Probe/anemometer for measuring flow speed and direction?

[MUfluids]

Hi,

I'm trying to measure the steady state velocity field behind the wake of a bluff body in the wind tunnel. My uni is equipped with 4-hole cobra probes which is capable of receiving flow within a 45 degree cone of acceptance. As the base of a bluff body dominated with re-circulating flow, cobra probe has difficulty measuring back flows (out of acceptance angle).

So my question is:
- Is there any probe/anemometer that measures the flow speed and flow direction (velocity) at a certain point in space. (capable of measuring back flow also)
- My first thought was the hot wire anemometer, but that only measures the flow speed and not the flow direction, is that right?

edited: Also I'd like to add, the probe need to be small.

Any suggestion will be greatly appreciated, thank you!

 

[Danger]

I know absolutely nothing about this, but that's never stopped me from chirping up before. Could you use multiple "hot wire" anemometers and some sort of device to measure the time differential between events to "triangulate" the direction?

 

[Baluncore]

You could use a mass-flow sensor.
http://en.wikipedia.org/wiki/Thermal_mass_flow_meter
I would probably use a bead thermistor and monitor the current needed to maintain a fixed resistance and so temperature.

You might also consider 3D Laser Doppler Velocometry
http://en.wikipedia.org/wiki/Laser_Doppler_velocimetry

 

[MUfluids]

I know absolutely nothing about this, but that's never stopped me from chirping up before. Could you use multiple "hot wire" anemometers and some sort of device to measure the time differential between events to "triangulate" the direction?

Hi, thank you for your response. Yes, that is a possible option, although the main concern with that is the reduced spatial resolution. As testing is conducted on a 1/15th sized train wagon model, I'm hoping not to sacrifice too much spatial resolution. But I will still look into this idea, thanks a lot!

You could use a mass-flow sensor.
http://en.wikipedia.org/wiki/Thermal_mass_flow_meter
I would probably use a bead thermistor and monitor the current needed to maintain a fixed resistance and so temperature.

You might also consider 3D Laser Doppler Velocometry
http://en.wikipedia.org/wiki/Laser_Doppler_velocimetry

Hi, thanks for the input! Having a quick read on bead thermistor, that is still an omnidirectional measuring device, is that correct? So it will tell me the magnitude of flow speed based on the current needed to maintain the temperature in the thermistor, but it will not tell me the direction of the flow?

In regards to the LDV, the setup cost for that will be quite high, as our wind tunnel currently isn't equipped for that. I am considering in using the water channel and the PIV equipment that is available. Although, I will then have to deal with the significantly lower Reynolds number in the water channel.

 

[Danger]

a 1/15th sized train wagon model

Yeah, that's pretty small alright. Hey now, though...
What about a "wind vane" like on a barn roof mounted on a mouse-type rotary optical sensor to determine which way it's facing?

 

[MUfluids]

Yeah, that's pretty small alright. Hey now, though...
What about a "wind vane" like on a barn roof mounted on a mouse-type rotary optical sensor to determine which way it's facing?

Yea, that is still a bit big. I should clarify, the cross section of the model is like 171mm by 438mm, so it's really small. The probe I was using before looks like this. which has really good spatial resolution, except it doesn't pick up flow beyond 45degree acceptance angle.

 

[Danger]

Okay then... hmmm...
I'll think on it some more, but that's probably all I've got.

 

[MUfluids]

Okay then... hmmm...
I'll think on it some more, but that's probably all I've got.

Thanks a lot tho, appreciate it.

 

[boneh3ad]

The problem is that no matter what, you will have a probe body blocking any reverse flow. A multi-wire hot-wire system can measure all three velocity components in the spatial resolution you need without any problems, but it will still not give you reversed flow since there will be a probe body there. The right tool for the job would be PIV.

The closest I think you will get is to use something like an http://www.aeroprobe.com/probes/view/omniprobes and maybe insert the probe from the side so that it can at least get you 360 degrees of measurements in the plane of interest.

 

[RandomGuy88]

Another possibility is a split film which is very similar to a hot wire. The difference is that rather than a conductive wire, you have a thin conductive coating on a non conducting cylindrical substrate. A split film simply splits this coating in half, so that half of the cylinder is one sensor while the other half is another sensor. You can arrange the split film to either measure velocity in the x-y plane, or by rotating it 90 deg you can measure forward or reversed flow. If I recall correctly I believe the split film in this configuration can only resolve whether the velocity is forwards or reversed but if you use a triple split film then you can measure any velocity on the x-y plane (all 360 degrees). This type of probe is expensive and requires a detailed calibration but it would give you better spatial resolution and frequency response then PIV (in most cases).

