Controlling a Cormorant Roaster with Artisan - Page 2

[davecx365]

Nice setup and report!
To me, the gas control is still the hardest part to control on the Cormorant. I find it hard to make adjustments for anything less than 4mbr at a time. The heat control (via Artisan) is about the only thing I miss from using the Hottop. Its good to know that someone has found a solution for gas control.

I would like to do something like as well but that will have to wait because of too many other projects. I do have a couple of questions.

Roughly how much did it cost you to accomplish the gas control part?

You also mentioned that a proportional valve as a possible solution but rejected it because of cost. What do those run if someone wanted to go that route?

In the photo of the completed cart, what is on the right-hand side of the roaster? (the big black box thing) An additional cooling unit?

[Brewzologist]

I still think I will go with RTD and hope that setup of the probes (installation into the roaster) and the Phidgets is as "easy" as everyone says. I am a retired IT person but not particularly handy like you are with building things.

You can source the RTD probes anywhere as long as they are the right length for the Cormorant. The diameter to get depends on the compression fitting included or sourced to fit the roaster. I suggest a ~3mm diameter RTD. If you search for "RTD" in the roasting forum you will find a lot of information about them. Honestly, the hardest part for me was finding a 3mm compression fitting to fit the thread pitch on my roaster, which may be a non-issue for the Cormorant. I bought my RTD's from Phidgets and highly recommend them versus standard thermocouples.

As for the value of Artisan and how to configure it, I'd suggest you start here to learn all you can: https://artisan-scope.org/docs/quick-start-guide/

[Brewzologist]

You also mentioned that a proportional valve as a possible solution but rejected it because of cost. What do those run if someone wanted to go that route?

If you are considering a proportional gas valve you may find the following thread useful. In the end, the proportional valve was not necessarily a slam-dunk, and some elected to use a needle valve and servo for gas automation instead. I haven't gone back and researched valves again since automating my roaster, but finding a valve that gives you the flow and precision needed for your roaster is certainly a challenge regardless of type.

Exploring North Gas Roaster Automation: Part 1

[bwren]

@ronncat, fantastic work especially on the gas valve control! Love the clean work (you even neatly mounted the 3.7V buck converter for the lamp!)

I am the guy who first identified the current limit issue on GitHub with Phidgets DC motor controllers and Artisan, and made the request to Marko to add the API code. How fortuitous that you came upon the GitHub site at the very same time the solution was created!

Two things that may be of interest to you:

1. I made a request to Marko about a month ago to see if it was possible to allow Artisan control outputs to be driven by external quantifiers. After a few back-and-forth tests and coding adds/changes, there is now a solution that should be in the beta version you just downloaded. You can now log fan and drum changes automatically while controlling the fan and drum from the roaster. I can still also send these commands from Artisan, but quite often I prefer to be hands-on at the roaster. I'm happy to share details if you are interested - the only parts needed are 2 potentiometers, which plug into the Phidget Hub. Marko is truly amazing, and I urge anyone on this forum who uses Artisan to please donate, if you have not done so already.

2. I see you are using DCC1002 controllers for both motors. The fans are 2-wire brushless, that do not play well with PWM signals (the DCC10xx series are mainly intended for brushed DC motors like the drum motor). Yes it works, but it will compromise the life of the fans. It is for this reason that Johan does not provide a PWM controller for the fans, unlike the drum controller board. The fan controller board from Cormorant is actually a variable voltage regulator, that limits the output to around 12V (12.5V is their max rating). Also, if you have not done so already, you should scale the fan Phidgets PWM output to around 86% to reduce the 14.1V from the power supply.

I first tried to filter the PWM output of the DCC1003 (2x 4A version of your 1002) with a custom low-pass filter. This brought the ripple down to a few mV, but it was a bit of a clunky solution. So I ended up using a Phidgets OUT1000 to provide 0-5V control that drives a custom voltage regulator board for the fans. I'm happy to share details, and will have a few extra custom PCBs which I just ordered from China.

I decided to house all the control stuff in an outboard 6x3x2" case with a 8-pin connector and thermocouple jacks. The roaster, sensors, and laptop get plugged into this box. The only electronics I have inside the roaster are the two high-current components: Phidgets DCC1003 and the custom fan regulator board.

I thought about gas control, but was a bit intimidated by the scope and cost, and felt I was a bit over my head going down that rabbit hole. But after seeing your incredible work, I may be inspired to look at it again! BTW, Phidgets has many stepper motor controllers, and who knows, Johan may be willing to incorporate them.

I am also working on a way to auto-log diffuser changes (open/close). I just got a $10 Phidgets PRX2300 that I'm going to try and use for this purpose.

Cheers, and congrats on the great work!

[ronncat (original poster)]

Gavin,
You are the man!!! If it wasn't for you, I would be out shopping for a toupee
Haven't had a chance to jump on the Facebook group of late, but it appears as if you've posted all the pertinent info here.

I WANT TO STRESS TO EVERYONE THAT GAVIN IS MUCH MUCH MORE VERSED IN THE ART OF ELECTRONICS THAN I WILL EVER BE. I WOULD TAKE HIS COMMENTS AND SUGGESTIONS TO HEART.

