filter inlet/outlet diameters

[SkyCatcher]

Hi,

I have just aquired of a large external filter (project) which has a larger
diameter inlet (3/4") than its outlet (1/2"). Would there be a problem if I
reduced the inlet down to 1/2" (back pressure??) so I can use the same rigid
pipe work to/from the tank?

Thanks,

Sky.

[NetMax]

"SkyCatcher" > wrote in message
...
> Hi,
>
> I have just aquired of a large external filter (project) which has a
> larger diameter inlet (3/4") than its outlet (1/2"). Would there be a
> problem if I reduced the inlet down to 1/2" (back pressure??) so I can
> use the same rigid pipe work to/from the tank?
>
> Thanks,
>
> Sky.


I think that might actually be preferable. Canisters are an airtight
system, which theoretically work better when the weight of the water in
the intake side is similar (comparable) to the weight of the water in the
return side. In practice, I don't know how dissimilar they have to be
before they have a negative effect, but there is a significant difference
between 3/4 and 1/2 diameter hose (2:1 perhaps?). Also your flow rate is
a function of the total resistance, and a narrow section would be your
largest constraint. Having a significantly larger section means you have
more weight to move, and the narrow section restricts the movement
(negates the advantage of the thicker hose somewhat). jmo
--
www.NetMax.tk

[Margolis]

I don't know about that. All of the pumps and filters I have ever used have
always come with warnings about restricting the flow on the intake side.

--

Margolis
http://web.archive.org/web/20030215212142/http://www.agqx.org/faqs/AGQ2FAQ.htm
http://www.unrealtower.org/faq

[NetMax]

"Margolis" > wrote in message
...
>I don't know about that. All of the pumps and filters I have ever used
>have always come with warnings about restricting the flow on the intake
>side.
>
> --
>
> Margolis


Good point, however it's not a flow restriction if the output diameter is
as big as the input diameter. I'm also assuming the 1/2 dia meets the
spec, otherwise change everything to a larger size hose.

In regards to restricting the input, I've read the same thing, however,
in real life, we restrict the intake on our filters all the time. It's
called a clogged intake strainer ;~). Even the design of a canister
filter typically places the filter media in the input side, with the
output unrestricted. Go figure.
--
www.NetMax.tk

[Margolis]

If the pump was explicitly engineered for the larger intake it could be
considered a flow restriction imo.. I am not certain, it is just a
thought.

--

Margolis
http://web.archive.org/web/20030215212142/http://www.agqx.org/faqs/AGQ2FAQ.htm
http://www.unrealtower.org/faq

[NetMax]

"George Pontis" > wrote in message
...
> In article >,
> says...
>> "SkyCatcher" > wrote in message
>> ...
>> > Hi,
>> >
>> > I have just aquired of a large external filter (project) which has a
>> > larger diameter inlet (3/4") than its outlet (1/2"). Would there be
>> > a
>> > problem if I reduced the inlet down to 1/2" (back pressure??) so I
>> > can
>> > use the same rigid pipe work to/from the tank?
>> >
>> > Thanks,
>> >
>> > Sky.
>>
>>
>> I think that might actually be preferable. Canisters are an airtight
>> system, which theoretically work better when the weight of the water
>> in
>> the intake side is similar (comparable) to the weight of the water in
>> the
>> return side. In practice, I don't know how dissimilar they have to be
>> before they have a negative effect, but there is a significant
>> difference
>> between 3/4 and 1/2 diameter hose (2:1 perhaps?). Also your flow rate
>> is
>> a function of the total resistance, and a narrow section would be your
>> largest constraint. Having a significantly larger section means you
>> have
>> more weight to move, and the narrow section restricts the movement
>> (negates the advantage of the thicker hose somewhat). jmo
>>
>
> The total weight of the water in the intake and output may be different
> but the
> pressure only depends on the height of the water column and not on the
> pipe
> diameter. Talking about the weight of the water can be confusing, but
> perhaps it
> is helpful to remember that it is a closed system. The weight of the
> water going
> into the canister per unit time had better be exactly the same as the
> weight of
> the water flowing out. Anything else and there is a leak.

