Synchronous vs Induction on a centrifugal compressor

[Lektrikman]

I have a 1000 Hp, 1800 RPM, synchronous ODP motor driving a centrifugal compressor through a 1:2 speed increaser. This motor because of its location needs to be removed and cleaned approximately every two to three years. The speed increaser has recently developed some strange vibration spectrums on our PDM analyzer and it appears the gearbox is going to need a rebuild. Due to the combined costs of having to rebuild the gearbox as well as the periodic cleaning costs I am contemplating changing the motor to a 2 pole TEFC design and driving it direct to eliminate the speed increaser as well as the need for periodic cleaning.
My question is: How do I calculate any energy savings, if there are any, between my current set up with the synchronous motor and speed increaser and a straight induction motor? There must be some losses in the speed increaser and I'm just not sure whether the synchronous motor is more expensive to operate than an induction motor, especially when you consider the exciter costs. I'd appreciate any thoughts you guys might have, Thanks, Lektrikman

 

[davidbeach]

A rule of thumb that has been posted several times is that motors with horsepowers greater than speed in rpm are good candidates for synchronous while motors with horsepower less than speed in rpm will likely be induction. So, a 1000 hp, 3600 rpm motor would most likely be an induction motor. On the other hand, it won't actually be a 3600 rpm motor, more like 3500 rpm. If you really need 3600 rpm, you would need a drive to run the motor at something over 60 Hz or a synchronous motor.

 

[aolalde]

Since the power factor (PF) of synchronous motors can be adjusted to 1.0, the operating efficiency (EFF) can be very high as compared to that of an induction motor. So the daily energy consumption for the induction motor will be higher.
The benefits of an induction motor directly coupled are the initial investment savings and service expenses for the gear box. As Davidbeach said check if your compressor accepts the speed reduction due to the slip in an induction motor.I assume this is a centrifugal compressor since reciprocant type could require an additional flywheel.
May be the overall expense after several (10) years will be similar for both options.

 

[waross]

If you're using a synchronous motor now and change to an induction motor, you may have power factor issues. You may have to include a cpacitor bank with the induction motor. If the synchronous motor is over-excited to improve plant power factor you will need even more capacitors.
Re the speed loss with an induction motor, I believe that the Horspower speed curve for a centrifugal water pump is a third power function. A speed reduction of 2% on a water pump may cause a 6% drop in work done. There are other factors such as change in dynamic head with a change in speed which tends to increase loading etc.
Can any-one out there tell us what the Speed-Horsepower relation-ship is for a centrifugal compressor?

 

[jraef]

The power factor issue is an important one to investigate fully before making a change. Many facilities put synchronous motors on large loads that are required to run continuously, because they use them as a synchronous condenser by runnig it over unity to improve the power factor for the entire factility. In many facilities that I have been in where a lot of compressed air is used, that synchronous condenser is the air compressor.

So should you switch to induction it may not be just a matter of adding a PFC capacitor bank to this motor in order to maintain efficientcy, you may be required to add capacitors to ALL of the motors in the plant!

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[motorspert]

A 2 pole synchronous motor at this output will be expensive. I'd seriously consider a 2 pole motor and a VFD. This may give you overall cost savings as you can control the pressure/flow with speed rather than with dampers. The removal of the gearbox will save energy. Modern gearboxes can be as high as 99% efficicient, the old rule of thumb for a gearbox was to use 98.5% as its efficiency. The removal of the gearbox reduces maintenance costs and has less power consumed by oil lubrication system

The VFD is expensive, but can incorporate power factor correction - if that is an issue, there may be other synchronous motor that can take up the burden.


As the correspondents state above, to make a decision you need to determine:
Is the load constant?
Is power factor an issue?





Waross
the speed HP relationship is a cube law for a centrif compressor

 

[oftenlost]

Four pole synchronous motors of recent design are wound on star punchings and tend to cost more to rewind the rotors as the poles are not removable and need to be rewound on an elevated fixture or with a shaft pit. The centrifugal forces are high and there are problems with copper extrusion and blocking. The two pole units are Vvvveeeerry expensive to rewind as they are normally wound on forgings and the manhours are high.
Two pole induction motors are problematic as ball bearings fail frequently, rotors are long with respect to diameter and tend to flex.
All things being equal and ignoring down time, I would go for induction at this HP and speed.

 

[waross]

Hello motorspert
Thanks for the information.
Lektrikman
Comparing a 3600 RPM synchronous motor with a 3500 RPM induction motor, the slip is 2.8%. When this is cubed, your loss of capacity will be over 8% {1-(3500^3/3600^3)}
If you have lots of capacity then there is no problem, but be aware that a small drop in speed is significant.
If you can withstand the capacity loss, you can make up the power factor with capactors. If you don't add capacitors, everything will work fine but you will probably get an expensive penalty on your next power bill.

 

[electricpete]

I'm not sure I fully understood the question and I didn't read the responses.

In the typical induction vs syncronous debate, I like to throw in my favorite overlooked aspect: sync motors are much more complex, more equipment to maintain, more equipment to possibly fail. In contrast the rotor of a squirrel cage motor does it's job reliably forever with no external support (excepting rare rotor bar problems).

My vote on that basis alone is to go induction. Also if you have a chance to get rid of a gearbox by putting in a higher speed motor, another great opportunity to reduce maintenance requirements and improve relaibility.



=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Another aspect of sync motors,they have torque oscillations during startup at a frequency twice slip. That frequency starts very high and ends very low. If there is a torsional resonance anywhere in that range, it will be excited during startup. Even with a brief time, this can be damaging to powertrain components.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[motorspert]

Oftenlost
Because of the high inertia the "star punching" type design would probably not have enough thermal capacity to dissipate the heat during start up and so solid pole rotors are often offered for compressor duties.

electricpete - i agree that the ocilllating torques seen during start-up need addressing - but the existing drive is synchronous, so the mechanical system would be able to withstand the oscillating torques.

There are many pole induction machines in service with sleeve bearings which have no problems with shaft flexing, so this should not be too much of a concern

 

[electricpete]

With the existing drive, they had to rebuild the gearbox due to "strange vibrations".

I am not saying by any means saying we have enough info to draw any conclusions or to say torsional resonances are at fault.. but conversely, we also can't use the existing installation to prove the system is working properly.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[motorspert]

Electricpete - well spotted, if the number of starts is quite high, that could be a reason for the box problems.

 

[mc5w]

If you do decide to go with a variable frequency drive and an induction motor do not forget to attach a universal harmonic filter to the branch circuit for the motor.

With a VFD you could run a 4-pole motor at 122 or 124 Hertz - a 4-pole or 6-pole induction motor usually has optimal design and usually has the best efficiency and mechanical ruggedness. However, a 2-pole motor started across the line has simplicity going for it and if your compressor can tolerated slightly less than 3,600 RPM that would be the best way.

 

[snox135]

Wouldn't removing the 1:2 speed increaser decrease the load on the motor? If this is the case couldn't he use a smaller hp motor? Thus increasing his savings.