Petting a pig, testing data loggers and a question...

[zeedubbya]

So I wanted to share an experience and ask a question.  I decided to buy 3 of the cheap Bluetooth Data Loggers and put them into action.  The Boveda and the Sensorpush have been delayed in shipping (for some unknown reason), but the Inkbird came in and after 24 hours of calibration at two separate points it seems to be pretty accurate.  The question I have for all the scientists out there.  Is it possible this thing is really able to read to 1/100th of a dF or percentage humidity?  If I had a system precise enough to calibrate it down to 1/100th, or 1/10th even, would it accurately measure this?  I find this a very hard idea to swallow.  

Secondly it’s well past time I gave @PigFish his due for the setup he made for me.  This system is unreal.  We cobbled together what he referred to as a “quirky” system and it’s working Upmann amazingly well.  I can only imagine what his newest generation system is able to achieve.  The data log for this one is just downright boring. Which I am sure is a good thing! The temperature is not controlled (future plans), but the level to which this system maintains my preferred 63% rH is stunning.  The only spot in the entire humidor where it fluctuates much at all is right next to the humidifier.  Over the past few days I’ve had ambient humidity above and below the set point and it stays pegged within .5% of 63 ALL THE TIME.  And this is in a wood box!  It’s truly unreal.  Below is just one days example, but they’ve all looked like this.  The humdity spike is where I moved it around.  

Also a thanks to @planetary for setting me up a spreadsheet to keep a longer term log and see the results side by side.  I will try to share these here over the coming months to see which data logger seems to be working the best.  

 

 

[Fuzz]

I highly doubt a small, cheap bluetooth device would be able to be that accurate. My thermocouple is accurate to ±1℃±5%, and that is a relatively cheap thermocouple (Paid AUD$200. I don't need it super accurate). There are other thermocouples that can read down to ±0.3°C± 0.04%, but you are paying a hefty price in comparison (AUD$500). 

[zeedubbya]

12 minutes ago, Fuzz said:

I highly doubt a small, cheap bluetooth device would be able to be that accurate. My thermocouple is accurate to ±1℃±5%, and that is a relatively cheap thermocouple (Paid AUD$200. I don't need it super accurate). There are other thermocouples that can read down to ±0.3°C± 0.04%, but you are paying a hefty price in comparison (AUD$500). 

I agree with you completely Fuzz—I can’t imagine it being that accurate.  I wonder if it’s just a made up number?  In the calibration it did let me calibrate it by 1/10ths however?  Here’s the specs listed for it.  

Temperature Range: -40~60°C (Inner sensor) 
-40~100°C (External probe) 
Temperature Accuracy: Typical: 0.3°C / 0.5°F, 
Max: 0.5°C / 0.9°F (Inner sensor) 
1°C (External probe) 
Humidity Range: 0~99%RH 
Humidity Accuracy(25°C/ 77°F, 20%~80%RH): 
Typical: 3%RH, Max: 4.5%RH 
Humidity Accuracy(25°C/ 77°F, 0%~100%RH): 
Typical: 4.5%RH, Max: 7.5%RH

[MrTANKEm]

Wow!!! That’s pretty awesome. Using a data logger really shows how well your system is doing. In your case flipping awesome!!!

I just spent the last hour trying to get excel to chart my log, with only a small amount of success. Can I ask someone to send me an excel sheet with the charting format already set up please?

The calibration of your logger maybe be off (the actual value, example 65%h) but it should show the change spot on (example: so if it says it went up 1.5%, it really went up exactly 1.5%. Verses it say 1.5% and it really goes up 2%).  Either way your logger should be close enough not to worry about for how you are using it. Plus using your taste buds to get your right RH is the most important here. I am not a temperature & humidity calibration expert, but I have had some training in the calibration area. 

I am looking forward to seeing your updates on your humidor. 

[Fuzz]

Maybe this Inkbird thing is using a thermistor? A thermistor is cheap, more accurate than a thermocouple or resistance temperature detector, and responds faster to changes.

[zeedubbya]

Just grabbed another data set from it and it changed the values displayed on its own....now it’s definitely more interesting....hmmmm...what a funny little device...

 

 

 

[Fugu]

20 hours ago, zeedubbya said:

So I wanted to share an experience and ask a question.  I decided to buy 3 of the cheap Bluetooth Data Loggers and put them into action.  The Boveda and the Sensorpush have been delayed in shipping (for some unknown reason), but the Inkbird came in and after 24 hours of calibration at two separate points it seems to be pretty accurate.  The question I have for all the scientists out there.  Is it possible this thing is really able to read to 1/100th of a dF or percentage humidity?  If I had a system precise enough to calibrate it down to 1/100th, or 1/10th even, would it accurately measure this?  I find this a very hard idea to swallow.

Difference between "accuracy" and "resolution". Gents you are about to put accuracy, repeatability and precision, relative and absolute errors, long-term stability ("drift"), and as such total measurement uncertainty, all in one pot. There is, regrettably, way more to it than just a simple comparison measurement. Starts with a required specification for ones calibration reference normals. Metrology is a science-discipline of its own....

Short answer - while you can always get any device calibrated by e.g. doing a simple x/y-comparison regression for a selected reference source, and numbers read to the last digit, this doesn't tell you much about the true system's measurement uncertainty. This figure, the absolute uncertainty, which is determined via a thorough analysis of errors (only of any value if accompanied by a statistical statement of the confidence interval covered) is the only authoritative parameter that would provide information on "accuracy". (mind you, 'accuracy' actually being slang - in metrology one uses the term measurement uncertainty for the absolute system error).

But - for our humidor purposes, a regular calibration and a certain redundancy of "cheapo" devices will just do fine for the task. The reading of your gauges might even resolve a fraction of a % and a hundredth of a degree, but don't expect that to reflect the actual measurement accuracy. Also to note, an instrument's accuracy is not just controlled by the sensing element in use (whether PRT, NTC, PTC, thermocouple etc. for temp), but also by the subsequent signal processing, the quality of the bridge and parts (Tk), wiring, compensatory circuitry etc. used therein. The final calibration rating depends on the actual calibration setup.

In practice, I found - if regularly and decently calibrated - you may expect up to about 0.2 K uncertainty in temperature in cheap devices (once aged, temp is actually pretty long-term stable even in cheapos) and up to 2% (perhaps better) in rH if you don't go to the extremes. This is absolute values, the relative errors of short-term changes (like displayed in your plots, Zane) can be smaller. And this is just fine for the purpose. What counts the most in our measurement task is long-term stability rather than the absolute figures. For our cigars it is pretty irrelevant whether they are stored at 70 °F or 70.69 °F. But the 70 °F of today, should preferably still be 70 °F tomorrow. Precision should be good (repeatability) and long-term drift should be low (but can be controlled by routine recalibs), absolute measurement uncertainty is less important...

And just an example to put instrumentation effort in perspective: For temperature measurement using a precision lab-thermometer with external high-class PRT probe, one has to expect in the range of 10k $$ to be able to achieve ~5-mK measurement uncertainty ("accuracy") over a certain, limited range. For 1-mK(1/1000 Kelvin)-equipment well north of that. And the actual outcome will depend a lot on its proper use!

Have fun with your new toy(s)!
 

[PigFish]

Mate, I am sorry I did not get to your email... It was not until I saw this post that i remembered you sent it. I need to start logging on during the week. I simply don't know where about 12 hours a day disappear to. It is called a job, I suppose!

Goo, really hit it. Terms ware being confused here. Your instrument company's "expected range" was given you when they say +- 3%. And frankly this is not much of a guarantee of anything.

I have an easy algorithm for everyone. Smoking is the final arbiter. There is no other way to do it. Instruments are guide, a means to establish a common thread between humidor and smoker. Yet it is taste and feel that correlate the data from instruments to what proper storage really means.

Data logging is not about accuracy, but more about precision and repeatability. Controlled humidors are about making the unstable, stable! The 'out of bounds,' in-bounds!

Data logging is the diagnosis, the test and in my case the 'tell' about when a humidor works or does not work. The word 'work' is subjective based on my experience and where I set the bar. Everyone will have their own definition and certainly not take it and set it where I set it.

So, here is my 'indoor' Aristocrat in June. It follows the temp of my house. If my house is stable, it is stable. The ambient then is part of the humidor 'system.' How well your system works, is define on how well you control your ambient. Stable ambient, stable humidor. Unstable ambient, unstable humidor. As a reminder, percent moisture content 'IS' cigar condition and represents the cutting edge of taste (to me).

I have matched the scales so that I am not accused of cheating.

This is my Gen13 controlled humidor 'outside.' This means it is in my shop and does not depend on me controlling the ambient. The weather controls the ambient. It is in the 'real world.' For giggles I pasted up the temp in the real world on the chart. The two loggers are time logged so they align. As you can see, I don't keep the outside at 76 top that I keep the inside!!!

My humidor may make 1000 climate adjustments for every 1 made in the Aristo. It has to... That is what it takes to convert 95F to 100F temp to 70F temp and +-1rH stability... where the mean is my set point.

My Aristo, and the cigars in it, would be trashed and warped in 48 hours in my shop! Bringing my humidor from 'outside' to inside... and it works even better!

Bear in mind that these are just maintenance and overall performance data logs. If I am really testing to see if something works, I sample the humidor once a second. A lot happens in a controlled humidor in 5 or 10 seconds, and you can miss it if you wait minutes to pick up a data point. My maintenance logs are about 14 second samples.

What is really going on in a precision controlled humidor by the second? Glad you asked.

... did I mention at 95F ambient!

You cannot possibly understand humidor performance without data logging. Once you get used to looking at them you can actually tell what your humidor is doing and diagnose why it is doing it.

There is not a problem I have not seen in building controlled humidors. Those problems can only really be analyzed by using a data logger!

Cheers! -Piggy

 

[zeedubbya]

Thanks for reply @PigFish, no worries on not responding to the email—I know you’re very busy.  I have determined this little data logger is good enough for me.  It’s showing me what I want to see.  As you have said before, I have settled.  It shows me a consistency.  A consistency I am ok with.  Better than I expected even.   I’m know it’s far far from a perfect device, but it works for my situation.  These points are recorded every 10 seconds, but it only gets 30,000 data points so that is another limitation of it.  But again, I’ve settled.  

My cigars are smoking better than ever, and I am very pleased with the precision of this system.  Next step is to get it in a sealed environment (large wine cooler) and work on temperature controls. Which I intend to do relatively soon.  Here’s the latest data from it if anyone is interested.  

And by the way...data logs used to bore me, but now I love to look at them.  

Z

 

 

[PigFish]

Nothing to frown at my friend. That is a very consistent humidor. My old controller?

[zeedubbya]

6 hours ago, PigFish said:

Nothing to frown at my friend. That is a very consistent humidor. My old controller?

Yes of course it’s your old controller!!  And it just keeps getting better!  Most recent log I have...I am beyond impressed.  I bow down to your excellence in humidity management.  It’s truly remarkable my friend.  I have it set on 63 and I am convinced the data logger is just off by about that much, but damn look at that consistency!  3 day spread of less than 1% in a wood box!

 

 

[zeedubbya]

What’s even more impressive is how much the ambient has been fluctuating.  Been 72% and down to 40% ambient rH.  So it’s been keeping steady regardless of ambient humidity.  It’s very impressive.  I did seal the box up pretty well, but always figured I would be more of a victim to the ambient.  When I move it into the big wine cooler I anticipate it will be even better still.  

