Low SG manometer fluid

[Warpspeed]

Rather than using a shallow inclined water manometer to measure very low pressures, I would like to try a vertical manometer. This would reduce the errors due to slight change of inclination.

For this to be effective it would require a suitable fluid that has a specific gravity only a fraction that of water.

I am a chemical ignoramus, so could anyone here suggest something that is reasonably inert and having fairly low evaporation rate?

 

[ihg]

You can purchase a manometer oil from mffr. of manometers - (eg Dwyer) 0.71 SG from memory. Anything lower than this is likely to be volatile. Using an inclined tube manometer AND oil is the usual path if you aren't into precision electronics. They do require a level bubble to set up, though

 

[Warpspeed]

Thank you ihg,

That is what I thought, The Dwyer fluid is certainly worth a try and is probably the best that can be found.

Actually I am an electronic engineer and it is because I am fully aware of the problems, and am just heretic enough to believe a fluid filled manometer is going to be more robust, and have far fewer long term problems than anything electronic ! ! !

Cheers, Tony.

 

[hacksaw]

manometers are trouble free until you blow the column. then you have the problem of insuring that the correct fill fluid is used, etc.

if you want reliable monitoring use a transmitter, one correctly installed.

 

[Warpspeed]

Have to agree, a sudden unexpected overpressure can easily cause an inconvenient fluid loss.

I plan to use a reservoir at each end of the manometer so that the fluid will just drain back. That is another advantage of the humble manometer. A pressure transducer often is never quite the same again after being over-pressured (or dropped).

 

[quark]

Why can't you try for magnehelic gauges from Dwyer? The advantage is that you won't have backlash errors because of the absence of internal gears.

Regards,

 

[Warpspeed]

A good suggestion quark, but I really require a manometer with a square law scale. It will be measuring pressure drop across an orifice plate. The reason for the low SG fluid is to make the scale as long and steeply inclined as possible for ease of reading and repeatability.

I have seen magnehelc gauges measuring low pressures in flue pipes and air-conditioning applications, but have not the faintest idea how they work.

 

[Assumptions]

What pressure of the measurement?

 

[Warpspeed]

A full confession is in order......

I am building a homemade automotive type (air) flow-bench, and require an orifice plate and manometer to measure airflow. I need to easily and repeatably measure flow variations of perhaps one percent.

Accurate absolute measurement accuracy is not required, only good repeatability for measuring and recording changes.

The traditional way, is to use an inclined water manometer with a very long scale, perhaps three or four feet long in order to have a nice open easily read scale.

However, the shallow incline means that leveling the manometer becomes critical.

Ideally a vertical manometer eliminates the leveling problem, but the scale then becomes very short, perhaps only six to twelve inches long over the working range.

If I could find a suitable working fluid that has a long term repeatable SG of only perhaps 0.2 or 0.3, it would solve my problem and make me very happy.

I can always build in an accurate leveling system, but if I can avoid this without creating other problems it would simplify things greatly.

 

[mtrehy]

Hi guys,

I'm at the start of building my own flow bench and I found some great instructions from popular hot rodding

http://www.ag.auburn.edu/users/gparmer/articles/flowbench2/

There is just a couple of points I hoped someone could clarify:

Basically I want to understand the maths, from what I understand you have 2 manometers, the first one is used to measure test pressure - this 1 is plumbed into or under the test stand so it measures test pressure which you wind up to 28" of water (which I worked out to be 4660lb/ft^2).
The second manometer measures the differential pressure across the orifice plate between the plenum chambers, what I would like to know is how the flow in CFM is calculated from this differential pressure. It seems as if the variables that I know are the test pressure and the orifice size.
Also how do I set up my inclined manometer to accurately display the percentage flow rate and how do I produce the scale required?

Thanks in advance for any help....

m_trehy@hotmail.com

 

[Warpspeed]

Hi mtrehy, the orifice plate and flow manometer are the heart of a flowbench, and are probably the most difficult part to find information on when beginning a project like this.

The flow manometer just measures the pressure drop across a plain thin sheet metal orifice plate. The relationship between pressure and flow will always be square law. So if you double flow, pressure goes up four times. Triple flow, pressure goes up nine times. This relationship is exact for any sized orifice.

So what you do is make yourself a a flow scale in percentage flow. First you need to decide at what pressure you are going to use as the 100% flow reference pressure across your orifice plate. It can be any pressure you decide to use. A lot depends on your available blower pressure and flow. It takes a lot of horsepower and electrical power to generate high pressures and flows, so usually the orifice plate pressure drop will be considerably less than the test pressure across what you are measuring.

If your test pressure is going to be 28" of water, and your flow manometer 12" of water at 100% flow, your air blower must be capable of at least 40" over the desired flow range of the bench.

Now a 12" manometer scale will be a bit cramped, so you incline a much longer manometer tube so one end is twelve inches higher than the other.

It does not matter how long your scale is, or the angle of inclination you end up with, or the final orifice flow pressures you decide. The same square law scale is required.

So suppose you decide that your scale is going to be two feet long. At 100% flow the fluid goes two feet along the sacale. At 50% flow the fluid only goes six inches.

Remember doubling the flow (50% to 100%) causes four times the pressure drop (6" to 24&quot. So you mark off your scale for all flows from say 40% to 100%.

The way you calculate the scale increments would go like this, (for 51%) 0.51 x 0.51 = 0.2601. And 0.2601 x 24" = 6.2424".

If your scale length was 500mm instead, it would be 0.2601 x 500mm = 130mm.

So your 50% division would be at 6" and your 51% division would be at 6.24". You just keep going up to 100% or even beyond. 101% = 24.4824".

