My 6 Input Mixing Hot End – Part 8

1. Introduction

Further to my last post (part 7), I have now made and fitted the part cooling solution, assembled the latest version of the hot end (although it is still lacking heat breaks), completed the fitting of the gantry belts and wiring, made and fitted a small circuit board for the Z probe/switch, and run homing tests.

2. The part cooling solution.

As I mentioned in my previous post, I don’t yet know what the final size and shape of the hot end will be, and there are likely to be different variants too. So I needed to design a part cooling solution which had plenty of adjustment options. I also mentioned that it is likely that I may also be printing with larger nozzles and/or at high speed, so a lot of cooling air might be needed.

I used some small 30mm blower fans with my Diamond hot end, but these were somewhat limited in flow by the ducting I made, so I changed them for 40mm blower fans. However, this new carriage/hot end combination doesn’t quite allow 40mm fans to be fitted in any position other than vertical or a maximum of about 20 degrees rotated from the vertical. So I elected to use the 30mm fans but without any (restrictive) ducting.

After a great deal of thought and several design iterations, this is what I came up with. Here are the parts I made. I’m using up odd bits of aluminium that I have lying around. The angle brackets were made from some square tubing that was left over from the filament storage rack I made, and the flat plates were made from old gantry plates that have been replaced with newer designs.

The flat plates have two M3 tapped holes for long screws and two M2.5 tapped holes to mount the fans. Here are the fans and screws fitted to the plates.

The long screws have lock nuts fitted and they slide through holes in another block with further lock nuts either side of that block. This will give me adjustment in the X direction (towards or away from the hot end).

The right angled brackets bolt to the carriage like so.

The slots give me height adjustment in the Z direction.

The long screws fitted to the fan plates go into the two holes in the blocks shown below. A bolt goes through each block and is secured to the slotted angle plate with lock nuts. This gives me further adjustment in the Z direction. The blocks are then secured to the bolt with grub screws. There is a small amount of adjustment in the Y direction by moving the block along the bolt, but they can also be rotated to position the fans at an angle.

Here are the fans roughly positioned for this particular hot end configuration. The air flow is biased to one side as it exits the fan so one fan will cool the part one side and in front of the nozzle, and the other fan will cool the opposite side and behind the nozzle. That’s the theory but I’ll need to run some tests and maybe make some positional adjustments. I’ll also need to run some tests to evaluate what speeds to run the fans at. These fans do work with PWM but I have to drop the frequency to a really low values (around 10Hz).

3. The latest hot end iteration

I’ve spoken before about the latest design revisions so I won’t go into details again. Here is what it looks like (without any heat breaks).

I mentioned that I have moved the heater and thermistor to make them more accessible (although that has added to the overall size and mass). Here is how the heater and thermistor are now retained.

For now, this is just a temporary assembly. When I get new heat breaks I’ll use Boron Nitride paste (that Slice Engineering kindly supplied) on the heater and thermistor cartridges. This is the other side of the hot end.

The screws at each end prevent the cartridges from falling out in either direction, similar to the method that Slice Engineering use for their Mosquito hot end. Doing it this way is, in my opinion, much better than clamping the heater and using single screws in the side to hold the thermistor in place, because that could easily damage them.

The block directly beneath the heater has just a single hole through the middle. This is the block which will be removed and replaced with a mixing chamber. But for now, I want to get the hot end working like any other “mixing” hot end (which simply combine filaments), and hopefully resolve the previous leakage problem that I had. Also, I want to evaluate what sort of melt rate can be achieved with these multiple, large melt zones. Then I’ll go back and revisit mixing.

4. Putting it all together

This is a picture of the hot end gantry assembly looking down.

The X end stops are fitted. Normally the Duet firmware will respect axes limits making it impossible to move beyond those limits. But for that to work, the printer must first be homed. So by default, the firmware only allows homing moves when the machine is first powered up. I find that inconvenient for commissioning and testing so I set the configuration to allow moves prior to homing. But sometimes I do something stupid and attempt to send the carriage beyond its’ physical limits, so I also have an X max end stop switch which will trigger an emergency stop if it gets pressed. I used this on Duet 2 but I haven’t tested that yet on Duet 3 because external triggers on expansion boards don’t work with the older beta version of firmware that I’m still using. I’m told that this is now fixed in later release candidate versions of firmware but I’ll wait for the stable release, then get around to updating it.

For the hot end gantry I use genuine Gates Power Grip belts although I can’t say that I ever had any problems using “generic” GT2 belts and I still use them on my other gantries where it’s far less important. Just to prove that these are genuine Gates belts, here is picture.

Here is a picture of the hot end in place. I still haven’t got the replacement hot end cooling fans so I’m using the high flow ones – hence the temporary, un-braided wiring. Also, the wiring for the heater and thermistor is longer than it needs to be because the firmware still does not allow PID tuning of heaters attached to expansion boards. So I have to move the carriage back to Y max, and temporarily connect the heater and thermistor to the main board to do that, hence the longer wires.

There isn’t much clearance. One thing I always forget about when I do my OpenScad designs is to include fixings and wires, so things always look like they have more clearance than is really the case. I had to change some 20mm long screws for 16mm long ones, and replace some cap headed bolts for button headed ones, but I got it all to fit and work as it should in the end. Here is another picture from below (the part cooling fans still need some positional adjustment).

I made this little board to hold the LED, resistor and a couple of Molex headers for the Z probe/switch.

I’ve temporarily fixed it to the gantry with a nylon nut and bolt and a piece of thin plastic to act as an insulator. I’ll make a better holder for it once I get a working hot end and can actually print some parts again, but here is a picture of it fitted (the red arrow shows where to look).

The switch works really well. I’m glad that I fitted the LED as it’s quite reassuring. I can’t see any movement whatsoever in the hot end when the switch triggers as it only needs to break contact. The following video clip shows the Z axis homing in operation. After heating the nozzle to around 140 deg C to soften any plastic that might have “oozed”, I drop the bed about 5mm, do a coarse fast homing move first, drop the bed again, do a slow homing move and finish by dropping the bed 5mm.

This is the first time I’ve “Embedded” a YouTube video with WordPress so please leave a comment if the video doesn’t play nice and I’ll fix it.

5. Next Steps

I’m hoping that it won’t be too long before I get funds together to replace the heat breaks that I stupidly broke. Meanwhile, I need to re-fit the silicone rubber strip that I use to wipe the nozzle. This needs to be re-positioned because the new hot end sits lower than the old Diamond hot end.

When changing from one colour to another, it’s sometimes (not always) necessary to purge some filament which I do simply by extruding into a “bucket”. Then after the purge, the nozzle is passed back and forth across the silicone rubber strip to wipe it. But this all happens at the back of the printer and now that it sits inside a “booth”, it’s difficult to empty the “purge bucket”. So I’m considering moving the bed further towards the rear of the printer and re-arranging things so that the purge/wipe happens at the front.

More to follow ……..

I hope anyone reading this is still managing to stay safe and well.

Ian