120V to 120/240V. Transformer or other ideas?

[gar]

190818-1604 EDT

ron:

I think the autotransformer is the better approach.

Why? If I parallel two Honda 2200s I get 4400 capability at 120 V. Adding a 120 to 240 autotransformer makes 120-0-120 available. Adequately sized I can supply up to a maximum of 4400 on either 120 phase or 240 in any combination that does not exceed 4400 total load.

No isolarion is required, and therefore with an autotransformer you can get the the 4400 capability in any combination with less transformer weight than an isolation transformer.

In a home application there is a continuously varying load between the two phases. If you parallel the two phase circuits, then any combination of loading to 4400 is possible at 120 V, but no 240 V.

.

 

[synchro]

However, I have been feeding my personal circuit at 120V on both legs for 20yrs. I know anything that is 240V will have a potential of 0V so its not like there is any risk.

That would be fine as long as you don't have any MWBC's where the neutral would be overloaded using only one phase on all hot wires.

What WOULD be nice to see is one of the transformer companies build a special transformer for this. Delete the secondary entirely, and provide a beefed up 180* winding on the primary side so the entire KVA could be used on the 180* leg. The only other option to get full KVA from that 180* leg is double the size of the transformer.

Both half-windings of a 2:1 autotransformer like a 240V /120V flow the same amount of current, just in opposite directions to and from the center tap. So the gauge of each half-winding should be the same in a 2:1 autotransformer. And therefore the 240/120V center-tapped primary winding of a standard transformer (which would have half-windings of the same gauge) is very suitable as-is for use as an autotransformer.

In general I agree with your comments and plans, just wanted to clarify a few points.

 

[__dan]

The auto transformer connection would only see the load of the current x the winding voltage, for either line or load. The transformer kVA rating need only be 1/2 the kVA of the load, for the 120 to 240 connection.

If it's a project for yourself and a surplus or Ebay unit is likely. the common buck boost autotransformer with 12 / 24 or 16 / 32 Volt secondary windings is what I have used (because I have a collection of takeouts and curio items in my basement).

You only need the primary 120 / 240 Volt winding and ignore the secondary. I am guessing it is very cheap amd easy to find on Ebay, because it already has a standard application for 280 to 240 buck boost. So many of them get put in and taken out, it's an Ebay item.

Something like this.

https://www.ebay.com/itm/GE-9T51B14...00049a3:g:OjAAAOSwfHpdOUhY&LH_ItemCondition=4

 

[ron]

190818-1604 EDT

ron:

I think the autotransformer is the better approach.

I agree. The current offering from Honda only parallels the output, that's why I mentioned "it is too bad it can't control the inverters to be 180 out of phase, rather than in phase"

 

[gar]

190819-1141 EDT

ron:

The interesting aspect of the two parallel inverters with an autotransformer is that either phase can be loaded to the full output of the two inverters so long as the total load does not exceed the total capacity of the two inverters. However, it requires the additional weight of the autotransformer.

If the two inverters could be phased 180 apart, then the maximum load on one 120 V phase is limited to the ouput of only one inverter.

.

 

[fastline]

The auto transformer connection would only see the load of the current x the winding voltage, for either line or load. The transformer kVA rating need only be 1/2 the kVA of the load, for the 120 to 240 connection.

I think we might need to clarify this so everyone is on the same page and I can understand why you say this. Why would the KVA rating be ok at 1/2 of the load?

If you take a 2:1 primary (yes, we are ignoring the secondary), that is 2 coils on the primary for say 120x240 and say 3kva. Typical wiring is series wired for 240v, or parallel wired for 120, to get full kva. That means each coil at 120V is 1.5kva.

Now if you feed 120V to only one of the coils, you can obviously pull load from this same connection and completely by pass the transformer so you can get full power from the source. If you pull 240V, you can get 3kva, or full rating from the transformer. However, when pulling 120V from the non-powered 180* coil in the primary, you are at 120V and still limited by the ampacity of the coil, which will be 1.5kva.

However, considering the source is 3kva anyway, all you need to do is slightly bias the "load balance" to favor the stronger side. In a perfect world, you would have a perfect balance between legs which would still equal the 3kva.

I was just "dreaming" about parallel of inverters to get split phase simply to avoid a transformer. However, as mentioned above, that would mean each leg would be limited by the capacity of only one generator. I guess inrush on harder loads could be better handled with the transformer arrangement.

 

[__dan]

I think we might need to clarify this so everyone is on the same page and I can understand why you say this. Why would the KVA rating be ok at 1/2 of the load?

.

I drew it out and i'm thinking it should work. If I have a 20 A line current at 120 V in. I could see how it splits as 10 A to half the winding that is powered and 10 A through the other winding that is in series with both the line side, the other half winding, and the load.

With the autotransformer connection, only half of the voltage needs the transformer, the other half is the line 120, in series with that other half winding, and the load.

 

[gar]

190821-0737 EDT

New possible idea on how to use an autotransformer and get the full power output of the sum of the 2 generators without the autotransformer having to pass that full power.

I have to believe that the second generator when added in parallel to the first generator knows how to synchronize with that first generator's output. If not, then a synchronizing patch cord is required. In either case what I am suggesting may work.

Do not connect the second generator in parallel with the first, but rather in series. Both generators will operate and supply 120 V loads with up to the maximum power of the individual generators. The outputs will not be locked in frequency and phase to each other. The common connection between the generators is not yet a neutral, but will become the neutral with the addition of my next item.

