BCS 2 Vessel No Sparge Garage Brewery Build

[poptarts]

Great job! Pretty sure I would have lost motivation to brew right about the time I saw 9 degrees and realized the pumps were frozen.

I wouldn't be able to brew half the year if that was the case haha. nothing some snowboarding gear and fire cant fix.

 

[iijakii]

Lol man that's rough. Time to get a wood stove installed in the garage! Maybe you can plumb its exhaust and the steam hood together.

 

[poptarts]

Lol man that's rough. Time to get a wood stove installed in the garage! Maybe you can plumb its exhaust and the steam hood together.

There is a wood stove, it just takes awhile to bring the temp up 60 degrees haha. And im just going to duct exhaust right out the wall behind the kettle, super simple like.

 

[BrunDog]

Congrats!!!! Been waiting patiently for this maiden brew! Glad it went well!!

I did a brew in my garage today too. The temp was also a bit uncomfortable - 86 degrees!!

 

[Carlscan26]

Got a return float for the mash and my WAP since I'm going WiFi on the stand since it's in the garage

What size tubing does that float accommodate? If it's 1/2" can you post where you found it? I've only been able to find ones for 3/8" tubing.

 

[Carlscan26]

I scanned through the thread and couldn't find the details on the snorkel - can you elaborate on that?

 

[poptarts]

I scanned through the thread and couldn't find the details on the snorkel - can you elaborate on that?

Yea I can get you pics next time im outside. Its basically a Tee made out of 1/2 copper pipe and has a stainless steel wool ball at the base, works and cleans up insanely good and just sits there no installation. The ball does fit 1/2 tubing, its a little big but hose clamps on either side of it keep it from sliding around on the hose. Here is the float and I can get you pics of it on the hose as well.

 

[poptarts]

No signs of Fermentation on the brew. I'm worried my temps were more crazy than I thought

 

[BrunDog]

No signs of Fermentation on the brew. I'm worried my temps were more crazy than I thought

You said you used a large CONAN starter? What temp is the fermenter at?

 

[poptarts]

2l starter fresh gold bag and at 64

 

[BrunDog]

That's weird. No life at all?? The starter was legit (made bubbles, smelled yeasty, etc)?

Did you check for conversion post mash? Probably just needs more time but if no life 24 hours from now I would be worried.

 

[poptarts]

That's weird. No life at all?? The starter was legit (made bubbles, smelled yeasty, etc)?

Did you check for conversion post mash? Probably just needs more time but if no life 24 hours from now I would be worried.

Yea starter was 2 stage with canned wort I make. When I added second liter it was fermenting away visibly and smell. The wort is sweet and shows up proper of on refrac so I assume it converted fine. Idk I have never had anywhere near this long lag.... I'm thinking it's wrecked. Wort og 36 hrs later is same and 0 signs of fermentation

 

[thekraken]

I assume you checked your temperature probes for calibration?

What was your setpoint? I wouldn't think 4 degrees above would be enough to completely denature... Hell, I've done that

 

[poptarts]

Was aiming for 153 highest I saw was 158. I haven't checked probes but all read the same more or less. They are in thermo wells so I'm wondering if they were reading low

 

[BrunDog]

Sweet wort isn't necessarily converted wort. But even if the enzymes were denatured there would be some sugars and some yeast activity. I would pull a sample and do an iodine test just for fun.

In the future you should use a thermapen or other good thermometer until you prove your probes are spot on. I assume you adjusted the BCS thermistor coefficients to match the probes you have?

 

[poptarts]

Sweet wort isn't necessarily converted wort. But even if the enzymes were denatured there would be some sugars and some yeast activity. I would pull a sample and do an iodine test just for fun.

In the future you should use a thermapen or other good thermometer until you prove your probes are spot on. I assume you adjusted the BCS thermistor coefficients to match the probes you have?

