Fermentation temp monitoring?

Perhaps it's been done to death but since I'm new to this...

Is there any interested in a data log showing:

1. Ambient temp
2. Fermenter surface temp (the plastic bucket type)
3. Wort fermentation temp

I'd set up a datalogger with an exposed junction thermocouple for ambient maybe a foot or two away from the fermenter, an adhesive thin-wire thermocouple on the bucket external surface, then a sealed junction thermocouple immersed in the wort. All thermocouples would be set for the same height relative to the center of the wort.

What do you think for logging rate, if interested?

(Note this wouldn't happen until next batch, likely at least a month away.)

I wanted to do something like this, but slightly modified for my setup. I use a swamp cooler (Igloo IceCube), and I wanted to monitor the ambient, the "swampwater" temp, and the wort temp. I was most interested in the difference between the water bath temp and the wort temp. I've heard on here that they should be about equal, but I wanted some data to back that up. Alas, I never found any equipment that was cost justifiable (read: cheap) for me to do the test. Do you have a setup for this?
I swap out my ice bottles for my lagers every 12 hours. I felt a good rate for data logging (in my setup) would be every 5 mins. I felt like you'd see the drop in water bath temp from the addition of the ice bottles, and then the rise as fermentation continues.

Point me in the right direction on equipment, and I'll add my data to whatever you collect.

I've done something similar with my swamp cooler here:
http://woodlandbrew.blogspot.com/2012/09/swamp-cooler.html
There are links to the equipment I used below the post. It's pretty cheep, but you'll have to figure out how to convert ohms to degrees Fahrenheit.

Oh, and I would love to take a look at that data log!

I'll check it out, Woodland, thanks.
I think folks would be interested in the temperature differential and a nice graph showing the 3 temps, especially over a few different brews. The equipment is a hurdle I think though. I've looked at USB temp probes/loggers (easy), but most the probes aren't long enough or aren't submersible. Once I find good equipment, I wouldn't mind doing this for all brews.

Problems I see is that yeast strains will do different things in different worts. Use the same yeast strain in two different recipes and you can have very different reactions/actions.

Personally, I have a thermowell going into the middle of my fermenters. I send a thermometer sensor probe down that, sealing the top/opening of the thermowell (so that ambient air doesn't impact the reading). I then set the reader off to the side where I'll be able to see it. The unit I use records the high, low and shows the current temperature of the sensor. I have another sensor that shows me ambient temperatures. IME, fermenation temperatures can be anywhere from 1-10F above ambient (or even more). At some point, I'll probably look for a wireless unit that I can send down the thermowells I have, and read the temperature from someplace other than the basement.

Golddiggie said:

Problems I see is that yeast strains will do different things in different worts. Use the same yeast strain in two different recipes and you can have very different reactions/actions.

Click to expand...

Very much agree, that's one of the reasons I'd like to do this, especially multiple times, because I believe we'd be able to at least see a trend.

The last time I saw someone ask about this was here
I looked at this as well, but I don't think it's the best solution. The OP hopefully has some equipment ideas, if so I'm all ears.

whoaru99 said:

Is there any interested in a data log showing:

1. Ambient temp
2. Fermenter surface temp (the plastic bucket type)
3. Wort fermentation temp

Click to expand...

The only one I really care about is No. 3 and so that's what I log. I control fermenter temperature using a differential controller hooked to an RTD in a thermowell. The controller reads out PV in response to a request from the computer. The software I use takes as input the desired inter reading interval. I usually use 30 seconds as an interval for the fermentors as the beer often stays in them for as much as a month. For mashing (I use the same model controller) I log at a much higher rate (every couple of seconds).

StoneHands said:

The OP hopefully has some equipment ideas, if so I'm all ears.

Click to expand...

Yes, I have all the stuff to do it at my disposal. Belongs to my employer but is stuff that's assigned to me. The more I use it the better I become with it so it's not only of interest here but fosters more effective use of the tools which benefits our customer, myself, and my employer. If a work need comes up I'd just have to pull the plugs and try again some other time.

What I'd be using is a 16-channel (type K, FWIW) input module that converts the thermocouple signals to CAN messages, then outputs the CAN messages which would be recorded on a CAN logger. The input module also exports a CAN database that has all the necessary stuff to ID the messages, decode the signals, and properly scale them from the log file. The CAN database and the CAN log file then go into a program that allows me work with the data in a vast number of ways. In this case it would be primarily a graphical representation of the thermocouple signals over time.

We're talking roughly $7,000 worth of gear between the two boxes and analysis software plus it's not inherently geared for control, just monitoring/data logging. So, yeah, not a likely investment for even an advanced home brewer.

That's one sweet setup! Makes my dmm with it's RS232 interface look like a child's toy.

For monitoring fermentation temperature where good accuracy is desired thermocouples are generally not considered accurate enough. RTD's are usually used for this reason.

Correlation is important in the experiment you are trying to do. If all three thermal couples read the same value in the same environment then you should be good even if you are off a couple of degrees from the absolute temperature.

If all he wants to see is correlation then they don't even have to read the same in the same environment as biases are removed in computation of the correlation coefficient:

E{(T1 - E{T1})*(T2 - E{T2}}/normalization

but I don't think that's what he's after. I believe he wants to know the temperature of his fermenting beer.

In either case if you go into a brewery you won't find thermocouples on the fermentors - you'll find RTD's. As far as implementation most controllers allow for RTD, themocouple or thermistor input so in terms of the configuration of a system it doesn't much matter which is used and you might as well go for accuracy.

This stuff is normally used where RTD level accuracy isn't necessary. That said, I typically order the supplies in the special limits grade which specs to (in round numbers) +/-1C in the temperature range this would be. We have a calibrated thermocouple generator so I can validate pretty high accuracy in the instrumentation. What I don't have is a calibrated hot cell and cold cell, but usually I'll use an ice bath and boiling water as simple red-face checks for the complete system.

Bear in mind I'm not an Engineer, but my general function is to investigate and provide information so that we can decide if Engineering or other involvement is necessary.

Whoaru99, For "not an engineer" you really know your stuff.

Temperature monitoring is very interesting to me and also part of my day-job for testing refinery and power generation equipment performance. Most test thermocouples or RTD's are special tolerance with calibration records at a select temperatures. Also, most test instruments are bought for a single project and never touched again.

I keep a 2-channel data logger on my fermentation chamber (2-relay controller) with type-K thermocouples - example graph below. I usually log data every 3 minutes.

Wow DSmith,

That's a very interesting plot. It looks like you can control the fermentation temperature to one or two degrees of stability. That's very nice. By "two relay system" do you mean that you have heating and cooling elements involved in the control? I would love to look at the raw data that made the plot. I can PM you my email address if you could send the file.

Yeah, that's basically what I'd be doing sans control from the wort temp/feedback. My control would be non-feedback/indirect through whatever is going on with the house's HVAC system.

Initially I was thinking sample at something between 1 and 5 minute intervals but in hindsight the slowest sample rate for my thermocouple module is 1Hz.