Who is here? 1 guest(s)
 Print Thread
Hello! (soon) Roasting From Oregon USA

I am joining because, this seems like a nice community, where I can ask lots of questions building my DIY Drum roaster. Currently, as of posting I don't have any roaster but am gathering materials.

Materials I plan to use:

A Butter Roller for the main drum off Ali Express

L Brackets and Flat Brackets for the inside of the drum

Random Motor that's surplus from an ice cream maker

6-90V Motor speed Conrol off Ali Express

Custom 3*100mm Ungrounded Thermoouples off Ali Express

ZGT-40DA SSR from the same company as the Thermocouples

R8-100 PID Again from the same company

I know I need to get Phidgets but I want to try and find a cheaper alternative before I buy them. (I will probably make a separate Thread about this topic elsewhere on this forum)

I plan on using a surplus camp flame at first but when I can I plan on getting heaters so I can get the most out of the PID

Power Supply off Ali Express (I'm still figuring out Voltage, Amps, Watts etc.)

Let me know what you thing!

Have a great day!
Welcome !

... and some comments about your choices:
"Butter Roller for the main drum" - beware, holes are too big for coffee beans !

Chinese PIDs aren't appropriate for coffee roasting control, imo.
Check for a digital power regulator instead, as this: https://www.amazo...ref=sr_1_9
... or as picture attached, if link expires.

renatoa attached the following image:

Edited by renatoa on 06/12/2024 3:57 AM
Thanks for the reply!

Yes! I am aware! I forgot to add the fact I'm basically tying metal mesh to the outside using large hose clamps that are surplus.

As for your PID comment, I think I will stick with the PID mainly since the thermocouple I am buying is built for the PID and I know for a fact that it supports it. I am simply playing it safe.

Do you have any recommendations on what I should use in place of the Phidgets? ATM it's wayy out of my tight budget range (100-130 is my goal but have some room to budge to go higher if needed) right now I'm sitting at roughly $104 for everything I need, the Phidgets would go over my limit by roughly $40. The two Phidgets alone is around $71 and I need something that can connect to my Laptop and Artisan for about $30...

As an alternative to build from scratch, you could evaluate this roaster: https://homeroast...post_78568
The barebone version looks worth the money, imo, as a good starting point.
There are versions with perforated tin drum instead glass, just do a bit of forensic.
LE: this was been posted before reading your budget Grin

There is no such thing as "thermocouple built for a PID", both are generic and should match.

Also, there is nothing safe to play with a chinese PID for this task, conversely, it's a recipe for disappointments.
Roasting coffee has different dynamics than boiling water ;) And a FUJI PID is different beast than a chinese counterfeit ;)

The cheapest known (by me) way to drive a temperature into Artisan, without building something around Arduino, is the Mastech MS6514 device. $64 the device + two TC, best price of the moment on Ali.
Edited by renatoa on 06/12/2024 4:35 AM
I enjoy a challenge and would rather build it from the ground up. it lets me be creative and acquire more skills for other projects and It's always fun just to tinker.

What's different about Chinese PID and the PID's used in the builds I've been looking at?

In case you're curious this is the PID: https://www.aliex...13423.html

As for "both are generic and should match" thats just what the Manufacturer told me. Buying a PT100 thermocouple get a PID compatible with PT100. But I suppose I could buy something else like the Inkbird Kyle Greenhaw used in his build on youtube. Or something else that you think would be better suited. I will also look into the power regulators you mentioned.

Seems like it would be easier for me to just buy the Phidget components... I've looked at the TC4ESP, but it seems like its better for Fluid Bed Roasters? Is there anything similar I could build for drum roasting?

I have seen this: https://www.dfrob...-1753.html while doing some research on my own saves about $15, but still expensive... Not even sure if it would work with Artisan? I don't really know what I'm talking about when it comes to electronics like this...


...and the PID's used in the builds I've been looking at?

Which one ?
I don't remember any roaster controlled by a PID in the last 8 years I am member here...
Tried one in my very early days and abandoned instantly, the power regulation is jumpy and somewhat random.
Even a bimetal thermostat looked for me more reliable those days.