LOL on weight per unit time comparison. I don't entirely agree with your
statement that the pressure only depends on the height of the water
column (and if you are a hydraulics engineer, I might be getting into big
trouble ;~). If we assume the water is flowing, then flow and pressure
are both linked to series resistance (pipe diameters+) and water column
height (head pressure). If we simplify the column height's head pressure
to a unit of resistance and add that to the series resistance, I think
you would get : Flow = Pressure / Resistance, or Pressure = Flow x
Resistance, or Resistance = Pressure / Flow, so any way it's sliced, at
its most basic level, it's a 3 variable ratio (until a variable
approaches zero, such as when you lift your hose up and the water stops).
I'm sure this all gets *much* more complicated, such as the series
resistance actually being a dynamic non-linear variable, but I'm trying
to keep myself out of trouble, by keeping to Usenet 101 levels ;~).

> The idea of not restricting the input to a pump is important in some
> applications.
> A strong pump with an intake restriction can experience cavitation at
> the
> impeller, due to the low pressure at the inlet. Cavitation can erode
> the impeller
> in a remarkably short time. I don't know if any aquarium pumps are
> powerul enough
> for that to be a problem. Probably not for the average canister filter,
> but maybe
> so with the 1000 GPH and larger pumps.

Thanks for that. I didn't know about cavitation's effect on impellers.
I wonder if this has anything to do with the impeller change that Fluval
introduced several years ago. They used to have a forward pitch blade,
and changed over to a neutral pitch. In atmospheric systems, a forward
pitch produces more pressure at the expense of more noise and a reverse
pitch is quietest producing the least pressure (the neutral pitch a
compromise). In this fluid application, noise is probably not a
variable. I thought it was changed because of breakage, or perhaps it
tended to capture debris, but if it was more prone to cavitation, then
breakage due to mechanical stress would make sense.
--
www.NetMax.tk

[George Pontis]

In article >,
says...
>
> LOL on weight per unit time comparison. I don't entirely agree with your
> statement that the pressure only depends on the height of the water
> column (and if you are a hydraulics engineer, I might be getting into big
> trouble ;~). If we assume the water is flowing, then flow and pressure
> are both linked to series resistance (pipe diameters+) and water column
> height (head pressure). If we simplify the column height's head pressure
> to a unit of resistance and add that to the series resistance, I think
> you would get : Flow = Pressure / Resistance, or Pressure = Flow x
> Resistance, or Resistance = Pressure / Flow, so any way it's sliced, at
> its most basic level, it's a 3 variable ratio (until a variable
> approaches zero, such as when you lift your hose up and the water stops).
> I'm sure this all gets *much* more complicated, such as the series
> resistance actually being a dynamic non-linear variable, but I'm trying
> to keep myself out of trouble, by keeping to Usenet 101 levels ;~).

I am an electrical engineer and not trained in hydraulics. But I did benefit from
a particularly excellent physics teacher, and later worked in biotech where fluid
flow was important so one had to have some conversational knowledge. But I could
be wrong and do not mind being corrected. My statement about pressure being
dependent only on the height of the column neglected flow. However, I am thinking
that the fluid resistances in the tubing are not the dominant ones. I think that
the pump itself is dominant. The pump designers specify a tubing size that is big
enough to not significantly restrict the flow in a reasonable length. So the total
resistance is really several resistances in series, and I think that the tubing is
fairly small compared to the resistances inside the pump.

Thus, if we need to increase flow we usually have to buy a bigger pump because
there is not much to be gained by reducing the friction in the already large
tubing. Maybe 10% or so if we doubled or tripled the tubing diameter. A siphon
would be the opposite case. There, the resistance of the tubing is the only
resistance to flow so an increase in diameter improves flow significantly.

> > The idea of not restricting the input to a pump is important in some
> > applications.
> > A strong pump with an intake restriction can experience cavitation at
> > the
> > impeller, due to the low pressure at the inlet. Cavitation can erode
> > the impeller
> > in a remarkably short time. I don't know if any aquarium pumps are
> > powerul enough
> > for that to be a problem. Probably not for the average canister filter,
> > but maybe
> > so with the 1000 GPH and larger pumps.
>
> Thanks for that. I didn't know about cavitation's effect on impellers.
> I wonder if this has anything to do with the impeller change that Fluval
> introduced several years ago. They used to have a forward pitch blade,
> and changed over to a neutral pitch. In atmospheric systems, a forward
> pitch produces more pressure at the expense of more noise and a reverse
> pitch is quietest producing the least pressure (the neutral pitch a
> compromise). In this fluid application, noise is probably not a
> variable. I thought it was changed because of breakage, or perhaps it
> tended to capture debris, but if it was more prone to cavitation, then
> breakage due to mechanical stress would make sense.
>

