How to wire up a 240VAC receptacle

[Richard Crowley]

Because we don't have an L1 and L2,

But we DO have L1 and L2. Those are the names of the two "hot" wires coming into your house.  Between L1 and L2 are 240VAC.

In North America, the center-tap between L1 and L2 is grounded and all the 120V loads are connected between the center tap (called "Neutral") and EITHER L1 or L2, so you get half the incoming 240V  And 240V loads are connected BETWEEN L1 and L2.

In most of the rest of the world, they use L1 and L2 (240V) directly.  In many cases "L1" is grounded and called "Neutral". That is the industrial standard and that is what your PDU is designed for.

In some parts of the 240V world, BOTH sides (L1 and L2) are "above ground" and not even ground-referenced at all. They must treat ALL the wires as "hot" except the green-wire safety ground (which they call "PE" Proective Earth)  Because of this, much (most?) 240 volt equipment is designed with the assumption that BOTH sides are "hot".  That is why I find it so curious that you claim that your loads (servers) won't operate if both L1 and L2 are "hot".

because the 240VAC is on the same phase, where do the electrons go if they're not hooked to a neutral or ground to complete the circuit?    Somehow they have to make it back to the breaker, correct? 

The electrons go BETWEEN L1 and L2.  When you say "because the 240VAC is on the same phase" it makes no sense.  That is like saying that a yard-stick is 36 inches long no matter if you hold it at one end or the other end.  Or if you hold it in the middle ("split-phase")

I use the word ground, but all I mean is a completed circuit.   I know and understand that a completed circuit doesn't need a ground or a voltage potential of 0V.   I could technically power a circuit where the supply is 24V and the return is 12V.   So long as the return is at a lower voltage potential than the supply, we can have the electrons moving.

Yes, for 1/120th of a second. And then 1/120th of a second later, the 240V has changed to the opposite polarity (which is why it is called ALTERNATING current.

Using the water analogy that everyone seems to like to use, we have our "pump" (the transformer I'd buy for the PDU, for example).   The "water" (electrons) flow out of L1 into the PDU.   But then what?

The current flows around the loop from L1 through L2 back to the transformer on the pole.  The center-tap is there to split the 240V in half so that our historic North American standard of 120V loads can get 120V and not the full 240V coming into your house.

They need to make it back to the "pump" (transformer) somehow.    I know you say ground / neutral is irrelevant, but in this case, I cannot imagine how those pesky little buggers would make it back to complete the circuit without using one of those other wires that are at 0V potential.

   
Neither L1 nor L2 are  "at 0V potential".  Even "Neutral" wouldn't be "at 0V potential" unless it was (artificially) grounded.  The current flow of a 240V load never sees neutral/ground.

The components inside the PDU would be the loads

No. The things (computers) plugged into the PDU are the loads.  The PDU is just a fancy, expensive power-strip.

and by the time the current / electricity had passed through the entire circuit and was ready to leave the PDU, it surely wouldn't just be traveling up the L1 wire again, would it? 

A 240V load current passes between L1 and L2.  It does not know or care whether either side (or the center-tap) is grounded or not.

We'd have to use one of the additional neutral / ground wires in that instance. 

NO!  240V loads use the current between L1 and L2. Neutral/ground plays no part here.  Neutral is provided only for 120V loads.

Just like if we're dealing with 120VAC receptacles.   We cannot just hook up the hot wire.   That wouldn't do diddly squat.   We'd still have to hook up the neutral so the electrons could make it back to the panel.

Yes. From the hot wire (which is either L1 or L2) to Neutral is 120V.  But between L1 and L2 is 240V.  That is what 240V loads use.

With the 120-0-120, I understand how it works now.   I also understand why if I were to not use a double pole breaker but tried wiring up a NEMA L6-30R using a 30-amp single pole breaker in slot 1 and slot 2 (opposites sides of the panel), it wouldn't work.   I'd have 120VAC, not 240VAC.

Huh?  There is no such thing as a "single pole breaker in slot 1 and slot 2". Each "slot" is a pole. Between each "slot" and Neutral is 120V. But between two adjacent "slots" is 240V because all the odd number slots are connected to L1 and all the even number slots are connected to L2.  Do not be confused thinking that the left side of the breaker panel is L1 and the right side is L2. The "slots" alternate between L1 and L2 so that you can use a double breaker to provide a 240V branch circuit from two adjacent slots.

Zoom in on this photo to see the details.  L1 and L2 come in at the top from the transformer out on the pole (or buried under the sidewalk). The Neutral comes in to that big terminal at the upper right.  And it is connected to the light-colored screw terminal strip on the right side.  If you look very carefully, you will see that one of the screws in the neutral strip is colored green.  That is where the Neutral is connected to Ground in the North American standard.

Then EVERY OTHER slot is connected to either L1 or L2.  So if you put in a single-pole breaker, you connect to EITHER L1 or L2 and the return path comes back through the Neutral.  But if you install a double-pole breaker one pole connects to L1 and the other pole connects to L2. Because between L1 and L2 is 240V and the load current comes from L1, passes through the load, and returns to L2.  At least for 1/120th of a second.  And in the next 1/120th of a second the current flows from L2, through the load and back to L1.

Notice that during the discussion of a double-pole, 240V circuit there was no mention of Neutral.  And I didn't have one up my sleeve, either.

[IanB]

I'm sorry, I haven't gotten a lot of sleep lately, but yes, you are correct.   The NEMA L6-30P / L6-30R only has 3 wires, not four.    So in Europe and other parts of the world that use 240VAC, it's just like ours, where 120V goes down one wire, 120V goes down another?

