Mixing different pressures of air using valves

[Guilty Spark]

Hey guys!

I've always been curious what would happen if I used air valves and regulators to mix different pressures of air together?

For reference I spliced a few diagrams into one here:

And I made a picture of the setup I'm imagining with some nice labels so we all know what part to refer to!

Assuming the valves are all open, I'm guessing you get 100psi out correct? You can't possibly have more than what you put in?


My main problem is combinations of valves- ie: what happens to pressure "G" if only valves "11" and "12" are open?
Yielding a supply of 40psi and 60psi./

Assume room temperature air, fixed pipe size.. (1/4") I guess..


Anyways, what do you think?

 

[mfb]

What do those big numbers in the diagram mean?
In general, air pressure somewhere in a system depends on the whole system. If you close the gap at "G" and provide some supply of 60 psi pressure, the pressure can rise up to 60 psi, if air does not flow out somewhere.

 

[Guilty Spark]

The black numbers? Those are the regulators in the first picture.. set to different PSI.

I don't really want to close it off, as the whole point of this would be to use 3 regulators to get 6 different pressures or something.. know what I mean?

Let's assume the output is going to an air tool, a rotary air piston vented to atmosphere upon press of a button..


But just for arguments sake what would happen if you mixed 40psi with 60? Do you end up with 50?

Maybe I'm not considering how those regulators function either.. hmm./

 

[russ_watters]

If the pipes are big enough that there isn't much loss and the 60 psi regulator can handle the load, the 40 psi regulator will just close completely.

 

[Guilty Spark]

If the pipes are big enough that there isn't much loss and the 80 psi regulator can handle the load, the 40 and 60 psi regulators will just close completely.

I think I know what you mean... having a hard time picturing it though.


Looking at the regulator's guts, it seems to me that a lower PSI setting makes it more sensitive to pressure drops. What with the bigger space as you unscrew the adjuster?

So isn't that backwards? Wouldn't the 40PSI supply all it could first?

 

[rcgldr]

Wouldn't the 40PSI supply all it could first?

It shouldn't if the pressure on the output side is at 80 PSI.

 

[russ_watters]

I made an error there, thinking all three regulators were open. I fixed it now...

I think I know what you mean... having a hard time picturing it though.


Looking at the regulator's guts, it seems to me that a lower PSI setting makes it more sensitive to pressure drops. What with the bigger space as you unscrew the adjuster?

So isn't that backwards? Wouldn't the 40PSI supply all it could first?

The 40 psi regulator wants to see 40 psi and the 60 psi regulator wants to see 60 psi. The 60 psi regulator will open until it sees 60 psi, at which point the 40 psi regulator also sees 60 psi and closes.

 

[Guilty Spark]

Ah! Ok.. gotcha. So they will equalize and not allow the lower values to open anymore..?

Two thoughts I have:

-Backflow. I am confused about backflow..

Would the higher pressure (say the 60psi on the 40psi one) allow the 100psi to sneak past? Maybe just until it hits 60?

-And also under usage the pressure will NOT get a chance to settle.. it will be a constant outrush to the atmosphere. That has to let the 40psi one open up at least a bit?

 

[cjl]

If the pressure on the output side of the regulator is higher than the regulator's setting, it should close completely. So, in your diagram there, assuming no losses and more than adequate supply, the output pressure should simply be the highest pressure of any of the open inputs. So, if only the 40 and 60 are open, then the output will be 60psi and the 40psi regulator will be completely closed.

 

[Guilty Spark]

Well, that is useful then.. no backflow/

Just that I was worried as they aren't normally meant to regulate from both directions are they? It's normally pretty linear.


Ok, so what about when the situation is as you say.. (40 and 60 open. All other valves closed)

\\\>>Btw.. it's the valves that are doing the open/shut. The reg.s stay as they are. \\\


Would the pressure drop from the air gun cause enough loss to open the 40 slightly?
That is the effect I am going for.. to end up with a middle value perhaps?

 

[russ_watters]

Would the higher pressure (say the 60psi on the 40psi one) allow the 100psi to sneak past? Maybe just until it hits 60?

Sneak past where?

-And also under usage the pressure will NOT get a chance to settle.. it will be a constant outrush to the atmosphere. That has to let the 40psi one open up at least a bit?

