So who here has tried to make a homemade wind tunnel? My first attempt

[kazx9r]

Hello all, i posted this on another forum but didn't get much response, I had an 18” centrifugal fan around around and wanted to see if I can push some real air to make a wind tunnel. I'm just a hobbyist, and like to build and modify cars, figured this would be something to look into.

Looks like this:

Trust me this thing is heavy and pretty big but is it big enough?

Spec picture Shown here page 107

18-13a, so 18 diameter with about a 18”x18” outlet.

Chart only goes to 723rpm but I got this up to just above 2000rpm, took 14Hp, I may add some more motors.
Excuse the mess just a test run for 2 minutes just a 12 second clip:



Crazy right? This thing really blows some air, my guess 50-70mph (need a wind meter) my plan is to test small winglets for downforce.
My question is the honeycomb, is it needed? Amazon has a few but they're 1/4" holes but only 1", i need probably around 3-6" for a honeycomb? any thoughts.

I would greatly appreciate if someone could share their experiences or critique this project.

 

[Baluncore]

The honeycomb can be made from short thin-wall tube sections, cut from pipe, then glued side by side in a hexagonal pattern. Select the material and tube size you need. You may be able to throw out many redundant elements from the array where there are double walls not needed for structure.

 

[kazx9r]

The honeycomb can be made from short thin-wall tube sections, cut from pipe, then glued side by side in a hexagonal pattern. Select the material and tube size you need. You may be able to throw out many redundant elements from the array where there are double walls not needed for structure.

would this work?

https://www.amazon.com/dp/B01AYIZ5MK/?tag=pfamazon01-20

 

[Dullard]

The manufacturer's maximum rating for that blower is 3 HP. I wouldn't let anyone that I love stand anywhere near it at 14 HP.

 

[kazx9r]

I hear you Dullard, i told them to stand back. Haha. Thing is heavy, I over built it, 1.75" solid steel shaft almost 26" in length, pillow block balls bearings, but still need to reinforce it, the cage is not that sturdy.

 

[Baluncore]

would this work?

That will depend on the length and geometry of the airflow, how much untwisting is needed, and how much variation in velocity across the section can you tolerate? Will you close the airflow cycle?

 

[kazx9r]

That will depend on the length and geometry of the airflow, how much untwisting is needed, and how much variation in velocity across the section can you tolerate? Will you close the airflow cycle?

Thank you

It’s pretty turbulent coming out of the fan, I don’t think I have the space to close it.

What are you thinking for measurements? I saw a few videos where people put the wing on vertical sliders, I was thinking of doing that but have calibrated springs to measure sure the distance moved. f=kx b but open to any idea.

And then for drag using the sliders but horizontally, just thinking out loud.

 

[Baluncore]

It’s pretty turbulent coming out of the fan, I don’t think I have the space to close it.

Then you may as well give up now.
You must control it, or measurements will be unrepeatable noise.

What are you thinking for measurements?

I would measure pressures with a diagonal water manometer, and weigh the forces with a digital scale.

 

[kazx9r]

Then you may as well give up now.
You must control it, or measurements will be unrepeatable noise.


I would measure pressures with a diagonal water manometer, and weigh the forces with a digital scale.

Thank you, by closed do you mean closed loop or build the contraction, chamber and diffuser? Of course I plan to do this but do I need to have a closed loop?


As for instrumentation, I’ll take a look at the manometer.

 

[Baluncore]

Thank you, by closed do you mean closed loop or build the contraction, chamber and diffuser? Of course I plan to do this but do I need to have a closed loop?

There are two possibilities: You could get out of the wind by shutting yourself in a fixed position, closed box, or you could control the circulating air by enclosing that in a box.

You need to keep your body from disturbing the returning airflow. Until you close the tunnel, your body is part of the experiment.

 

[kazx9r]

There are two possibilities: You could get out of the wind by shutting yourself in a fixed position, closed box, or you could control the circulating air by enclosing that in a box.

You need to keep your body from disturbing the returning airflow. Until you close the tunnel, your body is part of the experiment.

balan, thank you

Someone posted this,

https://www.frh.utn.edu.ar/media/empresa/laboratorios/03_LA_fluidosytunel.pdf

is this what you have in mind? This looks similar to what I had in mind. So in order Fan, contraction, honeycomb, test chamber and then diffuser.

 

[berkeman]

Someone posted this,

My Spanish class was decades ago. Maybe you could provide a translation of any part that confuses you?

Also, I don't know if this has been mentioned already (sorry for skimming you thread), but you can achieve a lot smoother airflow in your test section of the wind tunnel by using your DIY air pump at the outlet, instead of trying to use it as a very turbulent pusher at the inlet...

 

[kazx9r]

My Spanish class was decades ago. Maybe you could provide a translation of any part that confuses you?

Also, I don't know if this has been mentioned already (sorry for skimming you thread), but you can achieve a lot smoother airflow in your test section of the wind tunnel by using your DIY air pump at the outlet, instead of trying to use it as a very turbulent pusher at the inlet...

