[viper]

Thats what sucks, the Boliy is lighter, puts out more power, and has remote start.. but I just don't trust it yet. I have put a fair number of hours on it as I run my power tools from it. Its sure tempting to just leave the Honda behind. I'm lucky have teenage boys that I can "direct" to help me!

Ha. You want remote features for that Honda? I gotchew! I do this shi* and I am working that up for my own. I want a clean, qualified, automated startup, and ECO override. What I am working on is a delay system to force ECO off (WFO) before an camper AC compressor tries to kick. SOOOOO much easier with an inverter compressor, but most people don't have that. Across the line starts SUCK.

Problem with the Honda is it does not have an electric choke so I have to build that. As you know, they usually only want a whisper of choke, so choke parameters get complicated based on ambient, engine temp, etc. No one has actually really built a REAL genny control for that Honda, but that is one of the best small generators on the planet.

 

[Missouri Bound]

You have left out a few factors.
How well the building is insulated, the location, orientation etc.
A 12,000 BTU unit may be enough.
That being said, unless you know the ampacity required nobody can tell you if your plan would work.
Fortunately mini-splits do not have a heavy start up draw....that may be your savior.
But starting a compressor on a generator that isn't sized properly can damage one or both of them.

 

[seharper]

The first rule of off-grid life is "cheap, stupid, successful: pick two".

I see a disturbing number of off-grid projects fail, and the reason is real simple: they thought they could have all three. Not surprisingly (thanks to Dunning-Kruger), these people DO NOT scrupulously identify their failures and iterate on their mistakes; they just get frustrated and huck the whole thing and pay for the utility service drop.

So with that said, the cabin (more like a small house) will have a mini-split for AC and heating. Due to the size (16X24) I require an 18K BTU unit to adequately heat and cool.

(Before anyone asks about the AC size. It was sized by the AC manufacturer as the building is lofted with 22ft ceilings. We thought about 2 110V units but that increased cost and complexity.)

The problem is your presumption (16x24 therefore 18K BTU) does not work. You can't derive that from that.

The figures you're relying on there are just stock A/C salesman drivel... they want to sell an A/C unit and they don't have any data, so they take blind wild guesses on the basis of "err on the high side". And who cares since it's grid-powered. It's not like somebody has to rack additional batteries or buy a bigger generator LOL.

What you really need is good data on

• The shade, which reveals the primary enemy of your A/C unit: Solar Gain. Solar gain is about 100W (341 BTU/hr) per square foot (square on with the sun, that is). Solar panels or other shader devices, if they have airflow behind them, can greatly decrease the solar gain felt by the building.
• The insulation of the building, which varies dramatically and has a huge impact on HVAC needed. A tight modern building can run its A/C from midnight to 7am and hold its cool literally all day. My drafty old cottage, no way no how! This matters.
• The local climate, obviously. Seattle is a different deal than Phoenix.

Then you can size the unit properly. Size matters when you're off grid.

Of course all mini-split over 12K BTU are 240V (I bet you can see where this is going). Since all my current power options are 120V I want to use a transformer to run the mini-split. Here is where my knowledge lacks.

All your current power options are 120V because you are cheap, and chose to minimize the size of your system. That's fine and good, but then -- you must sacrifice "stupid" OR "successful". You can guess my recommendation LOL.

Now, when you see a 240V unit, that is a strong signal from the manufacturer that the unit requires too much power to be practicably powered from 120V. So watch out. We'd need the nameplate specs and I don't see where you've shared them.

I would like to use a standard transformer (like an ST-5000) to step the voltage up to 240 to run the AC.

That's not a "standard" transformer at all. That's a cheap Chinese transformer - typical, unsafe, illegal rubbish sold direct mail only to slip it by the FTC and other regulators. (Customs is simply too busy to intercept them, but they would if they could). Of course all the junk stores - wish.com, eBay, AliExpress and Amazon Marketplace are just glutted with them, to the point where you can't find a real one on those crap marketplaces. Also the low price twists your mind so you are unwilling to pay for a real one.

If you didn't know, Amazon opened their sales website to 3rd party sellers, so effectively it's eBay now.



Here's the problem. An HVAC listed for sale in the US is expecting 240V, yes, except each leg is only 120V from ground. The unit may not be built, tested or approved or 240V phase-ground voltages. On the other hand if it's also sold in Europe, it would be, since their power is 240V hot-neutral. (and a whopping 416V between 2 phases of hot - distribution is 3 phases, not necessarily to every home.)

And then NEC has rules relating to allowed voltage vs neutral - some rules prohibit >150V to neutral. It's unfortunate that you committed to 120V, since all this is easy with split-phase generators and inverters. But this is a typical problem as you grow an off-grid system - maybe it's time to admit you outgrew your initial build. That's hardly failure.

But here's another problem with some units: they run 240V compressors but have 120V head units. That means they need genuine neutral. Now you could synthesize that with a buck-boost transformer, by configuring it to "boost" on the opposite side of neutral, thus synthesizing a "-120V"*. That would also be safer. But those are not $169 for 5KW. (on the other hand they don't have to be so large, since only half the power goes through them).