No, there's no 120 V anywhere. Recall that voltage is measured as the difference between two wires. There are only two power conductors in the L6-30 and they carry 240 V. There is simply 240 V between the two power conductors. That's it. End of story. There is no 120 V in a correctly wired L6-30 plug or receptacle. Only 240 V. (Ignore the ground for this discussion, it is there for safety, not for power delivery.)

This is what my European friend was talking about I think when he said it was a language barrier and that Neutral was meant to be hot.   If that's the case though, and they have two wires carrying the 240V load, then my PDU wouldn't work over there either, would it?   Because we still have single pole breakers.   They'd need 1-phase 240VAC, not two-phase.

Breakers are about safety and system protection. If both power conductors are "hot" then each power conductor needs a breaker (a "2-pole breaker"). In North America the 208 V AC or 240 V AC supply typically has two hot conductors, so it needs the 2-pole breaker. This is what the NA/Japan models of PDU provide.

In other parts of the world only one power conductor may be "hot", in which case only the hot conductor needs a breaker (a "single pole breaker"). This is what the INTL or WW models of PDU provide. They are suitable for parts of the world like Europe, but they are not code compliant for use in North America.

Again, this is all about safety and system protection and code compliance. It is not about voltages or power requirements.

[james_s]

That's what I've been missing.   I wasn't talking about the transformer here, I was talking about 120-0-120 split-phase that comes into the house, and trying to see how it's equivalent to 240VAC over in Europe.   Because over there, they'll have L1 and then Neutral hooked to the heating elements, which would still give the 240VAC.   I think I understand now fully how the AC works.

Now, why would connecting earth ground to a 240VAC source that comes from 120-0-120VAC USA split-phase or whatever you want to call it cause a short there?   It's because L1 and L2 are reciprocals of each other, right?  When one is 120VAC, the other is -120VAC.   When one is 75VAC, the other is -75VAC.   When one is zero, the other is zero.   If we added a ground, when either where anything other than 0VAC, we'd have our direct connection to ground, which would be horrible.   Am I finally understanding this?

Look at the schematic of the center tapped transformer winding, you have 3 wires coming out of that transformer, L1 and L2 are the ends and CT from the middle. Now in the North American setup CT is connected to ground, so now what happens when you also connect either L1 or L2 to ground? Draw it on paper if you have to but you should see a direct short across half of the secondary.

[Spork Schivago]

With that in mind, if you were me, would you take the chance and send the PDU back and order the other two, plus the additional equipment to get it running?  Or would you just purchase the transformer and new breaker panel?

I would send the PDU back and order the other two PDUs. The reason being that that other PDUs are designed and specified for installation in North America and they comply with the electrical and safety requirements of the North American electrical code.

Your existing PDU is not code compliant and is not suitable for connection to a North American electrical supply. It is intended for another market elsewhere in the world. This is why it has a strange plug on it and why electricans don't want to touch it.

You could, theoretically, install the transformer and the sub-panel, but it's going to be really complicated getting it to pass inspection because it will be a strange and abnormal setup.

If you don't want to take the responsibility of ordering the other PDUs just on the advice of this forum, you should find someone in your line of business to consult who can offer appropriate expert advice. I do not think you have been well advised by the HPE sales consultants in your journey so far.

Yeah, I agree with you on the HPE sales consultants.   We had this issue with Microsoft as well.   They've gotten so big, they can't keep track of all the licenses and agreements.   For example, one sales person said we could just use Windows Home edition to run our business, another said Pro or higher, another said we needed VL editions, finally, a sales rep said we needed the CSP (Cloud Service Provider) agreement, which we signed up for.   Then, on Spice Works, a Senior Project Manager who works directly for Microsoft and has an account on Spiceworks, helping people with licenses and everything, says no, CSP is not meant for on-premise servers.   If we hosted our servers off-premise or in the cloud, we'd be okay.   There was some hassle between getting my one VAR who sold us the licenses to talk to this guy, but I don't think they realized how much pull he had.   And he made contact with them, letting them know he wasn't just some idiot on the net surfing the forums.    I guess still, him and the guy who sold us the CSP are arguing over it.

Hopefully, we get to keep the CSP stuff, because I like it and it's a hell of a lot cheaper than purchasing Windows Server edition, now that  it's Core-Based, especially with the number of VMs we need.

I think maybe the same happened with HPE, or perhaps, the technician saw a chance to up-sell me on the PSU and said hey, you need this 1400 watt PSU.    The only reason this PDU was picked was because it was supposedly compatible with the PSUs.    I hate having to have a whole bunch of PDUs in the rack, but if that's the way I gotta do it, that's the way I gotta do it.   I just really hope you guys are right about this being able to run off 120-0-120VAC and they (HPE) are wrong.   The only thing I worry about on the PSU is that intelligent connector for the intelligent PDUs.   The cables aren't normal PSU cables.   The input on the PSU, it's got inside of it, three pins to power it, and then on the bottom, underneath those three pins, it has four pads.   It's those four pads I worry about.   The 500 watt ones that were in it, before the upgrades, they ain't got those pads there but those were made for not the intelligent PDUs.

I'll have to wait until Monday, but I'll see if I can send the PDU back.   I've had it a while now, because I was working on getting the transformer money around and all that jazz.

[Spork Schivago]

slot 1 and slot 2 are on the same phase, from the panels perspective

No, slot 1 and slot 2 are on the opposite phase, from the panel's perspective. That is why slot 1 and slot 2 have 240 V between them.

