Not a member yet?
Click here to register.

Forgotten your password?
Request a new one here.
You must login to post a message.

10/15/2021 2:19 AM
merlot85, maycondelpiero and hoeltz, Welcome !

10/14/2021 10:06 AM
Thanks for the addition to the group. Seriously considering building a drum roaster along the lines of oldgrumpus's. Love the design and craftsmanship.

10/14/2021 4:00 AM
Morning, ar3mia ! and... coffee drink

10/12/2021 11:55 AM
Evan Slack and CupOfJoe, Welcome !

10/07/2021 3:24 AM
Larry140, Roastiva and fkrauthan, Welcome !

In Memory Of Ginny

Latest Donations
JackH - 25.00
snwcmpr - 10.00
Anonymous - 2.00
Anonymous - 5.00
Anonymous - 5.00
Users Online
Guests Online: 16

Members Online: 0

Total Members: 7,360
Newest Member: maycondelpiero

View Thread

Who is here? 1 guest(s)
I'd like to develop a program for this. I'd like to develop a program for this. 21%[4 Votes]
I'd want to use someone's program for this. I'd want to use someone's program for this. 21%[4 Votes]
I'd buy a programmed Arduino! I'd buy a programmed Arduino! 58%[11 Votes]
Total Votes : 19
 Print Thread
Arduino $35 USB programmable roaster controller
The Arduino is a cheap, simple to program, versatile device that is intended to control devices through input/output ports.

It is more powerful than a PID or PLC and doesn't require a computer to operate, but needs one to program it, hence the USB port.

The hardware is sold at cost, about $35, and the software is Open Source, so it is free.

Anyone with a knowledge of coding syntax could program one to control a roaster. The advantage here is that you can then share the program with others.

Developers have already configured one as a thermostat using a K type thermocouple, and I've heard that PID algorithms have been written for it, too. But why stop there? Why not program it to profile a roast?

You could conceivable program it with multiple ramps creating a roast profile. You could save multiple profiles and then select the profile you want to use.

Here is a link to the website. They have a forum for support and sharing ideas: Arduino
Dan attached the following image:

Edited by Dan on 01/13/2009 10:45 AM
Koffee Kosmo
Now this is interesting and a great next step for consistent and repeatable roasted bean profiles

Its got me interested
Great find
I home roast and I like it. Designer of the KKTO
Roaster Build information
Blog -

Bezzera Strega, Mazzer Robur Grinder, Pullman Tamper Convex,
(KKTO) Turbo Oven Home Roaster.
Do you use this, Dan? How many outputs does it have? Can it directly output to control a couple of SSRs?

Roaster: CoffeeAir II 2# DIY air roaster
Grinder: Vintage Grindmaster 500
Brewers: Vintage Cory DCU DCL, Aeropress, Press, Osaka Titanium pourover
Island Addict
This is what I use in my Poppery roaster, except that the microcontroller communicates with my laptop computer via serial port. User controls and data logging/real time graphing are done in an application running on the laptop.

Bill Welch has designed a system based on a Pic microcontroller.

The Ardunio has over a dozen input/output ports, not counting ones usually reserved for special functions. Each port can directly drive a SSR.

From what I can tell, it functions much like a PLC, so yes, it can drive SSR's etc. It can drive servo motors, too, which opens up some potential. For instance, I could use one output to control a servo motor to tilt my drum, dumping the beans. From the fellow I talked to, this unit is more programmable than a PLC, and easier to program than a PIC. They make different types and some, like the Wiring, have more I/Os.
Its a really neat piece of hardware, and I think it has a great deal of potential for home roasting. While I like to do most things myself just for the learning process, I would be likely to buy a pre-programmed set-up if available simply because I don't have the time to do everything!
Island Addict
I think the best reasons for building your own PID from scratch would be to get highly customized functionality not available off the shelf and/or if you are a total geek and love doing projects involving electronics and computer programming. Those were my reasons, at least.

Saving money is probably the least compelling reason. To get PID functionality you would have to add at least one thermocouple input that can be read by the microcontroller. Off the shelf solutions can be quite expensive. A homemade solution can be had for around $20, but that involves spending quality time with a soldering iron and a circuit board. You would also have to add user interface components: a LCD screen and probably a handful of switches and potentiometers or a small keypad. Let's say $50. Add in a project box, connectors, wiring, misc. electronic parts. You're quickly getting into the range of what a surplus PID unit costs. And that's before you spend weeks of time building the hardware and writing/testing the software.

Rob, It depends on what you want your controller to do. If you only want it to control temperatures for one profile, then a Fuji ramp/soak PID is the way to go. If you want it to do extra stuff like keep multiple profiles in memory, control the motor, etc. Then this unit is the solution.

