water reservoir level sensor

[trend]

as sucky as that sounds, you sound very interesting because of the knowledge you have gained.,.. One thing, since you are the expert, I woudl like to know your opinion:
I would rather know more exact numbers (but this isn't CRUCIAL.. but helpful), should I be looking more for a bobber like a car gas tank has? or are they less dependable.. or should i just stick tons of reed switchs? I mean, those 5 ml values in the previous post would do.. but just wondering

thanks soo much! Lee

 

[UberBlue]

as sucky as that sounds, you sound very interesting because of the knowledge you have gained.,.. One thing, since you are the expert, I woudl like to know your opinion:
I would rather know more exact numbers (but this isn't CRUCIAL.. but helpful), should I be looking more for a bobber like a car gas tank has? or are they less dependable.. or should i just stick tons of reed switchs? I mean, those 5 ml values in the previous post would do.. but just wondering

thanks soo much! Lee

The total volume of the res could be figure, then reed switches placed at the appropriate levels that corespond with those 5 ml levels. As far as what temperature those levels would correspond to in a monitoring program, I have no idea, but it would be obvious once thing are put together.

The problem with the gas bobber is there is no way to interface with the computer without resorting to complicated circuitry.

 

[UberBlue]

crap, i need 0ml, 14ml, 19ml, 25ml, 40ml and maybe another one to make sure i don't overfill the cylinder

i forgot to mention 19ml
thanks-Lee

So six points total including the overflow.

4.3K is a common resistor value and shouldn't be hard to find.

Using the 4.3's, six of them would give a total range of 25.8K ohm, which is close to the 26k range measured.

So now res dimentions need to settled upon, volume figured, and where at the 0ml, 14ml, 19ml, 25ml,and 40ml points are.

 

[trend]

where can i get the magnetic reed switches?
and should i mount them vertically or horizontically on the tube (what wouldbe more accurate?)


and also, PVC is the best material to use to make a cylinder?

 

[UberBlue]

MAJOR hitch!

It dawned on me in the shower that the only thing the temp sensor in the power supply is used for is to regulate the PS fan speed. None of the temp information is relayed to the motherboard, only fan speed is.

 

[UberBlue]

really? i have "intel active monitor" and it monitors temp and fan speed.. looking at it right now

Power supply temp? Or speed?

 

[trend]

ok.. well what i was thinking was.. hook up the device to the motherboard fan power supplies.. because don't they tell how fast the fan is running?

here is a link

how is that data gathered?

 

[UberBlue]

Are you sure that one of those "system zones" is the power supply temp?

If it is, I have no idea how the MOBO gets ahold of it.

 

[trend]

don't know... is there any add on hardware that i can use to hook this device up to? digital doc5 or such?

 

[trend]

hmm... well i have 3 pin fan connectors on my mobo, couldn't i use those? because do'nt they monitor rpms ?

 

[UberBlue]

hmm... well i have 3 pin fan connectors on my mobo, couldn't i use those? because do'nt they monitor rpms ?

The 3-pin fan connector consist of +, -, and rpm sense. Unless you've got another cable pluged into your mother board besides the 24-pin ATX, ATX 12V, and the 3-pin fan plug, I'm betting that neither of those "system zones" is the power supply temp.

Don't frett. It should be fairly easy to spoof the rpm sense on a 3-pin fan connector. Just use that instead.

Give me awhile, I'm rolling around some ideas.

 

[UberBlue]

don't know... is there any add on hardware that i can use to hook this device up to? digital doc5 or such?

DigiDoc 5 would work, but what about your computer interface?

 

[trend]

yeah, scratch digital doc5...


yeah on you helping me ! because i am stumped...

does the 3rd pin allow the mobo to tell the fan how fast to spin.. or visa verse?
I would think it would be omni directional and the mobo would send more power to the fan to tell it to speed up

 

[UberBlue]

yeah, scratch digital doc5...


yeah on you helping me ! because i am stumped...

does the 3rd pin allow the mobo to tell the fan how fast to spin.. or visa verse?
I would think it would be omni directional and the mobo would send more power to the fan to tell it to speed up

The rpm sense pin only tells the motherboard how fast the fan is spinning.