On my panel, double pole breakers go every other, slot 1 and slot 3, or slot 2 and slot 4.   So I'd think 1 and 2 would both be on L1, where 3 and 4 would be on L2.

[Spork Schivago]

Because we don't have an L1 and L2,

But we DO have L1 and L2. Those are the names of the two "hot" wires coming into your house.  Between L1 and L2 are 240VAC.

In North America, the center-tap between L1 and L2 is grounded and all the 120V loads are connected between the center tap (called "Neutral") and EITHER L1 or L2, so you get half the incoming 240V  And 240V loads are connected BETWEEN L1 and L2.

In most of the rest of the world, they use L1 and L2 (240V) directly.  In many cases "L1" is grounded and called "Neutral". That is the industrial standard and that is what your PDU is designed for.

In some parts of the 240V world, BOTH sides (L1 and L2) are "above ground" and not even ground-referenced at all. They must treat ALL the wires as "hot" except the green-wire safety ground (which they call "PE" Proective Earth)  Because of this, much (most?) 240 volt equipment is designed with the assumption that BOTH sides are "hot".  That is why I find it so curious that you claim that your loads (servers) won't operate if both L1 and L2 are "hot".

HPE claims the loads won't work if I try wiring this PSU up to 120-0-120.   But I don't know if this matters, there's more inputs than just the three wires on the 1400 watt PSU.   There's actually a total of 7.   4 of which are made to communicate with the intelligent PDUs (which I don't have, and didn't want, because I don't like them but might  have been able to purchase one of those).   This was the only reason I was thinking HPE might have been correct, those four pads inside the connector there....but I'm going to try and send the PDU back now and purchase the ones made for 120-0-120 and see how it actually works, see if they're wrong and you guys are right.   You guys seem to really know your shit, and everything you guys have said has made since.

because the 240VAC is on the same phase, where do the electrons go if they're not hooked to a neutral or ground to complete the circuit?    Somehow they have to make it back to the breaker, correct? 

The electrons go BETWEEN L1 and L2.  When you say "because the 240VAC is on the same phase" it makes no sense.  That is like saying that a yard-stick is 36 inches long no matter if you hold it at one end or the other end.  Or if you hold it in the middle ("split-phase")

I use the word ground, but all I mean is a completed circuit.   I know and understand that a completed circuit doesn't need a ground or a voltage potential of 0V.   I could technically power a circuit where the supply is 24V and the return is 12V.   So long as the return is at a lower voltage potential than the supply, we can have the electrons moving.

Yes, for 1/120th of a second. And then 1/120th of a second later, the 240V has changed to the opposite polarity (which is why it is called ALTERNATING current.

Using the water analogy that everyone seems to like to use, we have our "pump" (the transformer I'd buy for the PDU, for example).   The "water" (electrons) flow out of L1 into the PDU.   But then what?

The current flows around the loop from L1 through L2 back to the transformer on the pole.  The center-tap is there to split the 240V in half so that our historic North American standard of 120V loads can get 120V and not the full 240V coming into your house.

But we're not talking about the center tapped 120-0-120VAC source.   I'm talking about the transformer, if I bought that.   But after you saying we still have an L1 and L2, I believe I understand.   It doesn't matter, to the PDU, if L1 and L2 are both hot (120VAC different phases) or one hot (240VAC) and the other neutral.   It acts the same.

They need to make it back to the "pump" (transformer) somehow.    I know you say ground / neutral is irrelevant, but in this case, I cannot imagine how those pesky little buggers would make it back to complete the circuit without using one of those other wires that are at 0V potential.

   
Neither L1 nor L2 are  "at 0V potential".  Even "Neutral" wouldn't be "at 0V potential" unless it was (artificially) grounded.  The current flow of a 240V load never sees neutral/ground.

The components inside the PDU would be the loads

No. The things (computers) plugged into the PDU are the loads.  The PDU is just a fancy, expensive power-strip.

Yes, but each component inside the PDU is also a load.   And because there's circuitry that's running with nothing plugged into the PDU at all, the PDU is still considered a load itself.   If it wasn't, it'd be 100% efficient, instead of 80% efficient while running at x voltage, 85% efficient while running at y voltage, and 90% efficient while running at z voltage.

and by the time the current / electricity had passed through the entire circuit and was ready to leave the PDU, it surely wouldn't just be traveling up the L1 wire again, would it? 

A 240V load current passes between L1 and L2.  It does not know or care whether either side (or the center-tap) is grounded or not.

We'd have to use one of the additional neutral / ground wires in that instance. 

NO!  240V loads use the current between L1 and L2. Neutral/ground plays no part here.  Neutral is provided only for 120V loads.

You're saying if I kept the PDU, bought that transformer, neutral / ground plays no part?   L1 and L2 are still hot?    I thought the whole purpose of the transformer was to get one 240VAC L1....because we only have one-pole breakers inside of the PDU....If L1 and L2 both have current going through them, how would this transformer protect me if one of those internal breakers where to trip?   We'd still have current going through half the output.

Just like if we're dealing with 120VAC receptacles.   We cannot just hook up the hot wire.   That wouldn't do diddly squat.   We'd still have to hook up the neutral so the electrons could make it back to the panel.

Yes. From the hot wire (which is either L1 or L2) to Neutral is 120V.  But between L1 and L2 is 240V.  That is what 240V loads use.

With the 120-0-120, I understand how it works now.   I also understand why if I were to not use a double pole breaker but tried wiring up a NEMA L6-30R using a 30-amp single pole breaker in slot 1 and slot 2 (opposites sides of the panel), it wouldn't work.   I'd have 120VAC, not 240VAC.