I was talking with my Arduino buddy today and he said there is a touch screen accessory. It could be used to program this thing and certainly to perform profile selection and start/stop functions. I asked him if he thought I could use it to draw a profile curve with my finger (time versus temperature) and then have the controller follow that curve and he thought it would be possible! Try THAT with a PID!
Island Addict
Totally agreed. These things are just tiny computers, so with the right hardware you can program them to do anything you want. The best thing is that you can add new features just by writing code. I'm currently writing code for mine that will allow me to replicate automatically the exact temperature profile for any batch I roast manually. has a good selection of peripherals such as touchscreens and accelerometer. How about a Wii-like controller for your roaster! Grin They also have a great Beginning Embedded Electronics Tutorial (http://www.sparkf...ials_id=57) based on the same microcontroller as in the Arduino Diecimila.

Anyway, I would be happy to share any of the experience I've acquired, such as how to build the thermocouple interface. I would also be happy to share code I've written and/or help develop new code.
Kaffee Bitte
That certainly sounds like a great way of controlling a roast, but would it require only using an all electric setup?

"Some days it's spice, other days it's bitter dirt."
I don't see why it couldn't control a gas heater. Certainly it could control an on-off burner, but perhaps a proportional one, too.

Sparkfun has the touch screen I was talking about 128x128 pixels, in color, can hold 60 full screen images in memory.

This forum thread talks about an Arduino PID program! http://www.arduin...1226431507
Edited by Dan on 01/08/2009 9:18 PM
If anyone wants to work on this project I'll do what I can. I can't program, but I do know process control, and I am very good at graphical interfaces and can design the screenshots for the touch screen.
I need help understanding the difference between this --> Arduino <--- and this --> PID Temperature Controller <--.

I understand the graphical interface difference, as the Arduino must be programmed or have a graphic interface added and the PID has the keys and display already there. Anything else? Is there the equivalent of an Arduino inside this PID?

Why would someone buy one over the other?

Roaster: CoffeeAir II 2# DIY air roaster
Grinder: Vintage Grindmaster 500
Brewers: Vintage Cory DCU DCL, Aeropress, Press, Osaka Titanium pourover
On this kind of stuff, I am an idiot. But I am watching and listening with RAPT ATTENTION. I am REALLY interested to see how this thread and these ideas develop!

europiccola | yama + coryrod | chemex | AP | clever
wbp1 | wepp1 | bm/hg | co hybrid (still coming soon...)

Let me try to describe what I know.

On the PID side, you can read about my PID here

I like my PID, but it has one shortcoming, you can only control the ramp rate in increments of one minute.

This means, if I want to ramp the temp from 100 F to 300 F, I can do it in 2 mins, or 3 mins, or 4 mins, but I can?t do it in 2 mins and 30 seconds. This means I have to make compromises in my profiles and can?t try everything I want to do.
I would like to separate the ramps into ramp to drying, ramp to first crack, and ramp to end, and maybe a few sub categories in there, and be able to control the time of those ramps to the second, or at least in 10-15 second increments, instead of the minute increments I am forced to use now.

My PID, and most others can cycle the heater power in one second increments. Some of the better Fuji?s and Watlows can cycle even faster, I think. So, if I am running at 50% power, the PID turns on the heater for one second, then turns it off for one second.

My PID has a RS-485 interface, and came with free software for it. I bought a ebay RS-485 to RS-232 controller from ebay for around $5, and just plug it into my laptop.
I can save profiles to disk, recall them whenever I want, and send them to the PID. I keep an excel spreadsheet with the PID profiles, with a different worksheet for every bean I have. Whenever I can, I roast a profile I think I like, then do another roast with it slightly changed, with something like faster drying time, or faster time to first crack, and then cup the results a couple days later, and see which is better.

I have been thinking about getting a Fuji PID, the PXG series, that can control ramp time to the second, but it is expensive, over $300 with the RS-485 interface.

If I did get the Fuji PID, I could use the program by Arpi, you can read about it here

Looks like nice software. With my PID, the PID logs the temp data to a text file. I use the PID to control ET. I have a meter with a USB output from coffeesnobs, which monitors BT. If I want a graph, I have to bring both sets of data into excel. With Arpi?s software, if I had another Fuji PID, I wouldn?t have to do that, but it?s even more expensive.

One last think about my PID setup. I was using a variac for heat control, and no fan control before, and I was having trouble getting consistent roasts. That?s why I decided to get a PID, and it is obviously more consistent with a PID. Since I manually control air flow/fan speed, there is still a human element, so there is some inconsistency.

Now, onto the Arduino. First, you can?t just buy a $35 board and make a PID. I would want a display, I think they were around $10.

