Help me push a lot of air down a long tube

The Something Awful Forums > Discussion > Hobbies, Crafts, & Houses > Help me push a lot of air down a long tube

ANIME AKBAR: Jan 25, 2007; afu~

I have a project where I need to cool some electronics bolted to heatsinks, but for arbitrary reasons I can't put any moving parts nearby (like within 25 feet), including motors or fans. Water cooling isn't an option either, has to be ambient air, about the equivalent of a 120mm fan you find in desktop computers (airflow of 20-50cfm). My plan was to feed in fresh air via rubber tubing from a pump or compressor, and direct it over the heatsink fins with some sort of nozzle. Originally I thought I could push that much air out of a big aquarium pump over 30ft of 1/4" rubber tubing, but it seems that I'll need something much more powerful than an aquarium pump. But most of the other air pumps out there are compressors for air tools, which have insanely high pressure to deliver short bursts of air. I need a continuous stream.

After doing some research the best looking thing I've come up with is something like this High-Flow Low-Pressure Compressed Air Blower. But I don't know what sort of pressure I'll need to push the air to the destination over whatever tubing I use, or how to get the air out of the tubing onto the fins of my heatsinks. Any help would be appreciated.

ANIME AKBAR fucked around with this message at 13:08 on Mar 29, 2017

# ? Mar 28, 2017 21:22

Adbot: ADBOT LOVES YOU

# ? Apr 20, 2024 11:15

LogisticEarth: Mar 28, 2004; Someone once told me, "Time is a flat circle".

Is there a reason a small HVAC setup wouldn't work? Do you need a thin stream of air directed at the heat sinks, or just be able to move air across them?Like, if you want to kind of Micky Mouse it, you could hook up a moderately powerful blower to a rigid PVC or HDPE line. I think you don't really need a pump or compressor, just a blower.

How much space do you have to work with, budget, etc? It's going to be difficult to give you a real solution without these details.

# ? Mar 28, 2017 22:29

Bibendum: Sep 5, 2003; nunc est Bibendum

Ridged pipe with a smooth interior and the gentlest bends possible will have the least resistance but size will be king here.

This page should have some good reference
http://www.wpb-radon.com/Piping_pressure_drop.html

That blower you linked is probably overkill, Something like this mightmake more sense. and would probably directly attach to cheap PVC pipe.
https://www.amazon.com/Pactrade-Marine-Turbo-Blower-120VDC/dp/B01KY89N5K/ref=sr_1_19?ie=UTF8&qid=1490744805&sr=8-19&keywords=blower+120v

You shouldn't need to worry about getting the air directly into the heatsink fins, itf the heatsinks are sized correctly convection should do that for you, you just need to get the hot air out of the enclosure.

edit: but yeah more details needed to really be helpful.

Bibendum fucked around with this message at 00:54 on Mar 29, 2017

# ? Mar 29, 2017 00:49

LogisticEarth: Mar 28, 2004; Someone once told me, "Time is a flat circle".

Bibendum posted:

This page should have some good reference
http://www.wpb-radon.com/Piping_pressure_drop.html

What a goddamn coincidence that you linked this, since I worked with this guy on a bunch of sub slab vapor mitigation systems. I was thinking a radon system was the perfect example of a simple application where a blower would easily move air over 25-30 ft.

# ? Mar 29, 2017 01:59

ANIME AKBAR: Jan 25, 2007; afu~

LogisticEarth posted:

Is there a reason a small HVAC setup wouldn't work? Do you need a thin stream of air directed at the heat sinks, or just be able to move air across them?Like, if you want to kind of Micky Mouse it, you could hook up a moderately powerful blower to a rigid PVC or HDPE line. I think you don't really need a pump or compressor, just a blower.

How much space do you have to work with, budget, etc? It's going to be difficult to give you a real solution without these details.

