.0 when I can find where to buy it.
So, for the moment, I think I’ll stick with the SKR v1.2 which will require a splitter for the Z-Axis motors.
One option is to buy a board like this. But it’s amazing how long it takes to get this by mail. I’d probably wait a month… and for what… 3 connectors and a cable?
So, instead I decided to just make something for now.
I haven’t even clipped the resistors off yet
But I mounted it and while it’s the worst soldering job I’ve ever done in my life, the soldering iron’s tip wiggles around for some reason… and it’s time to replace it. But it verifies. There are no shorts and all wires are connected as they should be.
What I hate about this solution is that even though the Z-axis motors aren’t using much power as they barely ever run, I don’t like placing the higher load on the driver.
From my Ender 3 which I’ve added a BigTreeTech TFT24 to, I have this LCD panel left over. This is what I’ll use for the printer. I’m going to get rid of that TERRIBLE touch screen. There’s just no reason for it. Everything I need is already in OctoPrint except for configuring the Z-axis offset. I’m quite sure that’s an option on Marlin.
Now, to get going, I’ll download and print this Ender 5 Plus Bigtree SKR 1.3 1.4 Bracket by mjcopper over at Thingiverse.
Cable management
I need to work on cable management. One option I have is to start crimping extension cables. Another is to make a circuit board I can mount towards the top of the printer and run one big ribbon cable up there. This of course a far more attractive option.
Keeping it dry
Some time ago, I got a nice discount on a PolyMaker PolyBox
I’m not entirely sure what the difference between version 1 and version 2 is, but I have a version 2. There’s a major drawback to this design though.
It uses Silica Gel which is an extremely hygrophobic material and has to be changed regularly. It’s cheap, but it’s a pain. Not only that, but PolyMaker suggests keeping the material in the box while printing, but keeping the filament in the included resealable storage bag when not in use. I prefer to just keep the filament in the box at all times and this has ruined any material I had laying around. So I’ve had to spend quite a bit of time letting the filament cook in my food dehydrator to recover from filament left sitting all through corona.
So I’m considering the purchase of a Print Dry Print Dryer which looks like a pretty good and even reasonably priced solution for keeping two spools dry. The only real big change I would make to this box if it were up to me would be to add a serial port or USB port to allow it to be controlled from OctoPrint. Then you wouldn’t need to be present to deal with it. You could just tell it, set the temperature to 55 degrees and cook for 8 hours. I imagine you could even print the entire time while the filament is being dehydrated since the filament going to the printer would be on the outside of the spool and probably dries first.
Closing
Well there’s nothing more to say until my Chimera+ shows up in the mail and the printing of the brackets are done. Maybe I’ll work on building Marlin for this platform and test it on a spare power supply.
I don’t like using Visual Studio Code for this… instead, I’ll install platform IO on the Raspberry Pi where I’ll run OctoPrint and just compile it using sed and bash.