Add an autotransformer to the series connection of the two generators. Now the sync signal is provided to the second generator. You probably don't connect the second generator to the autotransformer until both generators are running, and possibly without some series resistance to begin. In this case the autotransformer is the synchronizing connection.

If a separate sync wire is required, then it may require a phase inverter.

If this power paralleling scheme will work, then full power of the two generators can go to either 120 circuit with an autotransformer 1/2 the size of the one in my previous post.

.

 

[fastline]

Don't guess I follow on the series wiring on the generators. A digital inverter is going to sync differently than conventional. It needs to see full wave form to sync because it synthesizes the wave. Yes, you have to start one, then the other. You don't have to use a sync cord.

If you connect two generators, I envision N on Gen1 to L1 on Gen2. What are we making with that? They won't sync in that way.

However, there are sync ports to plug into if you "want to do it" that way. I guess I can force them to sync, but if you then connect as above, its a dead short, so......?

Drawing maybe?

 

[gar]

190821-1104 EDT

fastline:

The autotransformer is the means to sync the two generators when series connected. At least that is my hope. You should be able to think this thru and possibly try an experiment.

Connecting two generators in series will cause no problems with 120 V loads. A 240 V resistive load could be tried, and I would expect without synchronization that would go from +240 to 0 to -240 at a rate determined by the difference frequency of the two generators. Here I used + and - to indicate phase relationship.

Since you indicated a sync connection is possible you might try it and see what happens, it must be isolated from at least one generator output. If it forces both generators to be in phase, which is expected, then a simple series connection of the generators will produce 240. Power on each phase will be limited to that of its generator.

.

 

[gar]

190821-1104 EDT

fastline:

The autotransformer is the means to sync the two generators when series connected. At least that is my hope. You should be able to think this thru and possibly try an experiment.

Connecting two generators in series will cause no problems with 120 V loads. A 240 V resistive load could be tried, and I would expect without synchronization that would go from +240 to 0 to -240 at a rate determined by the difference frequency of the two generators. Here I used + and - to indicate phase relationship.

Since you indicated a sync connection is possible you might try it and see what happens, it must be isolated from at least one of the two generators. If it forces both generators to be in phase, which is expected, then a simple series connection of the generators will produce 240. Power on each phase will be limited to that of its generator.

.

 

[synchro]

I think I understand what gar is saying. The neutrals of both generators would be tied together and connected to the center-tap of the autotransformer. One outer end of the autotransformer winding would connected to a first inverter generator, and the other end to a second inverter generator.
I would view this as equivalently connecting the generators in parallel, but with a 180 degree inverting transformer in between them. So the inverter outputs would sync up 180 degrees apart, with 240V RMS across them.
If the 120V loads are balanced on the 2 phases then the autotransformer conducts esentially no current. But the autotransformer will help supply current with unbalanced loads. If 120V loads are only on one phase then the autotransformer will deliver up to the rated power of the inverter connected to the other phase.
What's interesting is that if either generator goes out then you still have 120V power available on both phases.
​​​

 

[gar]

190821-1130 EDT

Website problems. Remove my post #30, #31 is a modified one.

 

[LarryFine]

I think I understand what gar is saying. The neutrals of both generators would be tied together and connected to the center-tap of the autotransformer.​​​

I would think you'd connect one genny's L1 to the other genny's L2, and use that point as the neutral.

 

[gar]

190821-1145 EDT

I don't think it matters what the lines are called the inverters in the generators self synchronize.

What could be a problem is if one of the output lines is tied to chassis, and therefore the EGC. But usually there is a removable ground jumper.

On the other hand find the output lead that is grounded, and on both generators connect those leads together. But you probably want to remove ground jumpers anyway. I would still connect together the two leads that were originally grounded.

.

 

[gar]

190821-1431 EDT

Interesting information:

http://www.arrl.org/files/file/QST Binaries/QS0608Kleinschmidt.pdf look at the clean waveform.

http://www.jkovach.net/projects/powerquality/ shows various waveforms

https://www.generatormag.com/best-inverter-generator-reviews/ has a limited comparison chart

.

 

[gar]

190821-2135 EDT

fastline:

Do you actually have one of the Honda's now? How likely are you to have two generators?

Even if you do not have any generator now you can do experiments with a transformer to understand what can be done. Does not require a large transformer with which to experiment. Any transformer with two separate 120 V primary coils is suitable. You have to learn how to correctly phase the windings. This can be done even with no labeling. For a 100 VA transformer or larger a series 60 W incandescent should provide adequate protection.

Phase the two windings in parallel incorrectly and it is pretty much a short circuit.

When you figure out how the transformer works, then you can simulate from the 120 line what would happen with a 120 V generator.

Give us some feed back.

.

 

[fastline]

I currently own a Honda inverter and I have an identical one available. I will probably own another. I have already tested on an autotransformer and can conform the obvious so far. L1 to H1, N to H2, connecte H2-H3, and H1 to H4 generates 240V.

 

[fastline]

duplicate.....

 

[LarryFine]

I currently own a Honda inverter and I have an identical one available. I will probably own another. I have already tested on an autotransformer and can conform the obvious so far. L1 to H1, N to H2, connecte H2-H3, and H1 to H4 generates 240V.

Cool, so we know that it works with one genny. Now, where to connect the second one?