----- Original Message -----=20 From: "Justin Richards" To: "Microcontroller discussion list - Public." Sent: Sunday, May 08, 2011 10:25 AM Subject: Re: [OT]:how are hydroelectric power stations synchronised >> I once saw a generator being switched on line at a hydro power station i= n > > I would like to see that. I used to develop control software for hydroelectric plants. The basic way = a=20 generator is synced in manually has been well described here: a device=20 called a synchroscope displays the phase relationship between the grid and= =20 the generator. The 'scope looks something like a clock with a single hand.= =20 When the hand is at twelve o'clock the generator is in phase with the grid.= =20 The user's adjust the generator speed (for a hydro generator this is done b= y=20 manipulating the 'wicket gates' that control the water through the turbine)= =20 until the frequency is nearly the same as the grid's frequency. This can be= =20 seen in the synchroscope because the hand will be rotating slowly around th= e=20 dial. Then, as the hand crosses twelve the operator closes in the circuit=20 breaker for the generator. Once closed in, the generator is forced to track= =20 the grid. Automatic synchronizers work on the same principal, except that a computer= =20 (or in the past an electro-mechanical rube-goldberg system!) watches the=20 phase relationship as it adjusts the wicket gates. When the frequency is=20 close and the phase is going through 0 degrees it again closes in the=20 breaker. Closing in a generator that is out of phase can be very exciting. A story... In the 1980's we were developing a control system for the Niagara Falls=20 Robert Moses power plant. The customer wanted us to implement a=20 semi-automatic mode for synchronization: the operator would use the compute= r=20 system to generate raise-and-lower pulses to the wicket gates, and, watchin= g=20 the synchroscope would then press a key on the keyboard to close in the=20 generator breaker. We tried to explain to the customer why this was a bad=20 idea (tm) but they wouldn't hear about it (more on that later). They=20 insisted that we implement the mode. During acceptance test the customer proceeded to test the semi-automatic=20 sync mode, with the operator pressing the keyboard key as the synchroscope= =20 went past twelve o'clock. Unfortunately what the customer refused to=20 understand was that there were inherent delays between pressing the keyboar= d=20 key and the actual command to close the breaker being sent out (see analysi= s=20 below). The net result is that a 160 MW generator was closed in=20 approximately 30 degrees out of phase. It was like a large bomb went off!=20 Luckily there was no permanent damage to the unit, although I understand th= e=20 breaker needed some work. Where did the delay come from? Well... The microprocessor in the keyboard had to get around to scanning the key... Then the keyboard had to send the keystroke over a serial link to the=20 display processor, which... Communicated via a multiplexed X.25 link with a communication processor,=20 which... Communicated over a bus interface with the main processor, which... Examined the request and decided to close the breaker. So it... Sent the breaker close command over a network to the front-end processor,=20 which... Communicated over a bus interface with a communication processor, which... Communicated over a multiplexed communication link to the RTU, which... Set a digital output to command the breaker to close. The upshot: The customer insisted that we remove the semi-automatic sync=20 capability from the software. -- Bob Ammerman RAm Systems --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .