I'm in the planing staged, and need tips / help.

My project so far is to take two cylinders in stainless steel, put a plate in each end which is attached to the outer(biggest) cylinder. The smallest cylinder will be fixed to shaft, so it can rotate free inside. Between the outer and inner cylinder the heating element will be placed. It's a 2000W heating element, does anyone have a clue if thats enough?

I'm thankful for every tip I can get!


Temporary sketch

Heating element, 2000W

Inner/outer cylinder compared to a A4-paper

rotation source


For ventilation i'm going to use a vacuum cleaner, this way I can roast the coffee in the comfort of my kitchen while the smoke etc, goes outside where the vacuum cleaner stands.

I'm also going to use a thermometer (see picture) to control the temperature.


Also i'm adding a window to see true, some bearings to the shaft, a hatch of course, probe tube.

It should be good up to about 500 grams.

The roast speed will be strongly affected by having good ventilation/heat transport. The theoretically most efficient system would use counterflow heat exchange:

-- Bring the air into the outer chamber from the front, let it flow over the heaters,
-- let it flow into the drum from the rear, then flow from rear to the front,
-- and then exit from the front/top.

In this scheme the warmest air on the inside of the drum, at the front, is matched to the coolest air entering the outside, also at the front, while the intermediate temperatures meet at the rear. The average of inner and outer air on the drum will always be the same, so drum temperatures will be even. The beans are exposed to air heated twice, once flowing towards the back in the outer chamber, and again flowing towards the front inside the drum; so the transport of heat into the beans is as efficient as it can get.

Use a damper or dimmer to control the airflow, and you'll be able to control the roast speed and maximum drum temperature somewhat independently, which is the goal of these controls.

As Jim already alluded to, ventilation and its control cannot be overstated here. The heating elements that you are considering are probably somewhat slow to react to control inputs (voltage variation). To provide some control over the roaster you will need to have a ventilation device that will give you enough flow and power to move in the cooler air when required to bring down or slow down the temperature rise.

2000 Watts is enough to bring up the temperature in the roaster for sure but if you intend to have a lot of air flow (what you get from a vacuum cleaner) to extract moisture and chaff, you either have to consider more heat power or some kind of recirculation. Otherwise this element will not be enough. It also depends on the insulation that you will put around the outer drum and most importantly the batch size that you are considering. For sure the thermal mass of the roaster will have a considerable effect as well. The more thermal mass the more heat power that is stored in and available to transfer to the beans.

Be aware that your vacuum cleaner will smell like mid-roast coffee every time you use it now!

I am assuming that you will install the elements outside of the drum, like most electric roasters. This means that roasting will be done mostly by conduction and a little by convection. The most air flow, the most convection roasting is happening. And in my book it is a good thing. More efficient, less tipping and surface charring.

This brings me to the next point. Incorporate stirring vanes! Lots of them and big ones! You want to mix those beans!

Finally maximum drum speed, to maximise mixing, should be, in rotation per minute, around (30/Pi)*(g/r)^1/2 (equation editted to correct a mistake). Where g = gravitational acceleration (about 9.81 meter/sec), r = the radius of the drum in meter. This was taken from a paper where it was stated that optimal should be between 0.9 and 0.75 of that speed.

Take all of this with a grain of salt. I have not been roasting in a drum for very long. So my experience is limited but I did try to build a drum roaster from scratch and miserably failed. I abandonned (or set aside) the project when I realised how many things I did not know and how expensive and incredibly long the learning by trial and experience would be.

I am not telling you off and I wish you all the best but be aware of what you are getting into.

Thank you for all this valuable info!

If i could roast 500g in one batch, that would be more than enough for me.

I will definitely take this in to account,

another_jim wrote:-- Bring the air into the outer chamber from the front, let it flow over the heaters,
-- let it flow into the drum from the rear, then flow from rear to the front,
-- and then exit from the front/top.

Do still need the hot air to re-circulate? Or can i take the hot air out, as you described, end replace it with air holding room temperature?

