2014 World Coffee Roasting Competition profiles - Page 4

Discuss roast levels and profiles for espresso, equipment for roasting coffee.
User avatar
antonio
Posts: 29
Joined: 15 years ago

#31: Post by antonio »

Probe is the same here. Thermometers and measurement circuits are different.

User avatar
MaKoMo
Posts: 846
Joined: 16 years ago

#32: Post by MaKoMo »

Oh, and all smoothing on the Artisan side is deactivated. Just raw data as received from the Phidgets 1046.

Advertisement
User avatar
antonio
Posts: 29
Joined: 15 years ago

#33: Post by antonio »

But right, the Giesen measurements compared to strict temperatures are delayed, shifted and attenuaited. Not because of bad placement of the probe nor because of slow response of the probe itself (however these problems also apply). It is because of data manipulation (DSP) in their system.

So all the time, when I am looking at the control panel of my roaster, I see the result of some funny mathematical calculations, not real time temperature measurments... And all profiles grabbed by Cropster during the WRC show us some funny mathematical calculations & transformations made on real time measurements.

The only way to have a strict navigation/measurments in the roast process with the Giesen roaster is to link it with the Roast Scope like Artisan, Cropster, RoastLog. And of course, use your own temperature sensors located in reasonable places.

User avatar
antonio
Posts: 29
Joined: 15 years ago

#34: Post by antonio »

During discussions with some engineers and from our investigations, we discovered such a unit as responsible for temperature measurements in Giesen roasters:


Click in the picture to see the documentation.

There are some Low Pass filters at each of the PT100 inputs. These filters have to be switched off, if we want to see right temperatures at the roaster's control panel. I wish the Giesen Coffee Roasters let us know how to do this little job...

In my humble opinion, presenting modified, malformed temperatures, far from any reasonable measurements is not fair. How about World Roasting Challenge? Who of you could understand these machines during a few roasts? Roasting coffee is somehow strict, scientific activity, so competitors have to have strict, logical, clean data at their tables.

Machinery, mechanical side of the Giesen Coffee Roasters is a different story: it is very good and reliable. We roasted thousands of great coffees with our W15, but always frustrated by the software side and the temperature measurements...

User avatar
antonio
Posts: 29
Joined: 15 years ago

#35: Post by antonio »

Since my last post, we studied the automatics of Giesen more. As I wrote before, input filters at the temperature measurments channels of the MFD-TAP13-PT-A unit realize a delay function. With these filters switched on, the delay is about 1 minute. Can you imagine? You roast your coffee, and the control panel of your roaster tells you how was the temperature of your coffee 1 minute before :).

We switched off the filter at the BT channel yesterday. What is the difference? Now the Artisan and the Giesen panel show almost the same temperatures and exactly the same timings. There is a little shift in value, about 4 degrees, but without any time delay in Giesen.

The question is: Why Giesen uses delay/filtering modules to smooth measurements? IMHO, it has not any reason in the circuitry and its stability. Important measurements should be as clean as possible, not malformed. We have not received any comment, any answer, any advice how to switch the filters off. The only reason I can imagine is to show users a very smooth, small amplitude wave in the auto mode... But the price of malformed information during roasts is too big.

popeye
Posts: 340
Joined: 18 years ago

#36: Post by popeye »

Wow. If these claims are true, that's a pretty big embarrassment for Giesen. Can anyone else back this up?
Spencer Weber

User avatar
antonio
Posts: 29
Joined: 15 years ago

#37: Post by antonio »

I have just removed the inertial filter from the ET channel (which is used for controlling and safety), and I can tell that the machine works better now. Soon, I will show some real, not malformed curves measured by the Giesen roaster circuitry.

I am sure, if the above story was about e.g. Coffee Tech rosters, my posts and questions would have immediate effect and response from the producer :).

Post Reply