Thanks for posting this very important information. I had wondered a bit if my fan was a bit wonky as it registers about 6V on the meter, but seems to be running at full tilt. I would love to have information on your custom voltage regulator board as that seems to be a great way to approach this. Also would love to know how you incorporated the potentiometers into the fan and drum speeds. Can you use the existing ones on the Cormorant and wire them to the hub? This would be very slick as it would still offer manual control to the roast.

I also have on hand a Phidgets stepper motor controller. I decided to go the Pololu Tic route along with the Phidget OUT1001 as it was already supported in Artisan, whereas the stepper motor controller was not. I would gladly replace the Tic and OUT1001 with the Phidget Stepper Motor controller, but I would have had to learn how to program it into Artisan. I have very limited experience doing this. I've only programmed a few times with Arduino for various projects over the years...

As far as the diffuser, I also considered doing something. Perhaps a linear actuator... But then I thought..is it really worth it??? Opinions welcome

Let me know if I can be any assistance with your gas hookup. At least you will know which valve worked for me. Still 4 of them left last time I checked ebay. Definitely could downsize the stepper motor as well.

[BoulderMike]

To ronncat and bwren: Sorry to ask this again but your posts are making me very, very nervous. If you two feel the need to do so much customization to the Cormorant, can someone like me expect it to be a good purchase/option when my goal is to use it "out of the box" and without customization. You two seem to be pointing out a lot of operational shortcomings of the Cormorant and it is making me wonder if a roaster like the Cormorant "out of the box" is a mistake for me. My goal is to roast good coffee without a lot of tweaking and customization. I am fine with tweaking the controls of the roaster during the roasting process, but don't want to customize.

[LBIespresso]

While waiting for my Cormorant I watched countless hours of YouTube videos on roasting, read both Scott Rao books, and took a 1 day roasting course. With that I was roasting drinkable coffee within my first 10 batches including the 4 seasoning batches that went to the compost pile. By drinkable I mean better than the local crappy coffee shop. Not too long after that I feel like my coffee is better than the good coffee places around me and on par with much of the mail order coffee I have always enjoyed.

The Cormorant will give back what you put in without much in limitation other than volume. Feel good about your purchase and get excited for delivery On the upside, with a wait list this long if you decide roasting is not for you there is a strong demand to sell into.

[bwren]

To ronncat and bwren: Sorry to ask this again but your posts are making me very, very nervous. If you two feel the need to do so much customization to the Cormorant, can someone like me expect it to be a good purchase/option when my goal is to use it "out of the box" and without customization. You two seem to be pointing out a lot of operational shortcomings of the Cormorant and it is making me wonder if a roaster like the Cormorant "out of the box" is a mistake for me. My goal is to roast good coffee without a lot of tweaking and customization. I am fine with tweaking the controls of the roaster during the roasting process, but don't want to customize.

BoulderMike, Artisan is designed as a simple roaster that works right out of the box, an in fact I don't think Johan is keen on any elaborate control mods, as he does not envision it as a hi-tech device. For my first few roasts he recommended I be hands-on, and not have my head buried in Artisan. I got surprisingly good results in my first few roasts without even using Artisan - I had a simple TC meter hooked up to the 2 probes, and logged with pencil and paper.
There will always be tinkerers like Ron and me that try to automate things, but believe me, you do not need to do so to get excellent results from the Cormorant and use it to its full potential. Just make sure you use TC or RTD probes as the analog gauge is not very useful.

Gavin

[JohnB.]

To ronncat and bwren: Sorry to ask this again but your posts are making me very, very nervous. If you two feel the need to do so much customization to the Cormorant, can someone like me expect it to be a good purchase/option when my goal is to use it "out of the box" and without customization.

I wouldn't worry about it. Received my CR600 11/2018 after roasting with a modified Hottop B for 10 years. Only mods I've made were going with 2mm TCs instead of the 5 or 6mm versions Johan was installing at the time, moving the ET probe to about where Johan eventually moved it & upgrading to the larger burner when Johan offered it (Fall 2019??). I monitor the Bt & Et temps with a Fluke 54-2 meter just as I did with the Hottop. All roasts are logged manually in my roast log; no Artisan.

Just roasted 5 1lb batches 2 days ago & had no issues with hitting the bt/development time (typically 2 minutes or less) I was shooting for. All my roasts fall within the City+ (approx. 20* bt after 1C takes off) to Full City (approx 30* bt after 1C starts) with a rare trip into early 2C once in a
great while. I haven't had the gas control issues mentioned here as I seldom find the need to make minute gas changes & I never turn it below 10 mBar. For a Full City roast I will usually come into 1C with a 20 mBar gas setting, shut off the gas completely after the first minute or soon after & coast right up to my desired bt at the end of the development time I was shooting for.

My only complaint with my Cormorant is that the fan control is touchy so it's hard to make small changes without some back & forth. That may have been remedied with the new fan control setup. It's a great roaster & very easy to control manually once you get used to it.

[BoulderMike]

Thanks Johnb and bwren, I truly appreciate your kindness in responding to my novice concerns. I believe I undertand more now that what you were customizing, at least I think I understand this, was adding the ability to control the Cormorant from Artisan. I don't need to do that. Anyway, I appreciate everyone who replied back and I am now comfortable and continue to eagerly anticipate getting my Cormorant. Now if the USD wasn't so weak ...