# Can I combine two separate 120V isolation transformers, tie the secondaries in series, but not tie the primaries?

I want to connect two separate 120V (7.5KVA) isolation transformers, tie the secondaries, but not the primaries. The input is 120V for each primary, with the goal to have 240/120V on the output. I guess you can call the primaries of the two transformers as floating and/or isolated primaries. I tried searching all of the world wide web for this type of application, but it seems very uncommon. Typically, the primaries are connected right at the transformers. The logic I had was that the transformers are technically connected at the neutrals upstream at a terminal block, but that distance is about 50 feet. To clarify my intention, I drew what the circuit would like below. I want to know if this is possible, what should I do to make this works, and what issues should I look out for. Also, how would the output look if the 2nd transformer below had Line 2 and the Neutral switched on the primary?

I did see issues online about non-matching polarity, insufficent insulation in the transformers, ease of mis-wiring, and others.

• Is the L1 and L2 US style 180 degree split phase or two 120 degree separated phases from a true three-phase system? Commented Oct 5, 2017 at 15:30
• winny is right. L1 and L2 should be in-phase. If they are 180° out of phase (actually I think they are) then you'll get 0V from combined output. Commented Oct 5, 2017 at 15:43
• In the 2-line US example, the secondaries could also be switched to give 240v. :) Commented Oct 5, 2017 at 15:49
• If it's US split phase then you'll have to invert one transformer's (opposite input OR output) terminals, which will give you the 240 V isolated you want. I assume that you are doing this to spread the power across the whole input supply? Commented Oct 5, 2017 at 15:49
• You might consider to put both primaries in series and connect this setup to L1 and L2. That way you have no need for a neutral. Commented Oct 5, 2017 at 16:09

simulate this circuit – Schematic created using CircuitLab

Figure 1. Three options - two of which actually work. Note the dot convention on the coils.

• (a) is the same as your schematic (but lower primary is inverted to keep phase convention the same). The primaries are in anti-phase so the output is zero (as pointed out in the comments).
• (b) is the corrected version of (a).
• (c) is an alternate working solution. Since the primary currents are the same what comes out the bottom of XFMR3 must go in the top of XFMR4 so there is no neutral current and the neutral connections can be omitted as shown on XFMRs 5 and 6.
• Umm, the OPs schematic looks like your b to me, not like your a. Commented Oct 29, 2018 at 20:01
• @Peter: Umm, I think I'm OK. Note the OP's schematic doesn't have the dots but identifies neutral on the bottom of the upper traffo and top of the lower traffo. I was constrained by the dots on the CircuitLab symbols so I had to invert the lower primary winding in (a) and (b). Read (a) again to see if I explained it correctly. (If I was right then toss me a vote!) Thanks for taking the time to comment. Commented Oct 29, 2018 at 23:05
• I would interpret lack of dots as being the same as both dots being at the top of the symbol (or equivilently both dots being at the bottom). The word "neutral" seems to be refering the the conductors in the wiring nothing to do with the phasing of the transformers. Commented Oct 29, 2018 at 23:22

My concern would be back-feeding.

I am guessing the reason the two transformers have separate neutrals is because they are going to be plugged into different sockets.

If that is the case then you have to consider what happens if one of the transformers is unplugged. Hazardous voltages are likely to feed through the plugged in transformer, through the 240V load and then back to the plug of the unplugged transformer.

Your circuit is good. The only thing you should pay attention to is in-phase secondary voltages (as @Jack Creasey said) to make them be added (and produce 240V output), not substracted (and produce 0V output).