Thank you, I have no idea what the pdf says I meant the setup they had in it

That looks like a pusher fan so I’m going with a similar setup so we need to straighten out this airflow, I don’t think I can go with a puller setup. I have seen other setups with similar setups but I understand the pusher will take a larger fan but I think I have the size and power to achieve this part. Some other setups

The above used straws for honeycomb.

Differences:

 

[kazx9r]

Stepping back for a second, I’m hoping someone can school me this, I understand for a lift and drag experiment the flow should be smooth and consistent however my setup is a downforce and drag experiment of a car wing or other car wing parts, I’m not trying to check airflow or vacuum under a car, only on top of a wing, a if I stuck a big ugly wing on my car with a strain gauge under it, and drive the car at say 70mph, would I never get a consistent strain gauge reading due to too much turbulent air?

It’s not like I can ask the flowing air to smooth out for me so I can get a reading.

Isn’t a turbulent air in a closed chamber a more realistic experiment than spending all this time trying smooth out the air with honeycombs and perforated sheets of metal.

 

[erobz]

I hear you Dullard, i told them to stand back. Haha. Thing is heavy, I over built it, 1.75" solid steel shaft almost 26" in length, pillow block balls bearings, but still need to reinforce it, the cage is not that sturdy.

He’s probably more concerned about the fan blades failing from spinning it a 3 times the rated rpm. The applied forces on them should increase as $F \propto v^2$ in each principal direction( radially, and tangentially)?

 

[Baluncore]

is this what you have in mind?

I think that system is a first approximation to an open system. I would enclose the system in two distinct counter-flow channels.

You are working at low speed, well below the speed of sound, where air is assumed not compressible. Anything you do downstream will be reflected, to contaminate your test space. The test space needs a smooth flow without turbulence.

Put the test chamber in a restricted section of the front channel, with orderly contraction, collimators, and then orderly expansion.

Put the fan in the return flow channel where there can be some vibration isolation, noise and turbulence reduction.

Turning the corners between the two channels or ducts, with internal curved sheet guides in the bends, will be calming, and easier than designing and building an inlet horn for smooth flow, then working in the wind.

Start with several standard sized cardboard boxes, and discover why it is called "duct tape".

 

[kazx9r]

He’s probably more concerned about the fan blades failing from spinning it a 3 times the rated rpm. The applied forces on them should increase as $F \propto v^2$ in each principal direction( radially, and tangentially)?

I get the thing about the safety and pushing this thing too fast, it needs to be checked for any blade damage or bending and also put some safety mesh.

thank you.

 

[kazx9r]

I think that system is a first approximation to an open system. I would enclose the system in two distinct counter-flow channels.

You are working at low speed, well below the speed of sound, where air is assumed not compressible. Anything you do downstream will be reflected, to contaminate your test space. The test space needs a smooth flow without turbulence.

Put the test chamber in a restricted section of the front channel, with orderly contraction, collimators, and then orderly expansion.

Put the fan in the return flow channel where there can be some vibration isolation, noise and turbulence reduction.

Turning the corners between the two channels or ducts, with internal curved sheet guides in the bends, will be calming, and easier than designing and building an inlet horn for smooth flow, then working in the wind.

Start with several standard sized cardboard boxes, and discover why it is called "duct tape".

thank you, I’m having a hard time visualizing this, any sketches of this setup online that you can link? All very interesting, thanks.

 

[anorlunda]

Normally, we are skittish about dangerous topics here on PF. You're lucky this thread has been allowed to continue.

May I suggest wood as an effective and inexpensive shield around that overspeeded fan. 4x4 beams, or old railroad ties, or sections of tree trunk, might do the job. Hardwood (not pine) is good at absorbing the energy of impacting objects.

When I worked at General Electric, we had a turbine overspeed test stand. It was shielded by steel plate, plus 6 feet (2 m) of railroad ties in all directions. One day, a blade broke off the turbine being tested. It penetrated the shield and the roof of the building, went high in the air, then came down through the roof again several hundred meters away and killed an engineer who was sitting at his desk.

 

[kazx9r]

Normally, we are skittish about dangerous topics here on PF. You're lucky this thread has been allowed to continue.

May I suggest wood as an effective and inexpensive shield around that overspeeded fan. 4x4 beams, or old railroad ties, or sections of tree trunk, might do the job well. Hardwood (not pine) is good at absorbing the energy of impacting objects.

When I worked at General Electric, we had a turbine overspeed test stand. It was shielded by steel plate, plus 6 feet (2 m) of railroad ties in all directions. One day, a blade broke off the turbine being tested. It penetrated the shield and the roof of the building, went high in the air, then came down through the roof again several hundred meters away and killed an engineer who was sitting at his desk.

Thank you, that’s the next step. Putting mesh and shielding.

If the mods don’t like it then I have no issue with them deleting it. Forum is pretty dead except for Balun helping out.

 

[berkeman]

You're lucky this thread has been allowed to continue.

Yeah, good point. I hadn't been checking this thread very closely, and just now realized it's based on an overspeed blower. That's not a good design.

Thread is closed on that basis.