GSU Relaying Questions 2

[Gwhiz]

I am upgrading the relaying at one of our generating stations from the old GE BDD15 to a pair of new Beckwith M-3310 relays. I want to take advantage of its features other than differential.

The transformer is rated at 145MVA 14.4/69 kV and connected in a delta wye configuration.
The generator is rated at 160MVA 15kV.
The two are connected in a unit configuration with two unit auxiliary transformers.

Question #1: In trying to set backup overcurrent relaying, I have found it difficult to coordinate with the inrush current curve. I have plotted the relay curves, transformer damage curve and inrush curve on a graph using Excel. The values for the inrush curve come from the Electrical Distribution System Protection book by Cooper. In reviewing relay settings for the same type of relay at another of our generating stations, I noticed that parts of the backup overcurrent curve specified for that relay lie to the left of the inrush curve for that transformer. (Similar configuration, transformer rated at 231 MVA) Are there any special considerations that need to be taken into account to modify the inrush curve for this transformer due to the unit connection?

Question #2: The old BDD relays had instantaneous elements on them that were set at 8X tap-which really makes me question the inrush curve (see above question). I want to set the new instantaneos elements in the relay to about the same setting. Do I use the 87W1 and 87W2 CT ratios that were calculated for the differential element?

Question #3: Is there any rule of thumb for setting the timing of the 51 elements? Another engineer suggested a pickup of 125% of rating and the curve set so that at the highest current level for a fault on the 69kV bus the backups will operate in 1.5 s.

Question #4: The backup OC's will not pick up for all faults on the 69kV lines running out on the station. I wanted to use a 46 element to pick up the unbalanced faults. Is there any advice on how to set this element?

Thanks in advance

 

[peterb]

Question #1:
With a unit connected generator/transformer, inrush is not going to be an issue. During unit run-up, the generator excitation is generally applied at a low value and increased gradually up to rated voltage - the transformer is connected to the generator and is not subject to a full voltage inrush.
You may need to check your curves, as the inrush point is usally taken as 10-12x rated current for 0.1 second. Unless you are using a very low time dial setting for the overcurrent element, operating time will be in excess of this. There are several excellent software programs available for time-current coordination studies, which I would recommend that you investigate rather than re-inventing the wheel; if all else fails, there is always the quaint old fashioned method of tracing the curves on time-current graph paper.
Question #2:
Checking on the BDD manual, the setting of 8x tap for the instantaneous is based on the AC value of current - note that the inrush waveform is heavily offset, with a high DC component. To quote from the manual - "This pickup value is based on the AC component of current transformer output only, since the differential current transformer in the relay produces only a half cycle of any DC (offset) component present." Check the Beckwith manual for specific data on the instantaneous relay characteristic when deciding on a setting - if the elements are not harmonic restrained, you will need to set them above any possible inrush (but see comments on Question #1 above).
Question #3:
Set the overcurrent pickup well above the rated current - this is fault protection and NOT overload protection. You don't want to restrict the ability of the unit to supply power under emergency conditions. I would regard 125% as a reasonable value. The relay operating curve would usually be selected as extremely inverse, to match the transformer capability curve. The operating time at the highest fault current would be set to coordinate with downstream protection - 1.5 sec sounds on the high side for this.
Question #4:
There is a variety of information available on the subject of negative sequence overcurrent protection. Check the papers available at

http://www.selinc.com

 

[peterb]

Gwhiz -
I missed out on pointing you to thread238-10885 in this forum, which discussed the question of transformer inrush in some detail.