Synchronous Motor controlled by a IGBT NPC Voltage Source Drive 2

[JBUDA54]

Hello everyone!

I am currently writting a technical proposal for a Project that I am trying to get Capital funding for. The proposal was reviewed in house and there was a couple of good comments that came from the review. Let's state a brief description on the project... We currently have an old Load Cummutative Inverter Drive Controlling a 1500HP Induced Draft Fan synchronous motor 2000V. This is 1980 technology and a retrofit of the controls or a drive replacement needs to be accomplished due to the fact that the Power Boiler can not run without these fans, thus the mill goes down. I have read from GE technical brochures that the LCI drive has poor PF at low motor speeds. The drive they are proposing to sell is a MV IGBT NPC Voltage Source Drive which they say has high true PF over all speeds. Here comes my questions... 1. Can you produce VARs when a synch motor is connected to a VFD? 2. Someone told me that VARs produced in a synch motor does not get pushed through XFMRs is this true and please provide some tech details concerning this. The question has been sent to GE but I wanted to hear from ENG-Tips.com as well. Thanks in advance!

 

[electricuwe]

1. Depending on exciation the synchronous motor can still produce reactive power, but the best use you can make from that in an inverter application is to adjust the exciation so that the power factor seen by the inverter at the output is 1.

Due to the fact that the DC-link is able to store energy input and output power factor are totally decoupled.

Regarding power factor at the input side it is important to distinguish between displacement power factor (cosphi) and true power factor lamda. Even with a simple diode frontend displacement powerfactor will be very close to 1, so that you can save on the utility bill if you are billed for VARs. A more sophisticated active frontend would also allow to get rid of harmonics and to get lamda close to 1. Unfortunately these kind of inverters are significantly more expensive and their efficiency is lower.

2. Of course a synchronous machine can also push VARs through transformers. Transformer impedance has to be added to synchronous reactance for evaluating such a setup. I guess sychronous machines (motor as well as generator) are connected to the grind via a transformer.

 

[JBUDA54]

Thank you electricuwe! Your remarks agree with the vendors response. The vendors did not advertise that the more sophisticated frontend would be less efficient though. They did emphasize that the more sophisticated frontend drives were much more expensive. As to my second question... Well duh... I wasn't sure what I was thinking when I sent that.

 

[RRaghunath]

electricuwe,
It is informative post. Thanks.
You wrote about True power factor Lambda. I am familiar with Cosphi but not the Lambda. Would you mind elaborating on that!
Another question is that Synchro motor can present UPF load to the inverter, is it an advantage from the inverter designer point of view?
Thanks.

 

[electricuwe]

In case of a nearly sinusoidal supply (and single phase or a balanced three phase system ), which is a good approximation for any grid voltage regarding power calculations, there are two reasons why a real power (W)might be less than apparent power (VA):

- phase shift between the fundamental of current and voltage
- harmonic currents

As long as there are no harmoncis in the current the following relation is valid (example for single phase):

P = U * I *cos(phi) with U and I beeing RMS values and phi beeing the phase shift between.

With harmonics in the current this equation is not valid any more and changes to:
P = U * I *lamda

because current distortion is beside displacement another reason for apparent power

As soon as rectifiers are involved, that are not desinged as PFC (Power Factor Correction) ar AFE (Active Frontend),
current harmomics are to be considered.