# Level sensor query

Discussion in 'Electrical Engineering' started by BIGEYE, Sep 1, 2005.

1. ### BIGEYEGuest

We have an application to measure the level of diesel (DERV) in a bunded
storage tank. The height of the tank is 1.83 metres. When the level of the
tank reaches 1.83 metres, the tank volume is 79000 litres.
Measurement of level will be by a submersible pressure transducer. As these
are supplied calibrated in metres H2O, how do I convert the equivalent
height of the diesel to give a volume of 79000 litres.
TIA

2. ### Guest

Divide the height in H2O by the specifioc gravity of the liquid in the
tank. If you assume diesel has a S/G of .84 <typical> a meter on the
transducer would represent 1.22 meters of fuel.

3. ### Scott SeidmanGuest

Height of water is a standard unit of pressure. I suspect it has nothing
to do with the sensor being meant for use in water.

4. ### Ignacio Simón YarzaGuest

Maybe you should use anti-explosive and antiflammable components. The diesel
shouldn't vary much it density with temperature, so the gfretwell reasoning
is precise:

p=ro*g*h, being ro= density, g=gravity constant and h=height.
Note that just dividing by diesel density is only valid in case you are
considering density of water to be 1kg/dm^3. If you were to use other
density units such as pounds/ft^^3.
Best regards.
Ignacio Simón Yarza
Mech&electronics and control eng.

6. ### John GGuest

Yes Height of Water is a standard of pressure but the important point
is -- Is a water sensor suitable to use in the explosive / corrosive
environment of diesel oil?

7. ### Jerry AvinsGuest

Is the volume proportional to height? Many fuel tanks have semicircular
bottoms and tops. Submerging the pressure transducer to the bottom may
be a problem. A good way is to read the pressure in the outlet pipe,
provided that the pressure drop due to flow can be neglected.

Jerry

8. ### Scott SeidmanGuest

Again, nobody said it was a water sensor. It is a pressure transducer.
"Water" only has to do with the calibration of the sensor. Of course the
sensor you use needs to be appropriate for the environment, but the fact
that the calibration is in units of height of water says absolutely nothing
about what the suitable environment is.

9. ### Scott SeidmanGuest

The pressure in the outlet pipe will be proportional to the height of the
fuel. Physics says your solution suffers exactly the problem you're trying
to solve.

I suspect the effects of rounded tops and bottoms will be largely
negligible, except at the almost empty and almost full levels.

If you need an exact calculation, you can use a pressure sensor to get
height, and use a lookup table to match the volume, or you can weigh the
tank.

10. ### Jerry AvinsGuest

What problem is that?
The diesel-fuel tanks for my emergency generators are about five feet
long horizontally, with substantially flat ends. The tanks are about 2.5
feet thick, with semicircular tops and bottoms being and flat
midsections also about 2.5 feet high, for a total height of about five
feet. The volume is the sum of a cylinder and a box, each 2.5 x 5. The
rounded parts are not a negligible part of the whole.

2000 gallons is the allowable limit for indoor fuel storage, so the
tanks are not completely filled. (They hold about 2100 gallons.) The
rest of the reserve is buried outdoors, with heaters to allow it to flow
in cold weather. The indoor tank allows time for the outdoor fuel to
warm, and provides gravity feed to the turbines (similar to jet
engines). We measure fuel on hand, both indoors and out, with a dip
stick and conversion table.

Jerry

11. ### daestromGuest

Well, I assume the transducer is designed for this hazardous environment.

To convert from meters H2O to another fluid, simply divide the reading by
the specific gravity of the fluid. Since diesel is less dense than water, a
full tank will not exert as much pressure on the transducer and the signal
will indicate a height less than 1.83 meters. But when you take the
transducer reading and divide it by the specific gravity of diesel fuel, you
correct for this density difference.

daestrom

12. ### Scott SeidmanGuest

ANY pressure sensor, whether sitting on the bottom of the tank as the OP
described, or at the outlet pipe, will provide an output proportional to
the height of liquid in the tank. The outlet at the tank will give you the
same reading, plus an offset for any difference in height, as the bottom of
the tank sensor.

13. ### Scott SeidmanGuest

Ooops, I got it. You're addressing the problem of getting the sensor in
the tank. I just thought that would be done through a sealed bung or
something.

Been a long week.

14. ### Jerry AvinsGuest

One of us misunderstood something.

It is easier to connect the sensor to a tee in the outlet line than to
put it into the tank.

If there is a pressure drop in the line due to flow, the reading will be
low. That can be avoided with a large diameter line between the tank and
the tee, or a separate line for the gauge in which there is no flow.

I understand the physics of pressure head and frictional losses, but I
don't know which problem you alluded to.

Jerry

15. ### Jerry AvinsGuest

...

I think that's the hard way, but in case someone wants to do it, how can
the weight/volume relation be nonlinear?

Jerry

16. ### Jerry AvinsGuest

"Level" is often used loosely, as in "What is your level of
uncertainty?" I would bet the price of a beer that the OP cares about
how much fuel there is, not about physical height.

Weight is linear with amount, and therefore a useful (though a bit
difficult to measure). If head is wanted, then pressure is the better on
both counts -- ease of measurement, and direct linearity.
If height above the floor is wanted, then pressure is linear (with an
offset that depends on the position of the gauge).
There are many options. The OP asked for practical help.

...

Jerry

17. ### BIGEYEGuest

Thanks for the info guys, but I never said the tank was non-linear. The tank
level is linear, and the answer I was seeking was given in an early reply.