The Space Saving XY Controller

XY Controller Decoding Matrix

I recently had to revisit an older control application, where the customer wanted to keep everything the same, but replace his older remote terminal unit technology with perhaps a newer one, yet keep the existing wiring and functionality. With some of the older equipment the limitations of space, both physical, and in terms of protocol address space, meant that engineering remote control solutions often meant optimization of space at the expense of urgency and time. The challenge in modernization almost always starts with No change in SPACE, no change in WIRING, and ends in trying to exchange modern high tech equipment into the same cabinets and racks which the older equipment occupies.



Remote Terminal Equipment for the past 40 years or more, in the electric power industry, has focussed on being able to provide TRIP/CLOSE contacts to operate Power Control equipment, such as Isolators and Circuit Breakers in Substations. The Remote Terminal Unit was never originally used for protection, but rather to apply supervisory control from afar, perhaps a control centre in a major city for example. Now Circuit Breakers generally do not operate often, but when they do, they take a bit of power. One of the key issues that faced the supervisory control engineer was the requirement to TRIP/CLOSE heavy current on many control lines.

OBJECTION

• "But heavy current relays take up a lot of SPACE in a cabinet, and there are more and more [100-200] equipment to control in large substations. "
• "AND the SCADA protocol has a limited address range, only handling up to approx 255 addresses.

The design and development of the XY controller was based on countering the above two objections. So how did the design engineer overcome the objections ?

ENGINEERS DESIGN RESPONSE

• "I know that you only want to do supervisory control on one or two breakers at a time."
• "I know that you have time to do any substation changeover, since the local protection gear takes care of the fast local switching."
• "I can cut the address range down to a factor of the square root of the total number of control lines if you don't mind relaxing the time constraint."
• "I can cut down the total number of high current relays to only TWO if you don't mind operating one or two circuit breakers at a time."

Taking a look at the engineer's response, the engineer was quick to realize that relaxing time constraints was the favoured solution path. This in turn lead to the ability to use an encode / decode technique to select which control lines to operate at any point in time. The engineer's approach was to translate the problem into an XY cartesian space, allowing for significant cost and space savings.

Imagine that you have about 400 control outputs which are installed, they are all high current, say for example 10Amps, and you want to be able to turn them on and off, with up to 3 of the outputs turned on and off simultaneously. You could have 800 x 10A relays which you control separately. Short story - too many cabinets of relays, most of which will rarely operate.

An optimized XY controller would be composed of an ARRAY in X Y space of control outputs, which allow an X column signal and a Y row signal to select an output, where each location in the array is one of the control outputs.  Then this basically ARMS the circuit which is fired by either a Master TRIP and Master CLOSE relay.

Here are the numerical relations..

APPLICATION SIZE

n = 400 {Number of control devices [e.g. CBs]}

DESIGN SIZES

x = 15   {Number of X control lines} 

z = 2      {High Current TRIP/CLOSE relay Masters} 

DERIVED QUANTITIES

y ==     {Number of Y control lines}

o ==    {Number of RTU low current control line outputs }

1.       y = n/x
2.      o = x + y + z

RESULT

y = 27 [rounded up]

o = 15+27+2 == 44 outputs [instead of 800]

ADVANTAGES

• My Remote terminal unit has less than 6% of the outputs and address space usage required
• I save a lot of space, needing only 2 heavy current relays instead of 800

TRADEOFFS

• Availability - if my master relay fails then I cannot trip or perhaps close up to 400 breakers
• Mean Time to Repair [can be quite long for very remote sites]
• Speed - It takes about 3 or 4 times as long to execute a TRIP or CLOSE because of the multiple protocol transactions required to setup the decoding XY matrix.

Yes, It turns out that it is economic to provide an XY control application, however because of the original design response, it does not suit all end users.  There are many end users which have only 10 to 20 control equipment per site, and in this case the savings are not worth the TRADEOFFS.  From experience, the use of the XY control arrangement tends to be suitable only for large substations in large cities, where the mean time to repair a failed Master relay is low, and the number of equipment to be controlled is very high.

In any case, the obvious contraction in cabinet size, having to be maintained throughout the modernization means that an XY control solution with more modern technology would still be required. Invensys provides modern distributed Power RTU equipment with it's SCD5200 range of Remote Terminal Units. These RTUs can emulate the XY encoder algorithm, and provide a more modern IEC61850 style interface to the entire substation, or convert the XY controlled equipment to a more open DNP3 protocol for transfer to a SCADA Master Station.

Chris J Smith

[Sydney, Mar 2012]

Seamless Texture Patterns - Patrick Hoesley