VFD on a Synchronous Motor

[Zoobie]

I have an existing 9000 hp/1200 rpm synchronous motor (compressor drive) that I wish to couple to a VFD. Someone told me that this was a very unusual arrangement. Does anyone have any experience with this type of setup? What are the pitfalls or things to watch out for? Problems with the exciter?

A new induction motor is a relatively small cost compared to total project (VFD + other compressor stuff). I would like to use the existing motor but am I better off scrapping the synchronous motor and putting in an induction motor?

I'm just a dumb chemical engineer so besides basic theory, synchronous motors are a bit of a mystery to me.

Thanks in advance for any advice.

 

[MVRockwell]

Dear Zoobie,
Synchronous motors with VFD are not necessarily unusual, however they are a little more complicated. What makes things interesting is that synchronous motors have a wide range of designs. For example there are two basic types, brush and brushless. This refers to how DC is applied to the rotor once the motor is near synchronous speed. Brush type have the DC source external to the motor. Brushless have its DC source contained within the motor rotor/shaft. Before you proceed in trying to get a VFD quote, you'll need the following information as minimum.
* Type of Synchronous motor (Brush or Brushless). Refer to motor datasheet
* Motor and Load torque speed curves.
* Type of exciter design (SCR, diode). Refer to motor datasheet if Brushless or existing motor controller drawings
* Type of regulation (current, voltage, or power factor). Based on application and preference
* Starting Mode (i.e. starting with DC applied yes/no). Based on application and preference.

If voltage drop is an issue, you'll need information on the power source you'll be connecting to.

The above information should be enough to get started, but you'll quickly find that more homework or discussion with a VFD vendor is required.


Background
Synchronous machines are generally made up of two motors. A synchronous motor, and a smaller induction motor. The smaller built-in induction mtor act as a starter motor to get the rotor close to synchronous speed. Once the motor is close to synchronous speed, a DC source is applied to the rotor either from an external source (brush type) or internal source (brushless) which is derived from a small DC or AC pilot field on the stator. This DC source than creates a magnetic field on the rotor, which is indepent of the slip, and follows the stators rotating magnetic field. An AC pilot field is more rare and usually on synchronous motor without a "starter" induction motor. During the starting of a synchronous machine the DC terminal are shunted with a resistor (Discharge Resistor).

 

[swgrmfg]

Zoobie:

What you're looking for is called an LCI, a Load Commutated Inverter. Siemens, GE, ABB, Alsthom, Toshiba manufactures them, among others. They've been around for a good 15 years now. Expect the cost to be around USD 500K.
Try calling one of them if your budget allows and your synchronous machine is in fairly good shape.

 

[mc5w]

You also need to be very careful because usually and inverter and a synchronous motor need to be built to match each other. An ordinary synchronous motor will not tolerate the pulse width modulation and high frequency unwanted current and voltage from a VFD.

Every AC servo drive that uses and inverter to drive a permanent magnet or externally excited synchronous motor has to have the motor and drive built specifically for each other. The only deviation is that 2 or 3 sizes of motors can use the same inverter design by changing the personality prom the customizes the drive to fit the motor.

You probably are better off to modify your autotransformer reduced voltage motor controller with some extra contactors that add starting voltage steps. Something like 65%, the 80%, then 100% voltage instead of 65%/100% or 80%/100%.

 

[TomG33]

Zoobie,
a question about the application first.
how has the need for the variable speed drive come about. Do you need to run the Compressor at different speeds to suit the Process, or are you looking for some type of "Soft Start"?
As mentioned by swgrmfg the LCI is probably the best way to achieve variable speed control of the Synchronous motor. e is right about the cost but a little short on the Length of time they have been around, More like 30 years or so rather then 15. I have been playing with them for 25 years and they were around before I started. )

With an LCI, The Stator current is controlled to match the load torque as well as to regulate the speed. So during a start, the stator current only ever goes up to whatever level is required to accelerate the load, no large inrush. It can also deliver full torque right throughout the speed range of the motor.

The motor field current is also controlled in coordination with the Stator current to achieve the appropriate motor excitation for any particular speed and load condition. the drive does not care weather it is a Slipring motor or a Brushelss motor, either can be accomodated by adjusting some "Parameters"

The other method of variable speed control of a synchronous motor is called a "Cycloconverter" This would not be the right choice for an application such as this because it is mainly designed for slow speed drives on Multiple pole synchronous motors where Synchronous frequency is way less then the line frequency.


A variable speed drive of this magnitude, no matter what technology you choose, is going to cost you a bit more then just a few "boxes of Beer" so take a look at the sites of the suppliers mentioned by swgrmfg, and I am sure they all will be interested enough to send an expert around to talk in person to you about it.

Incidently, for a large load such as this, The efficiencies of the Synchronous motor will be better then that of a similar sized induction motor, and at this scale even a tenth of a percent adds up to a lot of Dollars each year.

Tom Grayson

 

[Zoobie]

TomG33:

The synchronous motor drives a recompressor in a cryogenic natural gas liquids extraction plant. Operating this at variable speeds will help to operate the plant at higher efficiencies under different upstream and downstream process conditions (namely gas compositions and natural gas demand on the pipeline system on a daily and seasonal basis). The other thing we want to be able to do is to start slowly. If the compressor were to be brought on quickly (even under a lot of recycle), at best liquids production would increase too fast causing a process upset and at worst the plant would freeze off with CO2 solids.

Thank you very much (all of you) for your comments and suggestions. I have a quote from Siemens with three options which I am trying to digest on a Monday morning before my first cup of coffee.

Tom, thanks for the tip about the motor efficiencies. One of the options includes a new induction motor with an attractive price...I will look closely at the efficiencies so that operating costs are estimated carefully.

 

[motorspert]

Zoobie, there are a lot of othr issues you need to consider bofre leaping in to a VSD. The resomendations above are for LCI drives which put considerable current harmonics into the motor, I have suffered the pain of these where we ended up replacing 4 rotors to prevent harmonic overheating, also the harmonic currents cause pulsating torques which can wreck gearboxes, etc. I am sure the guys you are dealing with are familiar with the effect of harmonics I can point you to an excellent paper on the subject if you are interested.
As regard efficiencies, a synchronous motor on an LCI drive isnt very efficient as it has to run at UPF to work. Maybe look at a induction motor on a PWM type drive, or fit a drive which is not so arduious on the motor, i know Robicon amongst others do one.

Hope this helps

 

[dpc]

Historically, LCI drives and synch motors were for a long time the only viable option for large ac drive applications, so this technology is nothing new.

However, slapping a new LCI drive on an existing 9000 hp motor is a not a trivial matter. As mc5w mentioned, the synchronous motors used for these applications were typically a special design, made for LCI duty.

I would work with the drive manufacturers and motor manufacturer to get some assurances that this application makes sense. I'd also put the burden of proof on the drive supplier in terms of harmonics, performance requirements and an extended warranty on both the drive and the motor.

 

[MVRockwell]

Zoobie,

As you can see you have a few options and issues. As a drive supplier, like others, we have used VFD on various types of synchronous motors. For your application (9000HP) here are some drives that can work (granted discussion are required with the various vendors)

*

http://literature.rockwellautomation.com/idc/groups/literature/documents/pp/7000l-pp001_-en-p.pdf

*

http://www.geindustrial.com/products/manuals/GEH-6373.pdf

*

http://www.abb.com/global/abbzh/abb...e=us&m=9F2&c=5AB302C03E220BF9C1256EDF003CA63D