Author Topic: DPDT high current powerboard switch  (Read 855 times)

0 Members and 1 Guest are viewing this topic.

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
DPDT high current powerboard switch
« on: May 12, 2019, 11:24:36 am »
EDIT: possible solution in reply #7 -Not known if use of timers required..

Is there such a thing that could be used in a switchboard to allow for this configuration?
« Last Edit: May 12, 2019, 10:33:17 pm by phaseform »
 

Offline Gregg

  • Frequent Contributor
  • **
  • Posts: 535
  • Country: us
Re: DPDT high current powerboard switch
« Reply #1 on: May 12, 2019, 11:44:52 am »
What you are looking for is called a transfer switch.  Transfer switches come in many sizes: manual, automatic, semi-automatic, with different number of poles.  Depending on where you are located, the local electrical code, voltage and current requirements will help narrow the search. 
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Re: DPDT high current powerboard switch
« Reply #2 on: May 12, 2019, 01:47:47 pm »
Ahh awesome, seems to be called a changeover switch in oz. I found it... - (ebay version)

then, of course the ability to automate this comes to mind.
(Automatic transfer switch/Automatic changeover switch)

I have been looking around for a standardized solution for this, the closest I've come to so far is this


which appears to be a time delay auto changeover contractor configuration.

Seems like double throw contractors aren't so much a thing.
« Last Edit: May 12, 2019, 02:28:33 pm by phaseform »
 

Online MarkF

  • Super Contributor
  • ***
  • Posts: 1163
  • Country: us
Re: DPDT high current powerboard switch
« Reply #3 on: May 12, 2019, 02:23:42 pm »
This would be the simplest and most flexible

   
 

Online MarkF

  • Super Contributor
  • ***
  • Posts: 1163
  • Country: us
Re: DPDT high current powerboard switch
« Reply #4 on: May 12, 2019, 02:34:59 pm »
This is a second option

   
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Re: DPDT high current powerboard switch
« Reply #5 on: May 12, 2019, 03:07:16 pm »
Made a schematic for an auto changeover circuit.

Transfer switch is not required, only a safety precaution if the NC contactor failed. Not sure about AU compliance...??
 

Online MarkF

  • Super Contributor
  • ***
  • Posts: 1163
  • Country: us
Re: DPDT high current powerboard switch
« Reply #6 on: May 12, 2019, 03:31:21 pm »
For more relevant responses, please set your country code in your forum profile.

thanks
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Fully automatic transfer/changeover contactor circuit
« Reply #7 on: May 12, 2019, 03:34:49 pm »
Ahh yup thanks, will do.
Here is a schematic for a fully auto transfer/changeover circuit, with part references.

Parts: timer, NO contactor -ebay, NO/NC contactor -ebay.

Think that is solved!  ;D

Around AU$150 using standard certified DIN components.

There is a small time delay I am guessing is necessary due to the phase change between circuits.
Possible modification, I am thinking of a possible return current from the system, to add a ~2s delay off timer between the NO/NC contactor and the generator, to avoid the possibility of power in the system feeding back to the generator power inlet, which will just be a normal 10A plug. This delay off contactor is only rated at 8A though (2000VA) ((Generator is 3KVA)). Or this would require finding a different NO/NC contactor with 12v DC coil, and triggering it with the delay off contractor, which could be kept powered by super caps after power is lost...
« Last Edit: May 13, 2019, 01:52:09 pm by phaseform »
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 1625
Re: DPDT high current powerboard switch
« Reply #8 on: May 13, 2019, 08:49:30 am »
Here be dragons...

You really want a pair of mechanically AND electrically interlocked contractors to guarantee that in no condition does the generator back feed the grid even if one contactor fails closed due to contacts welding.

The usual approach is to leaverage a mechanically interlocked motor reversing contactor pair (with modified wiring) with two normally closed aux contact blocks to handle the electrical interlocking.

There is probably legislation and there may be certification requirements around transfer switches.

I do wonder why you don't just tee your heavy loads off the main grid feed before the changeover device, makes everything simpler, but if you go down the motor reversing contactor pair route you will find the contactors are all three pole so using the third pole on the grid contactor would work if you needed this for some weird reason.