As I understand it, twice per fan rotation, the sense lead is dropped to ground. The motherboard counts how many times it's dropped to ground, divides by two, and presto! Fan rpm. At least I *think* that's how it works.

 

[clocker2]

A level indicator out of a car fuel tank maybe?

They are normaly a long arm with a float on one end. I dont knw how big you res is but they could be shortend to fit.

Typically, fuel level sending units would be too large to fit a normal res.
Working on a similar principle however, and far smaller, would be the low brake fluid sensor from a master cylinder reservoir.
Most newish model cars have this and the added benefit is that it would operate on 12v.

 

[UberBlue]

I think I figured it out.

Instead of complicated circuitry to spoof fan RPM, just use a fan. Use the resistor network in the TLI to control fan speed between 5 - 7V.

It's not the most elegant solution, but it's the simplest.

And there's another little detail I thought of. What about when the float magnet is intermediary between two reed switches? The circuit will be open.

 

[UberBlue]

Typically, fuel level sending units would be too large to fit a normal res.
Working on a similar principle however, and far smaller, would be the low brake fluid sensor from a master cylinder reservoir.
Most newish model cars have this and the added benefit is that it would operate on 12v.

There's all kind of sensors that could be used. It's the interface with the computer that's proving to be the B****.

 

[squeakygeek]

I just found this thread, and I'm hopping aboard.

The RPM sensor for the fans is pulsed. This is out of the question... and c'mon, we can come up with a better solution than varying the speed of the fan!

Please explain exactly what you are doing, and include all details. How many computers? Why such the fluctuation? Why such a small res? Why does a computer need to know the water level? Don't leave anything out.

Your best bets are either a few of the reed sensors with a magnetic floaty-thing, or some of these: http://www.allelectronics.com/cgi-bin/category.cgi?category=search&item=MK-108&type=store

After reviewing the link, I conclude that the magnetic option would be more cost-effective, but I figured I would show the link anyway just so you guys are aware.

As for the resistors.... Why on earth do you want to turn a perfectly good digital signal into an analog signal, and come up with a way to interface that analog signal with a digital computer?

What I think you want to do is interface with the parallel port, and do a bit of programming. The parallel port has 8 data bits, and as long as you want 8 or less sensors, you don't even need any circuitry! With a bit of circuitry you could have up to 256 sensors, but you probably won't need this. Just hook up each sensor to a data bit, and the rest is software!

I'm not sure what you need to interface to a computer for, but I think you could easily rig up a Digital Doc 5 to do a basic fill operation, but the above idea is much cooler.

 

[UberBlue]

The RPM sensor for the fans is pulsed. This is out of the question... and c'mon, we can come up with a better solution than varying the speed of the fan!

A/D converters are spendy or complicated to build. That's the beauty of the fan. Use the analog output of the TLI (basicly a variable resistor) to control fan speed. The fan then sends it pulsed RPM output to the MOBO

What I think you want to do is interface with the parallel port, and do a bit of programming. The parallel port has 8 data bits, and as long as you want 8 or less sensors, you don't even need any circuitry! With a bit of circuitry you could have up to 256 sensors, but you probably won't need this. Just hook up each sensor to a data bit, and the rest is software!

Elaborate on this some more.

We still need an A/D converter. And the software is an issue unless somebody writes a program for him.

Another benefit of the ghetto fan A/D converter. plugs right in and uses existing software.

As for the resistors.... Why on earth do you want to turn a perfectly good digital signal into an analog signal, and come up with a way to interface that analog signal with a digital computer?

We're staring with an analog signal and thought we had an analog interface, but that fell through. Hence the fan idea, quick and easy A/D converter. If noise is an issue, cut the blades off.

 

[UberBlue]

I remember seeing something VERY similar to this a long time ago, only with photo diodes. For the life of me I can't remember where I saw it though.