Huh?  There is no such thing as a "single pole breaker in slot 1 and slot 2". Each "slot" is a pole. Between each "slot" and Neutral is 120V. But between two adjacent "slots" is 240V because all the odd number slots are connected to L1 and all the even number slots are connected to L2.  Do not be confused thinking that the left side of the breaker panel is L1 and the right side is L2. The "slots" alternate between L1 and L2 so that you can use a double breaker to provide a 240V branch circuit from two adjacent slots.

Zoom in on this photo to see the details.  L1 and L2 come in at the top from the transformer out on the pole (or buried under the sidewalk). The Neutral comes in to that big terminal at the upper right.  And it is connected to the light-colored screw terminal strip on the right side.  If you look very carefully, you will see that one of the screws in the neutral strip is colored green.  That is where the Neutral is connected to Ground in the North American standard.

Then EVERY OTHER slot is connected to either L1 or L2.  So if you put in a single-pole breaker, you connect to EITHER L1 or L2 and the return path comes back through the Neutral.  But if you install a double-pole breaker one pole connects to L1 and the other pole connects to L2. Because between L1 and L2 is 240V and the load current comes from L1, passes through the load, and returns to L2.  At least for 1/120th of a second.  And in the next 1/120th of a second the current flows from L2, through the load and back to L1.

Notice that during the discussion of a double-pole, 240V circuit there was no mention of Neutral.  And I didn't have one up my sleeve, either.

Yes, that's exactly what I was saying.   Every other slot is connected to L1 or L2.   If you where to hook single pole breakers to slot 1 and 2 (where slot 1 is hooked to L1 and slot 2 is hooked to L1), you won't get the 240VAC.   You'd get 120VAC.   You'd just be able to draw more current.   This is why those people that suggest making a DIY plug that just plugs into the two inputs on the same receptacle and are supposed to provide 240VAC will never work.   You need the supply from L1 and L2, not L1 and L1, or L2 and L2.

What I was saying was you take two single pole breakers, if you connect one to slot number 1 in the panel, and connect one to slot number 3, you've essentially created a double-pole breaker.   It wouldn't be safe, because if one breaker tripped, there's no guarantee the other would.    slot number 1 and 3 are L1, for example, 2 and 4 are L2 for example.   If you were to connect two single pole breakers, one to slot 1 and one to slot 2, and wire them expecting 240VAC between them, you wouldn't get it because both breakers would be on L1 or both breakers would be on L2.   This was my way of showing that I understood it, describing back what I learned, in my own words.

[IanB]

On my panel, double pole breakers go every other, slot 1 and slot 3, or slot 2 and slot 4.   So I'd think 1 and 2 would both be on L1, where 3 and 4 would be on L2.

Fair enough, I probably just used the wrong words. I'm just trying to say that for a 240 V circuit the double pole breaker connects between L1 and L2, and this gives 240 V single phase. In the house wiring L1 would be the black wire and L2 would be the red wire (or vice versa--it doesn't really matter). To wire a 240 V circuit from this breaker you would use a three conductor cable with red, black and ground wires.

[IanB]

This was the only reason I was thinking HPE might have been correct, those four pads inside the connector there....but I'm going to try and send the PDU back now and purchase the ones made for 120-0-120 and see how it actually works, see if they're wrong and you guys are right.

Please don't say "120-0-120" or "split phase" or anything like that to the sales people. It will cause confusion and get you into trouble.

What you need to say is "200 to 240 V AC single phase for use in the USA". A compatible PDU should be provided with 2-pole breakers for complete isolation of the supply, and the PSUs actually don't care. As long as the PSU is designed for "high-line" 200 to 240 V AC input it will be fine.

[IanB]

A quick summary.

Here are the PDU specs:



The P9S13A says "North America/Japan, single phase 200 to 240 V input, 24 amps input, fitted with two 2-pole 20 A breakers." Therefore it can distribute 24 A to servers or other equipment.

Here are the PSU specs:



It can run off 200 to 240 V AC, single phase, at 50 to 60 Hz. Running at full load at 240 V it will draw 6.5 amps. Therefore you could probably run four of these PSUs from one P9S13A PDU, assuming that the PSUs are not going to be running flat out. (Notice it also gives the heat rating in BTU/hr to estimate the required cooling in the server room. The air conditioning unit has to remove 5300 BTU/hr for each one of these power supplies.)

[Spork Schivago]

I'm sorry, I haven't gotten a lot of sleep lately, but yes, you are correct.   The NEMA L6-30P / L6-30R only has 3 wires, not four.    So in Europe and other parts of the world that use 240VAC, it's just like ours, where 120V goes down one wire, 120V goes down another?

No, there's no 120 V anywhere. Recall that voltage is measured as the difference between two wires. There are only two power conductors in the L6-30 and they carry 240 V. There is simply 240 V between the two power conductors. That's it. End of story. There is no 120 V in a correctly wired L6-30 plug or receptacle. Only 240 V. (Ignore the ground for this discussion, it is there for safety, not for power delivery.)

I think you're misunderstanding what I'm saying.   I correctly wired my NEMA L6-30P.  When L1 is referenced to L2 there's ~240VAC.   When L1 is referenced to Ground there's ~120VAC.   When L2 is referenced to Ground there's ~120VAC.   I understand the Ground is solely for safety purposes, trust me.   I'm asking, in Europe and other parts of the world, where most appliances are designed to run off 240VAC, not 120VAC, on their receptacles, when L1 is referenced to Ground, is it ~240VAC or ~120VAC?   When L2 is referenced to ground, is ~120VAC or 0VAC?   That's what I'm asking.   I know what it is in the USA with the NEMA plugs.   I know the 240VAC is acquired by referencing L1 and L2...