The heatsinks have short fins on their sides, so I want to direct thin jets of air between them. So I'll need some sort of manifold to break the main line into several smaller branches. The fins are only about 1/2" long, so my air streams need to fit in that. I'll try to get a picture of the heatsinks tomorrow.

I was hoping to keep the pipe/tube diameter at 1" or less, and most blowers seem to have much larger openings. The tube is going to need some flexibility to it as well.

ANIME AKBAR fucked around with this message at 04:29 on Mar 29, 2017

# ? Mar 29, 2017 04:27

ANIME AKBAR: Jan 25, 2007; afu~

Bibendum posted:

That blower you linked is probably overkill, Something like this mightmake more sense. and would probably directly attach to cheap PVC pipe.
https://www.amazon.com/Pactrade-Marine-Turbo-Blower-120VDC/dp/B01KY89N5K/ref=sr_1_19?ie=UTF8&qid=1490744805&sr=8-19&keywords=blower+120v

I've seen lots of blowers similar to this, but I'm assuming they won't work well if adapted to a smaller diameter tube...

quote:

You shouldn't need to worry about getting the air directly into the heatsink fins, itf the heatsinks are sized correctly convection should do that for you, you just need to get the hot air out of the enclosure.

The fins are optimized for forced airflow. I know from testing in open air that passive convection won't do much, even when surrounded by cool air.

ANIME AKBAR fucked around with this message at 05:14 on Mar 29, 2017

# ? Mar 29, 2017 04:32

babyeatingpsychopath: Oct 28, 2000; Forum Veteran

Get a normal air compressor and regulate the pressure down to get the flow rate you need. You could use a regulator for each heat sink if necessary, mitigating the need to correct for pressure drops at every stage.

So you'd have your compressor (say) 50' away, then a manifold near your equipment with a regulator at each port going to the different lengths of pipe/tube/whatever to your heat sinks.

Furthermore, if you put your high-pressure 1/4" nozzle inside a 1/2" tube, the venturi effect gives you a bunch more airflow at lower pressure.

code:


--------------------------------

===> HP AIR OUT HERE                LP AIR HERE

--------------------------------

# ? Mar 30, 2017 05:03

ANIME AKBAR: Jan 25, 2007; afu~

Pressure regulators are a new thing to me, can they efficiently drop a compressor output (>50psi) to a very low pressure (<3psi)? Also I don't think I'll be able think I'll be able to put the regulator near the output, so it would be compressor>regulator>long haul tubing>manifold>short haul tubing>heatsinks.

The heatsink fins won't provide too much resistance to the air (they're only about 4" long and 0.5" deep), so I need very little pressure to get a lot of airflow at the final output (I think?).

My basic problem is I don't know the theory behind compressed air and the various devices. My expertise is in electronics, but it seems that compressed gas systems are described in very different terms from electrical circuits.

So let's say I want my total output to be have pressure Pout at some volumetric flow rate Vout. What should my compressor be rated, and how can I account for losses in the tubing/regulator?

# ? Mar 30, 2017 14:49

asdf32: May 15, 2010; I lust for childrens' deaths. Ask me about how I don't care if my kids die.

Mattress inflator?

"Air flow booster" converts a compressor output into higher flow lower pressure with a wide range of available ratios. See McMaster.

# ? Mar 31, 2017 04:32

goodnight mooned: Aug 2, 2007

ANIME AKBAR posted:

Pressure regulators are a new thing to me, can they efficiently drop a compressor output (>50psi) to a very low pressure (<3psi)? Also I don't think I'll be able think I'll be able to put the regulator near the output, so it would be compressor>regulator>long haul tubing>manifold>short haul tubing>heatsinks.

The heatsink fins won't provide too much resistance to the air (they're only about 4" long and 0.5" deep), so I need very little pressure to get a lot of airflow at the final output (I think?).

My basic problem is I don't know the theory behind compressed air and the various devices. My expertise is in electronics, but it seems that compressed gas systems are described in very different terms from electrical circuits.