Yea I'm deftinatly going to compare against my other thermometers. And I didn't adjust but I believe they are set by brewers hardware by default. Again ill compare I suppose haha

 

[poptarts]

So there is about a 2" patch of bubbles filming the surface in the center today. Pretty sure its hosed. I will let it ride and see but Im going to assume its going to finish in the 30s and be a dumper. I think the temps must have been higher than reporting since it was swinging around and the thermowells have a delay I wouldn't doubt that it got much hotter and colder than reported. Any of you BCS guys have tips on how to tune the PIDs? I'm trying to understand it but I cant find like a real world example of like what each variable does, and I have no idea what the range on the is either, if there is one at all. All the things im finding are like PHD research for control system engineers and other crazy stuff.

 

[crane]

So there is about a 2" patch of bubbles filming the surface in the center today. Pretty sure its hosed. I will let it ride and see but Im going to assume its going to finish in the 30s and be a dumper. I think the temps must have been higher than reporting since it was swinging around and the thermowells have a delay I wouldn't doubt that it got much hotter and colder than reported. Any of you BCS guys have tips on how to tune the PIDs? I'm trying to understand it but I cant find like a real world example of like what each variable does, and I have no idea what the range on the is either, if there is one at all. All the things im finding are like PHD research for control system engineers and other crazy stuff.

Tuning a PID loop without calculus is a bit of an art form IMHO. Definitely use the PID tuning utility that is part of the recipe and configuration migration utilities. This will at least show you what each part of the algorithm is doing. For what we are doing you can set the derivative value to 0, or 0.1 since it won't accept 0. I also set the min integral to 0 which helps it settle quicker. I also reduce the max integral which helps lower overshoot.

There are some charts out there that help explain what gains to adjust when one or more parts of the step response are off.

 

[TrickyDick]

Tuning a PID loop without calculus is a bit of an art form IMHO. Definitely use the PID tuning utility that is part of the recipe and configuration migration utilities. This will at least show you what each part of the algorithm is doing. For what we are doing you can set the derivative value to 0, or 0.1 since it won't accept 0. I also set the min integral to 0 which helps it settle quicker. I also reduce the max integral which helps lower overshoot.

There are some charts out there that help explain what gains to adjust when one or more parts of the step response are off.

I don't pretend to understand that.

On my propane system I had PID capable temp controllers, but found that it was much simpler to set as a dumb controller with a differential at which it would begin heating if the wort/water cooled too much. Also, unless you are brewing the same thing each time, the amount of malt and water is likely to vary, which I believed would have made it difficult to train the PID, which had an automatic tuning mode.

Regardless, sorry to hear about the crappy fermentation. So sounds like you are suspecting that the mash temps were too high and you cooked the enzymes? Suggestion I have is to pitch some roeselare or other mixed strain and make a sour beer. I have never trusted a brewing thermometer unless I had verified it as reading the proper temp against another thermometer, such as a Thermapen (New model is supposedly water proof, but perhaps isn't immersion proof).

Keep us posted!

TD

 

[BrunDog]

I wouldn't worry about tuning right now. With enough thermal mass the default values should be fine. You need to make sure your probes are reading correctly. Check them against a legit thermometer. The thermowells shouldn't be an issue though they will slow down response slightly. Again, with enough thermal mass it should be ok.

Don't freak out - just debug everything in serial order. Dumping a batch is no biggie (even though it's a bummer). Get your system working and good beer follows.

 

[thekraken]

So there is about a 2" patch of bubbles filming the surface in the center today. Pretty sure its hosed. I will let it ride and see but Im going to assume its going to finish in the 30s and be a dumper. I think the temps must have been higher than reporting since it was swinging around and the thermowells have a delay I wouldn't doubt that it got much hotter and colder than reported. Any of you BCS guys have tips on how to tune the PIDs? I'm trying to understand it but I cant find like a real world example of like what each variable does, and I have no idea what the range on the is either, if there is one at all. All the things im finding are like PHD research for control system engineers and other crazy stuff.

I wouldn't think the thermowell's delay could be too detrimental but I'm no expert. On my system, at least, the temperature changes slow enough and the liquid passes over the thermowells quickly enough that I don't think it matters (for me).

