Circuit or Device to Close Relay at Specific Phase Angle

[OhmOracle]

Hello everyone,

I am looking to conduct some startup current surge testing on a set of 480VAC - 24VDC power supplies.  These power supplies are used in a 3-phase industrial environment, and are connected phase to phase between two of the phases. 

I would like to be able to control at phase angle/point in the AC sine wave power is applied to the power supply so that I can evaluate how this affects the startup current surge of the power supply.  I'm not sure what the best way to accomplish this might be.  I plan to use the circuit/device to either close a contactor or AC solid state relay to apply the power to the power supply being tested. 

I have a LabJack T7-pro that I might be able to use to recognize the zero crossing of the sine wave and activate one of its outputs,  but I'm worried it may not be consistent enough to vary closing between the 0 crossing, 45 degrees, 90 degrees, etc. 

I'm open to any ideas or tools that might be useful for this - a Raspberry Pi, a microcontroller, some other test instrument, an analog circuit, etc.  If anyone has any ideas that they might suggest and could point me in the right direction, I'd appreciate it!  Let me know if any other information may be helpful. 

[forrestc]

I would like to be able to control at phase angle/point in the AC sine wave power is applied to the power supply so that I can evaluate how this affects the startup current surge of the power supply.  I'm not sure what the best way to accomplish this might be.  I plan to use the circuit/device to either close a contactor or AC solid state relay to apply the power to the power supply being tested. 
..
I'm open to any ideas or tools that might be useful for this - a Raspberry Pi, a microcontroller, some other test instrument, an analog circuit, etc.  If anyone has any ideas that they might suggest and could point me in the right direction, I'd appreciate it!  Let me know if any other information may be helpful. 

WARNING:  In the following, I assume you know enough to keep yourself safe.  Some of the following suggestions are dangerous if not implemented correctly.

What you are describing is the way that AC lighting dimmer circuits work.  As a result, there are a LOT of circuit snipplets on the web that would work with minor changes.

Personally I'd do this as follows:

1) find a zero crossing detection circuit you're comfortable is safe and recreateable.   Hook the output of this circuit to an arduino input pin.   Note that some transition only in one direction (positive to negative), some transition in both directions.  You may want to only trigger in one direction so you can test your surge on both a 'positive' and 'negative' part of the cycle.

2) Hook an arduino output to your SSR or similar circuit.   

3) Additionally, if it makes sense, add buttons for things like trigger so you can operate the whole thing with a button.  Bonus for a small display.

4) Write code to wait for your 'start' trigger (from a button, etc), then wait for the zero crossing to transition, do a fixed delay, and trigger the output.   At it's simplest this should be a dozen or so lines of code.

One example of this entire process (but for a dimmer) is at https://www.bristolwatch.com/ele2/zcnew.htm

Note that an arduino can do microsecond-level delays.  Much more accurate than you need.   

[Retirednerd2020]

There are several ways to go.  Here is one way:

If I were just setting up a test I would use an "instant on" solid state relay.  This is the type that turns on as soon as a control signal is applied.
Any microcontroller could be used to do the timing.  Use an opto-isolator to generate a logic signal to the microcontroller.  Using a timer, triggered by the GPIO input or just a "Delayms" or similar, turn on the signal to the SSR at some time after the zero cross.  Look at https://www.gavazzionline.com/ for some SSR examples.  Your inrush current might be very high, depending on your power supply design so that is probably a more important specification than the average/RMS rating of the device.  Don't forget to heatsink it.

On the web, Look for "opto-isolated zero cross detector" or similar for the zero cross detection.  There are two main types, one is sensitive only for one direction of zero crossing and the other for both directions.  You would buy the opto type you need.  In the case of your power supply, without knowing your exact circuit, I assume it really shouldn't matter the polarity so I suggest a an opto-isolator with inverse-parallel leds in its input side.  If polarity does matter to you, an opto with a single LED with an external anti-parallel diode will work well.  You will need to provide resistors to drop the voltage/current.  Eg.  Say 10mA is desired.  R = 480*1.414/0.01 = 48 k ohms.  Power rating: 5 Watts, minimum.   Another way would be to incorporate a small transformer, say 480V to 24V to get the voltage down some, then a smaller value/power resistor will be needed.  This won't turn exactly at zero cross but it will be a small delay afterwards.  However, that delay is fairly predictable unless your AC waveform is very distorted.

So, 480V.  You need to take great care when working with a voltage like that.  You might walk away from a 120 or 240V shock but with 480V it is less likely.  Be sure to choose an opto-isolator with the proper isolation voltage.  Keep your HV and LV physically separated and isolated.  Cover the HV side with something so you CAN'T touch it.  I suffered a 480V shock when I was a young pup.  It was unpleasant to say the least and I'm lucky to be here to talk about it.

I personally would avoid the PC / Labjack in order to keep things simpler but you could use it if you prefer.

[tom66]

In my experience, many relays tend to have somewhat random response times varying by around 5-10ms, often combined with contact bounce before settling.  This will cause significant phase variation (at 60Hz, that might be as much as 180 degrees offset.)

A higher quality relay may be better, but I'd want to see how consistent the relay could control its load before considering this over something like a triac, SCR, or optoisolated MOSFET.

[PCB.Wiz]

Hello everyone,

I am looking to conduct some startup current surge testing on a set of 480VAC - 24VDC power supplies.  These power supplies are used in a 3-phase industrial environment, and are connected phase to phase between two of the phases. 

I would like to be able to control at phase angle/point in the AC sine wave power is applied to the power supply so that I can evaluate how this affects the startup current surge of the power supply. 

Is that surge causing problems ?

some background may help
https://electricalengineeringprofessional.wordpress.com/2019/03/18/transformer-inrush-current/
and with smaller numbers, single phase
https://sound-au.com/articles/soft-start.htm

[moffy]

Definitely an SCR pair or triac, they have very high surge ratings which would be ideal for your application and are phase controllable.

[Terry Bites]

Once you have a zero crossing pulse you add a timer to give a delay and/ or n-cycles delay. And then use the output to switch your power circuit
The timer can be a oneshot like 555 a clock/ a counter combo or analog phase shift at the input to the ZCdetector.
Its easy to get all the time functions you want (and much more) from an arduino. Add an SSR and thats about it.
Not the kind of SSR that has built in zero cross switching though.  Triacs are a low cost and robust alternative.
Here's one of my fave reliable ZCDs