So let's say I want my total output to be have pressure Pout at some volumetric flow rate Vout. What should my compressor be rated, and how can I account for losses in the tubing/regulator?

The pressure drop is proportional to flow rate and inversely proportional to your tube diameter.

e: I think using an air compressor and pressure regulators could work fine.

goodnight mooned fucked around with this message at 08:27 on Mar 31, 2017

# ? Mar 31, 2017 08:23

ANIME AKBAR: Jan 25, 2007; afu~

asdf32 posted:

Mattress inflator?

"Air flow booster" converts a compressor output into higher flow lower pressure with a wide range of available ratios. See McMaster.

Now this sounds useful, especially since it addresses another major concern I have, which is that the air from a compressor will probably be fairly warm. Mixing it with ambient air will help.

So combining this compressor with this booster should give me plenty of output, right?

Also with that compressor, will I need to remove moisture from the output even if I'm using a booster?

goodnight mooned posted:

The pressure drop is proportional to flow rate and inversely proportional to your tube diameter.

e: I think using an air compressor and pressure regulators could work fine.

Yeah but the way these components are specified is really obtuse, especially when they don't distinguish psig from psia, which matters since the pressure at my regulator/booster output is going to be pretty low.

I'm used to dealing with electrical components, which always come with their own datasheet with curves detailing what to expect under various operating conditions, but for this compressed air stuff they just throw a few numbers at you. I don't know what to expect in specific conditions, aside from conservation of mass and energy, of course.

For the booster above, it says it gives 155-250cfm of output flow with 90psi of input pressure. But isn't that dependent on the pressure at the output port? And my compressor can only deliver 3cfm at 90psi, and the booster has a 12:1 ratio, so I can only expect up to 36cfm at the output. So does that mean this booster is a bad match for this compressor?

ANIME AKBAR fucked around with this message at 15:36 on Mar 31, 2017

# ? Mar 31, 2017 15:33

goodnight mooned: Aug 2, 2007

ANIME AKBAR posted:

Now this sounds useful, especially since it addresses another major concern I have, which is that the air from a compressor will probably be fairly warm. Mixing it with ambient air will help.

So combining this compressor with this booster should give me plenty of output, right?

Also with that compressor, will I need to remove moisture from the output even if I'm using a booster?

Yeah but the way these components are specified is really obtuse, especially when they don't distinguish psig from psia, which matters since the pressure at my regulator/booster output is going to be pretty low.

Air pressure at your output is going to be lower than at the input?
Outputs of air compressors are usually defined at 0 bar.

ANIME AKBAR posted:

I'm used to dealing with electrical components, which always come with their own datasheet with curves detailing what to expect under various operating conditions, but for this compressed air stuff they just throw a few numbers at you. I don't know what to expect in specific conditions, aside from conservation of mass and energy, of course.

You should be able to get performance curves for compressors and regulators from the manufacturer, and you can estimate the friction losses through a pipe.

ANIME AKBAR posted:

For the booster above, it says it gives 155-250cfm of output flow with 90psi of input pressure. But isn't that dependent on the pressure at the output port? And my compressor can only deliver 3cfm at 90psi, and the booster has a 12:1 ratio, so I can only expect up to 36cfm at the output. So does that mean this booster is a bad match for this compressor?

That is an air compressor with high pressure but low flow (look at the size of the tank). You don't need 90psi to move air 25 feet.

Are you limited in the size tubing you can run? Because if you can run ducting you could use a centrifugal fan (they can move a lot of air) and avoid using a compressor.

# ? Mar 31, 2017 19:48

Neutrino: Mar 8, 2006; Fallen Rib

I would suggest not using tubing and instead using something like 1" PVC pipe as you air duct. PVC pipe can handle higher pressures and will be much cheaper and stronger for your purposes than vinyl or plastic tubing. I think using this as ducting will even simplify your compressor requirements.

