EcoRenovator

EcoRenovator (https://ecorenovator.org/forum/index.php)
-   Tools (https://ecorenovator.org/forum/forumdisplay.php?f=29)
-   -   Data logger recommendations (https://ecorenovator.org/forum/showthread.php?t=4923)

Geo NR Gee 10-02-16 04:51 PM

Data logger recommendations
 
What data logger is recommended for monitoring temperatures in geothermal boreholes and also in solar flat plate collectors? My plan is to have both monitored. Maybe a data logger is not what I am looking for?

jeff5may 10-03-16 01:10 AM

AC Hacker has had luck with these units:

MultiLogger

He includes a short description of them in this post:

http://ecorenovator.org/forum/geothe...html#post32670

The tiny board can be rigged as a standalone, battery-powered temperature logger or as a slave to another master, transferring data and obtaining power from a USB (or FTDI) cable. It can log over 100 OneWire DS18B20 temperature sensors at once. For under $50, it's an awesome deal.

There are lots of other plug-and-play temperature loggers available in many different configurations. Whatever flavor of sensor you like can be used, waterproof or not, that can transmit data over wifi or bluetooth radio channels or store data on whatever media you like. You just have to figure out a few things and find a compatible board. Most other devices are more expensive than the card cited above, once the capability to log 3 or 4 sensors (or more) is reached.

Geo NR Gee 10-03-16 10:56 AM

2 Attachment(s)
Thank you for the info Jeff5may. Looking at the links, it says one wire, but really this is what I came across... http://ecorenovator.org/forum/attach...1&d=1475509986 or this http://ecorenovator.org/forum/attach...1&d=1475509986

So if I understand this correctly, there is a ground wire, power and then the data wire? They say one wire because the data goes on one wire and each sensor is unique?

I have a 250' spool of underground solid core sprinkler system wire in my stash and wonder if that would work, or does it have to be stranded?

Since this is going to be direct buried, I could go with the sensor that are premade or make my own with pigtails and solder them up. I suppose the wires and sensor must be protected from moisture and the elements. Would enclosing them in epoxy and then attaching them to the pipe suffice?

jeff5may 10-03-16 11:14 PM

The 1wire interface is named that because it only needs one wire to interface with the microcomputer. It actually needs two wires connected to operate: the one wire, and ground. In parasitic mode, the sensors grab power from the one wire (named DQ) while there is no data traveling on the line. The easiest way to save cable is to wire a string of sensors on a twisted pair. The sensors all have unique addresses burnt into them during manufacturing, so telling them apart is easy for the microcomputer. All the DQ pins connect to the one wire, and the power and ground wires are all grounded.

http://www.tweaking4all.com/wp-conte...site-power.jpg

When running lots of sensors on one line, a bit of care must be taken to make sure the sensors don't all draw power at the same time. The most common time this happens is when they take a temperature reading. Rather than tell them all to work at once, they are called out in sequence.

To avoid this redundant process and possibly time consuming situation, you can either run a separate power wire to feed the sensors or rig a hard pullup transistor to the DQ line. The sensors only draw around 1.5 milliamps each, so even if there are lots of sensors on the line, all drawing power at once, the raw power draw is not significant enough to stress the interconnect wiring.

http://www.tweaking4all.com/wp-conte...rmal-power.jpg

These schematics are not completely accurate. The sensors can operate on as low as 3 VDC reliably. However, when running lots of sensors in parasitic power mode, it is desirable to have a little extra voltage "cushion" so the sensors don't starve the data line and corrupt it. Running a smaller pullup resistor or rigging a "hard pullup" transistor helps tremendously here.

When using the Dallas onewire library, it has different commands to use that correspond to parasite power or hard power modes. The parasite power mode commands use the PULLUP function built into the microcontroller to assist the hardwired pullup resistor on the data line pin. The hard power commands do not pull the line high by way of the microcontroller.

More in-depth info can be found here:
https://www.maximintegrated.com/en/a...dex.mvp/id/148

jeff5may 10-03-16 11:33 PM

I have done the sensor thing both ways. If you buy a bulk amount of sensors, the hermetic sealed ones are pretty cheap. Then again, the TO-92 bugs are much cheaper and fit perfectly inside a piece of 1/4" copper or stainless tube. The ones I built were potted up with either super black rtv or 5 minute epoxy. Both types held up well.

Geo NR Gee 10-04-16 01:15 AM

Thanks for clearing that up for me. However, what is a TO-92 Bug?


The intended use for the temp sensors and data logger is for a 200' deep borehole with a 1" geothermal ground loop and with the temperature sensors attached to it down the borehole. I plan on pumping in heated water (from flat plate collectors, thank you AC) through the pipe to test the different depths for the optimal heat sink storage. Im trying to determine at what depths there maybe loss from a water bearing area. Most wells in the area are from 60' to 120' deep. If I can predict the depth at which the water layer would be, I could stay above or below that to place the heat.

jeff5may 10-04-16 09:06 AM

TO-92 is a generic package standard.
http://www.secosgmbh.com:8080/packag...ions/TO-92.jpg

The 3 most common packages with pinouts:
http://www.tweaking4all.com/wp-conte...0-packages.png

AC_Hacker 10-04-16 11:54 PM

Mostly straight on the 1-Wire thing.

As you said, at minimum, you need Ground, and you need Data. It can be done, but the specifications are very strict, and if not adhered to, bad data will result.

It is SO MUCH EASIER to go with Ground, Data, Power. The layout is way more flexible, and tolerant of noise.

They say it MUST be done in a serial fashion, or if you have branches, badness will happen.

Rebel that I am, I did one branch (which would be fatal with 2-wire) and the data logging was very stable.

At 3 branches, I encountered a bit of trouble on start-up.

Each sensor has a unique number so the data logger can tell where data is coming from. If you don't know which sensor is giving you particular readings, you can run the system and warm up individual sensors, and watch the results.

I think that 1-Wire sensors are the greatest thing since Sex Lube.

-AC

P.S.: I am still very pleased with the Multilogger. It will work right away, as is. The source code is available on site. If you have the skills you can do other interesting things with it as well. There are better data loggers, but there is a substantial price jump.

Daox 10-05-16 09:23 AM

Does anyone have any example of how you program using the 1-wire sensors? I assume there is a library add on to use them?

AC_Hacker 10-05-16 02:49 PM

Quote:

Originally Posted by Daox (Post 51974)
Does anyone have any example of how you program using the 1-wire sensors? I assume there is a library add on to use them?

Sure.

Google: "Arduino 1-wire library"

Google: "Arduino 1-wire tutorial"

Then go to youtube and search: "Arduino 1-wire"


-AC


All times are GMT -5. The time now is 11:21 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Ad Management by RedTyger