I don't know much about the Fluval pump designs, but I can offer a bit more
explanation of the damage mechanism due to cavitation. When the pressure near an
impeller drops below the vapor pressure of the water, then the water flashes into
vapor. If we had a window into the pump it would look like there were air bubbles,
but this could be seen even in water that was completely degassed. So, now we have
these pockets of very low density vapor surrounded by normal density water. When
the pressure of the water then goes back above the vapor pressure there is an
instant collapse of the bubbles. The implosions are quite violent on a microscopic
scale, enough to rip tiny chunks of away from the surface. Even hardened metal can
be eroded by the implosions. The problem usually usually strikes near the outer
edges of the impeller because the velocity is higher there.

I have don't know if this ever happens in a canister filter pump. In other closed
loop system where it can be a problem, the pump is arranged to be pushing into the
high resistance path of the loop rather than drawing from it. Canisters usually do
the opposite - pull through the filter bed - so it would appear that the designers
don't seem too concerned.

[SkyCatcher]

Guys,

Very interesting discussion - certainly more than I was expecting!

To summarise though, do you think I could risk it?

cheers, Sky.

"George Pontis" > wrote in message
...
> In article >,
>
> says...
>>
>> LOL on weight per unit time comparison. I don't entirely agree with your
>> statement that the pressure only depends on the height of the water
>> column (and if you are a hydraulics engineer, I might be getting into big
>> trouble ;~). If we assume the water is flowing, then flow and pressure
>> are both linked to series resistance (pipe diameters+) and water column
>> height (head pressure). If we simplify the column height's head pressure
>> to a unit of resistance and add that to the series resistance, I think
>> you would get : Flow = Pressure / Resistance, or Pressure = Flow x
>> Resistance, or Resistance = Pressure / Flow, so any way it's sliced, at
>> its most basic level, it's a 3 variable ratio (until a variable
>> approaches zero, such as when you lift your hose up and the water stops).
>> I'm sure this all gets *much* more complicated, such as the series
>> resistance actually being a dynamic non-linear variable, but I'm trying
>> to keep myself out of trouble, by keeping to Usenet 101 levels ;~).
>
> I am an electrical engineer and not trained in hydraulics. But I did
> benefit from
> a particularly excellent physics teacher, and later worked in biotech
> where fluid
> flow was important so one had to have some conversational knowledge. But I
> could
> be wrong and do not mind being corrected. My statement about pressure
> being
> dependent only on the height of the column neglected flow. However, I am
> thinking
> that the fluid resistances in the tubing are not the dominant ones. I
> think that
> the pump itself is dominant. The pump designers specify a tubing size that
> is big
> enough to not significantly restrict the flow in a reasonable length. So
> the total
> resistance is really several resistances in series, and I think that the
> tubing is
> fairly small compared to the resistances inside the pump.
>
> Thus, if we need to increase flow we usually have to buy a bigger pump
> because
> there is not much to be gained by reducing the friction in the already
> large
> tubing. Maybe 10% or so if we doubled or tripled the tubing diameter. A
> siphon
> would be the opposite case. There, the resistance of the tubing is the
> only
> resistance to flow so an increase in diameter improves flow significantly.
>
>> > The idea of not restricting the input to a pump is important in some
>> > applications.
>> > A strong pump with an intake restriction can experience cavitation at
>> > the
>> > impeller, due to the low pressure at the inlet. Cavitation can erode
>> > the impeller
>> > in a remarkably short time. I don't know if any aquarium pumps are
>> > powerul enough
>> > for that to be a problem. Probably not for the average canister filter,
>> > but maybe
>> > so with the 1000 GPH and larger pumps.
>>
>> Thanks for that. I didn't know about cavitation's effect on impellers.
>> I wonder if this has anything to do with the impeller change that Fluval
>> introduced several years ago. They used to have a forward pitch blade,
>> and changed over to a neutral pitch. In atmospheric systems, a forward
>> pitch produces more pressure at the expense of more noise and a reverse
>> pitch is quietest producing the least pressure (the neutral pitch a
>> compromise). In this fluid application, noise is probably not a
>> variable. I thought it was changed because of breakage, or perhaps it
>> tended to capture debris, but if it was more prone to cavitation, then
>> breakage due to mechanical stress would make sense.
>>
>
> I don't know much about the Fluval pump designs, but I can offer a bit
> more
> explanation of the damage mechanism due to cavitation. When the pressure
> near an
> impeller drops below the vapor pressure of the water, then the water
> flashes into
> vapor. If we had a window into the pump it would look like there were air
> bubbles,
> but this could be seen even in water that was completely degassed. So, now
> we have
> these pockets of very low density vapor surrounded by normal density
> water. When
> the pressure of the water then goes back above the vapor pressure there is
> an
> instant collapse of the bubbles. The implosions are quite violent on a
> microscopic
> scale, enough to rip tiny chunks of away from the surface. Even hardened
> metal can
> be eroded by the implosions. The problem usually usually strikes near the
> outer
> edges of the impeller because the velocity is higher there.
>
> I have don't know if this ever happens in a canister filter pump. In other
> closed
> loop system where it can be a problem, the pump is arranged to be pushing
> into the
> high resistance path of the loop rather than drawing from it. Canisters
> usually do
> the opposite - pull through the filter bed - so it would appear that the
> designers
> don't seem too concerned.