This is what my European friend was talking about I think when he said it was a language barrier and that Neutral was meant to be hot.   If that's the case though, and they have two wires carrying the 240V load, then my PDU wouldn't work over there either, would it?   Because we still have single pole breakers.   They'd need 1-phase 240VAC, not two-phase.

Breakers are about safety and system protection. If both power conductors are "hot" then each power conductor needs a breaker (a "2-pole breaker"). In North America the 208 V AC or 240 V AC supply typically has two hot conductors, so it needs the 2-pole breaker. This is what the NA/Japan models of PDU provide.

In other parts of the world only one power conductor may be "hot", in which case only the hot conductor needs a breaker (a "single pole breaker"). This is what the INTL or WW models of PDU provide. They are suitable for parts of the world like Europe, but they are not code compliant for use in North America.

Again, this is all about safety and system protection and code compliance. It is not about voltages or power requirements.

That's what I was asking.   In Europe, are there two power conductors that are hot or just one?   In America, with 240VAC using our 120-0-120 split phase, both conductors, L1 and L2 are hot, each providing 120VAC out of phase from each other by 180°.   Why 180°?   Because 180° + 180° = 360° so when L1 is +120VAC, L2 will be -120VAC, give or take.   L1 and L2 will should always be the same voltage, but opposite signs of each other.   So when L1 is at +20VAC, L2 should be at -20VAC.   That's why they're 180° out of phase from each other.   If they where something like 120° out of phase with each other, L1 might be +120VAC, and L2 wouldn't be -120VAC.

[Monkeh]

In the majority of european countries (to my knowledge, I am aware some countries have three phase domestically), domestic properties are supplied with one phase and a neutral from a 3-phase 400V supply. So L is 230V to N, there is no L2, and N is nominally 0V to ground. This is what the PDUs you have are designed for.

[Spork Schivago]

That's what I've been missing.   I wasn't talking about the transformer here, I was talking about 120-0-120 split-phase that comes into the house, and trying to see how it's equivalent to 240VAC over in Europe.   Because over there, they'll have L1 and then Neutral hooked to the heating elements, which would still give the 240VAC.   I think I understand now fully how the AC works.

Now, why would connecting earth ground to a 240VAC source that comes from 120-0-120VAC USA split-phase or whatever you want to call it cause a short there?   It's because L1 and L2 are reciprocals of each other, right?  When one is 120VAC, the other is -120VAC.   When one is 75VAC, the other is -75VAC.   When one is zero, the other is zero.   If we added a ground, when either where anything other than 0VAC, we'd have our direct connection to ground, which would be horrible.   Am I finally understanding this?

Look at the schematic of the center tapped transformer winding, you have 3 wires coming out of that transformer, L1 and L2 are the ends and CT from the middle. Now in the North American setup CT is connected to ground, so now what happens when you also connect either L1 or L2 to ground? Draw it on paper if you have to but you should see a direct short across half of the secondary.

Yes, with North American setup.   But with the transformer I was going to purchase, I was under the impression if you connect L2 to ground there is no short.   I was under the impression that was the whole reason I needed to purchase that specific transformer that Mr. Crowley originally suggested, because of the single-pole breakers inside of the PDU that I was sold.   So L1 is the only hot wire, with ~240VAC in reference to ground or in reference to L2 (but only with that transformer Mr. Crowley suggested I purchase if I were to continue to keep the current PDU and wanted to maintain safety.)   That way, if one of the breakers tripped, it'd cut the path of L1, but L2 and N would still be active, but wouldn't be a danger.    I think you guys missed the part where I said the transformer that I was going to purchase and think that I'm talking about the transformer on the pole in North America.

With the transformer Mr. Crowley suggested I purchase if I were going to keep the PDU, there shouldn't be an L2 on the secondary side, if I understand everything correctly.   There's only an L1, an N, and an E.   With American transformers, yes, if we connect either L1 or L2 to ground, we get a direct short and the breaker (hopefully!) trips to protect us.   Or, with that transformer Mr. Crowley suggested I purchase if I were going to keep the PDU, on the PRIMARY side, if I connected L1 OR L2 to ground, we'd have a dead short and that'd be horrible and hopefully the double-pole breaker on the mains panel would protect us and trip.   But if we were to be doing that, we shouldn't be playing with electricity until we learned a lot more about it.

[Spork Schivago]

This was the only reason I was thinking HPE might have been correct, those four pads inside the connector there....but I'm going to try and send the PDU back now and purchase the ones made for 120-0-120 and see how it actually works, see if they're wrong and you guys are right.

Please don't say "120-0-120" or "split phase" or anything like that to the sales people. It will cause confusion and get you into trouble.

What you need to say is "200 to 240 V AC single phase for use in the USA". A compatible PDU should be provided with 2-pole breakers for complete isolation of the supply, and the PSUs actually don't care. As long as the PSU is designed for "high-line" 200 to 240 V AC input it will be fine.

Good to go!   Now, when you say don't say 120-0-120 or split phase or anything like that to the sales people, is that because it's incorrect or it will just confuse them?   Am I using the wrong terminology here?   I understand the difference between technicians and sales people and I learned a long time ago when I was working on PCs for customers to not get technical.   Keep it simple, no need to confuse them, just say it's fixed, you owe me blah.   Found a bad component or two (something along those lines), rather than saying the capacitors in the VRM circuit blew because of them being directly under the heatsink and not having adequate cooling or the proper heatsink, so I replaced them with high temp Panasonic's and purchased a taller, yet not as wide heatsink with a fan that could push more air per CFM through the heatsink then the previous one.

