Find the cfm of air compressors 2

[isutechman]

I'm doing a summer internship this summer, and one of my tasks is to do a compressed air study for the company to determine how much they are losing to leaks. The company I'm working for has very poor records of any of their equipment so the only information I have is what I have been able to get of nameplates of the compressors (we are talking about 1950’s 2-stage reciprocating compressors). They do not know what their supplies needs are other than it takes 7 or their 8 compressors to run the plant.

My 1st idea it is figure out what the total compressor capacity is, but I have not been able to come up with a good way to do this. I have heard a lot about draining air tanks and timing how long it takes to fill them up, but this is out of the question with the way the system is piped and the plant’s 24-7 operation. Are their any ideas of the best way to calculate the cfm of the compressor?

Here is a sample of the data I have from the nameplates:
Brand: Ingersoll Rand
Model XLE
Size: 20.5x12.5x8.5
Year: 1957
HP: 250
RPM: 514
And it has un-loaders, but almost never starts to unload.

Any other ideas about ways to determine the need for equipment and so on would be of great help. Thanks

 

[berkshire]

isutechman (Mechanical)
If you have the nameplate data, why don't you call the manufacturer and ask them? Ingersoll Rand are still in business. I realise that you may also have other brands of compressor, that some of them may be worn and not producing like they did when new, however the manufacturers are still your best bet.
If you can isolate compressors you can do a calibrated orifice check to determine CFM delivered.
B.E.

 

[TheTick]

I would start by monitoring input and output (CFM@pressure and amperage).

Hopefully there is some flow measurement device in place. If not, install one. The most "minimally invasive" device to install would be a simple pitot (make sure its on a straight run).

Measuring amperage for power is a bit rough, since you don't know the actual efficiency, but it's better than nothing.

 

[crjones]

You are starting at the right spot by cataloging your equipment. Do both the compressors and what is consuming the air. On the equipment using the air also note the pressure requirement too. An air tool is designed for 90 psig a drop in air pressure to 80psig loses 20% of its efficiency, so it works harder wears out and leaks. By the way an air leak .062" in diameter at 100psig is roughly a 10 cfm loss to give you an idea of what is ahead. As was said contact the manufacturer of the equipment or a rebuilder to determine air production and consumption. Take care of leaks and adjust the system pressures to the equipment. Once you've done that your ready to dig into the performance questions.

 

[MintJulep]

Tank pump-up is ok as a comparison, but as it is non-steady state it is difficult to work out an actual air delivered number.

Calibrated orifice is the standard method of test. If you are using 7 of 8 compressors, test #8, then put it on-line, test #7 and so on.

 

[Warpspeed]

I would just simply measure the air volume at the compressor air intake. Depending on how accurate you want it, various methods could be used.

The easiest is probably a simple hand held anemometer as used by the airconditioning people. This will read velocity in feet per minute, and if you know the area of the intake duct, air volume is simple to calculate.

Something like this perhaps ?

http://www.pat-training.co.uk/amemometer_airflow_meter.htm

 

[griffengm]

isutechman,
I agree with Berkshire about contacting the compressor manufacturer.
Second step is to catalog as many of your air tools as possible along with an approximate use rate (minutes used per hour or some similar number. If you look in most any catalog of air tools it will often have a use rate for these tools and you can work up a needs number from summing these.
Third step is getting some idea of peak demand, i.e. is there a time during the shift when all, most, or a maximum number of tools is in use.
One of the issues we had was how to determine when an air motor was worn out. For that, I bought a Hedland(sp?)flow meter that I set up so I could pop it into a line at most any location and check flow to an individual motor. Did we ever get a surprise as to how much air was being used!
As I recall, the meter was less than $200.
Hope this helps.

Griffy

 

[isutechman]

Thanks for all the ideas so far, it will be helpful, I called Ingersoll Rand, Sullair, and Gardner-Denver to get info about their compressors. So I hope they will come through with that. I still have 2 Worthington's and 1 Cooper Penjax compressor that I need specs. for, any ideas where to find info about them anymore?

The End user side is going to be a challenge; the factory was 1st build in the late '30's and was added onto until the '70's so piping plans are pretty bad. Also it is a tire plant so most of the equipment is very old and like the compressors most of the information has gone missing over the years.

One other question, would assuming about 85% efficiency be a good approximation for the compressor motors, they are all old and have been rebuild once or twice.

 

[budt]

As a general rule most compressors produce 4 CFM/HP at 100 PSI.