[NetMax]

"George Pontis" > wrote in message
...
> In article >,
>
> says...
>>
>> LOL on weight per unit time comparison. I don't entirely agree with
>> your
>> statement that the pressure only depends on the height of the water
>> column (and if you are a hydraulics engineer, I might be getting into
>> big
>> trouble ;~). If we assume the water is flowing, then flow and
>> pressure
>> are both linked to series resistance (pipe diameters+) and water
>> column
>> height (head pressure). If we simplify the column height's head
>> pressure
>> to a unit of resistance and add that to the series resistance, I think
>> you would get : Flow = Pressure / Resistance, or Pressure = Flow x
>> Resistance, or Resistance = Pressure / Flow, so any way it's sliced,
>> at
>> its most basic level, it's a 3 variable ratio (until a variable
>> approaches zero, such as when you lift your hose up and the water
>> stops).
>> I'm sure this all gets *much* more complicated, such as the series
>> resistance actually being a dynamic non-linear variable, but I'm
>> trying
>> to keep myself out of trouble, by keeping to Usenet 101 levels ;~).
>
> I am an electrical engineer and not trained in hydraulics. But I did
> benefit from
> a particularly excellent physics teacher, and later worked in biotech
> where fluid
> flow was important so one had to have some conversational knowledge.
> But I could
> be wrong and do not mind being corrected. My statement about pressure
> being
> dependent only on the height of the column neglected flow. However, I
> am thinking
> that the fluid resistances in the tubing are not the dominant ones. I
> think that
> the pump itself is dominant. The pump designers specify a tubing size
> that is big
> enough to not significantly restrict the flow in a reasonable length.
> So the total
> resistance is really several resistances in series, and I think that
> the tubing is
> fairly small compared to the resistances inside the pump.

Well my background is also in electrical engineering, so we might both
need to be corrected ;~). I think the series resistances account for
more drag than you are allowing. There are tables to calculate
resistance, and they use diameter, length, type of wall (corrugated or
smooth) and corner type (ie: 45 or 90 degree). I've done calculations on
air handler units where we literally ran out of power before the end of
the ducts were reached due to series resistance. I would expect a
similar relationship in aquatic applications, but really we have
essentially agreed, and it is only a matter of degrees now.

> Thus, if we need to increase flow we usually have to buy a bigger pump
> because
> there is not much to be gained by reducing the friction in the already
> large
> tubing. Maybe 10% or so if we doubled or tripled the tubing diameter. A
> siphon
> would be the opposite case. There, the resistance of the tubing is the
> only
> resistance to flow so an increase in diameter improves flow
> significantly.

A siphon might illustrate the effect of series resistance, as you can get
radically different results from the same height levels by switching hose
diameters ;~).