...thats just what the Manufacturer told me

On Aliexpress are sellers not manufacturers. Most cases they know almost nothing about what they sell.


Buying a PT100 thermocouple get a PID compatible with PT100.

They might, without knowing, have a point here... Grin
All these boxes are clones of an original japanese model, and the original knows many types of sensors, not only K TC or Pt100... a lot more... selectable from a menu. So you don't buy a different box for each probe type ;)
But the chinese guys cloned each a different set of features, what they been told by the customers of that moment.
I know at least five different versions of such boxes, each of them lacking a function or other, like for example AT mode (AutoTune)
You simply don't know what you buy, good luck with these.

Inkbird is indeed a brand... having available a detailed manual, among others...
Related to what I wrote above... if you download Inkbird PID manual, you will see the connections diagram, and support for both TC and RTD sensors, all in same box. So, a more complete and decent implementation... but still using on-off relay style power control, which is not what I want in a coffee roaster.


I've looked at the TC4ESP, but it seems like its better for Fluid Bed Roasters? Is there anything similar I could build for drum roasting?

Very strange idea... there is no hint in my posts to suggest such thing.
TC4(ESP) is good for any roaster, IF setup properly.
Even the drum roaster term is debatable... what I seen in the first post is rather an agitator, without any thermal characteristics to qualify it as a real drum roaster.

DFrobot board is MAX31855K based TC interface, that can be bought for a lot less from other sources. Under $10 typical.
It needs an Arduino to be used with a PC application, though.

Edited by renatoa on 06/12/2024 8:00 AM
Hello Mr. Himes
Thank you for your continued support.

I too enjoy roasting using a drum roaster.

I am wondering what temperature is used as the input PV for PID control of roasting.
I used to control the temperature of the beans directly by measuring the temperature of the beans, but I realized that it is difficult to control the gradation of temperature distribution toward the center of the beans, which I really need to control, because I can only measure the temperature of the surface of the beans.

Now we control the combustion intake air temperature on a time axis (ramp soak) and roast the beans while making fine adjustments based on BT (bean temperature), ET (exhaust air temperature), and CO concentration to predict the temperature at the center of the beans.

Of course, the gas flow is always PID-controlled to control the combustion temperature, and the exhaust fan is PID-controlled by an inverter to control the constant pressure inside the drum.

I prefer to use Fuji's temperature regulator.
I like this regulator because it has a "fuzzy mode" and when I use this mode, it gives me exquisite control when I change the SV value by time graph values, etc., and does not overshoot or hunch.

I think this mode is not available in general-purpose machines made in China.
I'll be honest Yasu, I have no idea, I don't even have a roaster yet, and have decided not to go PID route, as I was unaware of the flaws with using one. So I apologize for any incontinence but I have no idea how to help. Sorry.
I look forward to having a fun roasting machine.
Please introduce me again when you do.

So, I've done more research and I think I'm understanding things a bit better now, I have bought mostly everything I need, and that should be showing up soon.

I ended up deciding not going using a surplus camp fire like i said in my original post. I ended up buying a cheaper ceramic infrared heater of alibaba. (I've never purchased off alibaba so I don't know what I'm really talking about which you will later be able to tell{*}) I may change to flame heaters in the future however, I just want an easier way to control temperature for now.

I will also be building a frame around everything and insulating it, for a more complete look

As for PID. I misunderstood in thinking I could use a PID directly to artisan. so I settled with TCPESP and bought all the components, however I do want to test MAX31855K in the future just for fun. However, should I still get a PID to control the temperature of the heater or would a digital power regulator be adequate? If so how come I'm really only seeing 220V regulators? Or am I not understanding something?

As For power supply, I have a few questions. 1. do I need to power TCPESP off a power supply or will my laptop be enough to power everything? {*}2. I'm not sure what power supply I need to get because, I seem to only need to power the heater. The heater I purchased is 120V 600W and I'm really not sure if that's overkill or not, and I don't know what I need to power it, so any pointers would be greatly appreciated in that regard!

Anyway, any criticism is welcome, and encouraged. As I'm still a beginner, and don't really know what I'm doing.

