Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
Robb Senior Member Joined: 01/08/2007 Location: AustraliaPosts: 221
Posted: 01:20pm 16 Aug 2007
Copy link to clipboard
Print this post
With all the talk in this and other forums about useing AC induction motors as grid connected alternators I just had to give it a shot . First time I'v ever seen my meter go backwards Yes it does work
That said I wonder if anyone has made a practicle directional AC current detection circuit.
Its all very well to hope the field dies out with a power failure but given the capacitors in fluros etc I wouldn't want to bank on it so a fail safe cutout idea sounds better to me.
A directional AC current detector sound like a bit of an oxymoron but I think you get the idea.
thefinis Regular Member Joined: 23/08/2006 Location: United StatesPosts: 53
Posted: 04:42pm 16 Aug 2007
Copy link to clipboard
Print this post
Not sure if your wording is right but i know what you are wanting. AC is bidirectional hence the alternating part. There are meters etc for determining draw or feed(?). For using an induction motor as a genny the best way is to use an rpm switch. Below the rated motor rpms it is off and above it is on. I keep wondering if one could use a centrifugal clutch like on a go cart just not sure if they are sensitive enough.
The problem with wind turbines is matching the rpms needed in varying wind speeds to the very narrow range of rpms required when using a grid tied induction motor as a generator. It is from rated speed to 3-5% over rated speed. For say an 1800 rpm motor that is around 50 to 100 rpms over. If your turbine is able to overpower the motor and take it to higher rpms than that the motor will either burn up or lose its field and quit generating.
More later
Finis
Texas born and bred
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258
Posted: 09:56pm 16 Aug 2007
Copy link to clipboard
Print this post
Hi Robb
Most important it scrares me your talk of capacitors and inductance motors connected to the mains. capacitors will cause the unit to self excite, if the mains is lost it will still generate VERY DANGEROUS. The system will if connected to the mains obtain excitation from there and if mains lost no generation fail safe. A bi directional current meter i have worked with is simply a power meter, attach arm to the disc that comes against a micro switch if disc is rotated backwards. Also could fit switch thats operated in forward direction the NC contacts here could drop out contactor to prevent motoring etc
Robb Senior Member Joined: 01/08/2007 Location: AustraliaPosts: 221
Posted: 03:56am 17 Aug 2007
Copy link to clipboard
Print this post
Herbnz it scares me too thus my post. I dont think an induction motor will necesarily drop out if say the powerline is disconected 1 mile up the road given capacitors in you or your next door neighbours house eg flouro light power factor caps etc.
Centrifugal switches wouldnt be any good as if mains droped the system would just loose "lock" and run up in frequency and most likly voltage.
I tested the effect of self excition with caps on disconection of mains and we hit well over mains volts (~400 volts) proberbly at 400+Hz
Given fluro caps and the above test you cant bank on flux drop on mains failure.
The solution is what we are looking for.
The heading probebly should have read :"non islanding solutions - how?"
Robb Senior Member Joined: 01/08/2007 Location: AustraliaPosts: 221
Posted: 04:08am 17 Aug 2007
Copy link to clipboard
Print this post
By the way this was only a short duration experament. No way I'd be game to leave it runing at this stage .
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 05:45am 17 Aug 2007
Copy link to clipboard
Print this post
Robb, assuming the voltage remains in phase, the amplitude and phase of the current determines the strength and direction of power flow. In other words, as the power flow falls to zero and reverses, the current falls to zero and rises again, but 180 degrees out of phase.
This effect can easily be observed on an oscilloscope by triggering the horizontal sweep off the voltage, and monitoring the current in the vertical channel.
The relationship between voltage and current will not be precisely in phase in one direction (forward power), and 180 degrees out of phase (reverse power) due to reactive effects, but it should not be very far from it.
So what is required is a phase detection circuit to compare the phase of the current to the phase of the voltage. To determine which way power is flowing.
This could be a fairly simple circuit using a pair of voltage comparators. One turns the sine wave voltage into a reference square wave, and the other the current into a reference square wave.
Each of these square waves could be fed into the two inputs of an exclusive OR logic gate. The logic level output would then be logic high for power flow in one direction, and logic low for power flow in the opposite direction.
Because of reactive effects (power factor) the logic levels will not be absolute dc levels, but will almost be so. For instance a reactive load will definitely produce a sight phase error in the current flow, either leading or lagging.
So instead of having say a continuous logic low output from the exclusive OR gate, it might be logic low with short narrow positive going spikes. Or logic high with short narrow negative going spikes. The width of these spikes will be directly proportional to any reactive phase error.
But with an RC integrator circuit to time average the output of the exclusive OR gate, and a third voltage comparator, a reliable continuous logic level can be produced to indicate the direction of ac power flow.
It should not be difficult.
Edited by Warpspeed 2007-08-18Cheers, Tony.
