diy solar

diy solar

All-In-WTF (AIW) - AIO Design Disasters ...

50ShadesOfDirt

Solar Enthusiast
Joined
Jan 8, 2021
Messages
489
Why I'm still buying old-school LF inverters such as Magnum 40xx's, and why I'm not really buying into the market's current AIO product scheme ...

For decades, we had lots of wonderful LF inverters, from Magnum, Trace, Outback and many others. These things are big, heavy, and powerful workhorses. My magnum 4024 inverter-charger-mppt-epanel unit (4kw, 24v, basically an old-style LF AIO) powers our whole house, and has done so for over 5 years now. Sold by many vendors (mine was Northern Arizona wind/solar), it is all supported by Magnum (now Sensata) tech support folks, whom I usually pester with an occasional email, and through which I've always received the help I needed.

AIO's have come onto the scene over the past few years. These all-in-one designs, mostly HF, don't seem like they would be capable of replacing my LF workhorse inverter, for these reasons:

1. crummy website, horrible support, sales organizations vs sales engineers ... it's a red flag if:
- the website makes it impossible to get at support documents (manuals, warranties, tech advisories, etc.) for each and every model
- warranty is very iffy (1-year or less is a joke, 3-years or more is better but ultimately useless if you can't get warranty service)
- horror stories abound over the process to get support, replacements, repairs
- it's more a case of a "sales agent", vs a sales engineer

2. these AIO's just aren't where they need to be in terms of design ... the design is broken if:
- the brand/manufacturer is very proprietary such that each AIO is a black box and not modular
a.) there might be failure codes, but you can't figure them out, as you don't have a specific troubleshooting guide
b.) all components within the black box aren't discrete and FRU in nature. if one module dies, whole unit is down
- if the unit doesn't tell you what FRU has failed and needs replacing, and the vendor can't rapidly cross-ship the FRU
- if supporting documents, failure codes, and operations & FRU's aren't perfectly aligned

All of the above is a disaster in the making ... how can one buy these units, given these conditions? It's great for sales (of high-dollar units), but it's horrible for operations.

I had a LiTime LiFePO4 battery fail with a problematic BMS ... some troubleshooting, some pics, and they cross-shipped a replacement battery, and they ate the shipping costs ... that was good after-sales support and warranty service. It's why I still buy LiTime and SOK batteries ... different designs, but both very FRU-able.

I do have a test AIO ... I call it my All-in-WTF (AIW) model ... it's a pretty standard 2kw 24v AIO, and I've got it running a test workshop where power isn't critical. I'm trying to see how long it will last, what all the operational problems are, how well the vendor supports it (or abandons it), and so on. There's been more WTF moments than there have been "that's brilliant" ... but, it's being tested. It seems to be abandonware now, and I expect it to die any day as the warranty period is up (haven't yet used it a full calendar year). Years later, Magnum is still offering a 4024 LF inverter (it is made in Mexico these days), and they still offer to repair my existing one, if it ever dies. That's a stark contrast.

If you consider that the whole AIO is the FRU, that could be a design strategy ... not sure where that falls out in cost comparisons, as I might have to replace an AIO every year, but, it's kind of easy to do. I just haven't seen enough success stories of "my AIO died, and they replaced it under warranty", whereas, the opposite seems to be true.

I still hold out hope for the AIO market, as folks like MidniteSolar have at least put out trade-show pics of very modular & stackable AIO's that seem to be fully FRU. I'm hoping that when my Magnum finally dies, perhaps MidniteSolar will actually have that model out ...

But, I can't seem to get behind a single AIO that is currently being offered today. I can't see them replacing my old-school LF unit. The innards of these AIO's are now so complex that I couldn't repair it, or any piece of it. If the AIO doesn't have a well-executed FRU support & service model, I just can't bring myself to buy one.

Don't know what others think of new-school AIO HF inverters vs old-school LF inverters, but old-school LF inverters are still carrying the load on this homestead ...

Note that we're off-grid, diy'ers ... also note that Sensata has a very neat document out on LF-vs-HF inverter tech; any web search should bring it up.
 
