208v to 240v Buck Booster

[synchro]

Larry's suggestion in Post #21, along with using H2-H3 as the "neutral" would create the 3 new voltage points A', B', and N' shown in the diagram below. I just eyeballed the length of the new black line (which represents the autotransformer supplied by A and B).

I didn't write in all the voltages (given by the distance between two points in the diagram), but (A,B,C,N) are as usual for 208Y/120V. The new points have voltages to ground (to N) of A' = 134V, N' = 60V, and B' = 134V. But A'-N' and B'-N' are 120V as desired.

Cheers, Wayne

View attachment 2570412

Bingo! And as was said above, the system is grounded. Nothing says the load must use that conductor.

Yes, in that configuration N' would efffectively just be a tap on the primary of the buck-boost transformer.
I don't think it would be a good idea to put 120V loads across A'-N' or across B'-N', even aside from the restrictions of 210.9 and 215.11 on autotransformers.

If you put a load current I_L across A'-N', then a current I_L would flow through the 16V secondary winding A-A', 0.42 x I_L through winding N'-A, and 0.58 x I_L through winding N'-B, as a consequence of balancing the ampere-turns of those three windings. But the primary windings on A-N' and B-N' would only need to be capable of 16/104 = 0.154 times the rated current through each 16V secondary winding (which can fully support the rated L-L load crrents). Therefore, with a single 120V load from A' or B' to N', the allowable 120V load current would only be 0.154/0.58 = 0.265 times the rated current for L-L loads (and obviously would be further reduced by any 240V L-L loads).

 

[LarryFine]

So, viable or not viable? I'm not sure what your opinion is after reading your post twice.

Would it work for a load that needs a neutral as long as it need not be actually grounded?

 

[tortuga]

So, viable or not viable? I'm not sure what your opinion is after reading your post twice.

Would it work for a load that needs a neutral as long as it need not be actually grounded?

I'd say no don't do it in the real world. But it would be a fun 'lab' experiment.
For one if you create a ungrounded neutral you'd probably need a OCPD in it.
For developers that don't like the idea of 120/208 open wye and insist on 120/240 I'd do a scott-T transformer.
You still get one transformer and your 3 phase primary stays balanced.

 

[LarryFine]

For developers that don't like the idea of 120/208 open wye and insist on 120/240 I'd do a scott-T transformer.
You still get one transformer and your 3 phase primary stays balanced.

And you can run a dryer on one?

 

[wwhitney]

So for 26A @ 120V = 3120VA of load, your transformer would need to be rated at max(26A*32V, 15A*208V) = max(832 VA, 3952 VA) = 3952 VA.

Typo in the above, 15A * 208V = 3120 VA, so the transformer would have to be rated 3120 VA for 3120VA of unbalanced 120V load.

Now my results agree with synchro's, I just used 13A or 26A instead of I_L so that my results would be whole numbers instead of decimals. Obviously for other currents, you just scale the results accordingly.

So if you had a load that needed 120/240V where none of the 3 conductors is grounded, and it needed say 20A at 240V plus 5A for unbalanced 120V loads, you'd need to size your 208V:32V boost transformer at 20*240/7.5 + 5*120 = 1,240 VA. Almost half of that size requirement comes from the 5A of unbalanced 120V load. But still much smaller than the 6,000 VA you'd need for an isolation transformer.

Cheers, Wayne

 

[wwhitney]

For developers that don't like the idea of 120/208 open wye and insist on 120/240 I'd do a scott-T transformer.
You still get one transformer and your 3 phase primary stays balanced.

A Scott-T transformer can not take a single phase load and turn it into a balanced 3 phase load. The single phase load has time varying power delivery over one cycle, while a balanced 3 phase load has constant power delivery.

Cheers, Wayne

 

[tortuga]

Not as a autotransformer, but for the building.
Say for example a building has 480Y/277 service and a developer wants to convert some upper floors to say 10 residential type dwelling units, 5 per floor.
The developer wants 120/240 no 208.
You could tap two 120/240 single phase feeders off the scott-t and use one per floor.
look ma no 208!
and the 480 is balanced

 

[wwhitney]

and the 480 is balanced

With two separate 120/240V single phase systems, 90 degrees apart from each other, as provided by a Scott-T transformer, then yes when the two systems happen to be balanced with each other (i.e. the same current on all 4 ungrounded conductors, all currents in phase with the voltages), the 3 phase load will be balanced.