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258
Posted: 08:50am 17 Aug 2007
Copy link to clipboard
Print this post
Hi Robb
you raise a valid point if there is capacitors in the system somewhere. In normal situation any caps in equipment will be cancelled out by the job they are there for namely pf correction we dont correct above .9 . However a point that you make me realise is that somewhere before the first transformer ther could be a flouro turned on with a fault, ie no tube, with a pf cap idle across the lines.
One other protection would be to have a relay that was only energised when 50hz is present the windmill would be very unlikly to maintain 50 hz off line. Relays are available for this or could be made.
Tony are you triggering the scope from the voltage on the x axis or replacing the time base ? I have often measured phase angle by running the x axis on voltage y on current .
HerbEdited by herbnz 2007-08-18
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 09:20am 17 Aug 2007
Copy link to clipboard
Print this post
I does not matter if the X axis is the actual system sine wave voltage, or is a sawtooth triggered off the system voltage.
If the actual voltage, then a Lissajous figure will be displayed on the oscilloscope screen. If driven by a linear sawtooth, a sine wave will be displayed.
The original post was asking how to determine the direction of power flow between two parts of a circuit. There can only be one voltage and one current to measure, as two parts of any completed series circuit are joined by a single pair of wires. One part of the circuit is source, and the other part is load. The real question is, which is supplying which?
This is seemingly complicated by the constant regular reversals of voltage and current in any ac circuit. What determines power flow is the phase relationship between voltage and current.
Assuming one part is the grid, and the other part the "system", there will still be only an active wire and a neutral wire connecting both, through which the voltage and current are to be measured to determine power flow.
Power factor can only go to plus or minus ninety degrees, which corresponds to zero real true power. Beyond that the true power can be said to have reversed direction.Edited by Warpspeed 2007-08-18Cheers, Tony.
herbnz Senior Member Joined: 18/02/2007 Location: New ZealandPosts: 258
Posted: 08:02pm 17 Aug 2007
Copy link to clipboard
Print this post
Hi Tony Yea I agree I had posted that but on thinking deleted it within about 5min you must have been right there.
I dont see how a circuit that will detect power reversal will help with the chance of generating if the mains go down. your brillant efforts could I believe be altered to detect if at any time there was a leading pf (capacitors ) and raise alarms or initiaie a shut down.
The chances I believe of Capacitors on line that are not cancelled by the inductance on a mains supply are very remote, also it would have to create problems on the system as resonance is always close by in these situations.
Herb
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 10:10pm 17 Aug 2007
Copy link to clipboard
Print this post
One way to do that would be to use a phase detection circuit similar to which I suggested earlier. An exclusive OR gate followed by relatively slow integration will produce a dc output voltage proportional to the phase difference of current and voltage.
The system should normally operate within a reasonably small range of power factor if everything is working properly. Any large unexpected changes in power factor could be detected and used to isolate or shut the system down. Cheers, Tony.
Robb Senior Member Joined: 01/08/2007 Location: AustraliaPosts: 221
Posted: 02:08pm 18 Aug 2007
Copy link to clipboard
Print this post
Hmm interesting ideas
Anyone here actualy know what the actual deviations from 50 Hz are as far as the grid goes? I was thinking a couple of XTAL referenced PLLs just above and below 50hz may be one way to get a logic output to "kill" your alternator....
Is "that one" about the companys catching up cycles at night an old wives tale in this day and age?
I would have thought they would have everything locked to a rubidium standard these days.
Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406
Posted: 10:41pm 18 Aug 2007
Copy link to clipboard
Print this post
Typically the grid slows down at periods of peak demand, but almost immeasurably, and at periods of light load will speed up. So the phase of mains power compared to an external reference will vary quite a lot in an unpredictable way.
Many appliances use a mains synchronous motor to measure time, especially clocks !! So they try to keep the long term average mains frequency at exactly 50 (or 60 Hz).
Where remote generator set is in use, it is common practice to have both a mains driven clock and an accurate independent clock side by side on the wall. The power house operator will then sneak the motor governor speed up or down fractionally, to ensure that any clocks running on his system will also have their time corrected.
It will be appreciated that a large generator with a perfect speed governor will still gasp and slow for an instant if a very heavy load is connected. It will also try to race for an instant if a massive load is suddenly shed. All these small increments of sudden load change during the day can add up. The grid does exactly the same thing, but to a very small degree.
So yes, the grid is corrected, and the mains frequency is never perfectly stable. Another thing that can cause sudden unpredictable phase changes, is the connection and disconnection of reactive loads.
Any useful measurement of phase needs to be between voltage and current in the same part of the circuit.Cheers, Tony.
thefinis Regular Member Joined: 23/08/2006 Location: United StatesPosts: 53
Posted: 02:51pm 20 Aug 2007
Copy link to clipboard
Print this post
I knew I had seen this somewhere with info posted on what was needed to safely grid tie for induction generation. Hope it helps. It is mostly about what is needed for Texas laws and regs but covers much of what is needed to run one safely anywhere and gives info on disconnects and parts numbers for building interconnect box. I need to reread the whole thing again myself including the links.
Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
Print this post
. First time I'v ever seen my meter go backwards 
Yes it does work 
so a fail safe cutout idea sounds better to me.
. 