Ideas for dialing in my fermentor temp control?

[Brewsit]

Im looking for any ingenious ways to dial in my fermentor temperature control. Here are the specs:

BruControl software

27 gallon Stout stainless conical

Two thermistor probes, one in the top a couple inches in the top of the wort, one near the bottom.

ULWD 4500w heating element for heat (hooked up as 110v)

55 gallon Glycol chiller with stainless immersion chiller in the fermentor

Chiller set up with hysteresis for the chiller, reading the lower temp probe. 2 min delay and 1 degree offset.

Heater set up with PID and upper temp probe, 18/0.5/0.5 as the initial tune.

The picture should give an idea of what I’m seeing. It’s a 10 hour x axis and 63-68 degree y axis.

The upper probe shows about 1/2 a degree warmer than the lower right now, and this has slowly gotten closer through the last 10 hours. The heater has run about 3 or 4 times, I have it set at 15% max output and 50% max integral. It has run a couple of times through the day, the ambient air temp is low 60s.

I honestly have a 3rd grader’s understanding of PID so I was trying to be conservative.

It seems pretty stable but the fermentation hasn’t really kicked off yet.

Any thoughts on how to optimize this setup?

 

[BrunDog]

I would need to think about this. I’m not sure you should use PID for the heating but I don’t see PID being bad other than a need to limit the integral windup.

We are working on prediction for Hysteresis so slow moving systems don’t over or undershoot as much.

 

[Brewsit]

I think that predictive hysteresis would probably be ideal. It doesn’t seem to need much adjustment, it is pretty stable.

 

[BrunDog]

Probe location can make a big difference. I think there is an ideal location... too close to the liquid/cooling surface and it will respond quickly to temp changes, causing lots of cycling and apparent temp swings. Too deep in the liquid and it will be slow to respond to temp changes, causing over and undershooting. I *think* the ideal location is about a quarter of the distance between the heating/cooling interface and the center of the liquid volume.

 

[Brewsit]

I think with my system that ideal location will be a mystery. The cooling coil is just an immersion chiller sitting in the top half of the liquid, with the heating element below it. The lower probe is bent so it’s not too close to either, but it’s at about the same level as the heating element. The upper is just vertical pointing down into the top of the wort.

Last night I set the max output to 50 since it was holding a degree low. That brought it up to the set temp pretty quickly (probably too quickly, I would like to see it take a long time to adjust) but it held steady all night and only needed one small dose of heat around 1am. I think the yeast is starting to generate its own heat now which is another factor to add in...

 

[BrunDog]

Ok. Of course you want to avoid a situation where over or undershorts cause alternating heating and cooling!

 

[Brewsit]

So I’ve dialed it in pretty satisfactorily at this point. When I initiate a temp adjustment I keep the adjustments small because it’s really, REALLY good at getting there quickly. There’s a minor overshoot and then it kind of goes up and down over the course of an hour (by less than one tenth of a degree) and then locks in solid. I’m pretty happy with PID for the heater and hysteresis for the cooling coil at this point.