[Monkeh]

This was the only reason I was thinking HPE might have been correct, those four pads inside the connector there....but I'm going to try and send the PDU back now and purchase the ones made for 120-0-120 and see how it actually works, see if they're wrong and you guys are right.

Please don't say "120-0-120" or "split phase" or anything like that to the sales people. It will cause confusion and get you into trouble.

What you need to say is "200 to 240 V AC single phase for use in the USA". A compatible PDU should be provided with 2-pole breakers for complete isolation of the supply, and the PSUs actually don't care. As long as the PSU is designed for "high-line" 200 to 240 V AC input it will be fine.

Good to go!   Now, when you say don't say 120-0-120 or split phase or anything like that to the sales people, is that because it's incorrect or it will just confuse them?   Am I using the wrong terminology here?

The terminology is fine, but the people on the other end of the phone have absolutely no idea about the concepts of electrical installations. Even the techs you may speak to don't - they leave it all to the electricians and just know how to add simple numbers together for load calculations.

As I said earlier, all you'll get from the people you're able to contact is them reading '240VAC single-phase' off the datasheet and parroting it back. The only people directly involved in any of this hardware who know better are the ones designing the supplies - and they work for Delta and Flextronics, and speak Chinese.

[IanB]

That's what I was asking.   In Europe, are there two power conductors that are hot or just one?   In America, with 240VAC using our 120-0-120 split phase, both conductors, L1 and L2 are hot, each providing 120VAC out of phase from each other by 180°.   Why 180°?   Because 180° + 180° = 360° so when L1 is +120VAC, L2 will be -120VAC, give or take.   L1 and L2 will should always be the same voltage, but opposite signs of each other.   So when L1 is at +20VAC, L2 should be at -20VAC.   That's why they're 180° out of phase from each other.   If they where something like 120° out of phase with each other, L1 might be +120VAC, and L2 wouldn't be -120VAC.

Yes, in parts of Europe like the UK you have one live conductor at ~240 V referenced to ground and one neutral conductor at ~0 V.

Therefore in Europe, isolating the single live conductor and leaving the neutral conductor connected is safe (for example with a single pole breaker). Hence the "INTL" or "WW" PDUs have a single pole breaker.

However, using a single pole breaker in the USA where both conductors are hot is not safe. Both conductors have to be isolated (for example with a two pole breaker). Hence the "NA/Japan" PDUs are fitted with two pole breakers.

[Spork Schivago]

A quick summary.

Here are the PDU specs:



The P9S13A says "North America/Japan, single phase 200 to 240 V input, 24 amps input, fitted with two 2-pole 20 A breakers." Therefore it can distribute 24 A to servers or other equipment.

Here are the PSU specs:



It can run off 200 to 240 V AC, single phase, at 50 to 60 Hz. Running at full load at 240 V it will draw 6.5 amps. Therefore you could probably run four of these PSUs from one P9S13A PDU, assuming that the PSUs are not going to be running flat out. (Notice it also gives the heat rating in BTU/hr to estimate the required cooling in the server room. The air conditioning unit has to remove 5300 BTU/hr for each one of these power supplies.)

I wasn't even going to tell them I needed a PDU that runs off 200 - 240VAC made for North America with two-pole breakers.   My Account Executive is NOT a technician, he is the person that talks to the technicians.   So I was going to say on the phone or email on Monday, hey, is it too late to send that P9S16A PDU back and purchase a couple P9S13A PDU's?

I'll have to do the math real quick to determine how many PDUs I actually need.   I have more than just the servers in the rack, but it would be nice if I could divey up the servers between two different PDUs for redundancy purposes.

I had already looked at the heat they produce and calculated what size AC I was going to need.   I used those as a rough estimate for the other equipment, just to take an educated guess and overestimated a little.   It's a pretty big AC.

Anyway, when I ordered the plugs for all the equipment the PDU was going to power, I ordered a few extras, just for expansion later on and to have.   I came up with an idea.   I have that NEMA L6-30R wired up, and I wired the NEMA L6-30P myself for the BGA rework station.    I'm wondering if one of these servers could run with just one PSU, and I think it can right now, at least temporarily.    I'm thinking for an experiment, why not cut the end off one of those extra cords I ordered, unwire the NEMA L6-30P from the BGA unit, wire it up to the cord, plug the PSU into one of the servers, plug the cord into the PSU and see if it fires up?

According to HPE, it just won't start.   They didn't say it'd cause any damage or anything.   That way, I can just ease my mind, and then there's no doubts at all.....1400 watt PSUs....those plugs should have the correct gauge wire inside of them.

[IanB]

Yes, with North American setup.   But with the transformer I was going to purchase, I was under the impression if you connect L2 to ground there is no short.   I was under the impression that was the whole reason I needed to purchase that specific transformer that Mr. Crowley originally suggested, because of the single-pole breakers inside of the PDU that I was sold.   So L1 is the only hot wire, with ~240VAC in reference to ground or in reference to L2 (but only with that transformer Mr. Crowley suggested I purchase if I were to continue to keep the current PDU and wanted to maintain safety.)   That way, if one of the breakers tripped, it'd cut the path of L1, but L2 and N would still be active, but wouldn't be a danger.    I think you guys missed the part where I said the transformer that I was going to purchase and think that I'm talking about the transformer on the pole in North America.

