Recommendations for a Reliable Compressor from a Reputable Firm. 1

[MacMcMacmac]

I am in the early stages of seeking out a new compressor to replace our current 50 year old unit.

The compressor we have right now is the original 10 stage, 8MW, 32,000cfm (40lbs/sec) unit provided by GEC England, to provide air for our supersonic blowdown wind tunnel. While it has provided a supply of trouble-free, high quality air, there is a growing need to consider its successor. The compressor was installed at the facility when it was built in the late 1950’s/early 1960’s. It has been carefully maintained and operated, but the passage of time cannot be dismissed, and there are some issues with the integrity of the main castings that are signaling that its service life may be drawing to conclusion.

The following operating conditions are given just in case anyone would like to know:

Inlet Temperatures: +40c to -40c
Altitude: Approximately 375ft above sea level
Cooling water temperatures 10c to 25c
Relative Humidity: 2% to 100%
Discharge Pressure: 330psig

Our current compressor is powered by an 11,250hp synchronous motor at 7000VAC. It consumes 8MW under full load. Starting is by a reduced voltage arrangement using a Korndorfer auto transformer circuit.

I have put our feelers to several companies who build units of this class, but of course, they will always tell you that their machine is the best. So far, Cameron Compression Systems in Buffalo has given us the most detailed response. I know they have been in business for many years as Joy, Cooper-Cameron, and now just Cameron, so I would suspect that their machines have proven themselves over the years. I have also received general responses from Siemens, and Elliot Turbocompressor, and I am awaiting responses from Ingersoll Rand (Centac), Atlas Copco, and Aerzen, which is the only screw unit I have inquired about. Are there any other companies you would recommend?

I would also like to ask a few questions:

Does anyone on the board have direct experience with Centacs? They seem to be nicely packaged, but I am getting mixed reviews on durability and maintenance. IR has been around quite awhile, but with 15+ years of experience working with their smaller screw and piston units, I am not impressed with the direction their design philosophy has taken (i.e. smaller, faster cheaper). I don't know if this would apply to a class of machines like the Centac. You can only go so cheap on something like that before your rep turns to mud.

Do integrally geared, multistage, open impeller compressors pose any additional maintenance issues over a "train-style" horizontally split closed impeller machine? The compressor we have now is the latter type, driven by a 1200rpm motor. The RPM is increased through an epicyclic gearbox to turn the first 7 stages at 4500rpm, then through a second epicyclic gearbox to turn the last 3 stages at 10,000rpm. These seem like fairly conservative speeds compared with an integrally geared, open impeller unit, which has probably contributed to the longevity of our original compressor. That and the fact that this machine has been designed by the "throw more steel at it if in doubt" school of engineering. It just seems like slower is better to me, and the thought of a 20,000rpm impeller makes me a bit nervous.

Do you think a rotary screw unit capable of this flow and pressure is a good idea? I never knew one that size existed until I looked at the Aerzen website. We have had a small (1500cfm) diesel drive portable Kobleco oil-free screw providing air for smaller projects when our main unit was down for repair, and the temperatures it ran at was somewhat alarming. It was also a fairly noisy unit. I am not keen on the capacity for destruction a screw of this size poses if the rotor timing went off. It seems like major repair work would also be quite challenging, trying to open up such a large air end compared to a gear and rotor arrangement of a centrifugal unit.


Sorry for the long post. Any advice and direct experience/tales of horror you can offer is greatly appreciated.

 

[zdas04]

I wouldn't put that much hp into a timing gear (or chain). Dry screw's would be just about my last choice for this application (you'd have to do the job in at least 3 stages, so your control scheme would get really complex really quickly).

One thing to keep in mind is that nothing has the energy density of a centrifigul compressor. To get to your 8 MW with a recip or a flooded screw would be at least twice and probably 3-4 times the floor space. Even if you have room, you're doing about 23 compression ratios so with a flooded screw you'd want a 9.4 VI which I haven't seen in anyone's catalogs (probably doable, Howden has some amazing engineering capability, but not cheep or quickly available).

I'm thinking that in your shoes I would be looking exactly where you started--a dynamic compressor.

David

 

[jbattershell]

dear Macmcmacmac,
as the previous principal engineer for Cameron compression in Buffalo, I am a little biased towards that product line. I am no longer with the company and can speak truthfully.

a integrally geared centrifugal compressor for your application is an easy job for this type of compressor. the largest users of this type of compressor, the air separation companies, rely on this compressor to run continuous, 24/7/365 for decades. provided that you filter the incoming air, and do regular checks of the oil with sampling, and regular oil filter mainenance, this type of compressor will last for decades.

do not be afraid of the rpm of the pinions on an integrally geared compressor. it matters little in the scheme of things. the rotational energy is high in both types of compressors, whether it is a large single shaft compressor with large wheels spinning relatively slowly or smaller impellers on smaller pinions spinning faster.

I was not a fan of the IR machine, but you must read up and research your self on the reliability of the IR compressor.

in general, an integrally geared centrifugal air compressor will be less expensive than a single shaft centrifugal compressor.

no compressor of this size is going to be quiet. but, in general, un-attenuated screw compressors are noisier than un-attenuated centrifugals.

if your application is a wind tunnel, I would recommend only oil free compressors.

 

[pmover]

Sulzer . . . a CAES design i worked on had Sulzer compressor units. then again, the proposed units were driven by ~50MW motors. no slouch outfit! if i recall, there RBZ type compressor may be suitable . . it has internal coolers.

why not an AFC? i recall seeing a control system design for the large wind tunnel site in Ohio and it was 3 AFC's in series. may/will need cooling between AFC's.

always request for references & investigate. never buy serial #1; buy proven machinery! never listen to the salesperson - they only appeal to your senses.

good luck!
-pmover

 

[MacMcMacmac]

Thanks to all for the input. I realize everyone has to be impartial with their opinions on the board, so your first hand experience is gratefully acknowledged.

Oil flooded compressors were never an option due to the sensitivity of the dryer desiccant to contamination, and contamination in the test section of the tunnel. I never got a response from Aerzen at any rate, so a screw is not a serious option. Speaking of which, I sometimes wonder why companies bother to put a "Contact Us" tab on their website. After drilling down several layers to find the area sales rep, it is somewhat disappointing to not even get an email of acknowledgment that your request was received. I know in the grand scheme of things this might not be a particularly huge contract to go for, but I thought it would generate a bit more interest than it has.

Cameron came though with a pretty detailed response, and the price seems reasonable.

We also have a line on a very low hours (<100) Cameron unit sitting fallow. Our institute bought its twin a few years ago from the same company, and it has provided trouble free air for 7 years now. It is a somewhat smaller unit, providing about 35lbs/sec, but the potential cost savings may make it a worthwhile purchase.

 

[MacMcMacmac]

As far as axial-flow units go, I was under the impression that they were quite a bit more expensive and more prone to unstable operation. There is a constrained amount of floor space to work with as well, so I felt a centrifugal would be more space efficient. The plan to date is to keep our current unit as a back up, since its lifespan in that role would be almost unlimited, and it would prevent costly downtime due to mechanical failures or maintenance on its successor. It is also tightly integrated into the current piping, cooling and electrical systems. Scrapping it would be a costly endeavour in itself.