I prefer metric myself. If you make your scale say 600mm,or 900mm, it is a lot easier to find accurately 301.5 mm for example than mark off some strange fraction of an inch.

You can use this same scale for any orifice plate, that you finally decide to use, and any orifice reference pressure, by simply tilting it the right amount.

The flow through an orifice plate in CFM will come pretty close to 13.55 multiplied by the square root of pressure in inches of water, multiplied by orifice diameter squared.

So if you have a two inch diameter orifice, and your 100% flow reference point is sixteen inches of water:

CFM = 13.55 x square root of 16" x 2" squared. Your orifice should flow 13.55 x 4 x 4 = 216.8 CFM

The actual flow you get might vary quite a bit because the edges of your orifice might be slightly rounded or have burrs. Also the air upstream needs to be fairly undisturbed. Also measuring the static pressure of a moving airstream is not all that easy either.

There is a very good book, Practical Gas Flow by John Dalton, ISBN 0947981330 it is a good read.

Cheers, Warpspeed.

 

[mtrehy]

Thanks Warpspeed,

I have some issues with this however.

Looking at the MSD flow bench

http://www.ag.auburn.edu/users/gparmer/articles/flowbench2/

This uses an 1 3/4" orifice to flow 180 CFM at 28", using your calculation this orifice should flow 220 CFM.

Do you know how the following equation is derived?

"CFM = 13.55 x square root of 16" x 2" squared. Your orifice should flow 13.55 x 4 x 4 = 216.8 CFM"

Basically I think I understand most of it but I want to know how to design and calibrate my orifice plate I think that I will only be making a small bench (similar to the Superflow SF110/120).

Thanks again.

 

[Warpspeed]

Calibrating the orifice plate is THE big question for all of us.

The formula that I gave will give pretty good results if the air upstream is undisturbed. Now that is a very big if.

In most flow benches the air travels through the test piece first, and then through the orifice plate, so the orifice plate sees turbulent air. If the air is swirling, or blasting straight at the orifice in a jet, or blowing straight across the orifice at an angle, do not expect the flow coefficient to be the same as with undisturbed air upstream of the orifice.

Take the hypothetical case of two identical orifice plates both with two inch holes. The first is used as a test piece and has undisturbed room air entering, and after the air has passed this it becomes turbulent. It then flows through a second measurement orifice which is also two inches, then into the suction side of the blower.

You might be surprised to find that the pressure drops across each orifice are NOT identical. Rather a lot depends on the internal flow bench characteristics.

There absolutely must be a large plenum space up stream of the measurement orifice, and possibly other design features to prevent high velocity air from reaching the measurement orifice directly. The smaller your flowbench, the more difficult it is to do.

Another approach is to just use a calibrated test orifice proven on several other flow benches, and use this to calibrate your own orifice within your flow bench. You might find that your calculated 200 CFM orifice actually flows 172, or 213 or something.

I am now on my third design, and I will not be at all happy until I can get two orifice plates to flow exactly the same in either location. In my case bench size is of no importance, so there is about a cubic yard of plenum volume up stream of the measurement orifice, with baffles and mesh screens to break up turbulence.

But no matter how you end up doing it, you will need to do a lot of testing, and you will need to compare your flow figures to something you know is pretty accurate. Do not expect a paper design to be spot on first time.

There is a Forum with many like minded people that might interest you greatly. Recently it has had to shut down because of heavy spam. The original, now non functional forum is located at:

http://www.tractorsport.com/

http://wwwboard.html

All the old posts that contain a wealth of information have now been removed. However these posts will eventually be transferred to the new forum located at:

http://tractorsport.com/flowbench/

You can post there now, and ask these same questions. I post under the handle of "Tony". It will help to get the new Forum going, as there are only two posts there now, and it needs some input to really get going again.

See you there, Warpspeed, aka Tony.

 

[mtrehy]

Thanks Tony / Warpspeed,

I posted this stuff where you said and I also ordered the book you recommended (thanks for the help and advice).

I agree with what you say about the turbulent air affecting the flow etc. etc. but what I don't quite get is why the formula doesn't seem to work for this MSD flow bench and how they have come up with the orifice sizes that they specify.

What I really want to build is an improved copy of the SF110. I'm not exactly sure how this bench functions as it appears (externally at least) that the air is sucked through the inlet, then blown through the orifice at the top of the bench (do you know if this is correct?). The bench is tiny and I guess it only uses 2 vacuum motors which must be extremely tightly packaged and I can't quite understand how it changes from inlet to exhaust.. If anyone can offer a more complete description of the inner workings or even better - some pictures, I would be extremely grateful.

The MSD bench uses the following orifices at 28"

28 CFM = 1 x 5/16" dia
56 CFM = 8 x 5/16" dia
104 CFM = 1 x 1 1/4" dia
180 CFM = 1 x 1 3/4" dia
255 CFM = 1 x 2 1/8" dia
345 CFM = 2 x 1 1/2" dia + 1 x 2" dia
425 CFM = 2 x 1 7/8" dia + 1 x 3/4" dia

I would love to know how this is worked out mathematically.

 

[Warpspeed]

There has been a lot of discussion about the orifice sizes mentioned in that MSD flow bench article. Unfortunately the previous posts are not available, but should be re-posted.

The consensus being that the MSD design has a few problems. The main one being that the measurement orifice is directly below and inline with the cylinder head, and sees turbulent and unpredictable flow conditions.

The cure is to move the measurement orifice so it is no longer directly below the hole in the top of the flow bench. Also the orifice sizes given in the article are not right, as you have discovered yourself.

Anyhow, we can discuss this further, and others that have actually built this project will almost certainly join in. See you at the Flow Bench Forum.

Cheers.