 

[Baluncore]

A 3D ultrasonic anemometer is also a possibility. By using a MHz ultrasonic signal and a miniature array you could get low obstruction of the measured airflow and good 3D velocity measurements.

It might also be possible to implement a PRBS ranging, transit time or differential phase system where you have a single mobile transmitter in the flow with three or more external sensors in fixed positions. Travel time or phase differences would be due to the integral of flow velocities along the paths. Given receive data from many transmitter locations in the flow you could construct an image of the 3D flow field. Orientation polarisation of Tx and Rx would need to be sorted out.

 

[MUfluids]

The problem is that no matter what, you will have a probe body blocking any reverse flow. A multi-wire hot-wire system can measure all three velocity components in the spatial resolution you need without any problems, but it will still not give you reversed flow since there will be a probe body there. The right tool for the job would be PIV.

The closest I think you will get is to use something like an http://www.aeroprobe.com/probes/view/omniprobes and maybe insert the probe from the side so that it can at least get you 360 degrees of measurements in the plane of interest.

Yea, most likely I will be performing PIV in the near future. That Omniprobe you suggested looks interesting, I will look into that, thanks. I guess it really comes down to a compromise between do I want higher Reynolds number and test with a probe and suffer some blockage, or do I want to do PIV with lower Reynolds number. Thanks a lot for your advice.

Another possibility is a split film which is very similar to a hot wire. The difference is that rather than a conductive wire, you have a thin conductive coating on a non conducting cylindrical substrate. A split film simply splits this coating in half, so that half of the cylinder is one sensor while the other half is another sensor. You can arrange the split film to either measure velocity in the x-y plane, or by rotating it 90 deg you can measure forward or reversed flow. If I recall correctly I believe the split film in this configuration can only resolve whether the velocity is forwards or reversed but if you use a triple split film then you can measure any velocity on the x-y plane (all 360 degrees). This type of probe is expensive and requires a detailed calibration but it would give you better spatial resolution and frequency response then PIV (in most cases).

That's sounds like a possibility. I will go research into that. Probably will need to be really careful given how delicate it sounds. Wind speed at 30m/s with any impurities might be enough to damage it. Cheers.

A 3D ultrasonic anemometer is also a possibility. By using a MHz ultrasonic signal and a miniature array you could get low obstruction of the measured airflow and good 3D velocity measurements.

It might also be possible to implement a PRBS ranging, transit time or differential phase system where you have a single mobile transmitter in the flow with three or more external sensors in fixed positions. Travel time or phase differences would be due to the integral of flow velocities along the paths. Given receive data from many transmitter locations in the flow you could construct an image of the 3D flow field. Orientation polarisation of Tx and Rx would need to be sorted out.

Thanks, for the reply. It might be just me, but from some quick digging into ultrasonic anemometers, they all seem quite bulky and big? Do you know how small those devices come in?

 

[boneh3ad]

Yea, most likely I will be performing PIV in the near future. That Omniprobe you suggested looks interesting, I will look into that, thanks. I guess it really comes down to a compromise between do I want higher Reynolds number and test with a probe and suffer some blockage, or do I want to do PIV with lower Reynolds number. Thanks a lot for your advice.

You have a PIV system for said water tunnel it seems. Why can't you just use that system in a wind tunnel and just do the PIV in air?

 

[MUfluids]

You have a PIV system for said water tunnel it seems. Why can't you just use that system in a wind tunnel and just do the PIV in air?

Unfortunately, our equipment isn't quite capable of resolving the particles moving at higher speed in the wind tunnel.

 

[Danger]

Now you're just getting picky... :p

 

[boneh3ad]

Unfortunately, our equipment isn't quite capable of resolving the particles moving at higher speed in the wind tunnel.

What's the weak point in trying to do that? Laser power? Camera framerate?

 

[MUfluids]

What's the weak point in trying to do that? Laser power? Camera framerate?

Mainly camera's speed at capturing two frame fast enough. Water channel maximum flow speed is around 0.5m/s, in comparison, wind tunnel testing speed is at 30m/s. The cost of setting up PIV in the wind tunnel is the main hinderance really.