Let's Make Robots!

How to Make an All-Metal Heat-Chamber (Hot-End) for your 3D Printer

Maybe I should have given this a sub-title, "How to Save 40–50 Quid (give or take) by making your own."

[ This page is now complete.]


The old heating section worked well, so I've duplicated that.

I start with a 3/4" (that's ~19mm for you metric users out there.) square solid aluminum rod and cut a piece a bit less than twice that distance in length. Does not have to be exact. Then I mount that on the lathe to drill and tap it. You do not have to use a lathe, but I have one so that is what I will use. If you have a drill press that will work also. If you are using hand tools, then you have to be very careful about holding things straight.

I drilled all the way through using the tap-drill size for threading the block (halfway through) for the 1/4" thread. (1/4" = 6.35mm)

then I drilled only halfway through with a 1/4" drill to hold the nozzle. *(As it happened the nozzle was a little loose, so I put a threaded hole in the side of the block for a set screw to hold the nozzle in place securely. Sorry, it is on the other side.)

I drill the hole for the feed shaft and the nozzle. It doesn't have to be exactly in the center, but it should be straight, so the filament feeds vertically.

Next I thread the hole halfway down to screw the 1/4" threaded rod into it. Doing this on the lathe is just one method of making sure the tap goes in straight. Anyone who has tapped many holes by hand knows the tap can get started crooked and then you are screwed unless you sacrifice some threads to re-tap it.

The threaded rod is where the filament will go through to reach the heat chamber, so obviously we have to drill it down the center with a 3mm (or 1.75mm) drill. The filament I bought says it is guaranteed to be 3.0mm +/– 0.1mm. That means it could be as big as 3.1mm, so I should drill a hole that is a wee bit bigger than 3.1mm. I actually used a drill bit that was closer to 3.2mm.

For this work, a lathe really shines, but you could do it on a drill-press too, as long as you align the threaded shaft perfectly with the drill bit. I would recommend drilling a smaller pilot hole first, and then going to the larger one. 

Ok, below we see the various parts I have collected to put the new hot-end/cold-end together.

I did not snap a picture of that process, but I had to drill a hole for the filament into the nozzle. I did that on the lathe to be sure of accuracy. The copper pieces I used for nozzles are welding tips for a MIG welder. They come with a small hole in the center. I am using ones with a 0.6mm hole. I cut the nozzle off so it would only go halfway through the hot-block and meet the threaded rod.

I took a heatsink from an old video card and drilled and tapped it to screw it onto the 1/4" threaded core.  I did that part on the drill-press so it is not perfectly centered, but it is close enough. Functionality is the important thing.

I cut a piece of thin springy steel for a mounting bracket, and drilled three holes in it, —two at a measured distance for the screws that will hold it to the X-carriage and the Wade's extruder, and then a 1/4" hole centered between them where the 1/4" threaded core rod will attach using the nuts you see in the picture. Since the metal is thin, it cannot be easily threaded, so there will be a nut above and below to hold it firmly in place.

Finally, I salvaged the heating resistors from the old hot-end. They will need coated with thermal paste before inserting them into the new hot-block.

The old insulation was not impervious to the heat and partly burned away, so I am relying on kapton tape for insulation this time.

I put thermal paste on the resistors and insert them into the holes in the block which were sized for a snug fit.

Next I assemble the parts:

1) put the nozzle in place and screw in a set screw to hold it

2) screw the 1/4" threaded rod into the top half of the aluminum block until it is tight.

3) screw the threaded heatsink on the rod, but leave an air gap between the heatsink and the hot-block.

4) I added a washer here but that was optional.

5) turn on one nut and lock the heatsink in place.

6) slide the thin metal mounting bracket on and add the second nut, tightening the two nuts against each other.

   (a thin wrench or needle-nosed pliers works well here to hold the nut in the middle while you turn the other one on tight. 

Here is a second angle on it, showing how I put cardboard washers (Ok, it is not ALL metal) on the mounting screws so they are held in place while getting the heat-chamber in place. Since the steel mounting bracket is thin metal and iron/steel does not conduct heat as well as other metals anyway, it turned out the cardboard washers were not needed for heat barriers, but it helps to have something holding the screws in place while the head is mounted to the extruder.

[ Clue: In order to get at the screws, turn the heatsink 90° until you can screw them in and then turn it back, locking it against the nut. ]


As you see, I added more kapton tape for insulation to keep the resistor leads from shorting out against the aluminum of the block.

In my case I actually left the heatsink turned 90° to this because I had to replace the fan with a larger one which I mounted in back instead of where the little 30mm fans normally do.

I also found that I have to turn the fan on right away, so the heat does not follow up the threaded tube and make the filament melt up too high. With this 2 resistor method there is plenty of heat to melt the plastic, even if the fan is turned on all the time.

The size and shape of the heatsink is not important as long as it is big enough and makes good contact with the threaded rod in the middle (one reason I added the washer, –to have more surface contacting the heatsink.

I hope this gives people ideas of their own.

Since .5mm and smaller is common, some people wondered about the .6mm(+) nozzle I am using, so here are a few pictures to show how well the new heat-chamber works.

In fact in this next picture, I identified some of the important settings.


[ Some may wonder where I got the number 0.254mm for layer height. It happens that comes from Imperial measurement. One hundredth of an inch equates to 0.254 millimeters. One hundred layers that height will be exactly an inch.]

As I say in that picture the fit with a bearing was pretty amazing. Not too loose; it did not fall into the hole. And not too tight; I did not have to hammer it in. A little pressure with my thumb pushed it in and a couple light taps with the flat side of my needle-nosed pliers seated it all the way down.

If the prints do not get any worse than that, I can live with such a small amount of roughness.

I know, —That could be a mount for holding a screw-driver handy. (Ha Ha)


I must say, I am pretty amazed that the pieces are coming out so well considering all the problems I had with the machine along the way.

I guess it just shows that if you keep at something and don't give up along the way, that things will work out well in the end.

Good Luck with your own projects !


Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Amazing in detail as always, Dan! Thank you :)

Thank you for the compliment, in return.

Looks like you have the printer in pretty good shape. I would imagine your nozzle size will limit your layer thickness, but, I doubt it will be all that much of an issue.

Take a look. I added some new pictures at the end of the article to show how well they are turning out now.

I am pleased that after all the "fits" I had with it, that it is working so well now.