Been running rebranded SRNE AIO's for about 2 years. Other than some unit quirks and small errors on my part understanding the manuals (plus a couple of bone headed moves) they have just worked. I have not needed to use any Service or Warranty. Considering the 3kW units (4 of em, 2 to start and 2 upgraded after a years time) cost me ~$400ea. I can not complain.

My issue with the old LF inverters is they are costly and need several components to do the same thing my 1 AIO does. If the old line companies would bring out a low cost AIO I would certainly be interested in it as a future choice. Innovate or die. Mostly they look like they are dying.

But if what you have works for you than even if it is a Model T in a world of Teslas do not feel bad.
 
Thumbs up on the HF vs LF doc and other docs on the Magnum site.

I have a Samlex EVO-2224 and a bazillion years old Trace DR-1524 in daily use.

Regarding docs, check out the Samlex manual:


About a third of it is dedicated to general electrical, battery and inverter theory. Almost a handbook rather than a user manual. Check out the Definitions section on Page 8, which contains stuff like this:

+++++++++
Phase Angle, (φ): It is denoted by “φ” and speciies the angle in degrees by which the current
vector leads or lags the voltage vector in a sinusoidal voltage. In a purely inductive load, the
current vector lags the voltage vector by Phase Angle (φ) = 90°. In a purely capacitive load, the
current vector leads the voltage vector by Phase Angle, (φ) = 90°. In a purely resistive load, the
current vector is in phase with the voltage vector and hence, the Phase Angle, (φ) = 0°. In a load
consisting of a combination of resistances, inductances and capacitances, the Phase Angle (φ) of
the net current vector will be > 0° < 90° and may lag or lead the voltage vector.
++++++++++

Oh, yeah. Engineering porn in a inverter/charger manual.
 
I still hold out hope for the AIO market, as folks like MidniteSolar have at least put out trade-show pics of very modular & stackable AIO's that seem to be fully FRU. I'm hoping that when my Magnum finally dies, perhaps MidniteSolar will actually have that model out ...

But, I can't seem to get behind a single AIO that is currently being offered today. I can't see them replacing my old-school LF unit. The innards of these AIO's are now so complex that I couldn't repair it, or any piece of it. If the AIO doesn't have a well-executed FRU support & service model, I just can't bring myself to buy one.

Don't know what others think of new-school AIO HF inverters vs old-school LF inverters, but old-school LF inverters are still carrying the load on this homestead ...
I don't have any experience with LF to make any comment, other than the HF units (x3) run my home and business next door going on three years now. As to modular Field Replacable Units - the MPP's are quite modular inside, and when I blew up a comms board (my dumb mistake) a real tech contacted me and ran me through a few tests to confirm my assumption of the damaged part, priced a replacement module within two days and shipped it to my door within 15 days (from Taiwan), they sent me a repair manual electronically and it showed step by step how to dissasemble the inverter to get to other modules which I kept for future reference - so that makes me feel like the support is not as dismal as some have experienced.
I have thought about buying a spare of each internal module, for the MPP 6048's - but never did.
I sized the inverters to run below 80% capacity (generally) but have occationally run them right up to max without issue. But maybe they will not last, time will tell I guess.
All that said, I still watch and read the posts on the Outback and Midnight units to see what they are offering, and how the members like youself experience these units.
I keep dreaming about a pair of 12kW inverters, to replace my three 6kW, and free up some space for another big stack of battery packs...one day.
 
Has there been an objective analysis of the business continuity of the companies behind the old school designs? There has been several cases of corporate drama being aired in the open.

Also those old school designs may not be updated in capability (esp for hybrid folks), and even if the old design works well under new corporate management, said management may not prioritize product development on those
 
Apples & Oranges. Uncertainty & FUD. The AIO market is maturing and currently flooded with players. These designs tend to outperform the older ones in terms of efficiency, but I would agree many of the new players still have work to do to get the products up to speed, around the use case edges. I've found plenty of support info on the EG4 website, I poked around on the SolArk as well, seems to be reasonable. Some of the other vendors are not as forthcoming, I would definitely not go there again if I don't see some reasonable support. Since I have EG4 products, just looking at their manual from last year, there are huge revisions and clarifications as they get the products out in the market, find the issues and correct them, but to your point, they are not as mature as some of the products that have been on the market for 10+ years.