Cheers, Wayne

 

[synchro]

So, viable or not viable? I'm not sure what your opinion is after reading your post twice.

Would it work for a load that needs a neutral as long as it need not be actually grounded?

Well, 210.9 of NEC says:
"Branch circuits shall not be derived from autotransformers unless the circuit supplied has a grounded conductor that is electrically connected to a grounded conductor of the system supplying the autotransformer.
Exception No. 1: An autotransformer shall be permitted without the connection to a grounded conductor where transforming from a nominal 208 volts to a nominal 240-volt supply or similarly from 240 volts to 208 volts."
And so transforming to 120V is not mentioned in the exception.

Just from a technical point of view, the primary (higher voltage) winding of the buck-boost transformer would need to be protected.
But you can't protect this primary directly because of 450.4(A):
"An overcurrent device shall not be installed in series with the shunt winding (the winding common to both the input and the output circuits) ..."
I suspect this rule is because if an overcurrent device on the shunt winding is open circuited, then the voltage across the open primary winding might get boosted quite high when there's a load current passing through the low voltage winding that's in series with the load.

An overcurrent device on the supply to the buck-boost at a setting allowing the full L-L load capability of the transformer could allow the primary winding to be overloaded by a significant margin from a L'-N' load as I noted in post #41 above. If an overcurrent device for just 120V L'-N' loads was set to 0.265 of the maximum 240V L-L load current capability of the buck-boost, there would be no capability left for any 240V L'-L' loads. Now if you have very small, or always very well balanced L'-N' loads, then they could be accommodated.

 

[ptonsparky]

Hint. See post #2.

 

[LarryFine]

Hint. See post #2.

True, but we wouldn't have had all of this fun if we stopped there.

 

[ptonsparky]

True, but we wouldn't have had all of this fun if we stopped there.

Oh, I agree. Plus I learned enroute!

 

[kwired]

Good question, let's think it through. I'm going to assume a hypothetical transformer that gives us 240V from our 208V. I.e. the turns ratio is 32V : 208V, or 2 : 13. With both the primary and secondary center-tapped, so the post #21 scheme can be implemented.

Then for 240V loads, you would just size it as normal. 13A @ 240V would require 15A @ 208V, and of that 15A, 2A flows on the 208V primary windings, 13A flows in the 32V secondary windings. So for 13A @ 240V = 3120 VA of load, your transformer would need to be rated at 2A * 208V = 13A * 32V = 416 VA.

For a 120V unbalanced load of say 26A on A'-N', that would also require 15A @ 240V. The transformer coil currents would be 26A from A' - A; 11A from A - N' ; and 15A from B - N'. [The 11A from A-N' and the 15A from B-N' are of opposite sense (direction), so the average current over the primary coil is (15 - 11)/2 = 2A, while the average current over the secondary coil is (26 + 0)/2 = 13A, the correct ratio.] So for 26A @ 120V = 3120VA of load, your transformer would need to be rated at max(26A*32V, 15A*208V) = max(832 VA, 3952 VA) = 3952 VA.

For the mixed case, just add up the two results using the unbalanced 120V portion of the load, and the balanced 240V portion of the load. Definitely an inefficient way to provide for the 120V unbalanced load, but plausible if that load is small compared to the 240V balanced load.

Of course, this is all academic, I think, as connecting N' at 60V above ground to either a receptacle's grounded pin or an appliance's grounded conductor is surely a violation of something.

Cheers, Wayne

If you end up deriving an additional ungrounded conductor you very likely would need overcurrent protection on said conductor, at least if it were part of the premises wiring. If derived within listed equipment then listing requirements would determine what can be done with it.

Not as a autotransformer, but for the building.
Say for example a building has 480Y/277 service and a developer wants to convert some upper floors to say 10 residential type dwelling units, 5 per floor.
The developer wants 120/240 no 208.
You could tap two 120/240 single phase feeders off the scott-t and use one per floor.
look ma no 208!
and the 480 is balanced

How much will the Scott-T cost and how long of a time until you get it? compared to getting say three conventional 480x120/240 single phase units that hopefully are in stock and balancing the primaries across the 480 system and balancing the secondary as much as possible. If all dwellings are pretty identical then you can balance 9 of them and have one left over.