* Of course it's AC power, so +120V and -120V aren't really valid, or to be more precise, not valid half the time lol).

In case someone down the road reads this and want to understand the math. The unit I bought has a COP of 3.0.

Yeah, but it's not like that's its COP end of story. That's its COP in certain test conditions. Above those, COP will be better; below it worse. For instance, add freezing rain and it's having to do a lot more defrost cycles and the COP suffers as a result.

But it certainly is not a Honda and I don't care what any db test shows they are not as quiet.

Because they're measured in cheap Chinese decibels LOL.

 

[viper]

@seharper , did you get no action last night and on a bender..?

There are other ways to make split phase, but I asked and he said 240V is required. As for the 240 to ground issue, uh, I just cannot agree with any of that. The chit chat between any leg and ground should be in the Mohms sector. Not concerned.

As for HVAC loads, I have run them for a similar, even bigger envelope, and I am about 9k btu with 12ft side walls. Do I think it is overkill, OH YES, but he was not asking about any heat gain calcs, and the variable speed nature of minis does help to a degree and in a cabin, I can appreciate a little overkill because you open a door, you are probably losing half the envelope volume out the door! Then you have the "just got here, want it cool" factor.

No, I simply won't agree with a blanket statement of "100W solar gain" because that is far from accurate. Emissivity plays a huge part in solar gain and a white metal roof will perform quite different to a black asphalt roof. We have our own data on our test work there.

While I agree, HVAC requirements 'might be' off the mark here, there was not an indication of wanting that info. Don't know climate, insulation, or building specs. I probably don't care right now. He wants 240V to a mini, and I'd throw a fat check that a simple iso Xfmr will solve that problem, because I have done it 100s of times.

BUT, I didn't check the transformer he was referencing. Have seen this SO many times where China crap is referenced. OP, you NEED and actual "listed" transformer. Shop for a real one, they are out there. China junk will never work!

 

[seharper]

As for the 240 to ground issue, uh, I just cannot agree with any of that. The chit chat between any leg and ground should be in the Mohms sector. Not concerned.

It's a question of what the equipment is rated for.

As for HVAC loads, I have run them for a similar, even bigger envelope, and I am about 9k btu with 12ft side walls. Do I think it is overkill, OH YES, but he was not asking about any heat gain calcs, and the variable speed nature of minis does help to a degree and in a cabin, I can appreciate a little overkill because you open a door, you are probably losing half the envelope volume out the door! Then you have the "just got here, want it cool" factor.

With off-grid, overkill means a more costly battery bank, or a heavier generator to buy, lift and haul gas for. You don't want to do it unnecessarily. I'm assuming they take the generator home.

With off-grid, conservation > generation because conservation is cheaper. With grid, it's the other way 'round.

No, I simply won't agree with a blanket statement of "100W solar gain" because that is far from accurate. Emissivity plays a huge part in solar gain and a white metal roof will perform quite different to a black asphalt roof. We have our own data on our test work there.

I'd love to hear some data about the roof rejecting heat by radiating it or passing it to outside air. What I can vouch for is albedo. Of common roof materials that people prefer to use aesthetically, a whole lot of them are under 10% albedo, and virtually all are under 20%. Even silver "tin roof" has a sub 20% albedo.

The only way to get to even 50% is to really, really be trying, and you will notice immediately that it is a high-albedo roof because of being so white by comparison to every other roof. Some HOAs have a big problem with that, actually. The paint I use on metal has a 91% albedo, that's as good as I can get but it also depends on being kept clean.

Honestly I don't think much about emissivity because my 91% albedo roofs don't get hot enough to emit much lol.

 

[viper]

It's a question of what the equipment is rated for.




With off-grid, overkill means a more costly battery bank, or a heavier generator to buy, lift and haul gas for. You don't want to do it unnecessarily. I'm assuming they take the generator home.

With off-grid, conservation > generation because conservation is cheaper. With grid, it's the other way 'round.





I'd love to hear some data about the roof rejecting heat by radiating it or passing it to outside air. What I can vouch for is albedo. Of common roof materials that people prefer to use aesthetically, a whole lot of them are under 10% albedo, and virtually all are under 20%. Even silver "tin roof" has a sub 20% albedo.

The only way to get to even 50% is to really, really be trying, and you will notice immediately that it is a high-albedo roof because of being so white by comparison to every other roof. Some HOAs have a big problem with that, actually. The paint I use on metal has a 91% albedo, that's as good as I can get but it also depends on being kept clean.

Honestly I don't think much about emissivity because my 91% albedo roofs don't get hot enough to emit much lol.

I am not going to get in an argument over roofs, but I do believe some of your data is flawed, or understanding of it. We have built test roofs for other firms, and they were highly instrumented because this largely helps everyone. What you have to realize is approx (just go with it for now) 1000W/M^2 strikes a roof. That is energy that will be mitigated in one of two ways. Returned or absorbed. Those are your options.