With the transformer Mr. Crowley suggested I purchase if I were going to keep the PDU, there shouldn't be an L2 on the secondary side, if I understand everything correctly.   There's only an L1, an N, and an E.   With American transformers, yes, if we connect either L1 or L2 to ground, we get a direct short and the breaker (hopefully!) trips to protect us.   Or, with that transformer Mr. Crowley suggested I purchase if I were going to keep the PDU, on the PRIMARY side, if I connected L1 OR L2 to ground, we'd have a dead short and that'd be horrible and hopefully the double-pole breaker on the mains panel would protect us and trip.   But if we were to be doing that, we shouldn't be playing with electricity until we learned a lot more about it.

Broadly speaking this is correct--in theory. But in practice, if you installed the transformer Mr. Crowley suggested and created a secondary supply with a sub-panel, you would probably be getting into areas not routinely covered by the electrical code. You will be creating something that inspectors do not expect to see, and so they will be getting into difficult territory when trying to approve it. And the situation with your insurance company, should anything bad happen, would be similar. Bear in mind that insurance companies like finding ways not to pay claims...

So really the best strategy is to follow North American electrical practice as closely as possible, do everything to the letter of the code, and do not introduce strange and unfamiliar things into the system.

(Keep in mind that if you use the P9S13A that comes pre-fitted with an L6-30P plug, then all you have to do is to provide some correctly wired and installed 30 A circuits back to your breaker panel with L6-30R receptacles on the end of them for your PDUs to plug into. Everything after the receptacle will be off the shelf hardware designed for the North American market and all your inspector will have to do is to sign off on the L6-30R receptacles and circuits. Everything will be straightforward, recognizable, and to code. Your inspector will have an easy job and no worries.)

[IanB]

Anyway, when I ordered the plugs for all the equipment the PDU was going to power, I ordered a few extras, just for expansion later on and to have.   I came up with an idea.   I have that NEMA L6-30R wired up, and I wired the NEMA L6-30P myself for the BGA rework station.    I'm wondering if one of these servers could run with just one PSU, and I think it can right now, at least temporarily.    I'm thinking for an experiment, why not cut the end off one of those extra cords I ordered, unwire the NEMA L6-30P from the BGA unit, wire it up to the cord, plug the PSU into one of the servers, plug the cord into the PSU and see if it fires up?

Yes, you can plug a server PSU directly into an L6-30 receptacle as long as you have a correctly wired power cord between the PSU and the receptacle. As long as the PSU is designed to accept 240 V AC it should work fine.

[Spork Schivago]

Yes, with North American setup.   But with the transformer I was going to purchase, I was under the impression if you connect L2 to ground there is no short.   I was under the impression that was the whole reason I needed to purchase that specific transformer that Mr. Crowley originally suggested, because of the single-pole breakers inside of the PDU that I was sold.   So L1 is the only hot wire, with ~240VAC in reference to ground or in reference to L2 (but only with that transformer Mr. Crowley suggested I purchase if I were to continue to keep the current PDU and wanted to maintain safety.)   That way, if one of the breakers tripped, it'd cut the path of L1, but L2 and N would still be active, but wouldn't be a danger.    I think you guys missed the part where I said the transformer that I was going to purchase and think that I'm talking about the transformer on the pole in North America.

With the transformer Mr. Crowley suggested I purchase if I were going to keep the PDU, there shouldn't be an L2 on the secondary side, if I understand everything correctly.   There's only an L1, an N, and an E.   With American transformers, yes, if we connect either L1 or L2 to ground, we get a direct short and the breaker (hopefully!) trips to protect us.   Or, with that transformer Mr. Crowley suggested I purchase if I were going to keep the PDU, on the PRIMARY side, if I connected L1 OR L2 to ground, we'd have a dead short and that'd be horrible and hopefully the double-pole breaker on the mains panel would protect us and trip.   But if we were to be doing that, we shouldn't be playing with electricity until we learned a lot more about it.

Broadly speaking this is correct--in theory. But in practice, if you installed the transformer Mr. Crowley suggested and created a secondary supply with a sub-panel, you would probably be getting into areas not routinely covered by the electrical code. You will be creating something that inspectors do not expect to see, and so they will be getting into difficult territory when trying to approve it. And the situation with your insurance company, should anything bad happen, would be similar. Bear in mind that insurance companies like finding ways not to pay claims...

So really the best strategy is to follow North American electrical practice as closely as possible, do everything to the letter of the code, and do not introduce strange and unfamiliar things into the system.

(Keep in mind that if you use the P9S13A that comes pre-fitted with an L6-30P plug, then all you have to do is to provide some correctly wired and installed 30 A circuits back to your breaker panel with L6-30R receptacles on the end of them for your PDUs to plug into. Everything after the receptacle will be off the shelf hardware designed for the North American market and all your inspector will have to do is to sign off on the L6-30R receptacles and circuits. Everything will be straightforward, recognizable, and to code. Your inspector will have an easy job and no worries.)

Yes, I realize that, and that's why I was really upset when, after my exec talked to the techs, and after the power analysis, they claimed they didn't make a PDU that would power the PSU that worked in North America (which is just B.S. if you ask me).    I found an old power cord for a PC.   I saved them when I worked on it.   Three 16 gauge wires.   That should be able to handle the current running through the L1 and L2 of the NEMA L6-30R I'll be plugging it into.   Should be hard to wire up.   I'd feel a bit more comfortable with 12 gauge wire.   I'll see if I can find anymore.