Bud Trinkel CFPE
HYDRA-PNEU CONSULTING, INC.
fluidpower1 @ hotmail.com

http://www.fluidpower1.us

 

[jet1749]

Recently had similar problems regarding compressed air usage and available capacity. Our plant was early 1960's vintage and systems had been added without real concern regarding capacity. Maintenance of the distribution system was non existent, also maintenance of the consumers were patchy. Working on the basis that major investment was not an option, and we had to work with what we had, our motivation was primarily energy cost reduction. So we tackled the problem on a number of fronts.
1. Fit a temporary datalogger to the system to analyze air demand and compressor loading in order to get "before and after" verification
2. Track down the large leaks and repair
3. Use portable ultrasonic listening device to locate smaller leaks, tag leaks and collate the information.
4. Plan work orders to repair the leaks, use slack maintenance periods to chase the backlog of leaks.
Over a couple of weeks we had reduced demand to such an extent that our three compressors which were previously all running on load, were now running 1 loaded, 1 part loaded and 1 standing by.
Calculating compressor discharge capacity can be difficult as it varies according to discharge pressure, however piston displacement is easier.
CFM= HEA X S/12 X rpm for single acting cyl
where:
HEA head end area
S stroke (inches)
CFM= 2 X (HEA-RA) X S/12 X rpm for d/acting cyl
where:
HEA head end area
RA rod area
S stroke (inches)
As compressed air is not cost-free it may pay to maximise your capacity by reducing your leakage demand, and then see what capacity you have regained

 

[isutechman]

jet1749

Thanks for the equation, your problems sounds like the exact issue I'm having.
One question about that equation: If I have a two-stage compressor do I do the equation for the 1st and 2nd stage cylinders and then add them together? I ran this for one of our Worthington’s and if I don't do it that way the cfm comes out very small for a 200 hp compressor.

Thanks

 

[toothless]

With very old equipment I would avoid assuming efficiencies. Those efficiencies could be really bad after all that time.

Could you rent a compressor to allow you to take one or more of the other compressors out of service to fit them with flow meters?

 

[isutechman]

Toothless

I wish that was the case, but the company is cheep and about all I get for this study is what ever info I can find about are user side,and supply side and a leake detector! But for a 1st round study that might work out ok here, we think we have about 40% leaks, so just me going around with a leak dector is a 1st big step for this company!

 

[jet1749]

isutechman
use the equation for the 1st stage, as the capacities are not cumulative i.e. the 2nd stage only increases the pressure.
A good reference work is Reciprocating Compressors Operation and Maintenance by Heinz P Bloch published by Gulf Publishing

 

[griffengm]

One technique we used for finding leaks was to wait until production stopped for the day or weekend, pressurize the system and walk through the plant tagging leaks. Have maintenance fix them the next workday and repeat.
The cheapest meter we found was a squirt bottle of soapy water but many leaks were overhead so it was of limited usefulness.

Griffy

 

[berkshire]

isutechman (Mechanical)
I am posting a link to an article on compressed air management. It describes the situation from the facilities managers point of view. Hopefully you have this guy on your side. By now you will have already discovered some of the items mentioned and the others make interesting reading.
B.E.

http://www.mt-online.com/articles/06-99ca.cfm

 

[EngM02]

Hello,

I think you can put a Venturi nozzle at the output and measure the Pressure difference, which is directly correlated to the rate flow

Regards

 

[ThirdString]

This information doesn't address the question at hand, but may be useful for someone else who is reading this thread:
EngineeringToolbox.com has some compressed air information. A few pages I found useful there:

http://www.engineeringtoolbox.com/air-consumption-tools-d_847.html

http://www.engineeringtoolbox.com/compressed-air-pressure-loss-d_1014.html

 

[blfarrar]

If you can get the makers data for the compressors that would be a good start.
You need to access a UK government "Department of the Environment" department doucument
Good Practice Guide 126 - Compressing Air Costs

Contact
ETSU
Harwell
Didcot
Oxfordshire
OX11 0RA
tel 01235 436747

Cost was £3 for overseas customers + £6 postage

this gives a method for assessing the amount of air leakage that doesnt use data loggers.


Bruce L Farrar.
Works Engineering Manager
Marshalls Mono PLC.Brookfoot Works.
Halifax W.Yorks UK

 

[IRstuff]

Have you determined that there are leaks? Can you insert a digital pressure gauge and record the pressure readings during the day or night? A series of sharp sawtooths would indicate a pretty drastic leak, while flatter sawtooths would indicate slower leakage. Naturally, you'll need to distinguish between usage and leakage.

You ought to be able to infer something about the overall leak rate by noting the amount of droop down from the control pressure over a 1 min period.

I would think that there would be a way to determine the air consumption without even knowing the cfm of the compressors. What you would need to know is the capacity of the reservoir. Let's say that you have 100 psi as a control pressure with a 100 gal reservoir and a bunch of plumbing. If the pressure drops to 90 psi over a minute, you could crank the ideal gas law to determine the amount of out flow from the reservoir as a crude indication of the amount of leakage.

TTFN