Engineer/Programmer for Prototype temp/RH logging

Engineer/Programmer Needed for Prototype Temperature/RH Data Logging&Display

Objective: To provide cost effective real time display and logging of temperature and relative humidity in multiple vivaria as well as to log this data for other display purposes.

Project Requirements:

Basic:

  • A combined temperature and humidity probe that has a very small footprint, is protected from excess moisture, and has a long (10ft+) cable from the probe itself to the controller
  • Controller displayed readings of temperature and humidity based on probe number
  • Computer displayed readings of temperature and humidity based on probe number as well as the ability to log the data and make calculations from it (e.x. 24 hour high and low temp/RH, daily averages, weekly averages, etc)
    -The ability to connect 4 or more probes to the controller

Advanced:

  • The ability to turn on and off a device (e.x. a fan) based on probe readings
  • Based on temperature readings, have an audible alarm/tweet/e-mail notification/SMS message sent
  • The ability to access data via the internet (e.x. while on vacation or at work)

Pay: TBD


The only comparable system out there is the HerpKeeper (from digitalaquatics )which is way overcomplicated and expensive in my opinion. Additional temp/rh probes are $80 each. Also, I don't feel that this product supplies what the hobby wants.

There are many dart frog hobbyists that would use such a system (that I am requesting). Some hobbyists have entire rooms of their houses filled with dart frog tanks and have spent thousands and thousands of dollars on their frogs.

Some of your requirements may be difficult to meet.

A combined temperature and humidity probe that has a very small footprint, is protected from excess moisture, and has a long (10ft+) cable from the probe itself to the controller

Why not put the controller nearer to the probe? Long cables require additional processing/hardware.

  • Controller displayed readings of temperature and humidity based on probe number

How many probes do you figure?

  • Computer displayed readings of temperature and humidity based on probe number as well as the ability to log the data and make calculations from it (e.x. 24 hour high and low temp/RH, daily averages, weekly averages, etc)

Using the logged data is something that a spreadsheet on a PC is better suited for.

  • The ability to turn on and off a device (e.x. a fan) based on probe readings

One device per probe? Multiple devices per probe? Multiple probes per controller? Soon, it gets to be a nightmare configuring what to do when for what probe.

The only comparable system out there is the HerpKeeper (from digitalaquatics )which is way overcomplicated and expensive in my opinion.

What price point for the device do you have in mind? How is the data to be logged? SD card per controller? XBee on the controller sending data to a central location? Via ethernet on a server somewhere?

and have spent thousands and thousands of dollars on their frogs.

But a cheap system of monitoring the environment is needed?

A combined temperature and humidity probe that has a very small footprint, is protected from excess moisture, and has a long (10ft+) cable from the probe itself to the controller

Why not put the controller nearer to the probe? Long cables require additional processing/hardware.

It was my thinking that one controller would service multiple probes. Usually the tanks to be probed are set up on bakers racks (like metal bookshelves). This way one controller could service either one shelf of multiple tanks, or the entire rack. Often times there may be as many as 6 smaller tanks on one shelf, having longer probes would be useful for reaching from the top of the rack to the bottom to create a sampling of temperatures for that one rack/shelf.

Quote:

  • Controller displayed readings of temperature and humidity based on probe number

How many probes do you figure?

At least 4

Quote:

  • Computer displayed readings of temperature and humidity based on probe number as well as the ability to log the data and make calculations from it (e.x. 24 hour high and low temp/RH, daily averages, weekly averages, etc)

Using the logged data is something that a spreadsheet on a PC is better suited for.

I wanted a PC to be a major component of this project.

Quote:

  • The ability to turn on and off a device (e.x. a fan) based on probe readings

One device per probe? Multiple devices per probe? Multiple probes per controller? Soon, it gets to be a nightmare configuring what to do when for what probe.

I am definitely out of my area of expertise. The extent of programming I have had was on a TI series calculator.
Example: You have Probes 1-4 connected to controller A. Probe 2's temperature sensor hits 81 F, this triggers the fan assigned to controller A to run until Probe 2's reading is 78.
Generally if one tank(probe) reaches an undesirable temperature the rest are soon to follow. For simplicity's sake it shouldn't be a problem to just turn any and all fan's on.

Quote:
The only comparable system out there is the HerpKeeper (from digitalaquatics )which is way overcomplicated and expensive in my opinion.

What price point for the device do you have in mind? How is the data to be logged? SD card per controller? XBee on the controller sending data to a central location? Via ethernet on a server somewhere?

Around $100(?) for a Controller to PC, via ethernet (or some other hardline). Probes - maybe $60 each? Data should be logged by the PC. XBee (wireless right?) depending on cost, initially hard line

Quote:
and have spent thousands and thousands of dollars on their frogs.

But a cheap system of monitoring the environment is needed?

I don't mean 'cheap' when I say cost effective. Cost effective in that the temperature sampling across a wide range of tanks with an extra $10 to $20 off means more tanks can be sampled with a comparable price.

Examples (rack system and a very large collection room of over 300 frogs).

Maybe I'm the only idiot out there who likes to actually see a rough picture of things, but here goes:

Sounds like a job for a SHT15 - http://www.sparkfun.com/datasheets/Sensors/SHT1x_datasheet.pdf - in combination with a softserial library .

One Arduino can sample 4 sensors - once per 10 seconds or so - ; add an ethernet to post all the sensor data to a PHP page on a PC which stores it in a MySQL database or a logfile per sensor per day or so.. You can use this data to make nice graphs

Furthermore add 4 outputlines for solid state relays to switch PC-fan's for fresh air - they need a separate 12 Volt. (the automatic cooling is kept locally this way, independant of the logging PC..

A warning system can be built in the PC if the sensors are not logged, if temp or humid is too high or too low.

Please consider at least a master temp sensor that monitors the room temperature too.

my 2 cents,
Rob


SparkFun Humidity and Temperature Sensor Breakout - SHT15 - SEN-08257 - SparkFun Electronics - SHT15 - $38 if 10+
http://www.sparkfun.com/products/9950 - Arduino UNO - $27 if 10+
Arduino Ethernet Shield - DEV-09026 - SparkFun Electronics - ethernetshield - $42 if 10+
DC Brushless Fan - 80x80x15mm (5V) - COM-09649 - SparkFun Electronics - fan - $6 if 10+

So ex cables and power supplies ~70 + 4 * 50 = $270 for monitoring four tanks Adding cables (especially ethernet) etc will add extra bucks so in the end it must be makeable for $100 per tank for the hardware

My questions were more for getting you to think about the requirements, than to criticize anything. Writing a good requirements document is critical to successfully developing a project.

As Rob's price list and wiring comments shows, ethernet is not really going to be practical for a room full of tanks. A XBee shield and XBee at each tank may be all that is required. Each XBee can read sensor data (temp/humidity) and send that data wirelessly. The PC could be running a program that gets data periodically from each XBee, and tells the XBee to turn on or off a fan. There really doesn't need to be an Arduino at each tank.

Putting a XBee at each tank means that the temp/humidity sensor lines can be kept short, eliminating the need to deal with long lines.

Of course, as a project for you, there is one set of requirements. If you want to make and sell monitoring equipment/software, you may have different requirements.

Just some things to think about. Good luck.

Just curious, what is the temperature range your tanks are at, max, min, and what resolution are you looking to achieve.

For temperature alone, I've done a project with a few students using thermochromic liquid crystal sheets and a digital camera. FYI, a thermochromic liquid crystal sheet changes color at different temperature so you can attach a patch on every tank and just use one camera to get everything in one shot and keep grabbing frames. All depends on how accurately you want and the range of temp.

First of all thanks very much to everyone who replied!

robtillaart

Excellent information!

One Arduino can sample 4 sensors - once per 10 seconds or so - ; add an ethernet to post all the sensor data to a PHP page on a PC which stores it in a MySQL database or a logfile per sensor per day or so.. You can use this data to make nice graphs

So... one Arduino UNO can monitor 4 of the SHT15's and transmit the data wirelessly (thanks PaulS) via an XBee (of some sort) to a computer? Will the computer end of the transmission require another Arduino UNO, power supply (not through USB?), and XBee device(s)?

Is there a way for a solid state relay to operate something with higher voltage (regular household current) such as a: http://www.amazon.com/gp/product/B000WEIJ7K/ref=pd_lpo_k2_dp_sr_1?pf_rd_p=486539851&pf_rd_s=lpo-top-stripe-1&pf_rd_t=201&pf_rd_i=B000BKGOM2&pf_rd_m=ATVPDKIKX0DER&pf_rd_r=0C34ZQGDHZT2991M5TTC

PaulS

My questions were more for getting you to think about the requirements, than to criticize anything. Writing a good requirements document is critical to successfully developing a project.