This is my understanding of a simple PID algorithm:

sample = measured temperature
error = the difference between the setpoint and the sample
kP, kI, kD = PID *factors* for tuning

PID output =  kP * error
            + kI * (error + previous_error + error_before_that + ... + all_the_errors)
            - kD * (sample - previous_sample)

Take a look at the PID::Compute() function here:
https://github.com/br3ttb/Arduino-PID-Library/blob/master/PID_v1.cpp

Some time ago I used the code linked above and made a model in an excel spreadsheet. Playing around with this spreadsheet really helped me visualize and get a feel for what was really going on. It may not be the exact same approach to PID that BCS uses but it might help provide some intuition.

 

[poptarts]

I wouldn't worry about tuning right now. With enough thermal mass the default values should be fine. You need to make sure your probes are reading correctly. Check them against a legit thermometer. The thermowells shouldn't be an issue though they will slow down response slightly. Again, with enough thermal mass it should be ok.

Don't freak out - just debug everything in serial order. Dumping a batch is no biggie (even though it's a bummer). Get your system working and good beer follows.

Yea for sure, im a programmer so I know how this stuff goes. I'm thinking maybe I need to turn down max pulse on my mash element. I might toss some water in there tonight and go sit inside and try to get it working a bit more stable. I remembered that this thing keeps graphs so I got a picture of what happened.
Setpoint was 153 till then end when I set it to 168
Green is probe in MLT just under false bottom
Blue is MLT Return probe and the one I use for PID
Red is BK and the last spike was intended for my mashout process.

 

[thekraken]

Green is probe in MLT just under false bottom
Blue is MLT Return probe and the one I use for PID

Probes mixed up?

 

[TrickyDick]

From the looks, you shouldn't have baked your enzymes. not sure what the bottom timeframe is there. Wondering if you did an iodine test or not. If those temp probes aren't calibrated could be a different story. Any gravity readings indicating fermentation?

Good luck! Sounds like those bottom dump ports made cleanup a breeze.

TD

 

[poptarts]

Probes mixed up?

no element is in the MLT so MLT reads hotter since return gets cooled a bit by pumps and hose before it makes it back to the MLT

 

[poptarts]

From the looks, you shouldn't have baked your enzymes. not sure what the bottom timeframe is there. Wondering if you did an iodine test or not. If those temp probes aren't calibrated could be a different story. Any gravity readings indicating fermentation?

Good luck! Sounds like those bottom dump ports made cleanup a breeze.

TD

No I dont have iodine so cant do a test. I believe the probes are calibrated. They all read the same more or less and during boil the boil one was reading 198.6 - 199 and water boils here at 198.8 so they are definitely in the ball park.
And yea I dont feel it got hot enough to kill it, but its appearing it did and im confused as to why.

 

[thekraken]

no element is in the MLT so MLT reads hotter since return gets cooled a bit by pumps and hose before it makes it back to the MLT

Ohhh, sorry I thought you were running a RIMS tube, I must have had you confused with another build.

Out of curiosity what are your tuning values?

 

[poptarts]

Ohhh, sorry I thought you were running a RIMS tube, I must have had you confused with another build.

Out of curiosity what are your tuning values?

they were default to start, but I started tweaking them not knowing what I was doing obviously with no success. PID 20/.5/10
I think the plan is tonight to fill it up with like 10 gallons of water and hang out inside my toasty house and see if I cant get it figured out a bit more. Then brew again over the weekend hopefully with more success.

 

[thekraken]

I'm trying to understand it but I cant find like a real world example of like what each variable does

I just came across this, have you already seen it? It appears the BCS pid implementation is about the same as the one I was jabbering on about a few posts back:
http://wiki.embeddedcc.com/index.php/PID_Implementation

they were default to start, but I started tweaking them not knowing what I was doing obviously with no success. PID 20/.5/10

Did you tweak your integral clamps?

 

[poptarts]

I just came across this, have you already seen it? It appears the BCS pid implementation is about the same as the one I was jabbering on about a few posts back:
http://wiki.embeddedcc.com/index.php/PID_Implementation


Did you tweak your integral clamps?

I did see the wiki and am going to be mucking with it more tonight for sure. Can you explain what the integral clamps do? I don't understand them. I believe I understand the pulse ones but not the integral limits.