Flowmeter for compressor control (compressed air)

Has anyone ever used a flowmeter to start/stop air-compressors?
For example, let's say you have 3 compressors.
Compressor 1, 10 m³/hour
Compressor 2, 7 m³/hour
Compressor 3, 5 m³/hour

If you measure a flow of 9 m³/hour, start compressor 1
Flow of 13 m³, start 1 and 3 and so on, you get the idea.

I like to use a flowmeter but am not sure if it is as simple as this...

LittleOne,

It seems to me that you would want to monitor actual current pressure.

The idea should be to maintain a particular pressure in an accumulator of some kind.

If you see pressure fall below set point, bring in compressor #1.

If you see pressure falling quickly, bring in compressor #2.

If you see pressure falling radically, bring in compressor #3.

Now, bear in mind... you don't want these compressors starting and stopping and restarting too quickly. You will wreck the motor and the mags in short order.

Reasonable control depends on the inherent stability of the system (the loads) and the size of the accumulator.

If you find that your demand varying radically then you simply must get a larger accumulator.

Once a compressor goes on it should stay on for a certain amount of time... or at least, not be able to restart for a particular length of time.

Thanks Terry, that is how I worked it out now.

But I still wanna use a flowmeter, this gives us an easier way to start/stop compressors and we get a lot of feedback about the actual use of air. Our compressors have a modulating inlet valve so it is not possible to see how much air they deliver, that's why I like the flowmeter-concept.
However, I never used that before and I never saw it used for that kind af application.
So, if anyone used one for compressor control, let me know, those things are expensive and I don't wanna buy one to put in the 'drawer'

So... you are not using an accumulator.

We are using an accumulator...

You must be using some type of Rotary Compressor since you say the flow is modulated at the inlet.

I have never seen a flow meter used for compressor control but can't see why it is not possible.

The Flow Meter would have to be at the accumulator outlet to the system. If a demand for more than one compressors flow is reached it should not immediately turn on another compressor since it could be a short lived requirement that the accumulator should have no problem keeping up with. After some time at the higher flow a second compressor would be turned on and timed to run some minimum length of time so the electric motor is not over worked. The third compressor would have the same setup.

The length of time to wait for turning on another compressor would depend on the accumulator size. A large accumulator would allow for a longer time to see if the demand was short or prolonged.

Sounds like this could be an energy saver especially if the high usage is intermittent and for short periods.

enough

For every cm/m output I think you need at least 1 to 1.5 cm/m of storage even on a screw compressor. If your running more kw of compressor than needed by using your flow control it might save some $`s. By monitoring your flow and logging it you might be able to better control air leaks also, not that anyone ever has air leaks though. It amazes me how most everyone thinks air is cheap! I like your idea! ( Which probably means it want work) Let us KNOW

I wonder why a flow meter??

I may be wrong here and making a few gross assumptions but am willing to learn.

IF you know the volume of the receiver and you know the volume of the distribution system you should be able to calculate flow with a pressure gage and stopwatch (???).
dP/dT = dVolume/dt (??)
You would have to calculate the equivalent volume of free air at each of the pressures and I guess assume the temperature is constant (or calculate for temperature change).

This would yield SCFM -- uh would it not??

SO if you put in a pressure transmitter
AND the PLC has a clock
THEN the PLC could calculate SCFM free air and change compressors as needed / desired for the flow rate.

It would also give you a method of measuring air leaked ie measure SCFM during lunch when most equipment is idle??
It basically is what is done manually ie falling pressure start another compressor -- is it not??

It also seems important to me to have a good understanding of what your plants demand for air is throughout the day. Then you could bring whatever compressors you need on line a few minutes before you need them.

Dan Bentler

We have nitrogen generators that start/stop depending on flow. Air Compressors should work also. In our case we used a Rosemount Vortex meter with a pulse output going to a AB SLC. With the pulse output you can monitor rate (freq.) and total usage for cost tracking.

One downside to vortex meters:
they do not work very well at the low end of thier range, as you approach the lower limit, the error increases rapidly, if you have a certain fairly constant minimum usage, and size you meter correctly, vortex's are a good fit, an alternative, but much more expensive is to use some type of mass meter.

