Understanding the Reservoir System: Water Level Control and Lever Mechanism

[omka]

Hello,
Excuse my bad english.Could someone help me please to understand how does this systems works?
the role of the system is to maintain the constant water level in the reservoir.I don't understand how the opening and closing of the valve is controlled. How the lever system works. in the description of the system they talk about a comparison made between the height h (t) in the reservoir and the reference height hr fixed by the length of the bar connecting the float to the lever system.
Thank you and sorry for my english

 

[DrClaude]

What is represented at the surface of the water is probably a float. It goes up or down with the level of the water, and the vertical arm pivots. On the other side of that pivot arm is a mechanism that controls the valve, e.g., when it gets pushed down the valve reduces the flow.

 

[tygerdawg]

Further, this is a schematic of a simple mechanical feedback system.
The different pivot arm lengths l1 & l2 provide a means of setting the "gain" of the system (l1/l2).

• A short l1 (with long l2) will give small valve opening adjustments for large changes in tank level: low gain, slow response.
• A long l1 (with short l2) will give large valve opening adjustments for small changes in tank level: high gain, fast response.

 

[omka]

Thank you DrClaude and tygerdawg for your answer. L1 and L2 will change depending on the position of the float in the water by mean depending of the level of the water in the reservoir? how can we fix the input command (hr) in this system.

 

[Nidum]

Just set the float to a height where the valve gets shut off when the water level is correct . The box with the black blob is an adjustment mechanism for the height of the float .

The ratio of L2 to L1 sets the mechanical gain and the response sensitivity of the control system .

 

[CWatters]

L1 and L2 will change depending on the position of the float in the water by mean depending of the level of the water in the reservoir?

No. L1 and L2 would be set by the system designer...

The different pivot arm lengths l1 & l2 provide a means of setting the "gain" of the system (l1/l2).

So the equation for qi(t) would be something like... qi(t) = (L₁/L₂) * (h(t) - h(r))

Although not exactly that because there is a time delay due to the distance d.

 

[omka]

thank you Nidum and Cwatters for your help its start to be clear. Last question please why mechanic gain is L1/L2 and not L2/L1

 

[CWatters]

Edited: It's the way levers work. If you make L1 longer the valve moves further. If you make L2 longer the valve moves less for the same change in water depth.

 

[omka]

Edited: It's the way levers work. If you make L1 longer the valve moves further. If you make L2 longer the valve moves less for the same change in water depth.

THanks you CWatters

 

[omka]

hello
I still have question if you can help me please.
if in consider the Archimede's Law of the lever ---> i will have F1.L1=F2.L2
L1 and L2 are the pivot arm of the lever
F1 is the force applied to the valve
F2 is the force applied by the charge
What is the charge in this case. It is the float+water on the thank ?
Thank you again for you help

 

[omka]

No. L1 and L2 would be set by the system designer...
So the equation for qi(t) would be something like... qi(t) = (L₁/L₂) * (h(t) - h(r))

Although not exactly that because there is a time delay due to the distance d.

hello CWatter
the equation means that qi(t) will be proportionnel to the difference between the desired water level in the thank "hr" and the actual water level "h(t)". is the difference (h(t)-hr) or (hr-h(t))

 

[CWatters]

If I have understood correctly the flow rate from the valve depends on the position of the control not the force on the control.

The position of the valve control will be...

=-{h(t)-h_r} * l₁/l₂

That way if the water level matches the reference level the valve control is 0 and the valve off.

 

[CWatters]

hello CWatter
the equation means that qi(t) will be proportionnel to the difference between the desired water level in the thank "hr" and the actual water level "h(t)". is the difference (h(t)-hr) or (hr-h(t))

hr-ht or -(ht-hr)

If ht is higher than hr you want the flow rate to reduce or stop.

 

[Nidum]

Just for interest :

This is not really a proper proportional control system . For most levels of water in the tank the only information coming back from the float is a basic 'too high' or 'too low' indication . There is no proportionality and the valve is either fully shut or fully open . Only for the particular situation of the water level being somewhere near the set level does the system have any semblance of proportional control .

In reality the water level in the tank is mostly being controlled by what is more nearly a very crude 'all or nothing' type control system .

This system basically only has two states : Water too high = valve off . Water too low = valve on . Any narrow band of proportional control that may exist is largely irrelevant to controlling the tank water level .

 

[Asymptotic]

In reality the water level in the tank is mostly being controlled by what is more nearly a very crude 'all or nothing' type control system .

This system basically only has two states : Water too high = valve off . Water too low = valve on . Any narrow band of proportional control that may exist is largely irrelevant to controlling the tank water level .

Another aside: Successful operation depends upon a slowly changing, disturbance-free fluid level. In a real world system, surface ripples which may be caused by vibration or other mechanisms will tend to continually trigger the valve.

This case didn't involve a make-up valve, but is illustrative of a real-world float switch "gotcha". It was a hot oil temperature control unit that used a float switch for 'low reservoir level' sensing. When low level was sensed, the pump motor was turned off - so far, so good - but (as the result of an ill-conceived control system) it was allowed to turn on again immediately after the level switch indicated a 'not low' condition.

A slow leak developed in the system. Both reservoir and pump were part of a package unit, and as oil level gradually fell to the trip point it began (per operator reports) to intermittently pulse the pump motor on and off. I've never seen anything like it before or since, but what must have happened is oil level dropped to the point where vibration-induced wave action in conjunction with level disturbances from pump cycling was high enough to continually switch the float (reed switch based, 0.100 inch actuation travel). If you've never seen or heard a 2 pole pump motor switching on and off at nearly 60 Hz line frequency you haven't lived. By the time I reached the scene it had become a frighteningly loud door buzzer. :)

 

[CWatters]

Just for interest :

This is not really a proper proportional control system . For most levels of water in the tank the only information coming back from the float is a basic 'too high' or 'too low' indication . There is no proportionality and the valve is either fully shut or fully open . Only for the particular situation of the water level being somewhere near the set level does the system have any semblance of proportional control .

In reality the water level in the tank is mostly being controlled by what is more nearly a very crude 'all or nothing' type control system .

This system basically only has two states : Water too high = valve off . Water too low = valve on . Any narrow band of proportional control that may exist is largely irrelevant to controlling the tank water level .

Can you explain why you say this because it appears proportional to me.

 

[Nidum]

In a real version of this system the valve would only have a small range of movement between fully closed and fully open .

When the water level in the tank is somewhere near the set (ie required) level the float goes up or down with the water level and effectively sends a proportional control signal to the valve .

Where the water level in the tank is significantly higher or lower than the set level the float can only signal the valve to be fully shut or fully open and the valve will remain in one of those conditions regardless of float movement until the tank water level returns to being somewhere near the set level again .

The refill and level control system in a normal design of toilet cistern works essentially this way .