Autotransformers in solar energy applications

[Alex Torres]

TL;DR Summary

Safety concerns

Hi!!...There's a growing popularity on these transformers as they are being used to get 120/240 split phase out of a single 120v phase, read a recent comment about them not being as safe as an isolated transformer since their inner windings are not isolated....is an encapsulated general purpose transformer a safer alternative?..thanks

 

[berkeman]

TL;DR Summary: Safety concerns

.is an encapsulated general purpose transformer a safer alternative?

Yes, in general an isolated (traditional) transformer is safer than an autotransformer.

Having said that, it really depends on the application. Autotransformers ("Variacs") are most dangerous when used to transform an AC Mains voltage that is referenced to Earth ground, and the end user thinks that it is isolating the transformed Hot/Neutral from Earth ground. It isn't. You need the traditional isolation transformer structure to give you a floating transformed AC voltage.

But if the original AC Mains type voltage is already isolated from Earth ground for some reason, using an Autotransformer should generally be fine. Again, it depends on the application.

One other thing to consider is that a Variac Autotransformer can have its setting knob bumped or otherwise changed, so the ratio setting may not always be fixed reliably.

(Sorry for the dirty picture)

https://www.popsci.com/diy/article/2009-10/what-variac/

 

[Baluncore]

To convert 120V to 120V+120V=240V, requires only two 120V windings as an autotransformer. It uses less copper, is smaller, and lower cost, than the three 120V windings needed for an isolation transformer.
Safety is not a problem unless isolation is required, and not provided.

 

[Alex Torres]

Yes, in general an isolated (traditional) transformer is safer than an autotransformer.

Having said that, it really depends on the application. Autotransformers ("Variacs") are most dangerous when used to transform an AC Mains voltage that is referenced to Earth ground, and the end user thinks that it is isolating the transformed Hot/Neutral from Earth ground. It isn't. You need the traditional isolation transformer structure to give you a floating transformed AC voltage.

But if the original AC Mains type voltage is already isolated from Earth ground for some reason, using an Autotransformer should generally be fine. Again, it depends on the application.

One other thing to consider is that a Variac Autotransformer can have its setting knob bumped or otherwise changed, so the ratio setting may not always be fixed reliably.

(Sorry for the dirty picture)
View attachment 338677
https://www.popsci.com/diy/article/2009-10/what-variac/

Is there a general rule of thumb to size a general purpose transformer (wired for split phase output from a single phase leg) to an specific load??.Used a .5kva transformer for this purpose to power a 300watts single phase heater and the power loss was like 20% as the voltage dropped from 122vac to 103vac....thanks!!

 

[Alex Torres]

To convert 120V to 120V+120V=240V, requires only two 120V windings as an autotransformer. It uses less copper, is smaller, and lower cost, than the three 120V windings needed for an isolation transformer.
Safety is not a problem unless isolation is required, and not provided.

All autotransformers wiring diagrams that I've seen are for buck/boost applications, is there a wiring diagram out there to convert a single phase leg to split phase??..thanks!!

 

[berkeman]

All autotransformers wiring diagrams that I've seen are for buck/boost applications, is there a wiring diagram out there to convert a single phase leg to split phase??..thanks!!

Buck/Boost is a DC-DC power supply topology. I thought you were asking about straight AC transformers...?

A typical way to convert from single-phase to split phase with a transformer is to use a split secondary transformer. Is that what you are asking about?

https://www.electronics-tutorials.ws/transformer/multiple-winding-transformers.html

 

[Averagesupernova]

Buck/Boost is a DC-DC power supply topology. I thought you were asking about straight AC transformers...?

No, not necessarily. It's not uncommon to increase a 208 volt supply to 240 volts by wiring a transformer with the secondary in series (additively) with the 208 volt supply.

 

[berkeman]

No, not necessarily. It's not uncommon to increase a 208 volt supply to 240 volts by wiring a transformer with the secondary in series (additively) with the 208 volt supply.

Is that what he means by Buck/Boost? I hadn't heard that term used for that transformer wiring technique before.

 

[Averagesupernova]

Is that what he means by Buck/Boost? I hadn't heard that term used for that transformer wiring technique before.

I assumed that is was. @Alex Torres can you elaborate?

 

[Alex Torres]

Is that what he means by Buck/Boost? I hadn't heard that term used for that transformer wiring technique before.

These transformers have also become popular for grid tie applications, as electrical standards have placed tight restrictions on grid tie inverters while selling power to the grid, for example my xw6048 disconnets when selling above 128vac since the grid in my area is 126vac + - 3 volts the inverter immediately disconnects, so had to resort to one of these transformers to lower the grid input ac volts by 5%, they are very specific in terms of wiring instructions to buck/boost ac power supply by 5%, 10% 20% etc. but as i said before, haven't seen a single wiring diagram to convert a single phase leg into a split phase..

 

[dlgoff]

... ("Variacs") are most dangerous when used to transform an AC Mains voltage that is referenced to Earth ground ...

I know that's right. I recently got the crap knocked out of me.

 

[berkeman]

These transformers have also become popular for grid tie applications, as electrical standards have placed tight restrictions on grid tie inverters while selling power to the grid, for example my xw6048 disconnets when selling above 128vac since the grid in my area is 126vac + - 3 volts the inverter immediately disconnects, so had to resort to one of these transformers to lower the grid input ac volts by 5%, they are very specific in terms of wiring instructions to buck/boost ac power supply by 5%, 10% 20% etc. but as i said before, haven't seen a single wiring diagram to convert a single phase leg into a split phase..

Sorry, I'm not understanding this at all.

Grid ties from Solar Energy installations are regulated and inspected, no? Have you had your modified system re-inspected?

And can you say more about your use of the term "Buck/Boost"? In my experience as a power supply designer it usually refers to a DC-DC converter circuit where the input voltage may be above or below the desired output voltage. It could refer to some stage in your solar energy DC output converter/inverter circuit, but it's not obvious yet to me how it fits in.

Also, I'm assuming that your power converter circuit is a MPPT device, and I'm not sure that it will still be optimized if you are inserting an additional transformer stage in series with its output...

 

[DaveE]

Is that what he means by Buck/Boost? I hadn't heard that term used for that transformer wiring technique before.

Yes, those guys use it too. It really predates SMPS. More an electrician thing than an EE thing.

 

[Alex Torres]

Sorry, I'm not understanding this at all.

Grid ties from Solar Energy installations are regulated and inspected, no? Have you had your modified system re-inspected?

And can you say more about your use of the term "Buck/Boost"? In my experience as a power supply designer it usually refers to a DC-DC converter circuit where the input voltage may be above or below the desired output voltage. It could refer to some stage in your solar energy DC output converter/inverter circuit, but it's not obvious yet to me how it fits in.

Also, I'm assuming that your power converter circuit is a MPPT device, and I'm not sure that it will still be optimized if you are inserting an additional transformer stage in series with its output...

https://www.hubbell.com/acmeelectric/en/Buck-Boost-Transformers

 

[DaveE]

I'd vote for you hiring an electrician that can do a code compatible (i.e. safe) installation. Building codes are (mostly) your friend. There are good, sometimes not so obvious, reasons for their requirements.

If you want to know about the theory behind this, start with an EE course on circuits, magnetics, grounding, and such in a general sense. Learn in books, Labs, and online research; not in construction on your own house.

 

[Alex Torres]

You're right, safety comes first, sorry for the grid tie example and the link placed, it was just posted to clarify the use of auto transformers as an a/c buck/boost alternative... ...the main theme of the thread is safety related issues dealing with the use of auto transformers in off grid applications, and this has been clarified above...thanks!!

 

[berkeman]

https://www.hubbell.com/acmeelectric/en/Buck-Boost-Transformers

Okay, now I see the issue. That is Acme Electric, which has a well known, uh history in supplying things. That is probably where the strange use of "Buck/Boost" comes from with respect to transformers. Probably better for you to use more standard terminology going forward...

https://www.beaconjournal.com/story...-wile-e-coyote-akron-looney-tunes/6831400001/

 

[Averagesupernova]

Concerning creating split phase from a single 120 volt leg, yes it can be done. If you have some small transformers with center tap windings you can experiment with this sort of thing. For instance, do some experiments with small transformers with 12 or 24 volt secondaries. Drive one half of a center tapped winding with 12 volts. That one winding will act as an autotransformer. Be careful and don't forget you may be stepping up the voltage on another winding. Treat everything with care.

 

[Alex Torres]

Okay, now I see the issue. That is Acme Electric, which has a well known, uh history in supplying things. That is probably where the strange use of "Buck/Boost" comes from with respect to transformers. Probably better for you to use more standard terminology going forward...

