[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Needle valves and nutrients




> From: Dave Gomberg <gomberg at wcf_com>
> 
> At 03:48 AM 12/12/00 -0500, Jonathan Peakall <jpeakall at mcn_org> said:
> > > I do have a needle valve Tom, and it has never helped in an "end of
> > > tank" situation. In fact, I had a near tragedy.
> 
> There has been a lot of pontificating on how to set up a high pressure 
> system using a needle valve as a pressure dropping device.   A lot of it is 
> wrong.  Let me 'splain:

Your explanation, below, leaves me pretty puzzled, Dave, so please consider
my questions. Maybe I'm just being dense.

In the first place, I know of no one who thinks the needle valve is a
"pressure dropping" device. It's a flow-rate regulator, if designed and used
correctly. It basically doesn't *care* what the input pressure is, above
some minimum required for it to function. [And some maximum "blow-up"
pressure, of course!]

> 
> If you have a device (such as an Eheim diffusor) that runs at 15psi (more 
> or less), and you encounter an unprotected dump at end of tank, the input 
> to the Eheim will rise from 15psi to maybe 150 and a huge dump will occur 
> thru the Eheim.

First, the Eheim diffuser, itself, doesn't run at anything like 15 psi. Only
the spring-loaded check valve that comes with it requires near that much
pressure to get open. Any *tiny* increase in pressure rapidly increases
flow, as the check valve isn't a flow regulator at all -- it's a simple
pressure-dropping device -- a spring-loaded check valve. Once open, any
larger amount of gas can flow with minimal pressure-drop change.

Where did that 150 psi increase come from? 

I don't read any regulator specs that have *that* much increase when the
(closure) pressure on the high side gets too low. The usual figures I have
seen are about 10% of that or less. A 10-15psi increase will surely dump a
*lot* of gas through a wide-open check valve. It will barely change the flow
rate through a nearly-closed needle valve, I think. [Turbulence and friction
at work to create more pressure drop at nearly the same total flow rate
(bubbles/sec).]

> 
> If you have a device (such as an Eheim diffusor) that runs at 15psi (more 
> or less), and you encounter a dump at end of tank protected by a needle 
> valve which has been set to drop the pressure from say 40psi where your 
> regulator is set to 15psi needed by the Eheim, the input to the needle 
> valve will rise from 40psi to 200psi.  Obviously, this is a smaller 
> increase than without the needle valve.   The problem is we don't know what 
> the effect will be because the flow will change from laminar to turbulent 
> throughout the system.  

Where does that notion come from? 

The fundamental way flow regulators (needle valve or aperture) work is by
being highly turbulent, already. That gives the needed non-linearity to give
a nearly constant flow rate over a huge variation in input pressure. The
rest of the system probably stays essentially laminar because the flow rate
is so low.

> How this will affect the resistance of the valve or 
> the Eheim is not known.  I do know that the Eheim is very non-linear, and a 
> small boost in pressure makes a large boost in bubbles.  Raising the 
> pressure 10% from 15psi to 16.5psi can double the output.  So some 
> experimentation in this area is required if you wish to trust your system 
> to a needle valve.

What we need here is a more accurate value for the amount a beer regulator
goes out of regulation at end of tank, when set at a nominal outlet pressure
like 20 psi or so. My impression (and limited experience) is that it is no
where near 150-200 psi. I could be mistaken, though.

> 
> If you have a device (such as an Eheim diffusor) that runs at 15psi (more 
> or less), and you encounter a dump at end of tank protected by a secondary 
> regulator which has been set to drop the pressure from say 40psi where your 
> primary regulator is set to 15psi needed by the Eheim, the input to the 
> secondary will rise from 40psi to 200psi.  The problem is we don't know 
> what the effect will be because the secondary regulator may or may not 
> tolerate 200psi.  If it is a natural gas regulator it may break down.  Then 
> it may be worse that with no protection at all.   You must carefully quiz 
> the maker of your secondary regulator to ensure that the 200psi won't be a 
> problem.

I agree to the extent that *any* primary regulator failure could potentially
apply full tank pressure to the second regulator. If it is not rated for
1000 psi, then it *must* be protected by a safety hose that will blow off
between the regulators, and avoid any shrapnel situation. [Same thing for
any needle valve not rated for full tank pressure plus, BTW. Use a
sacrificial piece of hose between regulator and needle valve.]

> 
> I hope this helps clarify just how complex this issue is.  And I hope it 
> saves the fish of someone relying on an untested needle valve setup.  BTW, 
> you can borrow a high output pressure regulator (one that can go to 400psi 
> for example) and simulate a dump and test your system's performance under 
> those circumstances.  Be sure to take all necessary safety precautions if 
> you do this tho.

That's a good suggestion, Dave. I think I will do some playing around and
see what happens. After the Holidays? ;-)

Here are my basic questions that need solid answers:

How much increase of outlet *pressure* do our typical regulators actually
allow at end of tank?

How much pressure drop is due to what components in the Ehiem setup. The
regulator must be precisely set to overcome back pressures from a) water
depth; b) sintered-glass diffuser; and c) the check valve. It's a very
twitchy setting, because a and c are linear flow (i.e.,
constant-pressure-drop) devices and only the glass diffuser pressure drop
offers any turbulent flow-rate regulation (if it does).

How much does flow rate change through a needle valve set to a low
bubbles/sec and at modest pressure, as the regulator pressure setting
changes? 

Anyone with a regulator with outlet gauge and a good needle valve can test
that. Count the bubbles at some low-flow setting, say 1 per second, and
repeat at 5psi increments of regulator pressure. If one starts at 20 psi or
so, the flow increase should be very gradual, compared to the percentage
rate of pressure increase, if that particular needle valve is a good
flow-rate regulator.

I just picked up a two-stage oxygen regulator (mine-safety equipment) for $6
on eBay. I plan to also play some games with the low-pressure one removed
from those. The outlet guage reads from 0 to 60 cm of water! 2-stage
regulation, here I come! :-)

Wright


-- 

           Wright Huntley, Fremont CA, USA, 510 612-1467