[IanB]

Yes, I realize that, and that's why I was really upset when, after my exec talked to the techs, and after the power analysis, they claimed they didn't make a PDU that would power the PSU that worked in North America (which is just B.S. if you ask me).    I found an old power cord for a PC.   I saved them when I worked on it.   Three 16 gauge wires.   That should be able to handle the current running through the L1 and L2 of the NEMA L6-30R I'll be plugging it into.   Should be hard to wire up.   I'd feel a bit more comfortable with 12 gauge wire.   I'll see if I can find anymore.

Well the P9S13A can handle 24 amps at 240 volts (5760 W), which is (I think) the maximum continuous load allowed on a 30 amp circuit by the electrical code. So if you need more than 24 amps to run all your equipment you need more than one 30 amp circuit from the breaker panel and more than one PDU. Using a different PDU will not allow you to take more than 5.7 kW from your 30 amp circuit and still comply with code.

[Spork Schivago]

The three 16-gauge should be fine.   The wires that come with the PSU have three 17-gauge.   Unless those cords really are made for 240VAC running down L1...

How would I calculate the current draw that the PSU will be pulling from the NEMA receptacle for each wire?    Would it be 1400 watts / 120 volts or would it be 700 watts / 120 volts?

Or would it just be 1400 watts / 240 volts?   I'd think not the second, because L1 and L2 will each carry 120V.   But then again, we're still dealing with 240V so the current overall shouldn't change....but for the individual conductors, do we treat L1 and L2 as just having 240V total between them and the current draw per conductor (in the physical plug) will be roughly 5.9 amps, or would be roughly 11.7 amps?

[Spork Schivago]

Yes, I realize that, and that's why I was really upset when, after my exec talked to the techs, and after the power analysis, they claimed they didn't make a PDU that would power the PSU that worked in North America (which is just B.S. if you ask me).    I found an old power cord for a PC.   I saved them when I worked on it.   Three 16 gauge wires.   That should be able to handle the current running through the L1 and L2 of the NEMA L6-30R I'll be plugging it into.   Should be hard to wire up.   I'd feel a bit more comfortable with 12 gauge wire.   I'll see if I can find anymore.

Well the P9S13A can handle 24 amps at 240 volts (5760 W), which is (I think) the maximum continuous load allowed on a 30 amp circuit by the electrical code. So if you need more than 24 amps to run all your equipment you need more than one 30 amp circuit from the breaker panel and more than one PDU. Using a different PDU will not allow you to take more than 5.7 kW from your 30 amp circuit and still comply with code.

Right.   We will be drawing more than 5.7 kW from the server, therefore, we need at least 2 PDUs.

[Bratster]

The three 16-gauge should be fine.   The wires that come with the PSU have three 17-gauge.   Unless those cords really are made for 240VAC running down L1...

How would I calculate the current draw that the PSU will be pulling from the NEMA receptacle for each wire?    Would it be 1400 watts / 120 volts or would it be 700 watts / 120 volts?

Or would it just be 1400 watts / 240 volts?   I'd think not the second, because L1 and L2 will each carry 120V.   But then again, we're still dealing with 240V so the current overall shouldn't change....but for the individual conductors, do we treat L1 and L2 as just having 240V total between them and the current draw per conductor (in the physical plug) will be roughly 5.9 amps, or would be roughly 11.7 amps?

1400w/240v

So just under 6 amps.

Sent from my Moto x4 using Tapatalk

[Monkeh]

The three 16-gauge should be fine.   The wires that come with the PSU have three 17-gauge.   Unless those cords really are made for 240VAC running down L1...

How would I calculate the current draw that the PSU will be pulling from the NEMA receptacle for each wire?    Would it be 1400 watts / 120 volts or would it be 700 watts / 120 volts?

Or would it just be 1400 watts / 240 volts?   I'd think not the second, because L1 and L2 will each carry 120V.   But then again, we're still dealing with 240V so the current overall shouldn't change....but for the individual conductors, do we treat L1 and L2 as just having 240V total between them and the current draw per conductor (in the physical plug) will be roughly 5.9 amps, or would be roughly 11.7 amps?

They don't each carry 120V. There is 240V between them and that is all there is to it. The two wires make the circuit so the current running through them will be equal.

You won't be hitting the full power draw anyway. Assemble your lead for this test (only. it won't be suitable for continued use as the breaker protecting it is rather too large), and give it a go. It'll be fine, it'll power up, it will complain about lack of redundant supplies.

We will be drawing more than 5.7 kW from the server, therefore, we need at least 2 PDUs.

I really hope you don't mean one server.

Also remember they use redundant supplies and you're very unlikely to draw the full 1400W from a single machine.

[IanB]

How would I calculate the current draw that the PSU will be pulling from the NEMA receptacle for each wire?

I posted the specs in the table above (post #208). The nominal current draw at full load at 240 V will be 6.5 A. Look at line 5 of the PSU specification table. However, it won't be running at full load, so the current will be less than this. Any ordinary power cord can carry 6 amps for test purposes.

Would it be 1400 watts / 120 volts or would it be 700 watts / 120 volts?

Or would it just be 1400 watts / 240 volts?   I'd think not the second, because L1 and L2 will each carry 120V.   But then again, we're still dealing with 240V so the current overall shouldn't change....but for the individual conductors, do we treat L1 and L2 as just having 240V total between them and the current draw per conductor (in the physical plug) will be roughly 5.9 amps, or would be roughly 11.7 amps?

I thought we had been through this. There is no 120 V. There is only 240 V. There is no 120 V. It is 240 V. Please try to remember this.