Regards, Dan.
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Re: DPDT high current powerboard switch
« Reply #9 on: May 13, 2019, 06:40:55 pm »
Ahh cool thanks Dan.
I’ve made a schematic using mechanical & electrically interlocked contractors.

Auto switch to generator when Mains goes down, auto switch to Mains when back up. Option to kill generator when mains comes back up.

switching contactor such as this, timer as in above post. Still uses a couple extra DIN mount devices, specifically timers and a second contactor to disconnect the generator when mains comes back.

I put the RCD downstream of the contactors as here, although that wouldn't suffice. The Green Strobe indicates when the generator can be powered down, it should be paired to a NC switch connected to the Generator Contactor, assuming an NC accessory can be fitted next to the electro/mechanical interlock.

Just had a crazy idea, wire the generator ignition coil wiring through the Mains side of the electrical NC interlock, assuming that such a thing makes sense (not the actual spark plug wiring ofcourse).. that would make the strobe redundant.

Parts for the electro/mechanically interlocked automatic changeover contactor circuit alone are:
  • 3 pole NO contactor x2
  • suitable NC electro/mechanical interlock
  • 240v Coil NC contactor
  • 2s timer x2
  • MCBs for timers/contactor coils
« Last Edit: May 13, 2019, 11:15:59 pm by phaseform »
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 1625
Re: DPDT high current powerboard switch
« Reply #10 on: May 15, 2019, 03:38:43 am »
I suspect you are over thinking it, and don't forget you need to be switching the neutrals as well.

Also, in a TN-C-S scenario (which I think much of .au is?) you need to think carefully about exported earths from the equipotential zone to the generator....

I think if you search you will find all sorts of rules from the supply company about this stuff, particularly if you are in the northern territories where they get funny about the untrained changing a light fitting!

The usual approach is that you wire each contactor coil thru the normally closed aux contact on the OTHER contactor so that if one is engaged the other one cannot be pulled in.

Then you have a suitably butch HRC fuse from each of the input sides (PSC is an issue here) to supply power to the appropriate contactor and maybe a cam switch to forcibly drop one or the other out if you want to manually switch.

If you are feeling like it you can add frequency and voltage sensing relays to the generator side to prevent it from engaging the contactor if the frequency is too far off, and of course status lamps are easily wired across the appropriate contactor coils to indicate which source is selected.

Regards, Dan.
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Re: DPDT high current powerboard switch
« Reply #11 on: May 18, 2019, 08:19:42 pm »
ahh yup status lamps sound like a good idea. I guess would be good to plan this with a licensed electrician. In this diagram they mention that only the generator neutral can be switched. Not sure exactly how to think about earthing, my generator is just a simple 3kW Honda generator. Didn't actually think about this at all. Not exactly sure about HRC or PSC terms, they would be different from the RCD and MCBs downstream from the contactors? I would have put them up from the contactors, just going off that linked PDF..
yup it appears a certificate of compliance is required. Didn't see anything new for local Service & Installation Rules

Had to draw it up...
« Last Edit: May 18, 2019, 09:53:35 pm by phaseform »
 

Offline Gregg

  • Frequent Contributor
  • **
  • Posts: 535
  • Country: us
Re: DPDT high current powerboard switch
« Reply #12 on: May 20, 2019, 02:25:33 pm »
Far be it for me to discourage over thinking projects, but it may simplify your life to put all of the essential items that you want backed up to the generator on a separate sub panel (if sub panels are allowed where you are) and let everything else remain connected to the utility grid. The downside is that your transfer switch has to be capable of powering the entire sub panel, but that may be a good thing for future generator upgrades.

As Dan stated, you need to switch the neutrals even though they may both be ground referenced; you don’t want your ground wires conducting power and you do not want circulating neutral currents.  Also it is a very good idea to make sure your generator has stabilized via a frequency sensor to make sure you don’t damage anything; the time delay to transfer to generator should start only after the generator has stabilized.  The same goes for re-transfer back to utility; often after an outage there is a substantial time period before the line power is truly stable.  It is often cheaper to run the generator a little longer than to have it transfer to a bad power source.  In data centers we used 30 minutes delay before re-transfer, but it all depends on your location and tolerance for poor power quality.  I strongly suggest timers that can be adjusted.