View attachment 338711
https://www.beaconjournal.com/story...-wile-e-coyote-akron-looney-tunes/6831400001/

 

[Rive]

These transformers have also become popular for grid tie applications, as electrical standards have placed tight restrictions on grid tie inverters while selling power to the grid, for example my xw6048 disconnets when selling above 128vac since the grid in my area is 126vac + - 3 volts the inverter immediately disconnects, so had to resort to one of these transformers to lower the grid input ac volts by 5%

?
That limitation is there to keep grid voltage in limits.
With the described 'solution' to push out the generated energy anyway to the already congested grid the grid will be driven out of specs.
Is such practice even legal?

 

[Alex Torres]

Reference: https://www.physicsforums.com/threads/autotransformers-in-solar-energy-applications.1059115/

?
That limitation is there to keep grid voltage in limits.
With the described 'solution' to push out the generated energy anyway to the already congested grid the grid will be driven out of specs.
Is such practice even legal?

. these buck/boost transformers lowers or increase the voltage seen by the inverter, while the voltage and frequency seen by the grid remains the same

Reference: https://www.physicsforums.com/threads/autotransformers-in-solar-energy-applications.1059115/

 

[Rive]

The very principle behind the energy export of solar inverters is, that they are producing at slightly higher voltage than what can be seen on the grid => if too many inverters are feeding back to the grid then the grid voltage will rise.
To keep the grid voltage at acceptable limits, the inverters are expected to shut down when the grid voltage is above certain safe limits. This is a mandatory safety function.

What that trick with the autotransformer does in the quoted type of application is, that even when the inverter is already expected to shut down due grid overvoltage, it still tricks the inverter to operate, thus: bypassing this safety function => actually it may push the grid voltage above limits.

I do understand that individually it's not really nice to lose PV production due grid overvoltage, but I highly doubt this practice being legal and safe.

 

[Alex Torres]

The very principle behind the energy export of solar inverters is, that they are producing at slightly higher voltage than what can be seen on the grid => if too many inverters are feeding back to the grid then the grid voltage will rise.
To keep the grid voltage at acceptable limits, the inverters are expected to shut down when the grid voltage is above certain safe limits. This is a mandatory safety function.

What that trick with the autotransformer does in the quoted type of application is, that even when the inverter is already expected to shut down due grid overvoltage, it still tricks the inverter to operate, thus: bypassing this safety function => actually it may push the grid voltage above limits.

I do understand that individually it's not really nice to lose PV production due grid overvoltage, but I highly doubt this practice being legal and safe.

All grid tie installations must be approved by certified personnel and inspected by the local Utility Company, mine was certified by the same EE that suggested the installation of these transformers as he said there was a problem with the neighborhoods transformer that kept the voltage out of range triggering the inverter being disconnected from the grid more than 60% of the time. Remember he saying the same argument of yours, but the solution used by the Utility Company according to him was to limit the number of installations per area.

 

[Rive]

The whole thing feels really absurd.

there was a problem with the neighborhoods transformer that kept the voltage out of range triggering the inverter being disconnected from the grid more than 60% of the time.

Well, guess with new installations pushing the grid voltage further upward, the settings of older installations (for old, lower voltages) becoming inadequate...
...and then when they got their re-set, it's the turn of your installation to fail so often...

Makes one wonder where will the line voltage finally stop

 

[Alex Torres]

The whole thing feels really absurd.Well, guess with new installations pushing the grid voltage further upward, the settings of older installations (for old, lower voltages) becoming inadequate...
...and then when they got their re-set, it's the turn of your installation to fail so often...

Makes one wonder where will the line voltage finally stop

This extract is from Schneider modern inverters, the table default values are for the EU models but it also applies to the split phase US models XW+..... In contrast older versions of these inverters won't allow to change anti-islanding settings...

 

[DaveE]

The whole thing feels really absurd.Well, guess with new installations pushing the grid voltage further upward, the settings of older installations (for old, lower voltages) becoming inadequate...
...and then when they got their re-set, it's the turn of your installation to fail so often...

Makes one wonder where will the line voltage finally stop

Here in Palo Alto (yes, that's a utility, not just a city) you can't have residential solar that is an annual net generator of electricity. Or maybe you would have to meet an entirely different set of regulations, IDK. This is to insure that small solar will follow the grid, not drive it. I believe this is a common restriction in most places.