No, you're misunderstanding what equilibrium is. It doesn't mean that the pressure is the same everywhere in the system, it means that it is constant at any point in the system. As long as you don't change the piping configuration for a while, the highest setting valve will eventually find an equilibrium where it outputs exactly its rated pressure.

 

[russ_watters]

Would the pressure drop from the air gun cause enough loss to open the 40 slightly?
That is the effect I am going for.. to end up with a middle value perhaps?

What air gun/what is an air gun?

 

[cjl]

Would the pressure drop from the air gun cause enough loss to open the 40 slightly?
That is the effect I am going for.. to end up with a middle value perhaps?

The 40 won't open until the downstream pressure is below 40. It may have the effect of increasing the available flow volume once the flow rate is high enough to cause the pressure (including losses) to drop below 40 PSI, but that will depend on the details of the entire setup.

 

[Guilty Spark]

Right. Picture being a 1000 words and all that///

This ka-jigger is an air gun:

Basically an open valve to atmosphere..Re: sneak past -->
I thought maybe the higher pressure (say 60 acting on 40) backwards through the regulator would cause it to "Bounce" or not close as fast as usual.. I see now that the surface area of the diaphram is larger than the small bottom part. Meaning it will in fact close immediately upon overpressure.

CJ:
So you mean it will indeed open correct? If the 60 is not enough the 40 will "Help" out during a major pressure draw on the regulated side..?

----

Goal being to create 50 psi from 40 and 60- I can also toggle the 60 on and off to possibly allow the 40 to assist more or less.. with a controller chip//If I put a certain amount of 60psi air and a certain amount of 40psi air in a closed bottle.. I could end up with 50psi right?

 

[mfb]

Goal being to create 50 psi from 40 and 60- I can also toggle the 60 on and off to possibly allow the 40 to assist more or less.. with a controller chip//

You won't get that in a reliable way with your two regulators alone.

If I put a certain amount of 60psi air and a certain amount of 40psi air in a closed bottle.. I could end up with 50psi right?

If the container sizes have the correct ratio, sure.

 

[cjl]

CJ:
So you mean it will indeed open correct? If the 60 is not enough the 40 will "Help" out during a major pressure draw on the regulated side..?

Sure, but that'll only happen when the overall system pressure drops below 40 - it won't help pull the pressure back up to 50, or anything like that. If you need 50 psi, your best bet is a 50 psi regulator.

 

[Guilty Spark]

You won't get that in a reliable way with your two regulators alone.

If the container sizes have the correct ratio, sure.

It's ok, as long as it's some median value.. not really needing precision here.

If I had a 4L container for example, how do I figure out the proportions of each?

I recall the "ideal gas laws" but never seen a problem like that before... two different pressures of the same gas?

Sure, but that'll only happen when the overall system pressure drops below 40 - it won't help pull the pressure back up to 50, or anything like that. If you need 50 psi, your best bet is a 50 psi regulator.

I can't use more regulators unfortunately or I would- but for every one I use the valves multiply by the number of outputs.. the valves being the expensive part here.




=======

What I am thinking is that the downsteam pressure at the tool will be more influenced by changes in pressure. So with the software controlling the valve cycles it should be just a matter of dialing in the amounts of each? (Or frequency of open/close)

Everytime you shut a valve the air looses it's momentum, becomes a non-laminar flow and then is pushed again when you reopen.

The only difference here is in between cycles we have another source to play with.. so we can keep it somewhat smoother than without that second source/


Like PWM of led's ... but for air! :rofl:

 

[russ_watters]

Right. Picture being a 1000 words and all that///

This ka-jigger is an air gun:

Goal being to create 50 psi from 40 and 60- I can also toggle the 60 on and off to possibly allow the 40 to assist more or less.. with a controller chip//

Ok, I see. The air gun is a nozzle for creating a high pressure/velocity stream of air. But because of that it is a restriction and the pressure upstream of it will be relatively high.

Just think of it this way: air flows from areas of low pressure to high pressure. So there is no way to make air flow from the 40psi regulator toward the 60psi regulator or even just to the branch where they join. Air will flow from the branch where they join toward the 40psi regulator until the pressure reaches 60 psi.