 

[ruxton.stanislaw]

If you end up deriving an additional ungrounded conductor you very likely would need overcurrent protection on said conductor, at least if it were part of the premises wiring. If derived within listed equipment then listing requirements would determine what can be done with it.

How much will the Scott-T cost and how long of a time until you get it? compared to getting say three conventional 480x120/240 single phase units that hopefully are in stock and balancing the primaries across the 480 system and balancing the secondary as much as possible. If all dwellings are pretty identical then you can balance 9 of them and have one left over.

Nothing wrong with adding a fuse in-line there. That's a really good idea, in case someone connects an improperly wired dryer which bonds the ground and neutral, which is often an occurrence in my experience.

While I do admire the Scott-T transformer configuration plan, they may be more difficult to come by and more expensive than a hexaphase configuration.

 

[tortuga]

How much will the Scott-T cost and how long of a time until you get it? compared to getting say three conventional 480x120/240 single phase units

Thats a good question its a custom order that every transformer manufacturer knows how to build, not a custom design.
Penn transformer probably would be the best source, https://www.patransformer.com/
The other cost savings are realized by the fact you're just installing one transformer instead of three and you have one less panelboard and feeder. Less grounding and bonding etc.
Edit If the two single phase feeders went the same direction from the transformer say from a basement to some upper floors you would have savings on wire as you could run just one 5 wire feeder (shared neutral).

 

[ptonsparky]

I think we lost the OPr.

 

[tortuga]

I think we lost the OPr.

I think you answered his question in post#2 if he does anything else that would be silly.

 

[synchro]

Some time ago, I suggested that, if you select a unit so the secondaries will be connected in series, they could be wired to opposite ends of the primary, maintaining equal voltages to the neutral.

For example, looking at the boost diagram below, separate the secondaries and wire X3-X4 to the H4 end, and X1-X2 to the H1 end.

In other words, wire it like you would an open-delta setup, but with only one transformer, and no neutral connection to the transformer.

....

Larry's suggestion in Post #21, along with using H2-H3 as the "neutral" would create the 3 new voltage points A', B', and N' shown in the diagram below. I just eyeballed the length of the new black line (which represents the autotransformer supplied by A and B).

I didn't write in all the voltages (given by the distance between two points in the diagram), but (A,B,C,N) are as usual for 208Y/120V. The new points have voltages to ground (to N) of A' = 134V, N' = 60V, and B' = 134V. But A'-N' and B'-N' are 120V as desired.

Cheers, Wayne

View attachment 2570412

Bingo! And as was said above, the system is grounded. Nothing says the load must use that conductor.

I just had a thought about Larry's proposal above. The symmetrical nature of the A'-N and B'-N voltages looks like a good idea. I'm just wondering if the insulation between the two low voltage secondary windings X4-X3 and X2-X1 would be rated for 240V. There would definitely be an insulating layer between the primary and secondary windings, but there may not be one between the two low voltage secondary windings, because it would not normally be needed. For example, it's possible that these low voltage windings are next to and touching each other or they are overlapping, because of the small voltage difference they normally have. I think the manufacturer would need to be contacted to determine if this might be an issue.
Of course, two separate buck boost transformers could be used to avoid this potential issue (no pun intended).

 

[LarryFine]

Of course, two separate buck boost transformers could be used to avoid this potential issue (no pun intended).

Pun taken.

I'm sure the insulation that allows the secondaries to be line-voltage-plus to ground also insulates them from each other. However, as you said, it would be prudent to find out from the manufacturer.

 

[Jpflex]

120/208 incoming voltage
The dryer in question is 208/240 capable and has a 30amp rating.
I planned on installing a step up transformer, primary being 208v and secondary being 240v which should make the dryer heating element function better, right?
My question is about the amperage range, is 30amp the max the dryer would pull or would I be ok going 25 or 26 amp booster?
I had also read the conversion would change amperage or throw it off, anyone familiar with that, also, the booster will be very close to source and load so I shouldn’t have much loss in voltage due to distance. Thanks in advance for assistance!

Assuming the dryer is currently using 208 volts, and your boost transformer will boost this to 240 volts, then if your dryer is rated for 240 volts then the amount of current such as 30 ampere before will be proportionally less. In which case as explained by someone else there may be less voltage drop to the dryer and a better running dryer