logging ripple control signals?

In my area the utility uses ripple control signals (probably at 1050hz) to control off peak loads.

Are any of the voltage detection "building blocks" able to detect and therefore log (with sufficient sampling speed) these signals, over time?

More on ripple control: http://en.wikipedia.org/wiki/Load_management

An example of a fan maker dealing with customer issues from these ripple signals: http://www.hunterpacific.com.au/PDF/Ripple_Control.pdf


Robert Wall's picture

Re: logging ripple control signals?

I have never seen any mention of this here.

In theory, the standard sampling algorithm (as found in emonLib) should - just - be able to detect a signal up to 1250 Hz. I don't know whether that frequency would even get as far as the analogue input as I've no idea how the voltage sensor performs at that frequency. To find that out has to be your first step. If the signal does get through, I think you would need to branch out of the main processing path and add a software band pass filter and detector, and that might slow the loop sufficiently to take the frequency of interest above the Nyquist frequency - not good news.

The other consideration of course is; is the amplitude after it's found its way through the isolating transformer and dividers sufficient to detect reliably? The amplitude of the control signal is around 1 V according to the last but one post here: http://mybroadband.co.za/vb/showthread.php/134334-And-so-I-have-proved-t... (about what I'd guess anyway) and according what that poster found, a separate processor is almost certainly needed.

9fingers's picture

Re: logging ripple control signals?

Faced with this task,  would use a Phase Locked Loop tone detector as by far the easiest way. A single chip and a handful of passive components would give you a digital out put when the signal was present. If there is modulation to detect too, the PLL will demodulate FSK and PSK easily as well as OOK.

Older devices that I'm familiar with include NE567 - dunno if it is still around but there is bound to be something similar.

Look for PLL tone decoders.


If the bandwidth of the isolating transformer is too low to pass the signal, then you could *carefully* run the tone decoder on the hot side of the mains, referenced to neutral and then use an opto isolator to interface to your processor



dBC's picture

Re: logging ripple control signals?

For what it's worth, I just captured the local grid manager turning off my pool pump before the evening peak starts.  This is on the output side of an unloaded 9VAC wall-wart (nominally 9VAC, but with no load it's more like about 12VAC).  You can see the divots are just under 1 msec apart, consistent with 1050 Hz.  

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.