Not a member yet?
Click here to register.

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

08/20/2019 3:47 PM
Roasted the last of the SM Kenya Nyeri Katogoto this morning.

08/20/2019 3:47 PM
So, Allen gets notification. 'renatoa' gets notification. ... I do not. Funny stuff. But not earth shattering.

08/20/2019 2:01 PM
PM sent

08/20/2019 10:32 AM
snwcmpr, send me a pm so can see if I'm getting email notifications now. Thx.

08/18/2019 4:31 AM
Awesome I sent an email to AllenB and JackH The email to JackH failed. I had screenshots of my settings.

Users Online
Guests Online: 1

Members Online: 0

Total Members: 6,104
Newest Member: Rozak Kopi Sajang
In Memory Of Ginny

Latest Donations
allenb - 25.00
snwcmpr - 10.00
Groveland Hill Roas... - 25.00
renatoa - 2.00
JitterzZ - 50.00

View Thread

Who is here? 2 guest(s)
 Print Thread
TC4+ Arduino coffee roaster shield (TC4-compatible)
chmod755 wrote:


I hope someone sees this and can give me some guidance.

I just received my TC4+ board and I'm not sure how to make the connections.

I'm building an air blown, gas powered roaster. The air blower powered by an AC universal motor and is driven by a PWM AC dimmer circuit: https://www.tindi...50hz-60hz/

The dimmer circuit accepts 3.3V or 5V PWM signals up to 10kHz.

The propane gas is controlled by a Clippard EVP proportional valve and driven by Clippard's EVPD driver board:
The driver board accepts signals as follows: 0 to 5 VDC, 0 to 10 VDC, 0 to 20 mA, 4 to 20 mA, PWM @ ?2 kHz duty cycle

The TC4+ is able to control the dimmer circuit and the EVPD board on IO3 with IO3,x/n serial commands but I don't seem to be able to use IO2, OT1, or OT2.

I'm using the aArtisanQ_PID sketch and i've configured it to use CONFIG_PWM.

I've also tried installing a version of aArtisan_PID that enables high speed PWM on OT1 but it doesn't seem to work.

Any ideas on how I can use the TC4+ to control both the blower and the gas valve?

Hi chmod755! Good question. IO2 is meant for external interrupts, but you can do PWM on a number of other pins, for instance IO6. You should be able to use the command "AWRITE,D06,xxx" - could you give that a try and report back? It should work fine, it's just that aArtisan / aArtisanQ_PID don't have an extra command specifically for it. To connect to IO6, you can use 5V and GND from IO2 or from the power pins on the Arduino pin header, and IO6/D6 as the PWM pin (that's on the Arduino header near where the IO2/3 headers are).

What's that version with high-speed PWM on OT1? I hadn't come across that before, but happy to have a look at what might be going on there.
Edited by mg512 on 08/08/2019 10:23 PM
Using AWRITE you should also pay attention to the frequency of generated PWM, if it is appropriate for the equipment being controlled.


PWM speed on OT1 can be changed using TIME_BASE in user.h, and indeed it seems there is someone who reserved a placeholder for a future intended high speed versions:

// The faster frequencies below are for advanced users only, and will require changes to the PWM16 Library
//#define TIME_BASE 15 // approx. 977 Hz
//#define TIME_BASE 7 // approx. 1.95kHz
//#define TIME_BASE 6 // approx. 2.2kHz
//#define TIME_BASE 3 // approx. 3.9kHz

No idea if such PWM16 custom library was ever been released...
Edited by renatoa on 08/08/2019 10:47 PM
renatoa wrote:

Using AWRITE you should also pay attention to the frequency of generated PWM, if it is appropriate for the equipment being controlled.

Oh, that's a good point, thank you for catching that. chmod755, you should probably be fine if you put the valve on IO3, and the PWM dimmer board on one of the standard PWM pins (e.g. IO6) with the AWRITE command.
anyone got Artisan's 'follow background profile' to work with this? considering replacing the current fuji pxg4 with this.
Depends a lot about how well is tuned the PID system.
And also the machine build and working principle, especially the fact that the charging is made following a preheat phase, thus the machine having a turning point.
If the above is true, then profile following is very tricky to tune, you should have set a minimum heater level, roughly in the same ballpark as used for preheat.
When approaching turning point this level should be slightly bumped, else the PID logic will led to a massive undershot, in order to build an I term big enough to drive the roast, then will follow oscillation very hard to bring back the process on rails of the profile to follow.
So you need a bit of practice to manually handle the roast to ensure a smooth landing on the desired profile path in the TP neighbourhood, then turn on the PID from that point.
No wonder the best results testimonials are from FB users, that start the roast without preheat, from ambient temperature, so no TP and all the issues associated.
tyrtok wrote:

anyone got Artisan's 'follow background profile' to work with this? considering replacing the current fuji pxg4 with this.

Yep, follow-background-profile aka PID control works with the TC4+. I have pretty much always been using it exclusively that way, I don't think I've ever done a single roast manually. As renatoa points out, some people have been struggling getting the PID tuned properly with both the TC4 and TC4+ (same software, same PID algorithm), but for others including me it worked almost immediately. Once it's set up, it works pretty solidly, though I don't have any first-hand experience with the PXG4 so can't comment on how it compares to that.
Curious to see what means works immediately, and how looks the RoR for such roast.
Not trying to make a nitpick exercise, but the only way to make the curves looking nice is to apply huge amounts of filtering in Artisan and/or TC4.
These filtering will introduce delays in actual value reading, For example, the following TC4 filtering values 50, 66, 75, 80... means delays of 1, 2, 3, 4 seconds of the display value versus the process value.
The actual level of measurement methods used is simply unable to provide satisfactory data for the precision of observation claimed by some people when discussing about flic-flaks and other RoR somersault.
I can demonstrate mathematically that simply variations within the limits of a 0.1 C degrees temperature band, will produce RoR variations bigger than 10%, i.e. when we read a 10 C / minute RoR value actually this could be anything in the 8-12 band.
This is not a TC4 or anything other measurement method defect, but simple math, er... statistics, mean average and such...

Next factor that wasn't been discussed anywhere so far is precision of control.
1% step in power for a hot air machine is 6-7 C degrees variation for air temperature.
How much is the impact of this value on beans depends a lot on machine build and thermal inertia.
But surely this is not a smooth control... especially on IR machines, where radiation variation is instant, as is the light.
Edited by renatoa on 08/18/2019 8:01 PM
Jump to Forum:

Similar Threads

Thread Forum Replies Last Post
Hottop (Light) Roast Profile, Baked Coffee HotTop Roaster 1 08/20/2019 5:54 AM
CoffeeAir II Roaster Build Fluidbed Roaster 281 08/13/2019 6:29 AM
KKTO - My Humble Turbo Roaster Turbo Oven Roasters 172 08/09/2019 12:00 AM
Another 1.5kg or 3lb drum roaster from scratch Building a Coffee Roaster 99 08/07/2019 5:48 AM
My drum/bbq roaster Drum Roasters 4 08/06/2019 10:34 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 © 2019 PHP-Fusion Inc
Released as free software without warranties under GNU Affero GPL v3
Designed with by NetriX