the condensation has a moderately high resistance - around 100K, iirc. I mis-remembered the circuit - one probe (the lowest) goes to 5V and the other two form the high sides of 2 voltage dividers. the lower resistors are 200K to gnd. With no water, the comparators in the PIC see 0V, with water, they see around 3.2V. The reference input to the comparators is about 1.6V. this will work for water resistance up to 400K. I have had no problems on that front. Getting a pump to work in the shallow pan has been an issue - I wound up with a 12V windshield washer pump. works pretty well. the little giant submersables sucked...

 

Yes, the condensation will conduct to some degree as it willl be weakly ionic due to the gasses from the air (CO2 and suplher and nitrogen dioxides) disolved in it.

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez
And http://www.silicontronics.com, same screen name as here.

 

Well I saw the floater in the dehumidifier. Unfortunately, when the floater rises the switch extends and holds the floater in the upright position. So even if you pump the water out then the floater will still be upright. But, I think I thought of another way to detect the water level that I think will be fun

- Fairly strong magnet, I have many many neodymiums at work I can get
- PVC pipe
- small watertight bottle
- 2 magnetic reed switches. Can get these from work too and they're sealed if I remember correctly.

1. Attach magnetic reed switches to sides of PVC pipe, one high up and one low down
2. Insert magnet into sealed bottle
3. Put bottle in PVC pipe
4. Place PVC pipe vertically in water container

When the top reed switch closes, pump turns on. When bottom switch closes, pump turns off. This whole system could probably be controlled by adding a few other relays to control power... but she'd hate me if I did something stupid and filled her dorm room with water. So, I plan on still using the micro and use a current sense resistor.

I'm guessing (though I plan 100% on measuring several times to be sure I'm right) that the following is true:
1. If the pump is pumping ONLY air, current will be very low. Micro shuts off pump
2. If the pump is pumping water through the tube that slips off the counter onto the ground, current will be low (though not as low as pumping air)
3. If the pump is pumping water through an upright tube 4' high, then the current will rise as the water height rises.

So, if the micro doesn't see the current increase to a certain level, then it will shut off. Any comments?

__________________
A rectangular bear is just a polar bear after a coordinate transform. -- I dunno who.
A recent study shows that research causes cancer in rats. -- I dunno who said that one either.

 

I posted a simple 555 monostable circuit, which I used for a while. I hot glued a hollow plastic ball to the end of a metal rod. Hot glued a scrap of wood to the side of the pan, and the plastic tube from an old pen for the rod to side up and down. Cut the rod of so that it was about three inches short than the pen tube. Mounted a lever microswitch over the tube to trigger the timer.
Personally, I think you are really over complicating this thing. But then again, there is a woman involved...

 

I agree, seems overly complex with several points of failure. moving parts are going to be a problem area for any system that is expected to work long term (as in "forever"). especially home-made ones. gunk happens.

while my stainless steel sensors probably aren't 100% trouble free, they have been working just fine for quite a while. I just went and checked the resistance between 2 of the probes in the water in the catch pan and it was 121K. pretty stable over 12 months of use.

 

One thing I have discovered about washer fluid pumps... many of them must be submersed/wet at all times. This maintains the prime, provides lubrication/cooling for the pumping mechanism. They can have very short lives when run dry. They can burn out the pump/diaphram in short order due to friction.

 

yes, that is true for most of the small pumps I looked at. Run dry and die... I couldn't find a really small self priming pump. The WW pumps are low duty cycle devices as well. You need to build in a time out so that it doesn't run for an extended period of time, dry or not. It's one of the reasons I used a microcontroller but a 555 could be used to time limit the pumping action.

 

I recommend you use latching relays, the main advantage is that you don't need any semiconductors or even a low voltage power supply as you can get relays with mains coils.

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez
And http://www.silicontronics.com, same screen name as here.

 

hey noggin, umm...I have the feeling I already know ya from somehwere :>)

Keep it simple as 123, no point in re-inventing the wheel. I would suggest that you get a hold of an old tank sender unit outta the scrap yard, the kind used to indicate your fuel level in the tank of a car. Use the rod mechanism and float etc with a small paddle of some sort on the end and 4 cheap optical sensors, you can always cut/bend the rod etc to achieve the correct angle and arc. The sensors would be arranged with some discreet logic to detect failure and either shut the pump off, of empty the tank, either way, it gives you an opportunity to indicate failure of the control mechanism. You will need to be fairly inventive in order to strip it down to something useful to your application though.

Use a bistable or something equally simplistic, or a small micro if you must, to control your pump. You don't really need to pump it out as such, just prevent it overfilling, so maybe a 20-30 sec burst of the pump will be sufficient if the pump isn't rated for the duty. If the pump is capable of safely running long enough to completely empty the tank then use something state driven, much less chance of a failure or screw up.

You could always use a small micro and just read the resistance from the rheostat on the tank sender, if you can find a way to keep it as part of the mechanical setup, and just perform a table read/compare to control the pump.

Also, make sure your tray is sloped slightly and draw the water off from the deepest point.
in the case of a submersible/ flooded suction pump, you'll want this point to always have enough water in it so as not to lose the prime.

In any case, the more complex it is, the harder it will be to build, the more expensive it will end up, and the less reliable it will become.

Just my thoughts...

 

i've been using these for some years now and not one died yet.

http://www.cemeeng.com/index_en.htm

I believe the company Gorman-rupp also makes this type of pumps though these tend to be more expensive.
These pumps are self priming up to more than a meter, In my instruments they draw water from 1.25m without a problem. These also can lift that water considerably and here those cost from 15EUR.

However there is a drawback to them, running dry for a short time, a few minutes or so will not damage them although you'll emediatelly notice that it's dry, it gets really noisy.
Those work on 220V and there's a version for 110VAC. The flow is also not on a bright side, maybe 1-2l/min for this smallest one but then again do you need more?

Good luck.

 

How about a small boat type bilge pump? Or at least the factory made float/switch assembly?

 

you said it was a humidifier... right? if it was an ultrasonic humidifier, the fogger automaticly turns off if the water is low, so maybe you couls use that as a switch. or you could have a sensor to see if the fogger is activated. if i remember, i think they operate at about 1.6mhz?, if you can find a sensor that will be able to detect that frequency, it could tell you when it is off.

__________________
The 3 laws of science....
1. If it smells bad, It's chemistry.
2. If it's mushy, It's biology.
3. If it doesn't work, It's physics.
www.laserpointerforums.com A forum for disscusion of lasers and laser shows!
www.laserchat.org IRC chat for laser and electronic related stuff.

 

DEhumidifier.