Savings in compressed air systems

[Gauss2k]

Hello everyone,

I've just started to look at our compressed air system because we'll increase the number of production machines and we wonder if we'll need an extra compressor or not. Of course, we'd prefer not

I've just read some very interesting posts about leaks detection and how costly they are. As far as I know, there is no leak detection precedure in our plant so I would like to evaluate how must air we could save if we do find and patch leaks.

We have 6 Gardner Denver compressors: A 100HP Electra-saver, 3x100HP Electra-saver II, a 60HP and a 40HP (All screw compressors). @ about 100-105 PSI

I've seen easy procedures to estimate the leaks when there's only 1 compressor (check load and unload times when the production is off), but with more than one, what can I do to estimate the leak %? Simply shut them all down except one and see its load/unload time?

Also, I would like to know if there is any central control we could use to make our electricity consumption more effective. I've seem that the Auto-sentry ES or ES+ in the Electra-saver II allows to link the compressors, but since the ES and ES+ are not compatible (we unfortunately have 2 ES and 1 ES+), the best we can do is sequence 2 out of 6 compressors together..

Finally, I want something that allows me to shut down or start all compressors remotely (example: on lunch time)

Thanks you for your time!

 

[budt]

Try these sites for good info:

http://air.irco.com/asg/air_system_info/air_system_information.asp

http://www.knowpressure.org/content/library/sourcebook/index.htm

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

http://www.fluidpower1.us

 

[MintJulep]

Gauss,

Shutting down all but 1 compressor should work fine.

You could also shut them all down and watch the rate of drop in the main reservoir.

 

[pardal]

Hi Gauss2k.
The best clean place it is not the one most cleaned, it is the least dirted.
So allways repair the leaks.
First all , on a shutoff period , as early on a sunday , walk all over the plant and pay attention to leak noise, this are the big one.
Look at the walls and ceyling for spitterd zones , as air always carries oil it will be show.
Mark all place and program a repair action.
You can test how much you loose with only one comp. runing.
If it is not enough you shall have a big open on your piping.
Do a pre repair leak time test, and do a after test to know how much you gain.

To control all compressor by demand and by time or schedulle I use a LOGO by siemens with analog imput using a analogic pressure sensor, by this way you can change the way each compresor is started, changing starting order, keeping hours running, This is a economical way, a LOGO cost about 100 U$ , you will need a aditional module for 50U$ and the sensor wich should be the more expensive.
HOPE it help

Other more complex is to use a Programable Logic Controler, so you can log all variables and so on.


Pardal

 

[Romefu12]

Gauss2K

I totally agree with the leakage investigation and repair ideas expressed here. I have performed numerous tests similar to this and find a few other things you may want to evaluate including a moisture test to evaluate the water carry-over to the system for evaluating air dryer efficiency and any unnecessary cycling which may be occurring that can account for huge losses.

Another thing to look at that may sound a little far fetched is the compressor maintenance schedule. Your screw compressors are not meant to be started and stopped but rather loaded and unloaded using a type of modulating valve for a few very good and potentially costly reasons. Frequent starting and stopping of screw compressors causes unnecessary starting contact wear for the electrical maintenance crew as well as unnecessary ascending and descending of compressed air through the dew-point which creates water. Some screw compressors are infamous rain makers and the water created causes rust corrosion of the compressor separators, water traps, air receivers, piping and any instruments or tooling connected to the air supply from the increased moisture carry-over.

I have found that for screw type compressors, longer run times and a stable range of operating temperatures maintained in the compressors and air systems can greatly reduce costs of operating as well as maintenance in any air supply system. I also agree with the cycling study and smart controls stated in the above posts by pardal. As for a means of detecting leakage a good Ultrasonics probe sensor works well.

Romefu12

 

[griffengm]

Gauss,
Just as you can find total leakage by adding the sum of leaks, you should be able to devlop a cumulative load/unload time for your system and from this, a system leak rate. Time the load cycle through say three cycles at each compressor. Sum the averages to get a total load time for the plant.
This is just my system so take it for what its worth.
One tool I found indespensible was a small flowmeter set up with quick couplings on each end so that I could check flow at air motors and subsystems. It allowed me to set flows at motors to optimum rates, detect worn-out motors and develop an over-all snapshot of consumption in our plant. IIRC, the tool was less than $200. If you use a lot of air motors, pneumatic tools, air sparging systems, it will pay for itself in six months.

Griffy

 

[Gauss2k]

Thank you for all your answers, you brought several interesting points.

Romefu12 (and others): I've indeed read that starting and stoping a compressor too frequently (everyone seems to agree that it is around 5-6 times an hour) may damage it. But if I only shut them down only when the production is not running (instead of leaving them unloaded, which still uses quite a lot of energy), would it really shorten its life or increase maintenance? If I had a central automatic control or a compressor sequence, they would anyway start and stop several times every day, so aren't they supposed to be design for that?