[PigFish]

3 hours ago, zeedubbya said:

What’s even more impressive is how much the ambient has been fluctuating.  Been 72% and down to 40% ambient rH.  So it’s been keeping steady regardless of ambient humidity.  It’s very impressive.  I did seal the box up pretty well, but always figured I would be more of a victim to the ambient.  When I move it into the big wine cooler I anticipate it will be even better still.  

Dry is always better unless you have a sealed refrigerated unit. A wooden box will begin to take on water you cannot get rid of eventually if the ambient stays rH+. That problem is a reason for a large part of my business. Solid surface humidors need to dehydrate, or they won't work for long...

Cheers! -R

[zeedubbya]

6 hours ago, PigFish said:

Dry is always better unless you have a sealed refrigerated unit. A wooden box will begin to take on water you cannot get rid of eventually if the ambient stays rH+. That problem is a reason for a large part of my business. Solid surface humidors need to dehydrate, or they won't work for long...

Cheers! -R

Don’t forget I am running a bead silo in the humidor as well.  It seems to be doing what it’s supoosed to.  What makes me think this is when the humidifier has run dry the rH inside will drop down well below the set point even in the presence of a higher ambient humidity.  My only concern with it is I haven’t seen a long term rH+ environment yet, but I doubt I ever will.  The humidor is in my basement, but the air conditioning system in my house keeps the humidity down in the summer and it automatically stays down in the winter.  In fact I removed the bead silo in the winter when the ambient was steadily below the set point.  I think only if I add active cooling to the system (in the solid surface wine cooler) will I have the need for a better dehydration system.  Even then, I doubt I will see a big effect because the ambient temp is pretty steady in my basement.   We can cross that bridge when we come to it.  Right now it seems to be working pretty well.  

[PigFish]

Based on your ambient, I cannot see a reason to change. It really is about pragmatic cigar keeping and the heuristic humidor. I say don't change what works.

The silo is the only really effective way to use a desiccant bead buffer (MHO). They really 'force' the desiccant to do its job. Industrial refining proves this out.

Good job Zany!

-R

[joeypots]

Bead silo? I'd love to know more about this way to utilize beads.

[zeedubbya]

 

31 minutes ago, joeypots said:

Bead silo? I'd love to know more about this way to utilize beads.

I’ll let @PigFish explain more.  I’m not an expert.  I think it’s a very simple device, but if you aren’t using a proper dessicant with proper controls I could see it potentially creating more problems.  

[PigFish]

A bead silo is a patent pending invention of mine. The 'silo' simply forces air over a bead pack. You see beads become inert (relatively speaking) if they are not exposed to a continuous stream of air to be conditioned.

Imagine you set up an air conditioner in your room and have no fan over it circulate the air. You get a cool spot and don't utilize the potential of the cooling unit itself because you don't bring warm air to it, nor distribute cool air from it. Beaded desiccants work the same way. You pile them, isolate them, they largely become immersed in a surrounding that they create. That surrounding does little for your humidor unless you 'redistribute it' and get the beads working on a dissimilar space constantly. Beads work on the basis of 'dissimilar' space. There needs to be a delta to their conditioning to make them actually work (best).

I will post some pics if you are interested.

[joeypots]

14 minutes ago, PigFish said:

A bead silo is a patent pending invention of mine. The 'silo' simply forces air over a bead pack. You see beads become inert (relatively speaking) if they are not exposed to a continuous stream of air to be conditioned.

Imagine you set up an air conditioner in your room and have no fan over it circulate the air. You get a cool spot and don't utilize the potential of the cooling unit itself because you don't bring warm air to it, nor distribute cool air from it. Beaded desiccants work the same way. You pile them, isolate them, they largely become immersed in a surrounding that they create. That surrounding does little for your humidor unless you 'redistribute it' and get the beads working on a dissimilar space constantly. Beads work on the basis of 'dissimilar' space. There needs to be a delta to their conditioning to make them actually work (best).

I will post some pics if you are interested.

I'd like that very much, Ray. Thanks. Does a small fan in the humidor help beads do their job?