Then, you need to read in a thermocouple. Easiest way is to get a couple of the MAX6675, but they cost about $14 a pop from Digikey, and they?re surface mount. There are other ways to read in a thermocouple, but all are more work.

The MAX6675 is a surface mount chip, and the easiest way to interface it to the Arduino is to get a ?shield?. But you need a shield that can mount a surface mount chip. Also, if the shield can mount a 16 pin SM chip, then you can put two 6675?s on the same shield.

I was thinking about making a dual temp sensor/data logger with an Arduino at some point, which is why I know all of this.

There are guys are the net who have figured out how to read the 6675 with an Arduino, and you can look at their code, but I think it?s pretty straightforward.

Now, you can turn an Arduino into a PID. I never looked at driving a SSR with the Ardiuno, but looks like some of you have, and I?ll take your word that it will work, don?t see any reason why it wouldn?t.

That?s all I have time for now, but that should give you something to think about.

I didn?t go the Arduino route because it was going to take too much time to get something working, and I lucked into a good deal on my PID on ebay. It would be cool to make an Arduino that reads both ET and BT, and can control the heater and the fan power.

I probably have more to say on the subject, but it's late, so I will try to add some thoughts tomorrow.

Edited by seedlings on 04/15/2010 8:32 AM
Find yourself a old used Honeywell 3000. Higher numbers are better, but I would not go under 3000. Believe me they will do everything, multiple ramp segments at any speed, you can program by rate of rise, You can set off an alarm when a heat is reached. A new one cost more than roaster so hunt for a used one. The way you would profile would use multiple ramp segments. Honeywell calls there controllers "UDC" not "PID" , Its Universal Digital Controller. But its just a fancy PID. Is is the cadilac of process control in my opinion. I use Honeywell at work to control drying process in lumber kilns, some are 20 years old and still working. I have updated some to 5000 but they are not affordable for coffee roasters. Just a warning the program imputs can be scary, becouse it does so much there is lots of options. If you are slow and just work on one paramiter at the time you can work through it. I wish I knew how to otput to a computer with them, I know they will do it, can not get it to work.
Edited by Ringo on 04/15/2010 7:24 AM
Randy you've answered many of my questions, and raised others, thank you! I'll keep reading and shopping...

Ringo, what reservations would you have buying a used Honeywell 4000 knowing only that it "powers up"?


Roaster: CoffeeAir II 2# DIY air roaster
Grinder: Vintage Grindmaster 500
Brewers: Vintage Cory DCU DCL, Aeropress, Press, Osaka Titanium pourover


Ringo wrote:
I wish I knew how to otput to a computer with them, I know they will do it, can not get it to work.

That would be a deal breaker to me.
Having used a PID with a PC interface, I could not do without it.
It would be a pain to have to manually input the profiles.
We use Watlows at work, and the better Watlows sound like the Honeywell, lots of features. But, they charge for the PC interface program, and the ones I have seen on ebay usually don't have a 485/232 interface anyway.

I would, I have some that power up but the control relays are out. So there is a risk. But you do not want to buy one of these new. All boards inside of these are replaceable, little pricey. I have two used ones that I am going to put on my roaster. I know it would display tempature. The control I plan for my roaster a first is going to be. I will program in a ramp, The target temp will be displayed "set Point" and I will try to make my actual temp match that. So all through the roast I will have a moving target. Honywell have an Auxilary output that will output the reading from the TC. So if you can Data Log it will send a signal to you recording device. I will use a chart recorder, unless someone knows how to imput to computer. I will say it is a pain to imput profiles, takes some time. You can only save 3.
Edited by Ringo on 04/15/2010 9:24 AM
So the downside to this Auber PID is no PC connectivity:


Roaster: CoffeeAir II 2# DIY air roaster
Grinder: Vintage Grindmaster 500
Brewers: Vintage Cory DCU DCL, Aeropress, Press, Osaka Titanium pourover
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
"Well tempered roaster" more feedback Roasting Coffee 21 10/14/2021 6:26 AM
New Fluid bed roaster- Help with how to control it Fluidbed Roaster 5 10/12/2021 1:34 PM
Using a TC4 with Artisan Roaster Scope Dataloggers/Controllers/Rate of Rise Meters 28 10/10/2021 10:41 PM
Roaster build forced air and agitator Building a Coffee Roaster 9 10/07/2021 2:30 PM
Testing Arduino UNO with bitwisetech / popc and MAX6675 Popcorn Popper roasting 67 10/07/2021 4:32 AM
Homeroasters Association Logo, and all Content, Images, and Icons © 2005-2016 Homeroasters Association - Logos are the property of their respective owners.
Powered by PHP-Fusion Copyright © 2021 PHP-Fusion Inc
Released as free software without warranties under GNU Affero GPL v3
Designed with by NetriX