Have a great day!
The solid state regulators are not voltage aware, labeling them as 110/220 is simply a mistake, or rather a seller selling everything without knowing anything.
Even a power rating is useless without clearly stating for which mains.
Any regulator is capable to regulate double power on 230V than on 115V mains, because the amps are limiting factor.
What is really the most important for them is how accurate is related the heatsink size to the Amps they control.
A regulator as I recommended above, fitted with a heatsink sized about a matchbox dimensions , i.e. 6x4x2 cm is good up to 10Amps at 25C env. temperature, without forced ventilation.
This means 1150/2300W for 115/230V mains.
A fan helps a lot, could even double the capabilities.

If you go on TC4 way, then you need a SSR, not a (manual) power regulator.
The TC4 solution is capable of driving the heater using a SSR, via commands from a computer.
The same guidelines about heatsinks apply to SSR as above.

600W is close to insufficient power to roast even 250 grams, by most hot air methods, if not using heat re-circulation.
The most efficient machine I know capable of this performance is Skywalker, where the radiant bulb is inside the rotating cage, thus the closest possible to the beans, and least power losses in the ambient. It is capable of 400 grams using 700 Watts.
The next could be a TO lid based machine, using the smallest possible pot, perhaps 4 litres, well insulated.
I need 800 Watts in mine, in a 8 litres oven, to roast 250 grams.
Ahh I see, that makes more sense.

I'm assuming Since TC4ESP is essentially a clone of TC4-Sheild a lot of stuff can be interchangeable for when I'm troubleshooting? I considered maybe not doing it since I don't have too much electronic knowledge. However, I figured it could be a fun learning curve, and fun to troubleshoot. As there's only one way to find out so I bit the bullet, to find out.

What would be an adequate wattage for the heater? My drum size is rather small so I assumed 600 was enough (Diam. 3.5in x 6.9 length). the bag of beans i have is like 360 grams (I think) so I'll be doing relatively small batches as I learn. The heater is 120mm x 60mm curved, I know It might not get to the edges all the way, but I'm hoping insulating it will help with that?
Edited by Himes on 07/02/2024 1:33 AM
If you mean you have a heater like this: https://ro.rsdeli...ac/1988529
... well... I have no idea... never seen such approach for coffee roasting...

If what they state is true, i.e. 700C degrees, probably surface temperature, then it is very low compared with 2000C of a radiant element used in Behmor / Skywalker.
Even a heater fitted in metal element/tube is hotter, about 900C when glowing orange.

Before investing any penny on this path, I would try first to see if this element is capable to "bake" even a single layer of beans of his area. Probably 100 grams more or less...
I mean place a single layer of greens in the basket, make a temporary mount of the heater/basket assembly by placing the heater on insulate support (1-2 bricks) , then the basket above, as close as possible without heater touching metal. Remove the basket, preheat the heater to 200C at least, place the basket above and count minutes to first crack.

I know, no movement is not good, but at least we have an idea...
I've seen this done before with these heaters before. The Kyle Greenhaw videos, has been my main inspiration for this project. The 5 videos is the complete series and it's really quite interesting. He uses 2 heaters that are 800 watts. Apparently it can roast up to 450 grams but usually does 250 grams. So I figured smaller size = not as much wattage needed. I already purchased the heaters so we'll see! If all else fails I can use the surplus camp stove I have lying around which I won't really be able to control the temperature for but it is what it is until I figure something else out.
Kyle Greenhaw ... was never quoted in this site until you.... nor on other major coffee roasting sites.
So we have a series of 5 videos showing a build, without any word elsewhere if that build was a success, if he is actively using it, or an evaluation from other followers that what he did there was right.
Not exactly the definition of build to be used as a reference, imo...
He's still active, although not on this forum, as I commented on his most recent post and responded an hour later, he has a test roast in the part 5 and it seems to work pretty well? Of Alibaba where I ordered the heater it was only $21 and I am curious if it really does work so well just have to see. I see what you're saying though still not really a reason to base my whole build. But from what I've seen it's the only high quality video series out there, and it's more helpful for me to understand.
I remember someone not getting enough heat from a single heater but don't remember where I read this?

I do remember this video.
Kyle Greenhaw
Jump to Forum: