The Diamond 5 colour – part 3 (finally tamed)

This is the third part of my experiences with the Diamond 5 colour hot end. For those who missed the first two instalments, they can be found here The Diamond 5 colour (“Fullcolor”) hot end (and other related stuff). and here The Diamond 5 Colour – part 2.

To summarise where I left off, the 5 colour version of the Diamond seemed to suffer from heat creep issues which the 3 colour version did not. I suspected that this was due to the lower heat sink fins being a smaller diameter making them less efficient at dissipating heat. This was backed up by some temperature measurement tests that I undertook. I concluded that a bigger fan might offer a solution but the first fan that I chose was much too powerful.

I am unable to source a 50 mm 24V fan with a higher flow rate than 20cfm. I use this site https://www.mouser.co.uk/Thermal-Management/Fans-Blowers/_/N-axg88?P=1y9hrqvZ1yobr5uZ1z0xjmuZ1z0vn4m and filtering it by 24 V DC, I get 2,551 fans and blowers listed. Further filtering by size reduces that number but the one I have been using has the highest flow rate of all those listed. So I concluded that the only way to get a higher flow rate is to use a larger fan. Increasing the size from 50 mm to 60 mm gives a wide variety to choose from. I tried a 55 cfm fan but found that this was far to high a flow rate and prevented the hot end from getting up to print temperature. So I selected two more fan. The first was this one  https://www.mouser.co.uk/Search/ProductDetail.aspx?R=109R0624G402virtualkey57900000virtualkey978-109R0624G402 which is rated at 27 cfm. Also of note is the noise level which is rated at 39 dBa and compares well with the 50mm fan I was using and which was rated at 40.5 dBa. The second was rated at 37 cfm but I won’t bother giving the link as I didn’t use it in the end.

As I have mentioned before, it is necessary to keep the air flow away from the heater cartridge which is located centrally in the brass part of the Diamond hot end. So the mount has a tapered tube running down the centre. The inner diameter at the bottom is just slightly larger than the heater cartridge and tapers up to just less than the diameter of the fan motor. Here is a picture. Ignore my 3 point mounting system which is “tacked on” to the square fan mount.

stdMount

So, to fit a 60 mm fan, I made a simple adaptor but as the new fan would be 20mm higher, the top of the tapered tube would no longer be masked by the fan motor. Therefore the adaptor had a disc at the bottom which would effectively cap the top of the tapered tube. Here it is.

50to60adaptor

The two parts fitted together like this.

adaptorAssembly

Actually, the adaptor is oriented 180 degrees out in the picture but you get the idea.

My first attempt at using this was not a success. Basically there was too much restriction to the air flow causing back pressure and I could feel the air being blown back out of the fan. It was caused by the 20mm gap between the fan motor and the disc at the top of the tube. So I decided to do away with the adaptor and re-design the entire mount so that the tube extended right up to the fan motor. Here it is just printed.

60mmMount

Here it is looking down from the top with the 5 colour hot and assembled.

60mmMountFromTop

Here you can clearly see the tapered tube which just clears the heater cartridge at the bottom, and the top will now be almost flush with the fan motor. The airflow from the fan blades will now be directed down between the central tube and the heat sinks, with no disc in the path to restrict the flow.

Once the 60mm fan was installed, I then set up my experiment to measure the heat sink temperatures as before. Here it is

60mmTestSetup

As before, I fitted one thermocouple between the lower fins and one between the fins near the top of the heat sinks. Then I heated the hot end to 200 degrees and C and noted the heat sink temperatures. Here are the results

5Colour27cfmData

5 colour 27cfm graph

I’ll put the other results in here again for comparison. This is the 3 colour with 50mm 20 cfm fan

3 colour data

And this is the 5 colour with the same fan

5 colour data

As you can see, the results show that the temperatures for the 27cfm fan on the 5 colour are now very close to the temperatures with the 20cfm fan on the 3 colour version and some 10 to 12 degrees lower than when using a 20cfm fan with the 5 colour version.

With that in mind, I decided not to test the 37cfm fan as it seemed like the 27 cfm fan would be adequate and is less noisy.

So then I decided to fit this assembly and see if it actually improved the heat creep issues that I had experienced when using PET-G. Previously, I had found that after about an hour or more at print temperature (around 210 deg C) PET-G was swelling in the area of the heat break and “binding” inside the PTFE liner. This could be felt by turning the extruder gear by hand and required a lot of extruder tension to overcome. The increased tension would grind the filament away making printing difficult and very “hands on”.

Having re-tuned the PID parameters to suit the higher flow rate fan, I then loaded it with PET-G filament and heated the hot end to 215 degrees and just let it sit there. I can now report that with this higher flow rate fan, after an hour or more at print temperature, there is no increase it the effort required to extrude PET-G. Furthermore, fully retracting the filament shows no sign of the swelling that was previously noticed.

Here is a picture of the beast fitted on my machine. As ever, like all my other hot ends, it just drops into place with the 3 steel dowels meshing with the the 3 bronze bushes in the X carriage. The springs at the front hold it in place but allow it to tilt when pressed from below, to trigger the homing switch.  It’s not exactly the smallest, lightest hot end you’ll ever encounter.

5colour60mmFitted

I’ll put the Open Scad design on thingiverse but without the mounting arrangement that is specific to my printer. One thing to note is that assembly is very much easier if you fit the securing cable ties through the holes before clipping the hot end assembly in place.

Finally, by way of a torture test I printed this  https://www.thingiverse.com/thing:1330212 by Prusa research but scaled up uniformly to 360mm tall and in 5 colours. Print time would have been 28 hours but I had a layer shift issue (nothing to do with the Diamond hot end) at 26 hours so had to abort. Anyway, it was around 7 hours print time for each filament so the Yellow had cooked for 7 hours before it came into play, the Blue 14 hours and the clear 21 hours and there was no sign of the previous heat creep issues. Here is a picture

PrusaVase3

I printed it with FilaPrint PETG with the exception of the Black which was Taulman T-Glass. I printed it with 3 bottom layers and 3 perimeters and ran it through a little script to change the mixing ratio every 3rd layer so that it goes from Black to Red to Yellow to Blue to Clear. Previously when I tried to print the scaled up version of Lubie’s Aria in PET-G, I ran into all sorts of problems with the filament swelling and binding in the PTFE tube.

Here is a little video of it being printed.

 

So in conclusion, I have to say that I needed to use a 27 cfm fan to cure the heat creep issues that I encountered. I am unable to source a 50mm fan with that air flow so had to use a 60mm fan and modify the mount accordingly. It is perhaps worth noting that the “official” RepRa.me mount will only accept a 50mm fan. Furthermore, I did not need to use an 80 Watt heater as recommended and supplied by RepRap.me and found that a 40Watt heater is more than adequate and much easier to control.

Ian

 

 

 

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s