This is awesome! Thank you so much for the time spent creating this and your willingness to share with us all!!! I watched all 3 videos and my son and I are excited to set this up! You have saved us countless hours so I thank you once more!
Your videos are FANTASTIC! They have really helped me understand what is going on during a roast. I look forward to instrumenting my SR540 as well. I suspect that the reason the FC and SC waveforms appear so similar is that a major portion of the energy is above the ~12.5 kHz Nyquist limit on your recording. They are aliased/folded back to lower frequencies. It would be nice to have a 100+ kHz recording (or even a 1 MHz!)--if money were no object! This might allow us to see the spikes as true impulses rather than a spike with ringing for several ms. That's when we could probably differentiate the FC and SC by the characteristic shape of their impulse waveforms. Also, it would be awesome (and much cheaper) to look at the fan spectrum vs the fan setting. My guess is that a (software) notch or comb filter could take out the bulk of the residual fan noise. I know the point is to roast coffee, and I don't think any of these suggestions would help produce better coffee. 20 years ago I worked in a lab with access to everything needed to do these analyses except the broad spectrum mic/transducer. I'm just curious.
Thanks John for your feedback and good points on additional sound analysis work. I don't have the equipment for higher frequency recording so I don't think I will be able to help there. For my purposes the processed sound waveform gives me the clear indication I need to determine the timing of crack events in a more objective way. Something I may do in the future is to collect some statistics to see how these events depend on bean type and how much variability there is within the same bean type or when using different roast profiles on the same bean type. Thanks again for your comment!
Fantastic work. I’d like to use this in realtime, as I am hard of hearing. Any software or app suggestions? A sound engineer suggested that I get the Acon Digital DeNoise plugin and add it to the Audacity software. It is however $100+ USD, so I’d like to know if it would work. There are also real time spectrum analyzer apps (iOS and Android and PC), but I’ve heard that it is very difficult to use these apps to distinguish FC or even SC. Any help for a this hearing impaired person would be appreciated.
You are welcome! You can see more details of the Artisan configuration in the video ruclips.net/video/t1X0weUSZN0/видео.html, and in the description of that video you will see a link to download the Artisan configuration file I used. Let me know if this helps, or if you would like to see more details.
Thank you for your sharing. I had tried to duplicate what had been shown in this video. However, my normal amplitude had reached to 0.75 instead of 0.30 as yours, and unable to show the cracking sounds. According to your video, I should decrease the microphone level to reduce the amplitude in order to distinguish the cracking sounds. Is this correct/
Could you send me the sound file. I would like try to use FFT Fast Fourier Transform and see whether it is easier to determine the first crack stage automatically.
How did your thermocouple temperatures correspond to the SR540 internal temperature display? Can the SR540 internal temperature display be used with/without correction?
There is correlation but it is not simple, the dynamic response of the SR540 internal thermocouple is much faster than the others because it is a bare thermocouple junction measuring the air inlet temperature. It changes very fast after changing a fan or temperature setting, while the thermocouple inside the probe in the bean chamber changes more slowly and it is more representative of the bean interior temperature. I started trying to use the SR540 internal temperature sensor to adjust the settings, but I quickly grew frustrated and started exploring other options. Hope this helps!
Further to my comment below I am hearing impaired and really could use your help. Is there some way we can connect to chat further, other than using Comments?
Hello and thank you for your interest. I do not have much more to offer on the sound processing front, the Audacity program worked well for me and as shown in the video you can see the spikes for FC and SC without the need to hear them. Since I use the same beans and profile for my roasting, I no longer use the sound processing setup because the FC and SC always happen at approximately the same bean temperature (in my case ~ 400 F for FC and ~445 F SC).
This is awesome! Thank you so much for the time spent creating this and your willingness to share with us all!!! I watched all 3 videos and my son and I are excited to set this up! You have saved us countless hours so I thank you once more!
You are welcome! Glad it was helpful and that it will save you some time.
Thank you very much for these videos. I find them very informative and helpful.
You are welcome, glad you found them useful.
Your videos are FANTASTIC! They have really helped me understand what is going on during a roast. I look forward to instrumenting my SR540 as well. I suspect that the reason the FC and SC waveforms appear so similar is that a major portion of the energy is above the ~12.5 kHz Nyquist limit on your recording. They are aliased/folded back to lower frequencies. It would be nice to have a 100+ kHz recording (or even a 1 MHz!)--if money were no object! This might allow us to see the spikes as true impulses rather than a spike with ringing for several ms. That's when we could probably differentiate the FC and SC by the characteristic shape of their impulse waveforms.
Also, it would be awesome (and much cheaper) to look at the fan spectrum vs the fan setting. My guess is that a (software) notch or comb filter could take out the bulk of the residual fan noise.
I know the point is to roast coffee, and I don't think any of these suggestions would help produce better coffee. 20 years ago I worked in a lab with access to everything needed to do these analyses except the broad spectrum mic/transducer. I'm just curious.
Not my work, but looks like the kind of system that would work nicely.
Thanks John for your feedback and good points on additional sound analysis work. I don't have the equipment for higher frequency recording so I don't think I will be able to help there. For my purposes the processed sound waveform gives me the clear indication I need to determine the timing of crack events in a more objective way. Something I may do in the future is to collect some statistics to see how these events depend on bean type and how much variability there is within the same bean type or when using different roast profiles on the same bean type. Thanks again for your comment!
Fantastic work. I’d like to use this in realtime, as I am hard of hearing. Any software or app suggestions? A sound engineer suggested that I get the Acon Digital DeNoise plugin and add it to the Audacity software. It is however $100+ USD, so I’d like to know if it would work. There are also real time spectrum analyzer apps (iOS and Android and PC), but I’ve heard that it is very difficult to use these apps to distinguish FC or even SC. Any help for a this hearing impaired person would be appreciated.
Thanks for the great video, can you share more details how you set up in Artisan, such as curve, buttons, slides?
You are welcome! You can see more details of the Artisan configuration in the video ruclips.net/video/t1X0weUSZN0/видео.html, and in the description of that video you will see a link to download the Artisan configuration file I used. Let me know if this helps, or if you would like to see more details.
Thank you for your sharing. I had tried to duplicate what had been shown in this video. However, my normal amplitude had reached to 0.75 instead of 0.30 as yours, and unable to show the cracking sounds. According to your video, I should decrease the microphone level to reduce the amplitude in order to distinguish the cracking sounds. Is this correct/
Could you send me the sound file. I would like try to use FFT Fast Fourier Transform and see whether it is easier to determine the first crack stage automatically.
How did your thermocouple temperatures correspond to the SR540 internal temperature display? Can the SR540 internal temperature display be used with/without correction?
There is correlation but it is not simple, the dynamic response of the SR540 internal thermocouple is much faster than the others because it is a bare thermocouple junction measuring the air inlet temperature. It changes very fast after changing a fan or temperature setting, while the thermocouple inside the probe in the bean chamber changes more slowly and it is more representative of the bean interior temperature. I started trying to use the SR540 internal temperature sensor to adjust the settings, but I quickly grew frustrated and started exploring other options. Hope this helps!
Further to my comment below I am hearing impaired and really could use your help. Is there some way we can connect to chat further, other than using Comments?
Hello and thank you for your interest. I do not have much more to offer on the sound processing front, the Audacity program worked well for me and as shown in the video you can see the spikes for FC and SC without the need to hear them. Since I use the same beans and profile for my roasting, I no longer use the sound processing setup because the FC and SC always happen at approximately the same bean temperature (in my case ~ 400 F for FC and ~445 F SC).
We need Group watsapp
To share experiences