If it gives the best result to re-circulate air, I would need to redesign some parts.

another_jim wrote:Use a damper or dimmer to control the airflow, and you'll be able to control the roast speed and maximum drum temperature somewhat independently, which is the goal of these controls.

Do you think it's enough to make a hole in the tube going to the vacuum cleaner, and then adjusting the size of the hole, (Its nature to always pick the easiest route)

Whale wrote:Be aware that your vacuum cleaner will smell like mid-roast coffee every time you use it now!

I got one in spare

Whale wrote:This brings me to the next point. Incorporate stirring vanes! Lots of them and big ones! You want to mix those beans!

I was wondering if I had to make some sort of stirring vanes or not!

Whale wrote:I am not telling you off and I wish you all the best but be aware of what you are getting into.

Tnx!!

sversimo wrote:If it gives the best result to re-circulate air, I would need to redesign some parts.

Actually it does not.If you do not intend to do more than 500 grams batch the recirculation may not be necessary. Actually recirculation introduces problems of its own and is not desirable if you want to keep it simple. It main advantage is that it reduces the energy requirement. Most if not all small roasters do not use re-circulation because of the problems that it brings.

sversimo wrote:Do you think it's enough to make a hole in the tube going to the vacuum cleaner, and then adjusting the size of the hole, (Its nature to always pick the easiest route)

If you provide a by-pass to the air intake to the vacuum cleaner and have a means to control its effect, it would be sufficient I think.

sversimo wrote:I got one in spare

You have to also plan for chaff dust build-up There is a lot of dust that will collect on your vacuum filter that will change it efficiency. This is just a heads up because it affects all roasters that have fume dust collection, not just you design.

Looking at you latest drawing:Make the vanes bigger!

Seriously the parallel vanes that you show will not empty the drum when you open the exit door. So to properly empty the roaster and terminate the roast evenly for all beans, you will have to tilt the roaster forward. Or you have to incorporate angled vanes. This is more complicated to manufacture of course.

You are showing the trier and the view hole on opposite side of the rotation axis. Be aware that when the drum rotate the beans will be largely on one side of the roaster along with the rotation of the drum. Put the view window as low as possible, in the same direction as the drum rotation, such that you get good movement from the vanes of the beans in front of the window. A lot of roasters have the window in the exit door. This limits the size of the window which in turns limits the usefulness of it. On mine the window is large, with a good light shinning there, it gives a very good feedback of the beans color. But you have to make sure that the beans move.

Put the trier on the same side as the rotation as well. In my opinion, it is useful to be able to catch the beans as they fall off the drum side. These beans, for a conductive roaster, are the hottest most roasted. As well, this way the trier does not impede the movement of the beans to much.

Finally, make sure that you will have enough room to clean under the elements for any chaff and dust that will fall under the drum.

Silvain is right; recirculation is not needed, and hard to engineer (the fan would need to stand up to around 400F to 500F)

I'm concerned about using a vacuum cleaner fan -- for the air volume you need on this size roaster, a computer cooling fan is about right (I don't know the actual CFM numbers for roasting; does anyone?) In any case, you need dimmer controls on both the heat and air. This lets you control the profile and makes the size choices less critical.

IT looks not bad, but there is a big mistake.
The rotating drum must be shorter than the outer cylinder to let the airflow go through the drum.

this is my roaster

Very impressive design! I wish I had all these metal working tools!

Actually, the Quest M3, for example does not have a shorter drum. the air flow seems to be going through the small gap that exist. M3 owners please correct me if I am wrong.

Of course, this creates a restriction but if you are still planning on a vacuum cleaner for the air flow I would not think that this restriction should be too bad. Of course just a few little holes in the tail plate would increase the air flow a lot.

Please post more pictures. And let us know how it is coming along!

The M3 has a small gap at the front of the drum (so small that a bean cannot fit through), and this is enough to exhaust the air flowing along the outside the drum. It has a larger gap at the rear to allow air to enter and flow inside the drum. The vanes stir the beans towards the front, so beans do not fall out of the rear gap.

I do not know what proportion of the air flows inside versus outside the drum.

here is an other one