intrumentation for measuring pressure change in air ducts?

[USAeng]

In a power plant there is a fan that blows air into a header. After the header there are 3 6" air ducts. I need to get an intrument installed in each header so that if any or all of the ducts begin to become blocked off past a set amount then the switch will activate and shut off a fuel system. I am confused on if I should use a pressure switch or a pitot tube. It seems to me the pitot tube would be the correct one because it would measure velocity pressure which would be the difference in the ducts... the pressure switch would be measuring static pressure that would be equalized in the ducts and header if a duct were beginning to clog. Thanks for any input!

 

[USAeng]

Oh... I meant three 6" air ducts which looked confusing... also the intrumentation will be intalled each one of the three DUCTS not the header as mentioned above... sorry

 

[DRWeig]

Yes, a simple static pressure switch won't detect blockage in one of several ducts leaving a header. The static will rise some, but will be equal.

You can sense flow (velocity) with a pitot tube or a variety of other differential-pressure devices. You might also check what the normal velocity will be and at what level you want to cut off the fuel. Sail switches are a good alternative if they'll drop at an appropriate velocity for you...

Good on ya,

Goober Dave

 

[roydm]

If you need to measure the relative velocity consider putting a cabinet ventilation type fan (AC not brushless DC) in each duct. Connect the fan winding to a frequency to voltage converter and you have a cheap long lasting turbine flow meter.
The motor puts out a nice low voltage sine wave < 10 V P-P from memory.

 

[glowing]

I would recommend a thermal flow device. They work on the principle that of RTDs and a separate heating element. the differential in measured temperature is representative of a flow.
They are nice, because if they get really coated up then the differential temperature will approach 0, and it will look like no-flow.
They are available in 2 and 4 wire analog, and 4 wire discrete. I THINK i have seen 2 wire discrete as well.

There are lots of manufacturers out there. Kayden has economical options.

 

[roydm]

Glowing is right, thermal works great.
Just one thing if there is any moisture i.e. entrained water drops this can mess with them.
Roy

 

[glowing]

good point roydm. I have never really had issues, and in a 6" duct, I don't think there would be. Another option is a flow vein, like the Flowtect V4. These are however very finicky. I would stay away from them, but it is an option. There is one more...I am for some reason drawing a blank on it though. I think I have a manual on them at home. I will check tonight.

 

[IRstuff]

There are ultrasonic wind speed instruments:

http://www.electronickits.com/kit/complete/meas/ck112.htm

http://www.novalynx.com/200-windsonic.html

Most of the other possibilities have been mentioned, I think:

http://www.forestry-suppliers.com/p...=3803&title=Kestrel+2500+Pocket+Weather+Meter

http://www.forestry-suppliers.com/p....asp?mi=6479&title=Dwyer+Hand-Held+Wind+Meter

http://www.forestry-suppliers.com/p...=7479&title=Extech%C2+Cup+Thermo%2DAnemometer

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[roydm]

Personally I have had bad experience with any type of mechanical flow switch especially in a 24,7 operation where they sit in one position for months on end. For a critical situation where the switch absolutely has to work you need to think about what will happen if the device fails.
A sail type flow switch would be ok if it was on something where it gets frequent exercise but you should check in logic that it's working for example don't allow the fan to start unless the switch is indicating zero flow, Alarm if you shut down the blower and the switch doesn't indicate zero flow when you know the flow has stopped.

My idea of using a fan as a cheap flow turbine is one I have only used on low pressure air ducts (mining flotation) for balancing the flow between cells. There was no straight pipe after a butterfly valve and elbow. The fans I chose were almost pipe size and they spun about half the speed they would under power, where else could you get a turbine meter for <$100.

 

[USAeng]

Thank you all very much for the replies... I should have clarified that these are to be permanently mounted instruments that hope will last for years. The air flow is a constant approximate 50mph in the 6" ducts. Would a differential pressure switch work? I am guessing no because it would be comparing the pressure inside with atmosphere and since the pressure will rise slightly over all the ducts it will not be able to tell if one duct is clogged or all 3 are slightly clogged...

I will look into thermal flow device online today... I have never seen this decice before. So it will work better than a pitot tube in your opinion?

 

[glowing]

I have seen several installations using a DP setup.

You can do DP across the fan, and that is generally not too bad. This won't really work for you as you are using 1 fan and have 3 lines (unless of course that on fan is a primary and then each line has a booster). You can also do this if you have a damper inline of each vent, as long as people are not out there adjusting the louvers all the time.

You can use something with a very low sensing range and measure around a 90 or something. These instruments are generally not the most reliable measurement and I would avoid it. You will often need to introduce small tubing with these, and that plugs off.