Since you are really into this project, I also suggest you make a manual wraparound bypass so that you can take your creation out of circuit to work on it or manually transfer in case something goes horribly wrong. 

Also an automatic engine start contact and a time delay engine stop to allow the engine to cool down a bit after transferring to utility and to have it still running should the utility power become unstable partly due to your putting a load back on it or due to storm damage to the lines etc.

(edit:) You also don't want the transfers to happen too quickly especially when powering motors as the transfer can happen out of phase.  It is best to have a null time delay as well as the other delays mentioned.

« Last Edit: May 20, 2019, 02:37:46 pm by Gregg »
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Re: DPDT high current powerboard switch
« Reply #13 on: May 22, 2019, 04:16:01 pm »
A null time delay?
Half an hour timer for changeback to mains? Would require a Mains start override or some more complicated logic me thinks
 

Offline Gregg

  • Frequent Contributor
  • **
  • Posts: 535
  • Country: us
Re: DPDT high current powerboard switch
« Reply #14 on: May 23, 2019, 09:36:37 am »
Before I retired I did power design and engineering for huge data centers.  There were many, many large transfer switches, thousands of amps at 480 volts and generators to 2 MW each; generator paralleling, dual utility sources and really big UPS systems.  The generators all had block heaters to keep them ready to go and the typical time from power loss to transfer was 15 seconds.  15 seconds seems like a long time when wondering if it is going to work or if you will be spending hours writing reports as to why it didn’t work.
One site had a 3000 amp Zenith transfer switch that transferred so fast the 300KVA Liebert (junk) UPS tripped (into bypass but with an interruption).  We had a third party power testing company take lots of data and made a report that the UPS should have been able to handle the transfer.  There was no possibility to make the transfer switch slower or stop at a null point and it was cheaper to replace the UPS than the transfer switch.  Enclosed are a couple pictures of the transfer switch and controls

Here is a basic sequence with some parameters to consider:

1.   Utility power goes off
2.   You may or may not want a time delay to see if the power comes back on
3.   Generator start / run signal
4.   Time delay to make sure the generator is warmed up and frequency stabilized (frequency sensor may be a substitute here)
5.   Transfer the critical load to the generator (if there are things with large inrush, you may want to figure a way to stagger their transfer and not stall the generator or trip its breaker)
6.   Utility power comes back on; in many cases it is unstable for many reasons and sometimes it flickers on and off frequently due to branches blowing against the lines etc.)
7.   Time delay to make sure the utility power is stable
8.   Transfer load from generator to OFF position
9.   Short time delay (null) to make sure the transfer doesn’t happen too fast and partially out of phase.  Things like back EMF of motors can wreak havoc with a really fast transfer between two power sources.  5 to 10 seconds should be sufficient
10.   Transfer the power back to utility
11.   Time delay to stop generator to allow it to cool down plus if it is running and the power goes out again you can skip to step 5 above if you have it programmed for such occurrences.

If you are buying contactors, relay’s etc. it may be a good idea to get items that have a means of stopping in an off position; even if you don’t think you need it now it may be handy in the future for only a few dollars more.  Also make it easy to bypass your transfer devices and manually operate the system.  Don’t forget to label everything.
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 5112
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: DPDT high current powerboard switch
« Reply #15 on: May 25, 2019, 05:15:56 am »
Not sure how applicable it is for your load, but how come there hasn't been more investment into automatically switching loads into lower power states when running on backup power? I did try an experiment using a small MOSFET to pull down the PROCHOT line (active low signal used to indicate that the CPU or its voltage regulator is running hot) on a low power Ivy Bridge platform and if the CPU was at full load, the power draw was almost cut in half (from ~20W to ~12W) pretty much instantly. I'd imagine the power savings would be even greater on larger CPUs. A software approach would probably be able to do even better but would take longer to respond. (A combined approach would probably work the best.)

I can see that working with either some IP based protocol on a management network or a high frequency signal superimposed on the power lines that carries the current state of the power delivery infrastructure. In any case, cutting down on the backup power requirements to half or less is a big savings for a datacenter.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline phaseform

  • Contributor
  • Posts: 42
  • Country: au
Re: DPDT high current powerboard switch
« Reply #16 on: June 04, 2019, 09:05:10 am »
Thanks Gregg, very interesting!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf