Data Logging - another option

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

jebatty

Minister of Fire
Jan 1, 2008
5,796
Northern MN
I just got a couple of TEMPer1 temperature sensors, with accompanying software, and these look promising for inexpensive data logging (about $10/sensor incl s/h). These plug into the USB port and have a one meter cable to the sensor. Multiple sensors can be plugged into a USB hub, and the software will record a separate data file for each sensor. The software is reasonably configurable, such as how often to display data in the chart and how often to record data to the data file. The data file also can be sent by email to a remote computer, either automatically or manually. My graphing software easily handles the data files to produce customized charts. Other sensors and configurations are available. PCsensor

One thing I do not know is how far the cable can be extended. Extending the cable would involve cutting it (4 wires) and splicing in a longer cable section. I am going to experiment with that.

eBay is one source to get these, but there are others.

Has anyone else experimented with these?
 
For those interested -- I tried to splice in an extension wire, quite long, and got no reading from the sensor. A shorter wire might work, but I think the best is to leave well enough alone. I now have four of the TEMPer 1, all plugged into a USB hub. Since data logging is something I've had fun with, I also got an eight sensor unit. The software installed without issue, and all eight sensors are logged a minimum of once per second, settable to other periods. My laptop now logs the eight sensor unit, the four sensor USB hub, and my original four sensor Dallas one wire unit, for a total of 16 sensors, plus one additional that logs stack temperature. The data will transfer to my graphing software to produce charts "of many colors."
 
Very cool. Quick question - did the software come with the sensor or is it something available on that site you linked above? I milled around for a few minutes and didn't see any software. At these prices this seems like a pretty good candidate for Fall 2010 tinkering....thanks for sharing!
 
Software comes with the sensor. It does workable charts. I have separate graphing software that I use that is a little more elegant and flexible; uses the raw data to produce the charts. If you have done some charting/graphing with various text delimited files, it is easy; otherwise can be confusing.

As to the software, just have to remember to load the drivers first, then the software, then plug in the USB device. The stuff really does not come with very good directions, but again, if you have some computer experience it ends up being pretty intuitive. I made some early mistakes and had to roll back my system a couple of times and then do over. Not really a big deal, but resulted from lack of instructions.
 
Sounds good. Thanks again. I may have to get a few of these on order this week...
 
The USB 1 wire mini-LAN thermometer now showing up on the PCsensor website looks very interesting. They weren't being offered when I bought my 8 sensor unit or TEMPer1 units. Dallas DS18b20 sensors can be obtained for about $2 each, and then using audio stereo mini-jacks and cables (could use CAT5 or 6; or shielded audio) to connect to the adapter looks pretty easy to do. The software is elementary, but the txt data output file probably could be used by other software to do almost anything.
 
I ordered the USB 1 wire mini-LAN thermometer on the PCsensor website and it arrived yesterday. I think this will be the easiest way yet to assemble multiple sensors for data logging and charting. The drivers and software work as described on the website, and also provided is the Maxim 1 wire viewer. The provided software will log multiple sensors, perhaps up to 80 as the website states.

I got the near minimum offering, the 1-wire adapter and two sensors, although you can get the 1-wire adapter alone for $16 incl s/h. It is easy to make up your own cabling with your own DS18B20 sensor, although their cables with sensors are very reasonably priced in various package quantities with the adapter.

If this had been available two months ago, my life would have been much simpler. Fortunately, the bulk of my installed sensors are DS18B20's, and it won't be too difficult to wire in some mini-jacks to connect directly to the USB adapter. Re-work will be quite minimal.
 
The sensor wire is about 5' and terminates with an audio stereo mini plug. The 2 sensor setup that I got then has 2 sensors, an audio splitter Y plug/jack, and one 5 meter audio extension cable. Using stereo audio plugs/jacks is a simple and inexpensive way to rig up connections. 5V power is pulled from the USB port, and it is possible that with a variety of USB devices power demands could exceed what the port provides. Might consider a USB hub with external power if this is a problem.

The USB adapter itself has 2 input jacks, and so far as I can tell they operate in parallel and it makes no difference which one is used or how many sensors are placed on either line.
 
looks like $9 to buy 1 sensor with 5' wire, 3.5mm plug and metal probe containing the DS18B20 sensor. Can't quite tell how many sensors come with the various priced packages. Literature appears identical on all of them. Still don't think you could build these for much less that $9 doing it yourself. Jim does it appear that the probe end would work in a thermowell placed in my black iron piping?
 
It's a little confusing as to whether the probe is the same price as the regular sensor, if bought separately. Regardless, $9 for the regular (encapsulated) sensor, wire and plug is a good price. You can tell the # of sensors in the various packages. If the package ID ends in D#, the D is the regular sensor and the # is the number of sensors in the package. It the package ID ends in H#, the H is the probe sensor and the # is the number of probe sensors in the package. The packages with probes are more expensive than packages with regular sensors.

This afternoon I wired up a 3.5mm stereo audio plug to 8 x DS18B20 sensors previously fixed to my storage tank. I then plugged it into the adapter, and I immediately started reading and logging data from all 8 sensors. The default read and log rate is 1 read/log per second, meaning that with 8 sensors each sensor is read and logged once each 8 seconds. Both read and log rates are independent and can be set to longer times. If the read or log rate was set to 15 seconds, each sensor would be read or logged once each 120 seconds (8 x 15), or every two minutes.

Those small 3.5mm stereo audio plugs are a bit touchy to solder cable into. If a longer cable run is desired, it would be easy to start with the pre-wired sensor, cut the cable, and splice the extension into the cable rather than mess with the plug or jack. I needed to wire in a plug due to the fact that my 8 sensors already were fixed to my tank (under insulation and tank boxed in), and it would have been a bear to re-do these.

The more sensors in place, the more complicated it may become to pull out the data from the log file for any particular sensor or group of sensors. The log file is a csv file, which I found easy to convert to a txt file. I have the ability to program my database program to do any heavy lifting in pulling out particular data from txt files, and then I use a separate charting program to print charts as desired.

The charting program provided with the adapter, though, is pretty good, and charts can be printed. The charting program also allows selecting which sensors to show on the chart, so multiple charts can be printed showing different sensors.

I have a lot more experimenting to do, but the DS18B20 adapter setup is the best I have seen yet, especially for the price.
 
The charting program also allows selecting which sensors to show on the chart, so multiple charts can be printed showing different sensors.

I was wrong on this. There is some chart flexibility, but right now I don't think very much. As to logging periods, I might have been wrong also. I'm running a log right now and can report more tomorrow as to logging frequency.
 
The logging frequency appears to be exactly what the setting is. Yesterday evening I set "60" seconds as the frequency, and the data file this morning shows that every 60 seconds all 8 sensors were read and logged to the file.

After a few more test periods, I think I will have some interesting data revealing stratification in a 1000 gallon horizontal LP tank.
 
Will be good to see that 1000 gal tank data. If the output is .CSV format any database program should be able to read it easily and you could set up a spreadsheet (like Excel) to pull the data in also. Might not be as robust as a Nofossill or Arduino system but something most of us could do. Several "programs" I use at work are Microsoft Access front ends plucking data from various sources and displaying in an customized easier to view, point and click setups. I think I would still like to eventually have an cool Arduino or Nofossil type system but this would be a quick and dirty way to get up and running and the sensors could all be used down the road on the Arduino.

I said probes earlier but I meant the 1 wire sensors. I was wondering if the metal thing they are enscapsulated in woul be something that could be inserted into a thermo well directly in by piping. I was thinking it would give more accurate readings than just fastened to the pipe. Maybe it doesn't really matter that much.
 
I'm hoping all you electronic gifted computer types will go nice and slow with this project. I'd like to monitor my system temps with the pc too.
 
I was wondering if the metal thing they are enscapsulated in woul be something that could be inserted into a thermo well directly in by piping.

I think that would work. Insert the capsule, and stuff the remaining open space with conductive material. I surface mount all of my sensors with aluminum tape and/or cable ties; and then wrap with insulation.
 
I wish I had the ability to use sensors to control motor pump speeds via the pc. usb port. Some possibilites would be outdoor temperature reset, efficient storage tank charging, boiler return water temperature protection, etc.
 
That would be an elegant route to take. I'm now starting my 4th heating season with the Tarm, and also am quite satisfied with a simple plumbing system and a pretty good monitoring and temperature data logging setup. This is my first year with in-floor radiant after moving the Tarm to my new shop, and I may look into outdoor temperature reset for the future, but will wait to see how this year works without that.

Boiler return water temperature protection now is very easy with the balancing valve and digital panel meter temperature monitors on boiler supply, system return to Termovar, and Termovar return to boiler. Just a glance at the meters tells me everything is OK and the balancing valve is set correctly. Early season right now, system return is between 90-110F (from bottom of storage). At that return, and the Grundfos 15-58 set on Low, I set the balancing valve for 140-150F Termovar return to boiler, and boiler supply stays at about 160-165F. I haven't been charging the 1000 gal storage yet to more than about 130F at the temperature sensor point 12" above the bottom of the tank. That takes about a 6 hour burn and 5-7 days between burns. As it gets colder, I will charge the tank to higher temperatures, and I will have to close down the balancing valve some and may have to turn up the speed on the Grundfos to handle full boiler output and keep boiler return at around 150 and not more than 160F, at least until bottom of storage is below 160F. I can charge storage all the way to 190F top to bottom.

I also have a mixing valve and digital panel meter temperature monitors on radiant supply and radiant return. Again, at a glance I know exactly that radiant supply is OK and the mixing valve is performing properly.

Other digital panel meter temperature monitors tell me storage temperature 12" down from top, 24" down from top, and bottom. I don't need a meter to tell me top of tank, because on the high side it will be the boiler supply temperature and on the low side not lower than 12" down from top. The 12" and 24" sensors give a very good status report on storage charge status.

The final digital panel meter temperature monitor I have is on center of stack temperature. During a burn this tells at a glance whether the burn is satisfactory and without problems. No need to check the firebox or the gasification chamber. With my re-install in my new shop excellent gasification occurs in about the 330-400F stack temperature range, and I can push the boiler to higher output if I adjust the burn to the high 400F range. The meter tells the performance/output story.

I fiddle very little with any boiler, valve or pump speed controls. Everything is "set it and forget it" until I want to make a significant change in something. The meters provide the info that has eliminated the fiddling, and they immediately report the results of adjustment in anything. Almost too easy now. And the data logging tells performance over time and provides the evidence to backup most of the things I report.
 
I'd like to place an order with PCsensor.com.
I'm looking at 1 wire miniLAN thermometer (1W-D4) $34.00. I'm a bit confused with what all that comes with this kit. Will it come with everything needed to monitor 4 temperature sensors 15 feet long?
 
chuck172 said:
I'd like to place an order with PCsensor.com.
I'm looking at 1 wire miniLAN thermometer (1W-D4) $34.00. I'm a bit confused with what all that comes with this kit. Will it come with everything needed to monitor 4 temperature sensors 15 feet long?

looks like they changed their website since last week to show a picture of what the kit contains. JEBEATTY said the sensors are 5' themselves and the website indicates you get two 5 meter (15') extensions. Depending on your layout and where the PC will set this could allow you to cover over 30' of are or so.
 
I take it as 5 meters=aprox. 15 feet. I wonder how hard it would be to extend them further.
But with this kit I assume everything is included to read out 4 temperatures. Lets say bottom and top of storage tank. Supply and Return of Tarm boiler. At aprox. 15 feet from the usb port. Right?
 
It's very easy to extend. The simple way is to cut one of the supplied cables, splice in whatever extension you want, and you're done. Another way is to make up your own cable. I made one from a 50' length of 3 wire, 18g, thermostat wire. To this I hooked up 3 - DS18B20 sensors, one to measure outside temp, one to measure inside temp, and the third is pushed into a piece of pex buried in the concrete floor to measure floor temperature. All 3 are connected to one end of the 3 wire cable, the other end I soldered to a stereo audio mini plug, plugged it into the adapter, and it works just like it is supposed to, reading all 3 sensors and plotting a chart. The only "trick" is to get the right connections to the mini plug. GND is the tip, data is the middle, and VCC is the end towards the handle of the plug. Give it a try.
 
Great news Jim,
Earlier I thought you had trouble extending the cables. I have 2 sets of direct bury Cat5 with my 350' insulated pex run from house to barn. I was thinking the Cat5 should have the proper shielding to send the signal all the way to a PC in the house. Otherwise, I would have to have a second PC in the barn connected to the sensors and access it via an ethernet.

So I will be able to use 1 Cat5 line for the telephone and Internet to barn and the other Cat 5 will handle the Maxim sensors (3 wires) leaving 5 wires for some 12VDC contact closure things (storage aquastat on/off, boiler circ on/off, barn heat circ on/off, & possibly 1 more).

If you get a chance could you measure the diameter of the metal encapsulated portion covering the sensor? It looks like about 1/4" or 5/16" based on picture relative to the cable. If so it should fit into the thermowells nicely and allow for some grease for heat transfer. I was going to call the PCsensor directly but realized the number is in China.
 
I placed my order this morning. About how long will it take to get here from China?
 
The sensor is a bit less than 1/4" diameter, probably metric. Mail time was 1-2 weeks. Cat5 cable should be good. The trouble I had extending the cable was not with the 1-wire sensors but the non-1-wire sensors sold by PCsensor which I previously bought. Now I'll probably use those to make spider webs for Halloween.

Right now I'm doing a stratification test burn on my 1000 gal LP tank and will be able to provide results soon.
 
Status
Not open for further replies.