The term common today is "SRI" or Solar Reflective Index. It's pretty much like R value because no one can understand U value. Highly reflective and even 'reactive' coatings are a thing of the future for sure. I mean, sir, if you need some level of confirmation in any of this, all you need is a small sample of black and white barn metal, same gauge, whatever. Lay them out and see which gets hotter! It's not even close to comparable.

In one of my personal buildings, I actually have a black metal roof with Aluminum IR sheet right under it. Nothing else yet so I can walk right in there and test. It is nothing short of staggering! I will even admit, by the science, I was not convinced because the sheet makes contact with the roof steel, but I think all the corrugations gives enough air gap to do it's thing.

As for Emissivity. I replied in the garden area of a guy that I think has part of his yard burning up from Low E(Emissivity) windows. That is where you can mostly see this work. So yes, solar gain is very much NOT a fixed value.

 

[AllanJ]

Provided that it does not draw 120 volts for some of its components, neither half of the mini split system can detect the difference between split phase (240 volt) and "straight" 240 volt power including 240 volts hot to neutral from a 416 volt 3 phase system.. The term "split phase" is used when different AC voltages are derived from the same single phase source for simultaneous use (and will be in phase or 180 degrees out of phase with one another). Contrasted with, for example, a three phase system providing for some loads or applications two 120 volt supplies (that are not in phase or 180 degrees out of phase) together with a higher voltage supply that is only vaguely 240 volts (208 volts to be exact).

 

[travellerw]

Provided that it does not draw 120 volts for some of its components, neither half of the mini split system can detect the difference between split phase (240 volt) and "straight" 240 volt power including 240 volts hot to neutral from a 416 volt 3 phase system.. The term "split phase" is used when different AC voltages are derived from the same single phase source for simultaneous use (and will be in phase or 180 degrees out of phase with one another). Contrasted with, for example, a three phase system providing for some loads or applications two 120 volt supplies (that are not in phase or 180 degrees out of phase) together with a higher voltage supply that is only vaguely 240 volts (208 volts to be exact).

Thanks fella, I appreciate the straight answer and explanation. I have actually spent some of this weekend reading about 220V. Actually it really clicked when I researched multi-wire branch circuit. While this is not that, the theory behind multi-wire branch circuits helped me understand how the whole 2 hot thing works when the neutral is not used.

I will say the rest of this post certainly has its value in "Entertainment"! LOL.

 

[viper]

Thanks fella, I appreciate the straight answer and explanation. I have actually spent some of this weekend reading about 220V. Actually it really clicked when I researched multi-wire branch circuit. While this is not that, the theory behind multi-wire branch circuits helped me understand how the whole 2 hot thing works when the neutral is not used.

I will say the rest of this post certainly has its value in "Entertainment"! LOL.

It's 240......

 

[seharper]

Provided that it does not draw 120 volts for some of its components, neither half of the mini split system can detect the difference between split phase (240 volt) and "straight" 240 volt power including 240 volts hot to neutral from a 416 volt 3 phase system.

Depends what you call "detect". How about failing because voltage exceeds insulation rating ... does that count as "detecting"? LOL

NEC 110.3 requires equipment be installed according to its instructions. It is not designed for, nor tested for, voltages other than spec. If a Euro unit says "neutral must be near ground" that's that. If a domestic unit says "no phase wires over 120? Volts" then you can't run it Euro style or even off a 240V wild-leg delta system.

OP needs to use a quality isolation transformer to synthesize the other 120V phase.

However, I am concerned that after resolving the voltage problems, OP will smack directly into practical power limits of that generator and that inverter. I.E. the unit is too big a load for the generation available. So OP would have to enlarge inverter and generator anyway, and might as well just go 240V at that point (probably will have to).

 

[travellerw]

It's 240......

LOL.. OK!

 

[viper]

LOL.. OK!

I was just throwing you a bone there. Not trying to insult but references to "110" and "220" are a sure way to confirm to electrons that you are not in the know. There are many voltages that can be made, and we do wacky things in industrial to get where we need to go. Once you understand the "turns ratio" of transformers, you can do many things, and many have multiple taps, which just means a wire jumps off the coil train early.

It's all fun stuff and easy enough to learn. I am NOT on this train with needing split phase power to your device. It is not practical from a design perspective, and almost a certainty that design engineers considered many connection scenarios. Very super common would be the 240/120 delta 3phase, and a 480:240 single phase.

You will be fine.

 

[komondor]

You have left out a few factors.
How well the building is insulated, the location, orientation etc.
A 12,000 BTU unit may be enough.
That being said, unless you know the ampacity required nobody can tell you if your plan would work.
Fortunately mini-splits do not have a heavy start up draw....that may be your savior.
But starting a compressor on a generator that isn't sized properly can damage one or both of them.

I agree I would look at insulating as much as possible and spending that money once vs going thru all of the work of trying to get 240 volts.