>> > The idea of not restricting the input to a pump is important in some
>> > applications.
>> > A strong pump with an intake restriction can experience cavitation
>> > at
>> > the
>> > impeller, due to the low pressure at the inlet. Cavitation can erode
>> > the impeller
>> > in a remarkably short time. I don't know if any aquarium pumps are
>> > powerul enough
>> > for that to be a problem. Probably not for the average canister
>> > filter,
>> > but maybe
>> > so with the 1000 GPH and larger pumps.
>>
>> Thanks for that. I didn't know about cavitation's effect on
>> impellers.
>> I wonder if this has anything to do with the impeller change that
>> Fluval
>> introduced several years ago. They used to have a forward pitch
>> blade,
>> and changed over to a neutral pitch. In atmospheric systems, a
>> forward
>> pitch produces more pressure at the expense of more noise and a
>> reverse
>> pitch is quietest producing the least pressure (the neutral pitch a
>> compromise). In this fluid application, noise is probably not a
>> variable. I thought it was changed because of breakage, or perhaps it
>> tended to capture debris, but if it was more prone to cavitation, then
>> breakage due to mechanical stress would make sense.
>>
>
> I don't know much about the Fluval pump designs, but I can offer a bit
> more
> explanation of the damage mechanism due to cavitation. When the
> pressure near an
> impeller drops below the vapor pressure of the water, then the water
> flashes into
> vapor. If we had a window into the pump it would look like there were
> air bubbles,
> but this could be seen even in water that was completely degassed. So,
> now we have
> these pockets of very low density vapor surrounded by normal density
> water. When
> the pressure of the water then goes back above the vapor pressure there
> is an
> instant collapse of the bubbles. The implosions are quite violent on a
> microscopic
> scale, enough to rip tiny chunks of away from the surface. Even
> hardened metal can
> be eroded by the implosions. The problem usually usually strikes near
> the outer
> edges of the impeller because the velocity is higher there.

Thanks for taking the time to explain this.

> I have don't know if this ever happens in a canister filter pump. In
> other closed
> loop system where it can be a problem, the pump is arranged to be
> pushing into the
> high resistance path of the loop rather than drawing from it. Canisters
> usually do
> the opposite - pull through the filter bed - so it would appear that
> the designers
> don't seem too concerned.

Agreed, these canister pumps are actually quite small, so they probably
don't build up to cavitation threshold.
--
www.NetMax.tk

[Bill]

On 2005-02-27, George Pontis > wrote:
> A strong pump with an intake restriction can experience cavitation at
> the impeller, due to the low pressure at the inlet. Cavitation can
> erode the impeller in a remarkably short time.

Okay, maybe this is a silly question.

Would damage also occur from air bubbles getting sucked into the intake?
I ask this because my set-up has a bubble wand along the back wall of
the tank, directly below the filter intake. I can definitely hear the
bubbles getting sucked in regularly. Is this bad? I've unplugged the
air pump in the mean time. It's only been going since Sunday evening --
four days now. No fish in it yet.

Oh, and this is my de-lurk here. I've been reading the group like mad
for about a week or two now, trying to get caught up on the
not-yet-expired articles.

I said "my" earlier, but I really meant "our" -- my wife and I are doing
this together. I started wanting to keep fish about two or three months
ago, and we're just now getting things set up and going. I have other
newbie questions, but I don't want to hijack the thread.

[NetMax]

"Bill" > wrote in message
news:KCQVd.38950$Tt.22254@fed1read05...
> On 2005-02-27, George Pontis > wrote:
>> A strong pump with an intake restriction can experience cavitation at
>> the impeller, due to the low pressure at the inlet. Cavitation can
>> erode the impeller in a remarkably short time.
>
> Okay, maybe this is a silly question.
>
> Would damage also occur from air bubbles getting sucked into the
> intake?
> I ask this because my set-up has a bubble wand along the back wall of
> the tank, directly below the filter intake. I can definitely hear the
> bubbles getting sucked in regularly. Is this bad? I've unplugged the
> air pump in the mean time. It's only been going since Sunday
> evening --
> four days now. No fish in it yet.
>
> Oh, and this is my de-lurk here. I've been reading the group like mad
> for about a week or two now, trying to get caught up on the
> not-yet-expired articles.
>
> I said "my" earlier, but I really meant "our" -- my wife and I are
> doing
> this together. I started wanting to keep fish about two or three
> months
> ago, and we're just now getting things set up and going. I have other
> newbie questions, but I don't want to hijack the thread.

Welcome to the group. I've often positioned canister filter intakes to
collect some air bubble streams. It helps break up the surface tension
and that 'protein slick' that sometimes appears. Doing this has never
caused me problems, but on some older canisters, it may cause the filter
to break suction as the air collects in a pocket near the impeller.
Don't try this (collecting air bubble streams) with powerfilters, as the
bubbles block the filter media and push the sponge upwards through the
filter.

I recommend you put some ammonia in the tank to start a fishless cycle
(unless you have a friend who will give you some aged filter media).
Research fishless cycling under new tank syndrome. One of the FAQs is
http://faq.thekrib.com/

cheers
--
www.NetMax.tk