My question would be, 10 years ago were those products all that well documented and bug free? In 3-5 years some of this stuff is going to be like more like buying a TV or fridge. The AIO market will gobble up the bulk of new installs in the space, with a few people still wanting the component stereo instead. The bottom line is going to be cost and simplicity of install. Drop a box or three, cable up some busses (DC, AC) , connect up some PV strings and roll. It simplifies things dramatically, and the AIO is what finally persuaded me to take the plunge.
 
Why I'm still buying old-school LF inverters such as Magnum 40xx's, and why I'm not really buying into the market's current AIO product scheme ...

For decades, we had lots of wonderful LF inverters, from Magnum, Trace, Outback and many others. These things are big, heavy, and powerful workhorses. My magnum 4024 inverter-charger-mppt-epanel unit (4kw, 24v, basically an old-style LF AIO) powers our whole house, and has done so for over 5 years now. Sold by many vendors (mine was Northern Arizona wind/solar), it is all supported by Magnum (now Sensata) tech support folks, whom I usually pester with an occasional email, and through which I've always received the help I needed.

AIO's have come onto the scene over the past few years. These all-in-one designs, mostly HF, don't seem like they would be capable of replacing my LF workhorse inverter, for these reasons:

1. crummy website, horrible support, sales organizations vs sales engineers ... it's a red flag if:
- the website makes it impossible to get at support documents (manuals, warranties, tech advisories, etc.) for each and every model
- warranty is very iffy (1-year or less is a joke, 3-years or more is better but ultimately useless if you can't get warranty service)
- horror stories abound over the process to get support, replacements, repairs
- it's more a case of a "sales agent", vs a sales engineer

2. these AIO's just aren't where they need to be in terms of design ... the design is broken if:
- the brand/manufacturer is very proprietary such that each AIO is a black box and not modular
a.) there might be failure codes, but you can't figure them out, as you don't have a specific troubleshooting guide
b.) all components within the black box aren't discrete and FRU in nature. if one module dies, whole unit is down
- if the unit doesn't tell you what FRU has failed and needs replacing, and the vendor can't rapidly cross-ship the FRU
- if supporting documents, failure codes, and operations & FRU's aren't perfectly aligned

All of the above is a disaster in the making ... how can one buy these units, given these conditions? It's great for sales (of high-dollar units), but it's horrible for operations.

I had a LiTime LiFePO4 battery fail with a problematic BMS ... some troubleshooting, some pics, and they cross-shipped a replacement battery, and they ate the shipping costs ... that was good after-sales support and warranty service. It's why I still buy LiTime and SOK batteries ... different designs, but both very FRU-able.

I do have a test AIO ... I call it my All-in-WTF (AIW) model ... it's a pretty standard 2kw 24v AIO, and I've got it running a test workshop where power isn't critical. I'm trying to see how long it will last, what all the operational problems are, how well the vendor supports it (or abandons it), and so on. There's been more WTF moments than there have been "that's brilliant" ... but, it's being tested. It seems to be abandonware now, and I expect it to die any day as the warranty period is up (haven't yet used it a full calendar year). Years later, Magnum is still offering a 4024 LF inverter (it is made in Mexico these days), and they still offer to repair my existing one, if it ever dies. That's a stark contrast.

If you consider that the whole AIO is the FRU, that could be a design strategy ... not sure where that falls out in cost comparisons, as I might have to replace an AIO every year, but, it's kind of easy to do. I just haven't seen enough success stories of "my AIO died, and they replaced it under warranty", whereas, the opposite seems to be true.

I still hold out hope for the AIO market, as folks like MidniteSolar have at least put out trade-show pics of very modular & stackable AIO's that seem to be fully FRU. I'm hoping that when my Magnum finally dies, perhaps MidniteSolar will actually have that model out ...