 

[berkeman]

It would be nice if @anorlunda were still here to comment, but he's probably out sailing around the world somewhere right now...

It sounds like part of the problem comes from the existing grid infrastructure, especially the "last mile" or two into neighborhoods that now have residential power generation capability. If enough of those generators exist in an area, the existing power grid infrastructure may not be able to handle the reverse power back into the overall grid (resulting in local overvoltages for the customers in that area). Part of the solution will need to be to upgrade that local infrastructure (transformers and maybe some powerlines), and a question is who should pay for those upgrades?

 

[Baluncore]

All grid tie installations must be approved by certified personnel and inspected by the local Utility Company, mine was certified by the same EE that suggested the installation of these transformers as he said there was a problem with the neighborhoods transformer that kept the voltage out of range triggering the inverter being disconnected from the grid more than 60% of the time.

If the voltage taps on the transformer need to be moved, then that should be done by the power authority. Has anyone asked them for a change, why have they refused?

The transmission line network that supplies your local grid is limited in capacity, both as an exporter or importer of energy. Something must regulate the excess generation.

If your grid voltage is high, there is too much PV generation, or too little consumption. What is your grid voltage at night ?

You should not install an autotransformer, that will help you compete unfairly, against other more honest PV generators.

 

[Alex Torres]

If the voltage taps on the transformer need to be moved, then that should be done by the power authority. Has anyone asked them for a change, why have they refused?

The transmission line network that supplies your local grid is limited in capacity, both as an exporter or importer of energy. Something must regulate the excess generation.

If your grid voltage is high, there is too much PV generation, or too little consumption. What is your grid voltage at night ?

You should not install an autotransformer, that will help you compete unfairly, against other more honest PV generators.

The installation dates back to 2013, little or nothing did i know by that time about a/c buck/boost transformers, it was the EE's idea (who certified the system) to install them in order to taper down the grid input, the Utility Co. then certified it with them installed, please check on the Schneiders Manual extract posted above and see their inverters now allows to adjust these anti-islanding parameters (mine is an old model that won't allow for this) under a warning statement advising that the adjustments must be performed by qualified personnel and with permission of the local Utility Co. so to point a finger upon those installations sounds more of an overreaction, i think.

 

[Rive]

to point a finger upon those installations

I don't think that happened here. It's just the described system/method is really absurd on community level and triggered some dismay and disbelief here. It's not against you.

Regarding your original question, you don't really have many options. You may ask for a check and re-certificate, but a licenced EE (hopefully) will stick to the local code anyway: both to ensure safety and uniformity (so no electrician will meet any unexpected/unknown instrument there). So if it's about autotransformers there, then you will get an autotransformer anyway.
It's just as safe as any legit 'line voltage stuff' can be.

It sounds like part of the problem comes from the existing grid infrastructure, especially the "last mile" or two into neighborhoods that now have residential power generation capability.

Yes. Also, about residential PV: daytime most residents are at work, some already with an EV. The relevant consumption just escaped the area of the new production...

 

[Alex Torres]

Regarding your original question, you don't really have many options.

Sorry for not indicating, the growing popularity of these auto-transformers is mainly attached to off grid applications. The EU version of modern off grid inverters (220vac single phase) are way very cheaper than their US split phase counterparts, thus many DIY guys are posting videos on how use an autotransformer to get 120/240 split phase out from a 220vac line as a cheap solution. A safety issue raised as these european models has it neutral bonded to ground adding to that, auto transformers are not self isolated as a general purpose transformer is.

 

[berkeman]

Thread closed temporarily for safety review...

 

[berkeman]

Sorry for not indicating, the growing popularity of these auto-transformers is mainly attached to off grid applications. The EU version of modern off grid inverters (220vac single phase) are way very cheaper than their US split phase counterparts, thus many DIY guys are posting videos on how use an autotransformer to get 120/240 split phase out from a 220vac line as a cheap solution. A safety issue raised as these european models has it neutral bonded to ground adding to that, auto transformers are not self isolated as a general purpose transformer is.

I think we've covered the technical and regulatory issues pretty well, and since the thread may veer off again into discussions of not-very-safe practices, it's best if the thread remains closed now.

Thanks to all for the very helpful replies.