Finally today, I got to roast.

Basically, I changed my 1048 TC Phidget and the the stock MCR 5mm thermocouple BT probe and 2mm thermocouple ET probe for the following Phidgets:

VintHub
(2) TMP 1200s for RTDs
RTDs from Evosensors 3.25 inches
Ambient Humidity and Temp Sensor
Ambient Barometer Sensor
1136 Differential Air Pressure sensor

The RTDs were done for me in a custom size of 3.25 inches. I added my own transition protection to the probes. Evosensors is working on obtaining springs for people wanting something more substantial.

To set it up there are some things you have know. The 1136 is best put in Port0 on the VintHub and it set up as an Extra Device.
The BT 1200 is set up as the main device, and the ET is an Extra Device which is mapped to ET on the Symbolic Tab. The 1200s can be set up first using the Machine set up in the menu for Phidgets. NOTE this will erase all your other Extra Devices so make sure you save your settings. Also the Symbolic formula will change depending on the number of the extra device. Since my 1200 is the first device, the mapping is Y3. If it were the second it would be Y5. This is covered in Artisan blog posts.

Images below show the phidgets in an old wine box. For the air pressure connection to the roaster, the push to connect was cool to find. Previously there was an analog temp gauge there.

What I noticed is max ROR is higher. ET and BT are more aligned at start with the same size probes. Reaction time is quicker. You can see the ambient data in the statistics box on the right: temp, humidity, barometer.

The air pressure has given me some insight into what my fan dial really means. I truly believe the 1136 will help you learn how your roaster handles air flow, but I am not yet convinced it will help roast from a background. We shall see. For now I have chose to show the curve along with events I marked using alarms.



Top are ambient sensors and the 1136:

Holes in side make it easy to keep cords sorted:

Push to connect was a great find. 1136 recommends 3/32 ID hose so this was 3/16 OD.

Simple transition protection is heat shrink and a cord protector:


My Extra Device settings. There are other ways to set this up as well.

Mapping of the first Extra Device to the ET reading


If you want the details with links send me a PM with your email address and I will send my Word document.


Notes on what went wrong. Adapters didn't play well, so RTDs were too short. Make sure you understand your adapters before you settle on an RTD size. Then when the right ones were here, I discovered one of the TMP1200s was DOA so Phidgets had to send another. For the air pressure, I chose the wrong size fitting first time and had to wait for a correct one. Note Phidgets was great on the service. I just showed them the readings in the Phidgets control panel and they sent a new one overnight. But note they only accept returns in Canada so it's generally not worth the cost so order carefully. I had not realized the Humidity sensor did temp too, and so I ordered a temp sensor that is of no value now.

Finally a note on RTD insertion depth. Mine are 1". Here is what I was told about that topic:

The 2 inch minimum on a RTD is a misunderstood and in some ways outdated guideline.
In early years the RTD temperature sensing elements had small diameter platinum wire wound around a ceramic core and to achieve 100 ohms resistance in the circuit the length of the RTD element could be 1 more than 1 inch long. The placement of the element in the closed end tube or in a MI Cable construction could add another 1/2" as such the 2 inches. In fact today despite technology improvements with wire wound RTD sensing elements a common length is 25 mm (1 inch) plus the lead wires length.
The wire wound RTD element have mostly been replaced throughout much of industry with a thin film RTD temperature element, were the platinum is melted and deposited on ceramic wafer in sophisticated semiconductor type processing equipment and the most common length is approx. 2 mm (.078" or slightly more than 1/16 of an inch) as such the RTD temperature probes can provide accurate temperature measurements with insertion depths of less than 1/4".
The other reason of a 2 inch immersion being referenced is many temperature calibration equipment manufacturers indicate their highest accuracy is achieved when a temperature sensor is inserted at least 2 inches in their heated or cooled fluid, sand or metal block.

Looks great! Now you have my eyeing that 1136 for my Huky. Can you share more on the hardware you used on your MCR exhaust for this? I assume there is likely curved metal tube inside that points downstream?

P.S. FYI; I re-positioned one of my RTD's today in a compression fitting without a teflon ferrule and it worked great. As long as you don't tighten it down so much that it crimps the probe it will move easily. Fortunately in coffee roasting a light 1/4 turn will hold it without fixing the probe to the ferrule.

Steve,

Didn't use a tube. Not sure that would matter for a differential reading. Even if it did not sure it would matter for the relevant data here. BUT I am not an engineer.

What it has taught me is there is a more narrow range of fan dial settings than I thought. And it shows me more clearly how fan changes impact the roast. I think it's worth it so far. Now I am not sure yet it will allow me to replicate roast to roast. Time with tell.

Yes I did the very lightly tighten on the RTD adapter and it works fine.

Look at you go, Mike!! Love it. I'm upgrading to a digital gas manometer next week and then I'm going to tackle the air pressure thang.

Which pressure gage did you use?

Well I think I'm going to have to try another RTD - because my one dropped off rapidly after 150c when inserted to 30mm. Also you seem to be getting a TP point under 1min which is similar to my 2mm K Type.

My 3mm TP is about 1.30 mins. Not sure what's going on.

Phillip,

What do you mean dropped off? And your 3mm RTD has a tipping point of 90 seconds. That doesn't seem out of line. Different roasters will be different on that aspect.

Alan,

don't think it's appropriate for your beast, but it's the Phidget 1136 differential air sensor. I think these devices tell people a lot about their air movement/fans, even if they aren't precise enough for background roast profile. They will tell you if your adjustments are repeatable.

I think you may want to look into an Magnehelic indicating transmitter with a Phidgets DAQ1400. I don't understand the ins and outs, but I understand this is more of an industrial controls type level of usage that would be appropriate for your set up.

Keep us updated. This was a fun project. I enjoyed solving the puzzle and I think it will help in roasting.

So in playing with curve settings, I settled on the following to start.

I put a very small amount on Delta BT for smoothing. 1 and 2 did not impact the smoothing with Smooth Curves at zero. With 3 there is a small amount and based on my roaster I thought that for me it would give me visually what I need to start to compare sampling rates of 1,2 and 3 seconds.
I may ultimately reduce it to zero. I find that with the zero smoothing it's hard for me to judge if its just bouncing up and down or not doing what I hope it is.

If anyone has thoughts on sampling rates or curve settings including how you read them with a setting of 0, please share.




Setting 0

Setting 3

CarefreeBuzzBuzz wrote:Alan,

don't think it's appropriate for your beast, but it's the Phidget 1136 differential air sensor. I think these devices tell people a lot about their air movement/fans, even if they aren't precise enough for background roast profile. They will tell you if your adjustments are repeatable.

I think you may want to look into an Magnehelic indicating transmitter with a Phidgets DAQ1400. I don't understand the ins and outs, but I understand this is more of an industrial controls type level of usage that would be appropriate for your set up.

Keep us updated. This was a fun project. I enjoyed solving the puzzle and I think it will help in roasting.

I think the beast has a pressure gage built in. I'm looking for a solution for my 5kg roaster. It has a stepped detent damper which allows for a little repeatability, but with strict limitations. I'd love to get rid of the detent, but then I'd have to be able to measure air pressure to determine my settings.

Alan,

Can you explain more or show us how the detente work. I think one of the devices would be able to allow you to repeat settings within a measure of accuracy. Only you can decide if it's precise enough for you.