Low Air Flow Confirmation

P

Thread Starter

PaulKraemer1

Hi,

My company builds machines that have hot air dryers. We use a supply blower to push air through an electric heater on its way to the dryer, and we use an exhaust blower to remove heated air from the dryer. The exhausted air contains solvent vapors from the product that we are drying.

One of my requirements on these machines is that I need a means by which I can confirm that the blowers are moving air. I have been doing this by using differential pressure switches. In most cases, I use Dwyer 1910-00 (range: 0.07-0.15 inches WC). I have run into cases where my customer's blower speeds were fairly low and did not create enough of a pressure drop to make the switch change states. In one of these cases, I used a Dwyer 1638-0 (range 0.05-0.25 inches WC).

I am now working on a machine on which my expected air flows might be even lower, and I'm afraid I might have trouble even with the lower range of the 1638-0.

I was wondering if anyone here could recommend a sensor that might give me more flexibility in terms of being able to set it to change states at lower air flows. Ideally, I would like something that could withstand temperatures up to at least 100 degrees C and that would not be adversely affected by solvent vapors. Any advice will be greatly appreciated.

Also, if anyone has had success confirming low air flows using pressure switches, I'd appreciate it you could let me know if there are any "tricks" for maximizing your sensitivity. I have always used air tube to connect the "high" port on my pressure switch to the duct just upstream of my blower, while leaving the "low" port open to atmosphere. If anyone can suggest a different arrangement that might work better, I'd really appreciate it.

Thanks in advance,
Paul
 
C

Curt Wuollet

I've seen a setup that used a vane and a transmission photo sensor. It would seem that would sense flow regardless of pressure drop. And you could tune it to almost any flow.

Regards
cww
 
P

PaulKraemer1

Thanks Curt,

That sounds like a good idea, but I am having a difficult time envisioning how it would work. My initial thought is that I would have to insert a shaft into my duct. Inside the duct, I would mount a blade or some sort of a wheel that would rotate the shaft when there is air flow. On the outside of the duct, I would have to mount some sort of flag that I could look for with a photosensor. The flag would activate the photosensor at a frequency that would be related to air flow. Do you know of any photosensors that would allow me to set a frequency setpoint above which the sensor would activate a relay output (which I would use as my confirmation that there is air flow)?

Am I on the right track? Or am I totally missing the point of what you have in mind?

thanks in advance,
Paul
 
You need to know that air is moving. If for instance the line is blocked you could have a high DP with no air flow. So you are better off with a device that definitely indicates flow.

A vane assembly such as Curt suggests can be made quite simply. You could use a small rectangular piece of metal, twisted through 90 deg at its midpoint and mounted on an axle through the midpoint. The top part is parallel to the axle, and the bottom part is at right angles to it. Add a small weight to the bottom of the lower vane. If you blow across the assembly, it will tilt. A photointerruptor or similar can be used to see if the tilt is sufficient to show enough air flow.
 
R
If you cannot get a low pressure DP working here's something I have used in the past.

Mount a cabinet cooling fan in the duct, this spins in the breeze and produces a nice sine wave about 10V P-P from the windings,
Wire the sine wave into a frequency to current converter or direct into a frequency input and you have an inexpensive fail safe flow meter. The blade will not spin by itself and if it jambs for any reason it goes to the fail safe mode (no frequency).

I haven't tried this with a brushless DC type, just the 120 VAC variety.
 
J
Unless you really want to build something; there are a number of insertion turbine flow meters (including ours) that can read to quite low gas velocities. In addition there are "vane" meters (handheld or permanent mount) available from catalog houses that can perform this function, although the higher temperature might be a problem. Our unit can read down to 2 Ft/sec even on atmospheric pressure air and works to 400 F.

John Catch
www.inflowinc.com
[email protected]
 
C

Curt Wuollet

Kind of, except keeping anything spinning in the hot possibly contaminated air might be an issue. Imagine a strip of sheet metal about 5 " long. Twist one end 90 degrees. The long end goes in the duct like a paddle the short end goes up through a slot in line with the pipe. drill a hole in strip where i comes through the slot and put a pin through it. Now when there's no flow the strip hangs straight down. When there's flow the strip moves. Set a photo pair so it's interrupted by the top flag when its hanging down and not when the strip is rotated by the flow. Simple, almost foolproof, and the important stuff is outside the pipe. Weighting the paddle will vary how much flow it takes to swing it.