I like your idea! ( Which probably means it want work)

Click to expand...

Do you mean it probably will work or will not work? Sorry, English isn't my native language...

Leitmotif,
I could calculate the requirements but a flowmeter is easier and gives us a constant update on the demand. This is a real plus, air is expensive and monitoring the demand seems the way to go.

Thnaks Ken, that's what I wanna hear. I'll give it a try, I am gonna ask for the pricing on a mass meter, will be expensive but no pressure-fall compared to the vortex meters, I think... Have to check it out.
But on the other hand, a good system will pay back for itself over time, air is expensive. We had the same thing with our NH3 compressors. NH3 temperature at the low pressure size was set at -32°C, by changing this to -30 instead of needing 2 compressors online (one at 100%, one at 30% and going on/off) now only one compressor stays on all the time at 100%. This decision could be made only based on the results we got from our scada. Hope we can analyze the compressed air in the same manner and make a good decision.
Thanks all

Better check it out!

You are going to be surprised at the price difference of a mass flow meter compared to the vortex version. I would recommend that you use a vortex meter if you are dead set on controlling the air compressors by monitoring flow rate in your system. The insertion loss of the vortex meter is negligible, so you will not see any pressure loss that will affect your system. The choice is yours, but as has been recommended, the pressure regulation method is cheaper, but be sure to let us know how this experiment works out.

We do this all the time, but our "compressors" are actually "blowers". That is, the discharge pressure is low - less than 15 psig (1 bar). Flow is high - several thousand cubic meters per hour. And in our case the air is used for process purposes and the process performance is a function of flow, with pressure not being critical.

For the low volumes you are taling about a pitot tube or venturi and a differential pressure flow transmitter may be most economical flow measurement choice. That will still be more expensive than a pressure switch or even a pressure transmitter.

What is the air being used for? For example, shop air used for pneumatic tools and such has to be maintained at a certain pressure and pressure switches are used to increase or decrease the number of operating compressors. Also, if these are constant flow positive displacement compressors then your scheme won't work with stability. You will need a pretty good sized storage tank to dampen out fluctuations.

Some more details on the application and your objective will get better answers.

It can be done with flow or pressure. We have just installed flowmeters in our plant system and power monitors on every compressor. We are calculating the SCFM/KWHr and determining at what point our compressors become inefficient. We will not unload compressors below there effective operating range. You turn off a big one and trade it for a smaller one instead.


This has been quite an endevor, but with datalogging and good graphing we are able to see things much clearer using flow instead of pressure.

This is an elaborate approach but we run about 5000hp worth of compressors and it will save us a bundle to do it "right" versus however we want.

As stated earlier, you can also track leaks using flow. And yes everyone seems to think air is "FREE" at our plant too!

If your system is small and all you want is to stage a few compressors on and off. I think pressure makes more sense. You can do it either way in my opinion. If you use flow it just takes a little more calculating and thought to set it up than pressure.

Just an opinion.

I'll think I go with a mass flow meter, pricing will be steeper than a vortex meter but a mass flow meter only gives a pressuredrop of a couple of mBar compared to 300 mBar for a vortex. Also, it works also good at low flow. (good for leakdetection).
And a mass flow meter will give me an output that I can directly related to the stated output of any compressor ("free flow air")

Tom, the air is used in a french fries production plant. Air is important and we really like to get a good idea of the air used, therefore the mass flow meter. Price is important but we produce about 100.000.000 kg a year so "good air control" is more important, management doesn't like a shutdown due to poor compressor control
And yes, we have a large storage tank.

Tomorrow I am getting a price for both vortex and mass flow meters from ABB. The engineer told me that this was the perfect application for a mass flow meter -> compressor control and combined with scada software, a nice trending of our demands.

I'll keep you guys informed, thanks for all the help.

Yelp

Quote:
I like your idea! ( Which probably means it want work)

No I really thought it would work i thought about doing it
at our plant, but like alot of other things haven`t got around to it. I`m glad you started this post as I would have used pressure instead of flow. I see now after reading the other post flow is the way to go, because you`ll end up with some data you can trend even to the point of how well your compressors are preforming and when one of them may need attention. Sounds like you have an enteresting project ahead of you. Keep us informed.

:site: