single phase power redundancy with 3 phase

Discussion in 'Power Electronics' started by jmyoon7, Feb 27, 2017.

  1. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    I can't find any resources online regarding what we are trying to do,

    Basically we have a programmable controller (230VAC 1PH input) that is programmed to actuate the shunt trip coil of the main breaker (4 pole A,B,C,N) on a fault condition; but it also needs to remain powered incase of a phase loss (i e. if phase A goes it should be powered through B or C).

    While the controller power is 1PH, our system input is 400VAC 3PH. By implementing a basic circuit using relays we were hoping to create contingency even when a phase goes missing. see attached.

    The relay is a 400VAC rated relay with double pole (form-c). We did not use any timing relays, or any complex circuitry. When R1 is energized it will take precedence in powering the controller. When phase A is missing R1 will not energize, therefore R2 will take precedence. NOTE: all 3 relays do energize at the same time if all 3 phases are present. There is no induced voltage across the phases because there is a RFI filter at the 3 phase input.


    We tested several times with the setup as shown in the attached but results are no good, intermittently we would have these relays pop (blow up essentially) and it appear to have shorted which maybe the result of the fact that our setup basically compromises switching delay on each relays.

    Have any of you experts know anything about this? what could be an alternative solution to what we're trying to achieve? a 3-way redundant 1 phase power? is this even possible?

    Thanks guys, been scratching my head too much!

    Josh
     
  2. MaxHeadRoom

    Expert

    Jul 18, 2013
    16,380
    4,814
    It would seem to me that the PLC requires to be powered from an auxiliary 1ph source ahead of the main breaker, or from some other source not affected by the said breaker.
    Why cannot this be a solution?
    Max.
     
  3. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Yes, this is definitely a valid question,

    you may not agree with this answer but we have already manufactured a few of these systems (extremely compact enclosure) and rework involves not just wiring and adding bus bars but changing the physical dimension of the enclosure,

    so at this point i'm stuck with the controller power being brought from downstream of the main breaker,

    the relay popping phenomena is intermittent and hence we did not catch it at the time of design,

    I'm still curious as to what we want is achievable, and whether anyone can explain why i'm seeing what i'm seeing
     
  4. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Actually Max, I think with your suggestion; it still doesn't solve the problem of powering the controller in the event of a phase loss.

    If i lose phase A i still need some means to power the device through either phase B or C regardless of main breaker trip.
     
  5. LesJones

    Well-Known Member

    Jan 8, 2017
    1,573
    389
    I think the problem is that as the relays are switching that phases are shorted out due to the time it takes a relay to opperate. I assume that the PLC only requires relativly low power. (Say less than 300 watts.) I would power it from an inverter that ran from12 or 24 volts DC. I would power the inverter from three power supplies (One fed from each phase.) The negative of all the power supplies would be connected together and to the negative input of the inverter. The positive of each power supply would be connected via a diode to the positive of the inverter. The negative end of the diode to the power supply and the positive end to the inverter. You would need to add a warning circuit to warn you if any of the three power supplies failed.

    Les.
     
  6. tcmtech

    Distinguished Member

    Nov 4, 2013
    2,809
    2,703
    First off us knowing what the load is and why it needs to be kept running regardless of what phase is lost would help a lot.
     
  7. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    We can't use psu because not only we don't have space to add 3 psu in the enclosure but also the programmable controller input is 75-230vac. In addition, our relays and shunt trip voltage is 250 as well.
     
  8. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Load is minimal, only time you would see high inrush is when the breaker is tripped off the shunt trip. But that would be short pulse rather than a momentary hold, we may see about 10A for very short time, the relay contacts are rated for 400VAC 10A so it shouldn't be the problem.

    The reason why it needs to be kept running is because if a phase is lost we need to trip the main breaker, this can't be done if the controller isn't powered
     
  9. tcmtech

    Distinguished Member

    Nov 4, 2013
    2,809
    2,703
    If it has a universal input voltage rating of 75 - 230 VAC it will also work just fine on a 100 - 350 VDC input. All you need to do is run your three phase input to a three phase rectifier and feed the DC output of it to the PLC input.

    As for power dips or dropouts just putting a large enough capacitor on the DC link between the rectifier and the PLC input will buy you as much time as you need.

    Or you could just add a proper Phase Loss Protection relay to the system and use that trip the main breaker.

    Take your pick.

    https://www.google.com/search?q=pha...rome&ie=UTF-8#q=phase+loss+protection+relay&*
     
  10. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Thanks Tcmtech for the response,

    I will look into those options, to add to our problem it is extremely difficult to source parts that are rated for the environment, it has to have operation temperature rating of -40 to +70C.

    As for the relay phenomena, do you have any explanation of what causes the problem and why it is intermittent? Could it be that when relays are switching it's holding the arc between nc to no? and as a result as it switches it shorts with the other line? could this be solved by increasing the voltage rating of the relay ? say 600vac?
     
  11. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Yes, I cannot find any phase loss protection relay at this temperature range which is why we went with the relay scheme,
     
  12. tcmtech

    Distinguished Member

    Nov 4, 2013
    2,809
    2,703
    No idea. I just fix things with methods I know will work and not put too much thought into what clearly didn't work.

    More than likely when one phase drops out it isn't immediately dropping off to 100% dead zero voltage and power output but to a lower one that's still high enough for the active relay to stay on while a second one picks up thusly causing a phase to phase short.

    If you look at how a single core three phase transformer is made you will see that once a phase winding set loses it primary input power that phases secondary winding will then just drop its voltage some as its relative phase angle lines itself up with one of the others.

    Also if there are electric motors and other smaller transformers active in the system even if one of the incoming phase does drop off to zero input power those motors will run as rotary converters thusly now generating that phase themselves thusly keeping much of the equipment and load on that dead phase still very much active but at a reduced power level until the remaining active phases are overload and some over current protection device shuts things down.

    As for difficulty sourcing items if you have the system in place obviously someone had to know how to get the parts that made it there to begin with.
    Same with protection device cost. If you can afford to operate devices and equipment that needs three phase power to operate you can obviously afford to put in some basic form of protection for them.
    If you can't afford a $50 - $200 protection device you're operation is already dead and just haven't accepted it because obviously you cant afford to replace whatever motor or other devices that need protecting either. :rolleyes:
     
  13. tcmtech

    Distinguished Member

    Nov 4, 2013
    2,809
    2,703
    How often or realistically does the environment the actual device is working in ever see -40 C or +70C? (-40 F to +158 F) I don't know of any place on earth that ever sees that extreme of working temperatures on a regular basis let alone one where actively operating power or crittle control systems would see it.

    And if there is such a place anyone who can afford to operate such conditions obviously has the means to afford to properly design the necessary support systems to keep their equipment working in such conditions well.

    Simply put I don't buy the cost or climate extremes excuses. I live where -30 C is a common winter time temperature extreme and I know that +70 F is not a temperature that is found in any habitable place on earth which means that it would only be present in an enclosed cabinet or such location where obviously both heating and cooling countermeasures can be used to limit the temperature extremes.
     
  14. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Hi tcmtech

    whether it's realistic or not, it is the spec from the customer which cannot be deviated, i can't say what this is for but i can assure you it will be exposed under the environment circumstances,

    the challenge is not the cost, but we do have to meet the temperature rating as it needs to be environmentally tested in our test chambers in both low/high temperatures, it is not an excuse believe me. Actually if you did some research you will be quite surprised that there are many electronics that complies with the spec, just not phase loss protection relay.

    your input on alternative solution has been noted, i appreciate that
     
  15. LesJones

    Well-Known Member

    Jan 8, 2017
    1,573
    389
    I like tcmech's idea. Just a couple of points. I don't think that you could use a three phase rectifier straight from the incomming phases as the rectified voltage would be too high. (It would be phase to phase voltage.) If just three diodes were used from the phases then the voltage to neutral would be about right. It would probably work with just the loss of one phase without an additional capacitor but if only one phase remained then a capacitor would be required as the rectification would onle be half wave.

    Les.
     
  16. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    Thanks guys,

    we've tried the single diode per phase solution, measured voltage

    with all phase = 266vrms
    2 phase (1 phase missing) = 220 vrms
    1 phase (2 phases missing)= 150 vrms

    voltage range are acceptable input to the controller,
    the problem is with the shunt trip coil, vdc of the coil rating is 240 we would need to step the voltage down, we may add isolation tx's to drop the voltage then add in the caps to step up the voltage in case of a 1 phase.

    thanks for the input!
     
  17. KeepItSimpleStupid

    Distinguished Member

    Mar 4, 2014
    2,918
    510
  18. tcmtech

    Distinguished Member

    Nov 4, 2013
    2,809
    2,703
    Your excuses for this are just falling apart one after another being any customer with such a strict requirement for specifications would never accept some tossed to gether relay design as you have proposed and obviously tried to implement to begin with and then have the overall system set up in such a highly climatically variable exposed working environment. :rolleyes:


    I've done specially work on the side all my life and more than once someone has come to me with some daft request (because everyone else told them to go bugger off because their wants were unfounded) to having something done there was when it was blindingly obvious their expectations and understand were beyond realistic to point of just being stupid I too had to tell them that in order to meet their expectations that their project was going to be so outrageously expensive they could never afford it or that I simply would not do the work.

    My point is, 'Just because the customer wants it that way' doesn't actually justify anything in a design. Some customers are too dumb to know what it is they actually need and to what realistic or even extreme limits their wants have to work in.

    That's where working with your customer to find out if they have realistic goals and specification or not is key to doing a job and the having the guts to tell them whether what they want has any valid merit and justification or not and to me right now more and more or what you say is not adding up. :(
     
  19. jmyoon7

    Thread Starter New Member

    Feb 27, 2017
    10
    0
    i dont have time for this,

    thanks tcmtech, you got your great point across,
     
  20. tcmtech

    Distinguished Member

    Nov 4, 2013
    2,809
    2,703
    Yes, Being straightforward and honest about the intention and application of something is rather difficult for some so they will cut an run before accepting that what they want is not realistic.

    (Easier to blame the other guy, for not providing a solution than it is to accept that your requests of him were unrealistic and or unachievable, that way. ) :rolleyes:
     
Loading...