Data Logging - another option

[jebatty]

If you haven't noticed, I posted results of data logging use the 1-wire adapter and 11 sensors to measure stratification performance in a 1000 gal horizontal LP pressurized storage tank. Horizontal Tank - Stratification

The chart was not produced by the software that comes with the 1-wire adapter, although all data was collected by that software. The provided software, though, will produce a usable chart.

 

[mwk1000]

They look like a convenient way to add a couple of sensors into my tank. They are waterproof line and all an I can lower it into the tank all the way to the bottom ?

I would like to save the time in fabricating a submersible sensor and just drop these in for that money.

 

[jebatty]

Installing them in the tank likely would be better, but I don't know whether or not these would stand up to be submersed continuously in hot water. Give it a try and find out.

 

[rkusek]

I'm pretty sure water would penetrate the shrink tubing between the cable and sensor eventually. Better to assemble something out of copper or pipe with the little thermowells at various depths and insert into tank from top. I found a place that has brass 1/2" NPT x 3 1/2" thermowells for $3.50. Trying to see if i can find pipe fittings that will clear the entry hole in the tank.

 

[mwk1000]

Yes, shrink tubing does not inspire confidence. I'll just have to beak down and build something that will last inside the tank.

 

[Hunderliggur]

Can anyone recommend a 1-wire solution to 330-400F stack temperature range measurement? What parts from where? Alternative (US) sources? I am guessing it is a k-type thermocouple. What do you use for the A?D 1-wire device?

 

[kabbott]

Can anyone recommend a 1-wire solution to 330-400F stack temperature range measurement? What parts from where? Alternative (US) sources? I am guessing it is a k-type thermocouple. What do you use for the A?D 1-wire device?

I have seen circuits to use thermocouples. Some use a 1-wire battery monitor chip (I forget the number) that reads in microvolts. There are
also thermocouple amp ic's that take cold junction temperature into account and output a voltage that corresponds to temp C., that output could be sent to any 1-wire
device that reads millivolts.

here are a few links http://www.sensorsmag.com/sensors/electric-magnetic/the-1-wire-thermocouple-1086
and http://www.aag.com.mx/aagusa/index1.html

TAI8560 1-Wire Thermocouple Adapter Module from aag may do what you want.

Kris

 

[jebatty]

As to the 1-wire solution for the stack temperature, it may depend upon what your goal is. Mine was, first, to be able to visually monitor stack temp during a burn so that at a glance I could tell if performance was where it I expected it should be and, second, on occasion data log stack temperature over an entire burn to verify performance and identify any glitches that might occur.

I accomplished the visual with a K-type digital panel meter available for about $15. This is what I used to keep the Tarm in high burn for the data shown in the Horizontal Tank - Stratification post. As soon as temp started to drop into the high 300's, I knew that more wood needed to be added to keep the Tarm in high burn.

I accomplished the second data log goal with a stand-alone battery powered data logger, the Lascar EL-USB-TC, with a K-type probe into the center of the stack. This device has proved quite useful for other data logging, as it takes any K-type sensor and can be used anywhere. I've used it to do some test data logging before deciding to put in a permanent 1-wire device.

The long and short of my data logging is that I don't data log all the time. I undertook data logging to better understand my system and to tweak it to obtain improved efficiency and performance. I use data logging if I change something and want to know the effect. Even when things appear to be going well I data log, usually over a burn cycle, just to make sure that things are as they appear to be.

 

[mwk1000]

I was looking to do this myself. I found this http://www.arduino.cc/playground/Main/MAX6675Library for the MAX6675 that will work well for me but their were also some non-arduino strait USB interfaces on sparkfun.com. Just look under sensors->temp-> thermocouple (K type ) where they have a project example.

Right now i'm thinking that http://www.arduino.cc/playground/Main/RegulatedPositiveVoltageBooster is the ticket to driving a variable speed grundfos 2-10V controlled circulator based on my temps. Love this stuff.

 

[Medman]

Does anyone have a data logging setup running now that uses the arduino and more than 3 or 4 DS18B20 sensors? I have all the hardware now and am starting to assemble, but I would like to see code examples that are working in the real world.
I have the Arduino Mega board.