No timers were needed and it's low maintenance. If it's a dirty area you can use a mag prox. Believe it or not, a lot of airflow is sensed with a paddle on the arm of a microswitch. But I don't like mechanical switches for that. But once you get the concept, you can modify as needed.

Regards
cww
 
> any "tricks" for maximizing your sensitivity.

Connect the low side to suction side of the blower and the high side to the pressure side of the blower. DP goes up substantially. You might have to test with an indicating DP transmitter (could be water tube manometer) to measure the DP across the blower to know what switch range is needed.

See the article, "The air flow switch", here:
http://www.achrnews.com/blogs/17-opinions/post/85640-tech-page-the-airflow-switch
or (http://tinyurl.com/ohwmxpm)

snippet: "the low side would be piped into the inlet side of the blower wheel and the high side would be piped into the outlet side of the same blower wheel."

Note the caveat, that a blower wheel can spin on its own (unmotorized) due to upstream airflow, so the DP setpoint adjustment for proving airflow is critical; at a higher DP then unmotorized freewheeling but at a DP below normal motorized operation.

This is done for proving combustion blowers as an interlock to the flame safety controller. Fireye has a graphic (bottom right, page 3) showing their model PSAD DP switch piped across a blower in their air/gas pressure switch installation guide.
http://www.fireye.com/Documents/CO37.pdf

The principle is the same for all DP switches.
 
P
Thank you all for your suggestions,

I think I should be able to make a vane as described by Curt and Bruce D, or mount a cabinet cooling fan as described by Roy. I'll be able to give either (or both) of these a try when I have a machine in my shop that I can play with. At the moment, my main concern is a machine I already have installed at a customer. I already have a pressure switch mounted - I just have not been able to get it to change states with my air flows.

I am excited by David_2's post which makes me think that simply piping the low side of my pressure switch to the inlet side of the blower wheel might be all I need to do. I think I'll ask my customer to tap a hole on the inlet side and to measure the pressure with a manometer (I hope they have one). If the measured differential pressure falls withing the range of my pressure switch (or any other pressure switch that would be a quick swap), problem solved.

If possible though, I'd like to see if David_2 (or anyone else) could help me understand something I feel confused about in the tech-page-the-airflow-switch article that David_2 sent a link to. In the explanation of using an airflow switch (AFS) to determine filter status, the article recommends piping the low side of the AFS to the inlet side of the filters and facing the tube end into the flow of air. It recommends piping the high side of the AFS to the outlet side of the filters and facing the tube end away from the flow of air. It seems to me that if a blower is blowing air through a filter, the filter would create resistance to the air flow and that this would cause the pressure to be higher before the filter than after it. I also do not understand the reasoning behind aiming the tube end one way or another.

In my current application, I do not have filters, so my question about that is really just for my own understanding. As I am hoping to use an AFS as a fan proving switch, I was wondering if anyone can tell me which way I should aim the tube ends at the blower inlet and the blower outlet.

Thanks again for all of your help,
Paul
 
C
Away from the blower at the inlet and outlet. There is a deep physics answer, but think of it like a hood scoop, the flow packs a little more air in and draws a little more out. I'm sure one of our PE's will fill us in on the principle. It's what makes pitot tubes work.

Regards
cww
 
> the article recommends piping the low side of the AFS to the inlet side of the filters and facing the tube end into the flow of air.

<snip>

> It seems to me that if a blower is blowing air through a filter, the filter would create resistance to the air flow and that this would cause the pressure to be higher before the filter than after it. <

Good catch !! You are absolutely correct. (I didn't read anything but the blower part). The era of proofing trade journal articles died more than a decade ago. The high side of the gauge/switch is connected upstream, the low side is connected downstream as you indicate.