# ? Mar 31, 2017 20:22

ANIME AKBAR: Jan 25, 2007; afu~

goodnight mooned posted:

Air pressure at your output is going to be lower than at the input?

I mean at the output near my heatsinks, after whatever booster/regulator I use.

quote:

You should be able to get performance curves for compressors and regulators from the manufacturer, and you can estimate the friction losses through a pipe.

I guess that's one disadvantage of McMaster then. Never seen such documentation from them.

quote:

That is an air compressor with high pressure but low flow (look at the size of the tank). You don't need 90psi to move air 25 feet.

Right, I just need a few psig, I think. But the only thing I could find to directly generate that was the $500 blower in my first post.

quote:

Are you limited in the size tubing you can run? Because if you can run ducting you could use a centrifugal fan (they can move a lot of air) and avoid using a compressor.

I can use maybe 2" for most of the run, but few feet will need to be 1" or less.

Neutrino posted:

I would suggest not using tubing and instead using something like 1" PVC pipe as you air duct. PVC pipe can handle higher pressures and will be much cheaper and stronger for your purposes than vinyl or plastic tubing. I think using this as ducting will even simplify your compressor requirements.

Unfortunately there are a lot of bends in the air's path, and this is something that will not be installed permanently. I will have to be able to set it up and tear it down in minutes, so flexible hose/tubing is going to be necessary.

ANIME AKBAR fucked around with this message at 00:24 on Apr 1, 2017

# ? Apr 1, 2017 00:20

goodnight mooned: Aug 2, 2007

ANIME AKBAR posted:

I mean at the output near my heatsinks, after whatever booster/regulator I use.
I guess that's one disadvantage of McMaster then. Never seen such documentation from them.
Right, I just need a few psig, I think. But the only thing I could find to directly generate that was the $500 blower in my first post.
I can use maybe 2" for most of the run, but few feet will need to be 1" or less.
Unfortunately there are a lot of bends in the air's path, and this is something that will not be installed permanently. I will have to be able to set it up and tear it down in minutes, so flexible hose/tubing is going to be necessary.

Seems to me you're overthinking this then if you're just aiming for something cheap and temporary.

Go back and read babyeatingpsychopath's solution because he nailed it.

# ? Apr 1, 2017 03:16

Unsane: Jul 16, 2003

Why not just grab a small squirrel cage fan, and route some 3" duct into the case. You're way over thinking this.

# ? Apr 1, 2017 05:58

Neutrino: Mar 8, 2006; Fallen Rib

ANIME AKBAR posted:

Unfortunately there are a lot of bends in the air's path, and this is something that will not be installed permanently. I will have to be able to set it up and tear it down in minutes, so flexible hose/tubing is going to be necessary.

Sounds like all the more reason to use pvc piping. To make a way of keeping flexible tubing from flopping all over the place under pressure is going to require a more involved system than using something unflexible.

# ? Apr 2, 2017 05:08

Adbot: ADBOT LOVES YOU

# ? Apr 20, 2024 11:15

ANIME AKBAR: Jan 25, 2007; afu~

goodnight mooned posted:

Seems to me you're overthinking this then if you're just aiming for something cheap and temporary.

I gave the wrong idea. It's something which will need to be set up quite frequently for several hours, then completely torn down.

quote:

Go back and read babyeatingpsychopath's solution because he nailed it.

It sounds good, assuming the regulators can actually work properly with a very high input:output pressure ratio. I'm wondering how it compares to the compressed air booster...

Unsane posted:

Why not just grab a small squirrel cage fan, and route some 3" duct into the case. You're way over thinking this.

That would get it 95% of the way there, but I'm not just looking to bath an enclosure in air (there is no enclosure actually). I need to force air into several narrow channels just 1/4" wide.

ANIME AKBAR fucked around with this message at 00:15 on Apr 3, 2017

# ? Apr 3, 2017 00:11

The Something Awful Forums > Discussion > Hobbies, Crafts, & Houses > Help me push a lot of air down a long tube