If I put a certain amount of 60psi air and a certain amount of 40psi air in a closed bottle.. I could end up with 50psi right?

I suppose, but why? If you try to fill the bottle that way, the pressure will rise until it reaches 40psi, then the 40psi regulator will close and the 60psi regulator will keep pressurizing it until it reaches 60 psi or you shut it off.

Why can't you just get a variable regulator?

 

[Guilty Spark]

Ok, I see. The air gun is a nozzle for creating a high pressure/velocity stream of air. But because of that it is a restriction and the pressure upstream of it will be relatively high.

Just think of it this way: air flows from areas of low pressure to high pressure. So there is no way to make air flow from the 40psi regulator toward the 60psi regulator or even just to the branch where they join. Air will flow from the branch where they join toward the 40psi regulator until the pressure reaches 60 psi.

==
I suppose, but why? If you try to fill the bottle that way, the pressure will rise until it reaches 40psi, then the 40psi regulator will close and the 60psi regulator will keep pressurizing it until it reaches 60 psi or you shut it off.

Why can't you just get a variable regulator?

You mean high to low for the airflow right? I will assume that was backwards, or else we are rich! :)

Indeed, when both of them are on the situation prevents the 40 from doing anything.. but when the 60 is shut off then what?

That's when the pressure drops 59..58...51... 48! Then blam! We turn the 60 back on for a fraction of a second.. 55.. 54...51.. ---Etc.



==The bottle scenario is totally imaginary.. as if i could teleport two volumes of air into a sealed container full of nothing but vacuum.


Reasons for this:
I was going for cheaper parts, the ready made variables are WAY too much money. And anything I could rig up with stepper motors and such would be too unreliable. Plus the cheapo regulators aren't designed to be adjusted constantly.. so that's not safe.

Although the lowest regulator needs no valve I guess..

 

[russ_watters]

You mean high to low for the airflow right? I will assume that was backwards, or else we are rich! :)

Yes, sorry if I confused anything with that error.

Indeed, when both of them are on the situation prevents the 40 from doing anything.. but when the 60 is shut off then what?

That's when the pressure drops 59..58...51... 48! Then blam! We turn the 60 back on for a fraction of a second.. 55.. 54...51.. ---Etc.

You may be able to do some pulsing with the 60psi regulator to make it appear to give less, but it will be very difficult to pulse it fast enough and under enough control to get a steady 50 psi. And unless you do a really bad job of it, the 40 psi regulator will never open. If it does, you'll just be swinging back and forth between 40 and 60 psi.

 

[Guilty Spark]

That's ok, I was completely confusing myself with mixing and all that.. much simpler now. Well sort of..

They do make small "air capacitors" kind of like a very small air tank to smooth out the flow.. maybe that's one way to get a lock on a desired value?

Here is why I can't just keep adding regulators:

Every output block will need new valves for each additional value after 40. If more outputs are needed that will complicate things fast!

Valves=X*(n-1) Where X is the outputs and n is the regulators minus the first one which needs no valve.

Ugh. That formula looks expensive even with a fixed "n".

 

[cjl]

Wouldn't it be simpler (and cheaper) to get one single adjustable regulator (which you could adjust to 40, 50, 60, or any other desired pressure), rather than a whole bunch of fixed regulators and a complicated PWM type circuit?

 

[Guilty Spark]

Wouldn't it be simpler (and cheaper) to get one single adjustable regulator (which you could adjust to 40, 50, 60, or any other desired pressure), rather than a whole bunch of fixed regulators and a complicated PWM type circuit?

Of course... but I need more than one pressure at a time in a different place, there's just no way with one regulator. Besides topping up air tanks for each pressure maybe..

Also those things cost hundreds! For a decent one that could handle constant adjustment//

Plus I'd be worried it'd fail from use and the whole setup would be unusable.



It might be complicated but if you watch surplus auctions all this stuff can be had for well under the cost of just 1 of those things..

Although admittedly if I was rich I would just buy several. Lol.

 

[cjl]

With the circuit as you've shown it so far though, you won't be able to pull off multiple different pressures at different places - the pressure downstream of the regulators will be pretty much the same everywhere, since they all join into a common air system. If you need a bunch of different pressures, you'll pretty much need to keep them isolated from each other.