2-phase phase angle control

[Dryjoint]

Hi All. I'm familiar with and have built plenty of single phase ramp & comparator type leading-edge phase angle controllers, but currently have an application that requires the voltage to be derived from two of the three phases (UK supply). My maths is not up to creating an excel graph to represent voltage throughout the cycle of two 120 degree separated signals and I'm really struggling to visualise what a 2-phase signal would look like graphically. I guess the question at the root of this is "what do I sync to?" or is even this question too simplistic? Thanks.

 

[ScottyUK]

If you're looking at the voltage between two lines of a three phase supply then it's just a single phase supply. If you are using neutral or earth as a reference then that's different: measured from neutral (or earth) the three phase voltages are three sinusoids displaced by 120 degrees in phase, or 6.67ms in time on a 50Hz system. The voltages between phases are also sinusoids. I have an Excel sheet somewhere with these which I can upload, but I'm away from my own computer and can't get to it at present.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Dryjoint]

Thanks, Scotty. I was going to 'scope two phases in differential mode, but I thought in advance that if I saw a sinusoid, I'd just put it down as an artifact and not the real thing(!) So, if the voltage across any two phases is a sinusoid, is it the case that all I need to do to zero-crossing-detect for a two-phase control circuit is to supply the control circuit from those same two phases? I think I know the answer

 

[ScottyUK]

Yes, if you are intending to connect your circuit across two lines without using any other reference point then it will behave like a single phase controller connected line-neutral. Obviously the voltage will be [√]3x line-neutral. Fuse both connections to the lines; you probably only fused the live and not the neutral on a single-phase circuit. If you are working with inductive loads then the interaction between phases can lead to some very peculiar-looking waveforms and you can get into trouble with firing pulses. There are some good power electronics texts in faq238-1287 which would be worth borrowing or buying.

Please don't connect a scope directly to the line without an isolating probe - isolating probes are much cheaper than buying a new scope but more importantly they protect you from the high energy circuits: I've always considered being dead as damned inconvenient.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Dryjoint]

Thanks for that, Scotty. It's a resistive (heater) load, so it should look reasonably clean. &BTW, I've got a box of assorted high voltage thin film R's that I use for things like this. I've never burned-out any gear, but have had a few carbon deposits on my fingers... I think my "block" in trying to visualise the nature of the waveform was in not just seeing it as the same a 1-phase one, referenced to a point that is "moving" with it. My thought process -for what it was worth- was still tied to ground.

 

[ScottyUK]

Glad to clear that up, I understand how you had confused yourself.

re. probes: I was thinking more on the lines of

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&ci=13415&cs=psu&lc=EN.

I've spent too much time taking measurements on the kind of plant where a major fault would leave a glowing smoking hole in the substation floor to take use homebrewed methods of connecting to equipment. Just my opinion!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[davidbeach]

In excel, mathcad, matlab, or anything else of the sort of your choice, develop and plot two sine curves with a 120[°] phase shift (throw the third in for good measure if it helps) and call them line to neutral. Now plot the difference between the two curves, that's what you will see line to line.