They were good questions. I didn't mean to sound defensive. Sometimes it's hard for me to communicate clearly what I have in mind to others.

As Rob's price list and wiring comments shows.....really doesn't need to be an Arduino at each tank.

I guess I just do not really have that firm of a grasp of how all this stuff works. All I need are an SHT15, XBee, power supply, and sensor wiring/casing at the tank, and with appropriate receiving equiptment (?? another XBee?) at the computer?

My assumptions that I can't possibly do this on my own are being very strongly confirmed! :-X

My intention is not to sell a system like this just to make money. If I could somehow manage to put the components together I would sell at near cost to other hobbyists. I think people would be interested in the basic function of temp/RH logging.

liudr

Nominal temperature ranges are usually between 70 F - 78 F during the day along with a drop in temperature to around 62-65 F at night. As for resolution, to the nearest tenth. I'm sure people would get their jollies from as accurate a reading as possible (hundredth).

The experiences I've had with thermochromic liquid crystal sheets is that it can be hard to differentiate between the reading as the temperature of the object it is mounted on (i.e. glass) or the ambient temperature (which is what I need)

Anyone want to make one for me? :cry:

I would really like to learn about how to put something like this together, but I would be basically starting at square one zero. I have no experience soldering, reading wiring diagrams, deciphering very complex data sheets, programming in whatever language the Arduino uses, etc.

I don't mean to sound childish, I guess I can just be impatient.

So... one Arduino UNO can monitor 4 of the SHT15's and transmit the data wirelessly (thanks PaulS) via an XBee (of some sort) to a computer?

Yes.

Will the computer end of the transmission require another Arduino UNO, power supply (not through USB?), and XBee device(s)?

The PC end could consist of an XBee and a USB explorer (or there is a dongle version).

Is there a way for a solid state relay to operate something with higher voltage

That is the function of a relay, solid state or conventional.

All I need are an SHT15, XBee, power supply, and sensor wiring/casing at the tank, and with appropriate receiving equiptment (?? another XBee?) at the computer?

That is one way of doing it. It has advantages in that the sensor is close to the sender.

It has disadvantages in that the XBee has a smaller number of digital and analog pins, compared to an Arduino with XBee, and the XBee can't make decisions. The Arduino would be able to handle more inputs and make decisions, rather than the computer.

Anyone want to make one for me?

I'd be willing (PM me if interested). I have XBees and Arduinos and all the shields. The only things I don't have are the temperature/humidity sensor and solid state relay to control the fans/heaters.

o... one Arduino UNO can monitor 4 of the SHT15's

Yes, but if you don't mind plenty of cables, one MEGA - http://arduino.cc/en/Main/ArduinoBoardMega2560 - has 70 I/O lines so it can handle approx 20 temphumid sensors and fans. It sells for ~$65 ...

I like the XBEE architecture proposal as it keeps the room clean of cables. But is it part of the requirements?

If you do XBEE I think you want one XBEE per tank, that has one sensor + one fan.

XBEE - $30 (including some board thingie)
SHT - $38
FAN - $6
Power supply - $10

$85 - so almost equal in price but no wires and maximal modular. So the price we can say the price will be approx $100 per tank no matter what architecture. So which architecture to choose depends on the other requirements of the system than price

So Steven, can you sum up the requirements for us (single sentence per requirement please)

Here are some as a starter, some may apply some not up to you Steven.

  • the system must monitor temperature per tank

  • the system must monitor humidity per tank

  • the system must activate a fan at XX degrees F/C per tank

  • the system must deactivate a fan at XX degrees F/C per tank

  • the system must have settings per tank

  • the system must have a manual override for all fans [ON/OFF]

  • the system must sample temperature every 10 seconds

  • the system must sample humidity every 60 seconds

  • the system must apply a warning when temperature is outside the range [XX..YY]

  • the system must apply a warning when humidity is outside the range [XX..YY]

  • the system must have a local feedback about its working condition (might be a blinking LED (without delay :slight_smile: )

  • the system may not be noisy at night (not explicit enough)

  • the system must be save for the frogs (not explicit enough)
    ==> - the system must work on 5Volts

  • etc ...