I don't know of anyone who uses right angled sensing tubes except on air velocity pitot tubes for HVAC tests. I suppose you could, but a hole into the pipe/duct with a nipple/tube/barb fitting works for every thing I've ever seen.
 
P
Thanks David_2,

> I don't know of anyone who uses right angled sensing tubes
> except on air velocity pitot tubes for HVAC tests.<

I'm starting to feel like as long as I pipe the "lo port" to the duct at my fan inlet, keeping the "hi port" piped to the duct at my fan "outlet", I have a good chance of getting the results I want. My next step will be getting my customer to measure the differential pressure between these points with a manometer so that I can make sure we get a pressure switch with an appropriate range.

Just in case I don't get the results I want, I am not familiar with the type of "air velocity pitot tubes for HVAC tests" that you mentioned above. Is this something I should look into as a backup plan?

Thanks again for your help,
Paul
 
It doesn't take much to determine the DP.

Do a Google images search for
orifice plate with water tube manometer
to see a number of images on a how a length of vinyl/tygon tubing with some water in it which is connected to a couple taps/ports. The DP is the difference in water column height. $5.00 of material.

Dwyer has those L shaped static pressure sensor tubes
http://tinyurl.com/o9n9wkp

I've not worked with air duct flows, so I don't know what the difference is between measuring with open taps or static pressure probes. Both give a DP, I just don't know what the difference is.
 
> I'm starting to feel like as long as I pipe the "lo port" to
> the duct at my fan inlet, keeping the "hi port" piped to the
> duct at my fan "outlet", I have a good chance of getting the
> results I want.

All this will tell you is the output pressure is higher than the inlet pressure.

True if you have differential pressure you should have flow but what if the pipe is blocked, you still get DP

The fan in the duct actually measures flow, it's before the heater so the air should only be warm not too hot.

Flow switches are the hardest things to keep working, if they sit there 365/24 for years in the closed position chances are they will stay there if the flow stops. I have seen the expensive oil flow switches in a grinding mill at a power failure, at least half still showed normal flow.

Perhaps the twisted vane type suspended on a strain gauge, for sure no pivot points.

As for reliable low DP switches Dwyer Photohelic spring to mind
 
C
The right angle tubes would actually give you some indication of flow in addition to static pressure. That's what pitot tubes are used t measure.

Regards
cww
 
Paul,

I edited your post to get at the salient points
You indicated the pressure switches may not work, certainly you will get more DP if you measure across the blower but it will be reading differential pressure not flow. If this is good enough to meet your requirements that's ok.

>customer's blower speeds were fairly low and did not create
>enough of a pressure drop to make the switch change states.
>In one of these cases, I used a Dwyer 1638-0 (range
>0.05-0.25 inches WC).
>
>I am now working on a machine on which my expected air flows
>might be even lower, and I'm afraid I might have trouble
>even with the lower range of the 1638-0.
>
>I was wondering if anyone here could recommend a sensor that
>might give me more flexibility in terms of being able to set
>it to change states at lower air flows. Ideally, I would
>like something that could withstand temperatures up to at
>least 100 degrees C and that would not be adversely affected
>by solvent vapors. Any advice will be greatly appreciated.
>
>Also, if anyone has had success confirming low air flows
>using pressure switches, I'd appreciate it you could let me
>know if there are any "tricks" for maximizing your
>sensitivity.

I cannot quite visualize the difference between Curt's bent sheetmetal switch and a regular vane switch perhaps a simple sketch would be in order.

A Pitot tube will produce a tiny amount of DP, they really come into their own at high velocities and have a straight run requirement. If you use pitot tubes to sense the pressure on both sides of the blower the head gained will be insignificant, you might as well not bother.
Be careful when running your tubing to the switch, it needs to be self draining as a few drops of water will disable your low range switch.

Another method of measuring low flows is with a thermal sensor, a pair of thermistors or RTDs in a duct, one is heated the other not. The breeze cools down the heated sensor and the temperature is compared to the other. These do not work well with wet gas.
Google thermal flowmeter.

The nice thing about having a rotating sensor the frequency is almost proportional to velocity, instead of a discrete signal you get something you can trend and the set-point can be set by software.

Whatever you do Good Luck
Roy
 
Top