But, I can't seem to get behind a single AIO that is currently being offered today. I can't see them replacing my old-school LF unit. The innards of these AIO's are now so complex that I couldn't repair it, or any piece of it. If the AIO doesn't have a well-executed FRU support & service model, I just can't bring myself to buy one.

Don't know what others think of new-school AIO HF inverters vs old-school LF inverters, but old-school LF inverters are still carrying the load on this homestead ...

Note that we're off-grid, diy'ers ... also note that Sensata has a very neat document out on LF-vs-HF inverter tech; any web search should bring it up.
^This X100. ^


Some of these AIO units remind me of the days when you could get a tv with a built in vcr. I think overall the concept will be short lived.
 
25 years of FLAWLESS operation with my pair of Trace SW4048, if there's something they won't start around here I haven't found it. That includes a 1 HP well pump. I've installed and continue to maintain dozens of systems using them that are of similar age.

Something tells me there won't be a single EG4 make it to it's 20th birthday. I'm sure a few will make it to thier 10th birthday, not the majority though, and how many odd bugs will the owners still be suffering?

Then again.... I'm soon going to be replacing them with "The One" from Midnite. If I choose to sell them I'll be able to get almost as much what what "The One" is going to cost me. I might even buy another ONE to keep as a spare because to be honest at this point in my life I don't care about component level repair, just total annual uptime. I can swap out an entire AIO quicker than I can troubleshoot and acquire replacements.
 
Last edited:
@50ShadesOfDirt , I really appreciate your post. I often have the opposite opinion when I see recent installs using LFs, AIMS, or Outback inverters. They typically miss simple homeowner needs like connecting a generator with multiple parameters, setting a charge profile for LifePo4, and enabling remote monitoring with notifications, and offering an easy-to-use mobile app the home owner can use themselves.

I agree that LF inverters are workhorses. They perform exceptionally well, and I don’t see them as "old school." Instead, I view them as solutions for specific problems. We all know LFs excel with inductive loads like welders, drills, and big pumps. There's no reason not to use them for these purposes if they work best.

However, for the general market—people looking to run their entire home off-grid who aren't machinists, welders, or mechanics... AIO systems are fantastic. They offer compact designs and robust monitoring capabilities.

AIO systems have significant momentum and substantial industry R&D investment. As time goes on, they may surpass LF inverters, potentially leading to hybrids that combine the best of both worlds.

As a DIY enthusiast, choose whatever meets your needs. But as someone who has clients and must stay current with market offerings, I can’t ignore AIO devices and only install more expensive LF inverters when they're not necessary for typical household use.

Every problem has it's solution and just like LF inverters, HF AIO also have a purpose and they've not yet been intended to fully replace a scenario in which only a LF inverter will suffice.

Great read, I enjoyed all your points and I appreciate all the thought you put into your post. (y)
 
^This X100. ^


Some of these AIO units remind me of the days when you could get a tv with a built in vcr. I think overall the concept will be short lived.

Depends entirely on how the UL9540 DC ESS stuff goes. If that actually ends up being common, and we can use whatever batteries we want that have it, with whatever inverters / chargers we want that have a UL listing, then separates might come back into vogue.
 
How do we know ?
In that specific line you quoted, I was attempting to say that it's not generally in dispute that LF inverters are excellent for inductive loads. It was an assumption about the experience level of forum members and solar folks. If "we" don't all know, there's plenty of information available online to help us understand if we take the time to look it up. (y)
 
I have not seen any evidence supporting it.
Perhaps the "we all know" is what we all assume.
We "know" that certain "LF" inverters have massive surge ability that lasts longer than me in bed. If you have a relatively small normal load, and an occasional need to start something big, an "LF" inverter can be pretty handy.

Your alternative to handling surge load with most "HF" inverters is to simply stack more of them up to where you can cover your surge load with your normal running load rating. Certainly nothing wrong with that approach..
 
