Artisan, Arduino and thermocouple woes.

#1: Post by **marcism** » November 16th, 2018, 4:49 am

Hello all,

Seriously long-time lurker and avid forum searcher here. As with a lot of the threads I trawled through, I am another user with thermoprobe issues on my popper set up and with the huge amount of knowledge here, I'd love to pick some brains to see if there's a solution!

A picture is worth a thousand words!

This chart was my second roast of the day and was significantly worse than roast one. Sometimes it spikes earlier, sometimes later. The spike reads as "-2.5c" or "0" for the BT in Artisan before returning to its read temperature.

Set up is as follows:

Popper
Split circuit. Fan controlled by 24v dc laptop power supply via a PWM to vary fan speed. Connects from PWM to fan terminals. Heater controlled with a 2000w SCR. Everything works fine on this side of things!

Temperature - Arduino
Laptop (running on battery or plugged in does not change the results) -> Arduino UNO -> MAX6675 -> k-type thermocouple, currently just hanging inside the roasting chamber and is NOT drilled into the unit.

So unless I'm being daft, the probe is in no way touching the electronics from the popper. There is something I'm missing here, however, which is why I want to see if anyone out there has some input.

Troubleshooting:

I have tried putting a capacitor across the terminals where the probe connects to the MAX6675.

Tried some different filters inside of Artisan but perhaps I need to make more changes.

Took the prob out and tapped it, wiggled it around and it was a clean signal. However, when I hung it back inside the roaster with the fan running, it spiked hard and then settled.

One thing I just realised that I haven't tried yet is using the Serial Plotter in the Arduino IDE and seeing if it spikes there as well, which would identify this as a software issue. I will test this.

For reference, here is my Arduino Sketch:

#include <SPI.h>
#include "max6675.h"

// Default connection is using software SPI, but comment and uncomment one of
// the two examples below to switch between software SPI and hardware SPI:

// Example creating a thermocouple instance with software SPI on any three
// digital IO pins.
#define DO   5
#define CS   3
#define CLK  6
MAX6675 thermocouple(CLK, CS, DO);

// Example creating a thermocouple instance with hardware SPI (Uno/Mega only)
// on a given CS pin.
//#define CS   10
//Adafruit_MAX31855 thermocouple(CS);

void setup() {
  Serial.begin(9600);

  // wait for MAX chip to stabilize
  delay(500);
}

void loop() {
  // basic readout test, just print the current temp
   int incoming = 0;
   double c = thermocouple.readCelsius();
   if (isnan(c)) {
     Serial.println("Something wrong with thermocouple!");
   } else {
     incoming = Serial.read();
     Serial.println(c);
   }

   delay(100);
}

and my .py script for the External Program for Artisan to read the temps:

import serial
arduino = serial.Serial("COM5", 9600)   
datas = arduino.readline()
output = datas.decode('utf8')
print (output)
arduino.close()

So! If the laptop is on battery and the thermoprobe is not touching the roaster metal anywhere, only sitting inside the bean mass, could this still be an electrical problem? Is my thermoprobe toast? Should I return to my purely sense-based stovetop drum roaster??

Any input would be highly valued and perhaps this can help other people with similar issues in the future.

Kinds regards and a shout out to everyone here for helping all the users who hide in the shadows yet read here daily

Marc

#2: Post by **summer** » November 16th, 2018, 7:09 am

When I had spike issues on my former roaster, it dissapeared when I connected the chassis of the roaster with the chassis/ground of my pc.

November 16th, 2018, 10:42 am

I wonder if it isn't helping that I'm in Europe and everything is two pronged....

November 16th, 2018, 10:52 am

I dont know... Im in europe too, and the chassis/chassis connection did the trick for me.

My roaster did not have any 'real' grounding.

November 16th, 2018, 11:08 am

Interesting! Can you share how you did it? My laptop is currently on battery so I wonder where I would connect things...

November 16th, 2018, 12:41 pm

I solved this problem by removing the housing of the thermocouple, wrapping the thermocouple a couple times with teflon tape, and then inserting it back in the housing. This prevents the thermocouple from contacting the housing and creating a ground loop. Not my idea, but worked well for me.

#7: Post by **summer** » November 16th, 2018, 6:13 pm

marcism wrote:Interesting! Can you share how you did it? My laptop is currently on battery so I wonder where I would connect things...

I found out that the chassis on my macbook was accessible through the power plug. So with my macbook on battery, I connected a wire to the the roaster chassis (and actually also to the external fans, it was a Huky500), and connected the wires to the ground of the power plug (plugged into the laptop but not plugged into the wallsocket).

If you are using a PC, maybe you can connect to the chassis like this:

November 17th, 2018, 12:26 pm

Why not just unplug the laptop? Or get a USB isolator?

#9: Post by **ira** » November 17th, 2018, 1:14 pm

Connect the chassis of the roaster to the ground of the thermocouple interface box or the laptop or both. The metal shell of a USB or video cable should be ground so you can just use a one of those to connect to the laptop, no idea what the interface is so can't help with that. Do make sure that the chassis of the roaster is isolated. An AC voltmeter between the two devices should read chose to zero before you connect them, if it doesn't, fix that first.

Ira

November 20th, 2018, 7:39 am

CarefreeBuzzBuzz wrote:Why not just unplug the laptop? Or get a USB isolator?

That's what confused me the most! It IS unplugged!

Thanks for all the replies. I haven't had a minute to myself to sit down and figure this out, but you've all given me some very useful troubleshooting to work towards so hopefully this weekend I can update with some good news and a nice clean profile!

The biggest confusion here is that with the thermocouple IN NO WAY being attached to the roaster, it still produces noise. It is as if I am hanging it into the middle of the roast chamber and the only thing making a connection between probe and popper is the beans themselves.