Another option is adding a fixed restriction plate into the line, if your ventilation engineer will allow it. Depending on power plant type, and its inherent regulatory bodies this could be a major pain to gain permission on. It is something I considered in the past, but the red tape was not worth it in that situation. it does not have to be a huge restriction...remember, you are not making medicine, you just want to see if is plugging. They don't have to be bang on accurate.

That being said, consider annubar installations. They are a little more costly, but they are pretty accurate, and will likely only have to be pulled 1 - 2 times a year and be wiped off. The downfall is that they do require a fair amount of straight piping. If you decide to go this way, note that is you are a little short in your piping that it will through off the accuracy, but it will still be repeatable. I have actually done fairly good testing on this.

I like annubars because they are very repeatable and pretty easy installs. You can also use it for reliable trending. This means that you can chart flow loss and help anticipate ducting cleaning, meaning you can be proactive, not reactive...bonus.

for a 6" line a Rosemount would be around 7000 with mounting hardware. I am sure there are cheaper ones, but the Rosemount are a solid unit. They also have some advanced algorithms now that can be programmed to account for installations that do not meet the full straight pipe requirements....a HUGE bonus.

 

[IRstuff]

I assume "thermal flow" also refers to hot-wire approaches:

http://www.control3.com/4330p.htm

http://www.instrumart.com/ProductLi...medium=cpc&utm_campaign=Air+Velocity&Phone=10

http://www.omega.com/ppt/pptsc.asp?ref=hhf42

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[glowing]

not hot wire in the tradition handheld sense you refer to

Kayden makes a model:

http://www.kayden.com/

so does kobold (KAL-K, KAL-A, KAL-L, etc models)

Sage makes them

there are piles of makes and models

 

[USAeng]

I am still confused on if to use a DP switch or a thermal switch... the DP switch I can prob. get from Dwyer at an affordable amount. I got a quote today from one company for the thermal and they seemed more pricey.

I guess I dont understand how a DP switch could work in this application... if the thermal switch is what we need then the price is not as big of a deal... we also need a very good quality one that will last a long time. The company that I got a quote from was called Intek. Thanks again for the help with this.

 

[IRstuff]

I think you would want positive indication of air flow. This can be easily determined by any of the proposed air speed instruments, be it hot wire, impeller, or ultrasonic, and they also tell you the actual air speed, and not just that there might be flow.

Anything short of knowing that the air is actually moving raises the possibility of an alternative interpretation, which would require you to do a thorough analysis of all possible ways that the measurement could be interpreted.


TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[USAeng]

I talked to a technical person at Dwyer Instruments today and he thought a pitot tube in conjunction with a differential pressure switch would be the route to go. Personally I like the RTD idea better, but I do not have enough experience to judge. So far, it will prob. be decided by budget between those 2 options as both seem to be a way of measuring velocity/air flow in the duct. Thanks for the input!

 

[USAeng]

Glowing, have you ever used one of the Kayden flow switches? I called today and the guy said they are adjusted using a logrithmic scale... why wouldnt it be linear? I was figuring you would put the unit in then tell it that what it saw was 100% flow then tell it to trip an alarm at 75% flow or whatever... has anyone ever used one that you could setup simple like I just mentioned?

 

[USAeng]

maybe he meant the millivolt signal compared to velocity would be logarithmic? has anyone ever set one of these up?

 

[JLSeagull]

Instead of any switch, I agree with your initial preference of a flow element and transmitter.

 

[glowing]

JLSeagull - a "flow element" is a very broad statement. anyything that measures flow has a "flow element." An orifice plate is a flow element, as is a hot wire, as is the vortex shredding bar, as is a coriolis tube, etc etc etc.

What model was the Intek meter? I am familiar with some of their stuff.

The Kayden switches are pretty solid. I prefer them over the dwyer anyways. As for the logrithmic scale, I think he is meaning the correlation between the pipe diameter, and the temp dif that equates to a velocivty, and thereby the flow. That is a non linear calc. If memory serves correctly, you zero the unit, install in duct, get flow to normal velovity, enter that velocity (may need a velocical or something like that) and then walk away.

Any instrument you stick in there is going to need to be pulled AT LEAST 1 x per year for a quick wipe down, more likely every 6 months.
DP cells will plug up.

Another option is a dopple - u/s

http://www.automation.siemens.com/w1/automation-technology-sitrans-fug1010-gas-18675.htm

I think they make models for square pipe. You would have to talk to a Siemens/Milltronics rep.

If pulling the unit out for 2 minutes 2 x per year is not an issue, I would go with the rtd. DP, in my experience, can be a huge headache. it doesn't take much debris or buildup to render your measurement useless in an application like this.