 

[mwk1000]

I do. I posted a full tar file in the wiki that has the whole thing last year:

Arduino sketch, Logfile parser (ttymon) for the log the sketch makes, Mysql database setup, website with charts pulled from database.

https://www.hearth.com/econtent/index.php/wiki/TankControl_Information/

 

[Medman]

Thanks, I saw that. Is there any way you could put up the sketch only? I am using Windows right now, because my PLC programming sw is Windows, so I can't access the sketch in your compressed file. Also, I am encountering errors with the DallasTemperature library when compiling. Can you point me to the version you are using?

 

[mwk1000]

https://www.hearth.com/econtent/index.php/wiki/TankControl_Information/

Just the sketch is there now

 

[Medman]

Thanks, I have it. I will try it now!

 

[Medman]

Great! Your code works very well. I will make the necessary adjustments for my own system. Thanks so much for your help - this is a great start for me.

 

[mwk1000]

Glad it works. One note - I ran into a situation where the loadsource /return numbers were not allowing the tank to start, (I really need to get sensors IN the tank.) You will see when the pump stops the temperature readings on the side with the stopped pump become useless when they are directly attached to the HX as mine are.

Example - stop the tank circulator and the Loadsource temps climb very fast because the water is just sitting in the HX on that side. They are no longer representative of the true tank bottom temps.

Long story short - my boiler was at 200 and the circulator was not running --- I quickly added an extra start override for high temps that is NOT IN the version posted on the wiki. ( I'm on version 7 now )

You may not have this problem if you have better positions or more sensors to rely on.

 

[Medman]

I am not going to be using this system for control, so that shouldn't be an issue. I have a dedicated PLC for that, but it does not have a datalog option(picked it up used from the paper mill, which shut down last year). I want instead to log boiler data vs weather data to try and track temps and consumption, heat loss, etc. I am going to interface my weather station data to the log through the arduino as well.

 

[chuck172]

I ordered and received the four sensor package from www.PCsensor.com
I can only nickname 1 sensor. The others all I keep getting is setting failed. Anyone know the trick?
Also, if I want to add more splitters and sensors, can I get them from somewhere in the States?
It took two weeks shipping from china.

 

[jebatty]

I didn't try the nickname feature of the software, but I will give it a try and see what happens. As to the shipping time, I've learned to be patient. BTW, I found another China site that has the sensors for $5.99 each incl s/h in multiple quantities, and a site for splitters for $0.99 each. For sensors search "digital thermal probe" and for the splitters search "audio Y adapter" on eBay. The sensors do not have the audio plug, so you need to wire one or splice multiples of these together to a plug.

If you use the advantage of the 1-wire and splice multiple sensors into one plug, then you don't need many splitters at all. I have 8 sensors going to one plug/jack, for example. With just 2 splitters, you have 4 jack inputs, and that's likely all I will use.

 

[jebatty]

I can only nickname 1 sensor. The others all I keep getting is setting failed. Anyone know the trick?

I was able to nickname all of my 13 sensors successfully. Here's what worked. While reading the sensors (although I think it also will work while stopped), I clicked on a sensor, specified the nickname, and did not click "OK". I clicked on the next sensor, and then the nickname for the first sensor showed. Did the same until all sensors had nicknames. I then clicked stop (might not have to do this), and closed the program. I reloaded the program, and all sensors showed with their nicknames.

 

[jebatty]

Forgot to add, in the data file the nickname for a sensor shows as the last entry before the number of the next entry. There is no "," between the nickname and the number of the next entry.

 

[chuck172]

It doesn't work for me. I tried everything. I even completely uninstalled the programs, then re-installed. For some reason , I know it doesn't make sense, it looks like you only get one shot to nickname the sensors.

 

[chuck172]

Using my digital sensors, I'm amazed how just a small throttle of the bypass/ball valve will affect the boiler return temp past the tarmovar. I also throttle the main boiler circulator pump with a motor speed control. Again a big change in the boiler supply temp.
I wonder how well the termovar loading unit accomplishes all this automatically?

 

[jebatty]

I was amazed, as you, as to how monitoring information enables a person to better understand how each part of his system actually works, and how it allows a person to tune for improved performance. Continue to share you insights.

 

[Hansson]

I agree
Logging is a god way to see how your system works.
I using this one now.
http://translate.google.se/translat...=38&osCsid=g9h2eovapvhc1p6unb9n2324r2&act=url