I think the Pre-wired/ E panels are the legacy LF answer to AIO's but it seems like they didnt catch on, i assume due to pricing.
 
From what i have seen High frequency inverters will power inductive motor loads, they just need to be 2 times the size of a Low frequency inverter.
 
Last edited:
I think much of the HF vs LF debate here in this thread (and elsewhere), makes a lot of assumptions, which are not necessarily accurate. While it is true that many of the current "crop" of HF off shore inverters have had issues with large/inductive loads (as well as other issues), it does not have to be that way. For example, the Midnite Rosie, while not an AIO, is a HF design but it can handle surge loads very well. Why, because it was designed to do that. Other HF inverters could do this too, it just wasn't an important design consideration for them (at least not as important as making a very low price point it would seem).

When you look at the current offerings of HF and LF inverters, don't assume that the general characteristics they each have are cast in stone. They are not. The inverter choices we have now are changing, even as we speak. AIO vs separates, HF vs LF, High battery voltages vs low. Each will have some appeal to certain market segments. The solutions you see in the marketplace were just someone's best guess at a marketable product. And as always fads and trends will come and go.

My point, I guess if I have to summarize it, is don't assume that the products you see available now are in any way defining the limits of inverters. They aren't.
 
Operationally, I have heartburn over the current AIO's ... still seems like I'll need at least two on the wall, either running, or one in standby. If running, they'd have to support taking over if other dies. As I place a high priority on being up and running (actually, wife & daughters enforce that priority), it's why I went the magnum route. I'd have to have two AIO's to match the operations side of things.

The above post about "I was able to fix mine, and (chinese) manufacturer helped me troubleshoot, and a few weeks later, I had parts in hand" is good news from repair category, but not so good in the operations category.

My magnum is no better, in terms of how long it will take to fix it, but at least they lead me to believe it is both fixable (send it in and we will repair it), or I could swap another 4024 in (seems like I can still buy them). The thing has been running for over 5 years, with (hopefully) no end in sight.

Operations-wise, I keep a 12-volt inverter on the shelf, and can quickly reconfigure battery-bank to get power back on, in an hour or so. There's the fuel generator behind that ...

Testing continues on the AIO front, but operations still seem iffy ...
 
Operationally, I have heartburn over the current AIO's ... still seems like I'll need at least two on the wall, either running, or one in standby. If running, they'd have to support taking over if other dies. As I place a high priority on being up and running (actually, wife & daughters enforce that priority), it's why I went the magnum route. I'd have to have two AIO's to match the operations side of things.

The above post about "I was able to fix mine, and (chinese) manufacturer helped me troubleshoot, and a few weeks later, I had parts in hand" is good news from repair category, but not so good in the operations category.

My magnum is no better, in terms of how long it will take to fix it, but at least they lead me to believe it is both fixable (send it in and we will repair it), or I could swap another 4024 in (seems like I can still buy them). The thing has been running for over 5 years, with (hopefully) no end in sight.

Operations-wise, I keep a 12-volt inverter on the shelf, and can quickly reconfigure battery-bank to get power back on, in an hour or so. There's the fuel generator behind that ...

Testing continues on the AIO front, but operations still seem iffy ...
I would tend to agree with you 50Shades. But the problem is not limited to just the current crop of AIOs, it is pretty much everywhere, appliances, software, etc. Quality and testing have really gone down over the years and been replaced by low price and quick to market. We really have no one to blame but ourselves, if we refused to buy junk it wouldn't sell, at least not for long.

The problem is, in the environment we have, it is hard for a company to buck the trend. As a company, you may want to test and perfect your product before release, but that takes time and money. Will your customers support the higher price and the delays it entails? A few maybe, but quite a few will jump ship. All an end user can do is find the manufactures who still are producing quality products, and support them.

My current system (soon to be upgraded, hopefully) is based on an old Trace SW4024. Been running without a glitch for 25 years. In today's market, there are not that many inverters out there that will likely have the same track record for the next 25 years. But I am looking.
 

diy solar

diy solar
Back
Top