 

[Romefu12]

Gauss2k
It's difficult to state wheteher you would hurt these compressors by cycling as I do not have enough information to make this a definitive statement. There are many variables it would depend on including the type of screw compressor, delivery pressure, delivery volume, cycles per day, average ambient air temperature at intake etc, type of compressor fluid used (either glycol based or ester/oil based), storeage capacity, dryers in use etc. This is why a usage and maintenence study is necessary before making any capital investment in a solution

You are correct in stating that leaving the compressors running unloaded unnecessarilly would consume and waste a great deal of energy. However I can give you one example where running unloaded was more cost effiient

3 Ingersoll Rand HP 900 screw compressors (1 compressor alone delivers 320 psi @900 scfm, uses a ester based compressor fluid) when cycled 3 times per hour 1 compressor causes such a large amount of water and moisture to form I re-engineered the system with electric-timed blow-off valves, A heating element (like those used to warm a 55 gallon drum to keep the seperators above dewpoint) for the seperators and tuned the dryers and receivers to blow down on a schedule to control rust and corrossion which was causing compressor outages every 3 months that was extremely costly to maintenence in parts, cleanup and downtime.

Without these simple fixes (and other little fixes not mentioned here) these machines were better off running 24/ 7 unloaded. After the fixes, the compressor outages were scheduled at 1 year or greater intervals while still being able to cycle 2 to 3 times per hour.

This is 1 good example that a study of usage, leakage, and maintenence is highly recommended. The numbers may surpirise you when you figure out all of the costs.

Romefu12

 

[Cyclefrog]

Regarding compressor efficiency:

Both Grundfos and Goulds have controllers for demand water systems that 1) integrate multiple pumps, 2) adjust the pump's motor speed according to pressure drop, and 3) communicate between controllers so that primary duty is rotated between pumps. The result is even pressure output and highly energy efficient operation

I suspect that something similar exists for air compressors. Can some pneumatic wiz out there direct us to it?

 

[rmw]

Gauss2k,

An old ex partner of mine (ME) had previously been a power superintendent for a large paper mill. They had a large air compressor, and a spare.

Everytime his people would come to him and complain that they were out of air, and needed to run the spare along with the primary unit, he refused, and made them go find and repair the leaks.

For several years he managed to run on only one pump. Upon his departure for greener pastures, they started the spare, and soon had to have another spare.

Go Figure.

You may have plenty of capacity out in your plant already installed and just furnishing air to the leaks.

rmw

 

[pennpoint]

Gauss2k, et al.
The manangment of the company for whom I work, think that because you are compressing air then the result must then be free, and to use any way you choose. How this is I cannot fathom. However, They continually use air tools as if this sourse of motovation is free and without regard to leaks in the system. I cannot seem to get the point across that leaks equals cost.
Good for you souls that can make the point. Do not dispair there is an end to this madness.
Maybe!!

pennpoint

 

[Superskis1965]

A company called APT Honeywell makes a specialized resevior system to eliminate the need for extra compressors. We built them for them at Materials Handling Systems (MHS) 240-568-9898. They could give you the best information. Tell them Raymond sent you, I worked there for 5 years designing systems.

 

[Gauss2k]

Thanks for the info.

Fow now, I have scheduled a meeting with a guy from CompAir about their multiple compressor control "SmartAir" (

https://info.compair.com/dl.asp?section=lit&file=Smartair.pdf)

Gardner denver also seems to make some kind of multiple compressor control called the "multi-pilot" but it seems to be offered only in europe, I haven't seen any reference in US or canadian sites.

http://www.maskinteknikk.no/kontroll_system.htm

If the controls cannot be installed here, I'll continue to go through the other possibilities (Like Siemen's Logo or new reservoirs)

And just to let you know, our test for the leaks resulted in an estimated 15 to 20% leakage (changed slightly every time we did it). We'll establish a procedure to reduce that to *at least* 10%

 

[quark]

http://www.conservair.com/

These guys are US based and have very good systems for compressed air management. They have plenty of installations in India and I extensively checked them before installing their pressure controllers for our systems.

You may find some good posts about leak management in this forum. 10% is the ideal leak rate, in my opinion, without sacrifying your precious time. As already noted by you in your earlier posts, it is better to check the leak rate in the distribution system with one compressor operating.

PS: Always try to operate your plant at the lowest possible correct pressure.

Regards,


Eng-Tips.com : Solving your problems before you get them.

 

[pascalvp]

You could install one or more flowmeters in your main air supply lines and measure the actual compressed air consumption of your plant.

The air consumption can be logged with a data logger. After a few weeks you will have an exact picture of the consumption during production. During lunch time, you can see the total amount of air leakage.

Regards,

pascalvp

http://www.vpinstruments.com

Mass flow meters for compressed air