Here's a thought for a probe/controller board that goes in each tank and reports to a central base station. Prices are for single quantities:

1 x SHT11 (cheaper and slightly less accurate SHT15): $25
1 x SF1 (protective cover for SHT series): $5
1 x Atmega32: $4
1 x AT86RF230 (2.4 GHz 802.15.4 tranceiver): $4

This, along with a power supply and all the random ancillary stuff should fit on a very small board. Add a couple of transistor outputs to drive relays and you're good to go on most of your requirements, and for a lower retail price than a probe along from the Herp Keeper folks. Those ICs are rated for operation up to 85 deg C, which is more than enough temperature headroom. The center hub can be any 802.15.4 device that can easily be plugged into a PC, along with an LCD and a few buttons. I'm working on a shield-compatible Arduino clone with an onboard 802.15.4 radio, the Zigduino (www.logos-electro.com/zigduino) that would work well with a datalogger shield and an LCD shield like the Freetronics LCD shield (16x2 LCD Shield Kit | Freetronics) to fulfill your display requirements.

PaulS

Thanks for the offer. I will definitely take you up on the offer once I get a system in mind that would work for what I need/want. I will PM you soon.

robtillaart

I love this idea! It's pretty much exactly what I had in mind.

The lack of cables is not a requirement at all, only a convenience. Probe to controller connection being wireless is not essential whatsoever. Controller to PC connection being wireless would be nice.

More on this a little later in the post.

rocketgeek

I like the idea of a probe to wireless central base.


Main Requirements:

  • the system must monitor temperature per tank
  • the system must monitor humidity per tank
  • the system must have settings per tank
  • the system must sample temperature every 10 seconds
  • the system must sample humidity every 60 seconds
  • the system must apply a warning when temperature is outside the range [XX..YY]
  • the system must apply a warning when humidity is outside the range [XX..YY]
  • the system must have a local feedback about its working condition (local feedback on control side, much like indication lights on an ethernet connection)
  • the probe must be able to sample data accurately while its electrical components are protected from a high humidity environment (80% RH+)
  • the system (minus PC & any future fan expansion) must be less than 30db loud
  • the system must be easily expandable
    ==> - little difficulty when adding addt'l probes and controllers
  • the system must have a graphical PC interface for customization of data representation and overview of output data
  • the system's data must be easily accessible remotely by persons of little to no computer savvy
    ==> - i.e. hourly tweets or e-mails when function is activated
  • probe and probe connection to controller must be easily installed into tanks that are already inhabited
    ==> - i.e. small probe and probe connector/wire

Since I am a very visual person, here are some more visual aides to go along with requirements and further explanation.

Probe

  • the sensor must be able to take readings and be protected from excess moisture while being in a housing with the smallest footprint feasible.
  • the probe's wire connection to the controller must be as thin as possible
    ==> the midway connector is only needed if the controller's input socket (from probe) is of an unacceptable size
  • it is not necessary for the probe to be wireless
    ===( Any ideas of max probe wire length? )

Controller

  • the input socket from probe to controller must be uniform
    ==> - this way addt'l probes can be added/removed easily and without any tools
    ==> - status indication lights (red/green) for a go/no go on sensor

--Things that would be nice--

  • wireless communication between controller and PC
  • controller mounted small LCD display with 2 button scroll through sensor information

PC Interface

  • wireless/hardline connection from controller
  • remote accessibility of data
    ==> - tweet/sms/email
  • remote repetition of alarm signal
    ==> - audible/tweet/sms/email
  • ability to customize output display
    ==> - custom tank display names, choose which information to display, C or F (maybe K for s&g's)
  • the ability to save a log of sensor readings
  • must be easy to operate for people without extensive computer skills
    ==> - maybe a basic and advanced mode?

It could be a lot fancier than this, but sometimes simple is better.


Well that's it for now.

10,000 thank you's from me yet again!

It occurred to me that if the probe board doesn't have to drive mechanical relays, it should be possible to make it completely wire free by putting a solar panel and a battery/ultracap on each probe board. It will then siphon off otherwise wasted light from your habitat lamp(s). It would necessarily make the probe larger and a touch more expensive, but would make installation stupidly fast and easy.

Thanks to everyone who helped out. A prototype is in the works. Hopefully should be done in 3 weeks or so.