A piezo element can deliver high voltage spikes that a capacitor in parallel should reduce and enable you to adjust sensitivity with your potentiometer. You'll have to experiment to find the correct value - an oscilloscope would help you visualize what is happening.

The photos you posted show the parts that probably work the display, conditioning for sensors and what looks like two relative humidity sensors. If you can find the specification sheets of the actual display components, the RH sensors (or better yet - the input required and output specs on the board that conditions these signals) it would be straightforward for an Arduino to use them to control the display and the steam generator. I didn't see any temperature sensors (they may be mechanical thermostats that use a liquid or gas) or the relays that turn on and off the compressor, heater, or steam generator. The last photo shows the controller and it does have relays but their size indicates that they turn on small loads, like a larger relay.

Are there much larger relays located near the compressor, heater or steam generator that have larger wires?

Are there probes inside the proofer volume that transmit temperature?

Have you looked into using as many of the parts of the existing controller as you can? It probably already has relays hooked up to the components they control, it has switches or other input devices and a display that you may be able to program an arduino to interpret and control what's already there. Since all these interfaces are integrated into your proofer, replacing the dead controller with an arduino would be the most elegant and probably least expensive way to approach fixing it.

A balloon cannot take very much pressure inside but pneumatic tires have pressures in the range you mention. There are battery operated tire pumps that could be controlled to make the pressure go higher, with a solenoid valve to let out air to decrease the pressure. An Arduino can activate either of these depending on the pressure setpoint you want to maintain (measured with a pressure transducer).

If your system is always above ambient conditions, all you need is a small heater to raise the temperature to your desired temperature. Filament type lightbulbs can readily heat small areas to the temperature you mentioned.

To create a defined density of gas in a confined insulated volume is straightforward:
1. Monitor vacuum (or pressure) with pressure and temperature transducers that are rated for the ranges you are interested in;
2. Control a vacuum (or pressure) source such as a gas sampling pump based on the output of the vacuum/pressure transducer;
3. Heat/cool the volume with thermoelectric (Peltier) array(s) that heat or cool depending on the direction of current flow - using the temperature transducer for feedback.
A small DC powered pump (2 watts = 2 l/min.) can reduce the pressure in a 10 liter volume in a short time and can be shut off (or slowed down to maintain the desired pressure if there are leaks or the heated volume expands) - the built in check valve usually maintains the vacuum quite well.
The thermoelectric array has to be sized to the task and has to have a heat sink on the outside of the insulated space to reject heat (or provide it if the volume is to be kept above ambient).

Thanks everyone! It's now working great!

Thank you for your response Nishant.

I only have one USB cord and when I procure another, I'll try what you suggest.

I've been  using an UNO with a DIP processor for a few weeks - going very well. Acquired a new UNO, SMP edition, that blinks the "L" LED but will not talk with through COM 3 that works well with the other. When first plugged in, Win-XP recognized the new peripheral and I clicked on "install drivers" and Win-XP now doesn't flag when the SMP UNO gets plugged in.

I tried all the COM baud rates and have tried the procedures that work with the other UNO: IDE simply can't find COM3 when the SMP UNO is plugged into it, with its "L" LED blinking. Restarting the IDE & SMP UNO over and over that works with the other UNO hasn't worked for this one. Under the IDE Tools: Board there is only one Arduino UNO. Am I missing something?

Suggestions?

An ant, travelling at a maximum speed of 6 inches per second, would take 42 milliseconds to shadow the second of a pair of sensors 0.25" apart. So if we look at the two sensors every 30 milliseconds, we should be able to catch which sensor was the last sensor that gave a signal. If the last sensor is outboard of the ant hill entrance we can increment an “ant going out” register, if inboard, an “ant going in” register. To gather a bit more information, it would be nice to quantify how long a series of ant/sensor interactions lasts: are ants faster going in, going out, when it’s cool or hot, when it’s early or late?

What I think will work:

•   Create interim registers: for “elapsed time”, “elapsed time going in”, “elapsed time going out”, “ant was here time” “counts going in”, “counts going out” and “ant was here count”;
•   Look at two sensors;
•   Start “elapsed time” stopwatch when either sensor changes state;
•   Do nothing until neither sensor sees an ant;
•   If ant was last seen by the outboard sensor, increment “going out” count and transfer elapsed time to “elapsed time going out”; do the same for  going in for the inverse; and, if both sensors see IR again at the same time – add the elapsed time to “ant was here time” and increment “ant was here count”;
•   Every five minutes write the totals for the above to the SD card with date/time stamp;
•   To reduce spreadsheet work: before writing to the SD card it would be nice to divide elapsed time totals by counts to get average times of ant/sensor interface instead of just totals: 134 ants went out averaging 52 milliseconds makes more sense than it took 6,970 milliseconds for 134 ants to go out ;
•   Clear all registers;
•   Start looking at the two sensors for next iteration.

Once ants test this code in the lab, and maybe after a season of field work, it would extend logging time on a battery if we turn on the IR source and logging shield only when needed.

There seem to be many ways to accomplish the above in the Arduino language references and playground but most of these descriptions leave me with many questions: if…else, switch case, etc.? And then there is timing and options that use libraries. Although I’ve made small changes to working code that others have written, my eyes glaze over when the slate is blank.  It would probably be best to determine that the ant counting code works by watching progress over a serial stream, then adding the elapsed timing code before adding the logging code.

Any suggestions on how to begin?

cyberteque, that is a very good suggestion: although I'm staying away from sensors that require visible light (because these setups would be harder to hide if deployed where they may be noticed at night if someone sees a blinking light - strobed to minimize power). I'm currently trying a linear array of three IR phototransistors tied to digital pins. We'll have to see if ant shadows give strong signals as my Popsicle stick.

Any idea of how to get an Arduino to tabulate "ant going out" or "ant coming in" after looking at a series of  1-0-0, 0-1-0, 0-0-1 that may have a dozen repetitions of any, or all, of these patterns? To say nothing of throwing in a few 1-1-0, 1-0-1. and 0-1-1's. I haven't a clue! We may have to live with "ant activity" between 0-0-0's and any of the above. The actual signals are the inverse of the above since the detectors pull the output high when they "see" the IR source.

For higher power gearmotors and different wheels:

http://www.surpluscenter.com/sort.asp?catname=&byKeyword=yes&search=gearmotors
http://www.surpluscenter.com/sort.asp?catname=&byKeyword=yes&search=wheels

Spambake, To see if the  code works I'll swing a Popsicle stick through the sensor tunnel (upside down) in both directions at varying speeds, then try and confuse it by stopping in the middle and going back, eclipsing both at the same time with a piece of cardboard etc. Success will be measured in getting good counts and times recorded every minute, along with averaged temperature, humidity, light level and voltage reference with a time stamp for a few hours on the SD card. Then I'll send the setup to the ant kennel so their caregivers can hassle the poor ants by feeding them on the other side of the tunnel, while recording activity on video. I think it will make YouTube only if the ant carry protest placards.

I don't see any need for "sampling" ants - they'll be hassled enough without having to worry about being spirited off somewhere.

TND: the setup you described is exactly what I had in mind.
gardner, Graynomad: we do have the luxury of having harvester ant colonies living in a lab that, once I get code working on a bench here in NY, I can send for testing there (in CA).

Thank you all for your suggestions and observations.

In trying to follow ladyada's code: http://www.ladyada.net/learn/sensors/ir.html I realized that it may be okay to simply start counting microseconds when the IR beam to either sensor is interrupted, and stop when the other is occluded. Add one count to a register depending on order, and even a running total time in two other registers. Then - each minute divide respective totals by the number of counts in that direction and log both counts and speed. (If any over the speed limit, issue tickets? Or flash a warning to slow down?) One may also ignore or record "bad" count totals for the cases when both sensors are occluded at the same time or either sensor gets blocked without ever involving the other. I hope legs, antennae and body segments don't give us too much trouble. I'm minimizing the current to the IR source to just above the detection threshold both to prolong battery life and make it easier for an ant to interrupt the beam. I started playing with strobing the IR source ala ladyada's example for the same reasons. Her sensor example that uses a side-looking IR detector that is primarily sensitive to a 38 khz IR source and delivers digital signals also is a much better configuration for incorporating in a tunnel than the T 1-3/4 LEDs and source (LEDTECH Electro IRLED: Jameco part number 106526, IR source - part number 373001) I'm using that are analog.

New information: ant speed during the summer on an overcast morning is around 4 inches per second (derived from a video recording, not a radar gun). It was noted that under hot sun they seem to move much faster.

Because working with ants in the lab can be readily recorded on video, the Arduino data taken there can be correlated with actual activity. And some number of hours of recording in the field can be likewise visually corroborated to insure that each setup is working.

Now that I have ladyada's code as a guide (thank you very much! ladyada) I'll report any success with a setup that seems to work okay. It'll take a few days for me to find the time.

Skyjumper,

The Arduino-based counters are supplements to weeks of actual daybreak to sunset personal observations, which has been the primary mode of taking data  at the site for many decades. But folks need R&R and you can't carefully watch more than a few colonies at a time. By watching and taking notes with a few instrumented colonies and a few pristine, of matched age, one can get an idea of the effect of partially blocking an entrance. Larger, and older, colonies have multiple entrances/exits so that they can be monitored before placing a sensor stick to see if traffic changes (the sensors are buried in a broken branch that looks natural so that passersby won't notice them and want to pick one up to examine it when the scientists are not there - the Arduino and battery will be buried a foot or two away and the cable covered - we chose IR sensors so that they don't attract attention at night).

Very good advice, Arduino Rocks - I'll try it .

What I'm trying to prove at the moment is a circuit similar to the Honeybee Hive Traffic Counter:  http://www.discovercircuits.com/PDF-FILES/beectr.pdf  that uses the Arduino instead of flip-flops and counting module. Once I have the Arduino record a series of signals that represent an ant going first in one direction, then the other, I'll post them for suggestions on how to reduce the swings in analog voltages of the two side-by-side sensors viewing an IR transmitter to a single "in" or "out" count. If we don't succeed, we can always just record "an ant went by", like the bee counter. It would be nice to make an Arduino smarter than that.