diy solar

diy solar

Should the neutral wire from an autotransformer be bonded to ground in the breaker box?

Tony Scott and Supervstech:

I watched the videos before I bought the equipment. I don't recall DavidPoz mentioning anything about ground-neutral bond, nor does he discuss the risks of a flying neutral if the circuit breaker between the service panel buses and the transformer itself breaks (at least not in the videos I saw).

I'm only 99% certain I should drive the bonding screw in on my panel. I'm 100% certain that a disconnected breaker to the transformer will put my 120V phases in series on 240V, which I find really dangerous and scary. As I see it, if there is a fault in the transformer itself, the INPUT to the service panel needs to be disconnected, not just the transformer.
 
My updated diagram. Scrutinize heavily!
 

Attachments

  • pic_20.jpg
    pic_20.jpg
    102.4 KB · Views: 63
Tony Scott and Supervstech:

I watched the videos before I bought the equipment. I don't recall DavidPoz mentioning anything about ground-neutral bond, nor does he discuss the risks of a flying neutral if the circuit breaker between the service panel buses and the transformer itself breaks (at least not in the videos I saw).

I'm only 99% certain I should drive the bonding screw in on my panel. I'm 100% certain that a disconnected breaker to the transformer will put my 120V phases in series on 240V, which I find really dangerous and scary. As I see it, if there is a fault in the transformer itself, the INPUT to the service panel needs to be disconnected, not just the transformer.
The thread I linked above has a lot of discussion of neutral bonding and lost neutral and using contactors to break the circuit if the Neutral is lost.
 
I have a general question about Autotransformers.

Without being concerned about any inverters in the system, if a dual-pole 25A or 30A breaker is used to feed a 5kW Autotransformer like this TX-5000: https://drive.google.com/file/d/0B_.../preview?resourcekey=0-DNU_MnXJ8b2WbzEhpX8gEw

Will this mean that up to 5kW of load imbalance will automatically be corrected and the grid will see a balanced 240V AC current draw?

Just to be clear, I’m not talking about having anything connected downstream of the Autotransformer. It is just ‘dangling’ on it’s own dedicated 30A dual-pole circuit with L1/N/L2 extended to it from the main panel through the breaker.

What level of efficiency do Autotransformers deliver? How much energy consumed / heat generated when balancing 1kW of imbalance?
 
Last edited:
I have a general question about Autotransformers.

Without being concerned about any inverters in the system, if a dual-pole 25A or 30A breaker is used to feed a 5kW Autotransformer like this TX-5000: https://drive.google.com/file/d/0B_.../preview?resourcekey=0-DNU_MnXJ8b2WbzEhpX8gEw

Will this mean that up to 5kW of load imbalance will automatically be corrected and the grid will see a balanced 240V AC current draw?

Just to be clear, I’m not talking about having anything connected downstream of the Autotransformer. It is just ‘dangling’ on it’s own dedicated 30A dual-pole circuit with L1/N/L2 extended to it from the main panel through the breaker.

What level of efficiency do Autotransformers deliver? How much energy consumed / heat generated when balancing 1kW of imbalance?
The video posted by Tony on page one addresses a lot of those questions.

The solar edge transformer is specced for a 25A imbalance between 120V sides.
The video demonstrates the balancing capability of the transformer, with equal current from each leg even though there is only a load on "one side" of the transformer.
In the video, the standby consumption was 0.3% of 5000W = 15W.

 
The video posted by Tony on page one addresses a lot of those questions.

The solar edge transformer is specced for a 25A imbalance between 120V sides.
The video demonstrates the balancing capability of the transformer, with equal current from each leg even though there is only a load on "one side" of the transformer.
In the video, the standby consumption was 0.3% of 5000W = 15W.

Great - thanks.

0.3% efficiency is darned good and translates to a worst-case of 15W, so I suppose cooling is not a huge concern?
 
Looks like the transformer core and windings warm up from about 16-18 watts of losses.

If the transformer can handle up to 25 amps continuous of mismatch, that's 3000 watts. You might be delivering 100 kilowatts down L1 and 103 kilowatts down L2 and the transformer would only run current from the difference.

Of course, few inverters have that kind of power, but I've seen people using 2 or 3 of my growatt inverters but only 1 of these transformers. So long as your loads are more or less evenly spaced you shouldn't run into problems. Growatt allows up to 6 inverters to be combined to give up to 30kw.

A new question that comes to my mind is - what do you do if you think you're going to often have unbalanced loads on your phases and are likely to go well outside the 25A current capacity of the transformer? Can more than one of them be "paralleled" to give you 50, 75... amps?

I've looked for larger transformers like this but the only results I find are for industrial use; they're huge, heavy, and expensive.
 
This photo on the Signature Solar site shows five of the Growatts paralleled with two of the transformers.

MINI_44_360x.png
 
Looks like the transformer core and windings warm up from about 16-18 watts of losses.
Meaning cooling is a concern and something like fans need to be added, or not? I’m in a basement that only very rarely exceeds 65F…

If the transformer can handle up to 25 amps continuous of mismatch, that's 3000 watts. You might be delivering 100 kilowatts down L1 and 103 kilowatts down L2 and the transformer would only run current from the difference.
Thanks for the explanation. Kind of mystifies my why it’s labeled as a 5kW Autotransformer then - does that mean when converting 120V to 240V the maximum is 5kW or 2.5kW per leg?
Of course, few inverters have that kind of power, but I've seen people using 2 or 3 of my growatt inverters but only 1 of these transformers. [b{So long as your loads are more or less evenly spaced you shouldn't run into problems.[/b] Growatt allows up to 6 inverters to be combined to give up to 30kw.
I’m interested in an Autotransformer for the exact opposite problem (and 3kW of maximum imbalance capacity will be more than enough).

I’m currently running 2 120V 1kW GTIL inverters, one per leg, for load-shaving (offsetting consumption).

One of the big attractions of the GTILs is they just plug in like any appliance (a toaster oven that pushes some power into the socket rather than sucking some out) so no modification to existing wiring is needed (though I did add a new dedicated two-phase breaker wired to a new dedicated 2-phase outlet).

Everything works like a charm but I’ll often have one GTIL inverter maxed out at 800W (they only have 80% efficiency) while the other is putting out under 30W.

So first, I’ve obviously got some serious imbalance during the typical nighttime usage when it’s mainly fridges consuming, and second, I’m often underutilizing my total inverter capacity and leaving some serious self-consumption I could be offsetting on the table (over 45%).

So I’ve been thinking about tracking all the loads involved down and swapping legs on those circuits to better balance my load, but what a huge PITA.

So I’m considering wiring one of these 5kW (25A) Autotransformers into a dedicated circuit just to more easily balance load to ‘unlock’ the additional ~45% power capacity of my underutilized GTIL.

The imbalance I can often see is under 800W / 6.7A but because the one inverter is maxed out, it is masking whatever additional imbalance may reside beyond that level. But if a 20A breaker protecting the Autotransformer trips, I know I’ve got some big loads I need to track down and redistribute.

I’m on the fence as to whether spending ~$350 to avoid swapping a bunch of circuit breakers around makes sense, but the Autotransformer is more universal (always rebalancing, no matter which loads) and as a bonus it will guarantee that my 2 GTIL inverters are also always balanced in terms of output - that will reduce noise from cooling fans while also probably extending the lifetime of the overutilized unit.

Plus, if/when I add more inverter capacity in the future, I can consider going 240V single-phase…
A new question that comes to my mind is - what do you do if you think you're going to often have unbalanced loads on your phases and are likely to go well outside the 25A current capacity of the transformer? Can more than one of them be "paralleled" to give you 50, 75... amps?
I would think multiple Autotransformers on multiple dedicated 20A or 25A two-pole circuits should work in parallel to add balance capacity just fine…. Each has their own neutral being run from the main panel, right?
I've looked for larger transformers like this but the only results I find are for industrial use; they're huge, heavy, and expensive.
I’m looking for lower capacity / cheaper Autotransformers but I don’t think something like this 3kW 240/120V step-transformer Amazon sells as an ‘Autotransformer’ is the same thing, right?

 
From the specs
25A at 120VAC is 3000W
The transformer can handle a total of 5000W load.
Either phase can handle up to 25A, and it can handle a max delta of 25A, so 25A on one phase and 0A on the other.
Total combined load for both phases is 41.7A of 120VAC

So 25A on one phase would mean a limit of 16.7A on the other phase.
 
Nice Diagram. Best I have seen so far as my curious side gets me studying about auto transformers, and some guys using 230vac single phase GroWatts (like David Poz) and connecting em to 240vac split phase circuit breaker panels; OR other folks taking some 120vac OUTPUTs, and making some 240vac Split phase output options out of 120vac single phase. Interesting options to me.
 
Wondering: What does GTIL stand for?
Grid Tie Inverter Limited, I believe.

Output is limited from the use of a clamp-style CT sensor so there is no export.

Another term I’ve started to see for this feature is ‘zero export to CT’
 
Grid Tie Inverter Limited, I believe.

Output is limited from the use of a clamp-style CT sensor so there is no export.

Another term I’ve started to see for this feature is ‘zero export to CT’
I guess it’s Deye that’s coined that term:


It’s also Deye that manufactures the SUN GTIL inverters: https://www.amazon.com/Inverter-Lim...ocphy=9032080&hvtargid=pla-523294429130&psc=1
 
Grid Tie Inverter Limited, I believe.
Output is limited from the use of a clamp-style CT sensor so there is no export.
Wow: While I have a smart meter at my California home base/ that would not allow any grid tie experiments unless I go through full permit process with better equipment. ... I do have another base in Costa Rica with the older type mechanical electric meter (w reader record keeper coming by once a month) ... that might be something to experiment with. WOW: to reading about " ... use of a clamp-style CT sensor so there is no export". I made record of one guy (Chuck Henderson) from a Will Prowse youtub comment sections saying: (avoid) ... produce more grid tie solar than you use, for even a minute, and the power company will see you sending them power and could potentially have a problem".

I am in high interest mode to learn more about this arena of clamp-style CT sensor so there is no export (options) ... Any links to specific equipment or diagrams would be more than welcome. ;+) ... Bill
 
I guess it’s Deye that’s coined that term:


It’s also Deye that manufactures the SUN GTIL inverters: https://www.amazon.com/Inverter-Lim...ocphy=9032080&hvtargid=pla-523294429130&psc=1
Very Interested: Is anyone in this DIYSolarForum using one of those? Look like a step down in quality as compared to favored MPP brand or GroWatts. ... I am Interested in any type of "auto Power Limiter" options for grid tie experimenting. like on an old mechanical meter type connections (outside usa) ... like in Costa Rica.
 
Last edited:
Desert_AIP is taking the more correct approach to thinking about this.

Electricity is best understood with mathematics.

DC is easy enough. AC becomes complicated when you start to take into account such things as inductance, capacitance, resistance, reactance, back EMF, phase relationships between current and voltage, power factor, all kinds of messy stuff involving angles, trigonometry, and calculus. The hard work has been done, we just need to understand the results and follow the safety rules.

KVA calculations are based on things like how much magnetic flux can you build before you saturate the core, current capacity in the wire, and other things. There is really no "clear" answer for how much this transformer can handle, more of a "X will be ok so long as you don't expect Y" like - "this will power the lights but don't turn on the microwave, blender, and garbage disposal at the same time" kind of thing. Go well outside of basic load recommendations and you'll probably see smoke.


Also, just an update on my question regarding bonding. When I measured the two hot buses before bonding the transformer's neutral to ground, I had 140V on one leg and 100V on the other, with no loads. I shut everything down, installed the bonding screw which clamps the neutral bus to the breaker panel chassis - and now I have 120V on both legs even under unbalanced loads and my ammeter shows very little current out of the inverter on either leg so as far at 60Hz AC, the transformer's impedance is high without a load (I was afraid the inverter might see it as a ground) but now that I think about it, this is expected. (I'm no expert just dangerously self-educated)

This gets me to thinking how secure my ground is. I tested AC voltage of my new ground against the mains AC ground installed with my home, and it's 0V. So far so good...

Before I sit in my chair and relax I'm going to have to be sure that when my electrician friend has a look at it he can't find anything to bitch about. Until then I'm looking for mistakes...
 
Also, just an update on my question regarding bonding. When I measured the two hot buses before bonding the transformer's neutral to ground, I had 140V on one leg and 100V on the other, with no loads. I shut everything down, installed the bonding screw which clamps the neutral bus to the breaker panel chassis - and now I have 120V on both legs even under unbalanced loads and my ammeter shows very little current out of the inverter on either leg so as far at 60Hz AC, the transformer's impedance is high without a load (I was afraid the inverter might see it as a ground) but now that I think about it, this is expected. (I'm no expert just dangerously self-educated)
Where is your primary bond from neutral to ground?

How long was the neutral wire from wherever the primary bond is to this panel?
 
Back
Top