Hello from Maine

Welcome to the forum, yes, we have a good time over here.  Kind of a dedicated forum base to Genetry Solar / Power Jack inverters, but there's also a lot of solar system experience here 😉

2 hours ago, BlackWaterPark said:

A robust (supposedly) low frequency inverter between 5 and 7kw,

Should be quite robust.  We have several customers that have run multiple megawatts of power through their Genetry Solar inverters (myself included).  With the strong & clean FET drive, they should be quite reliable--though to be completely honest, we HAVE had some manufacturing difficulties that some customers have run into (poor soldering on a particular connector, ugh!)  And then we've had things that I simply didn't think of...that customers tested (of course they would 😉).  Any design fault/manufacturing issues are covered under warranty of course. 

Obviously, I learn from every inverter difficulty that a customer has--and as I've designed it from the ground up, I have thus far always been able to pinpoint exactly why a particular inverter failed--and take steps in future revisions to ensure that that cannot happen again.  So we learn 😉

2 hours ago, BlackWaterPark said:

120v and 240v option

You did see that the new Rev. C control board can automatically switch the input to handle 120v or 240v input, right?  Changing the output from 240v 6kw to 120v 6kw does require an internal rewiring (and reconfiguring the system setup through the LCD)...but it is possible.

2 hours ago, BlackWaterPark said:

a plethora of options to change a multitude of parameters built in

This is my biggest bugaboo with pretty much any commercial product.  For example, I KNOW "such and such" is completely possible--but the manufacturer simply didn't feel like making it possible.  AND if you ask them, you'll find out that they either don't know how to do it, or don't care to do it.  And you're out of luck.

Case in point: I actually sold the very first inverter I ever bought (a MeanWell TS-3000 HF inverter, pretty decent if I say so myself) simply because there was no way to adjust the OVP shutdown value.  Specsheet said 60v...but it would shut down at 58v.  And there was NO WAY to change it.

Yes, there's PAGES of settings on the GS inverter.  And if you need a feature that should be technically possible...ask!  I've added several features to the GS inverter per customer request.  For example, you can turn the power switch LED off from the configuration screen--that was a customer request.  (Living out of a van, and the Power Save flashing light was lighting up their van, making it difficult to sleep.)

2 hours ago, BlackWaterPark said:

with a WiFi function and upgradable firmware

Check and check.

I am planning a major redesign of the WiFi configuration and local server--once I get enough other stuff done.  mDNS local network discovery was another customer suggestion; I am currently open to ideas for the best way to provide inverter stats/control via local network.  Considering Modbus over TCP if I can find a small enough library that isn't thousands of lines of bloat...

To your questions:

2 hours ago, BlackWaterPark said:

1) and forgive me if this was documented somewhere that escaped my scrutiny, but what's the efficiencey numbers on the 48v GS6kw (assume all following queries are in regards to this unit)? 

Yah, it's not documented in the manual (yet)!  Efficiency is a bit of a tricky thing to measure, but measured/expected efficiency should be somewhere in the mid 80s.  About par for the course on low-frequency inverters; yes, high-frequency inverters can get into the mid 90s on efficiency--but they're terrible with imbalanced loads, surge loads, or well, ...dare I say, "loads in general"?

Most of the losses are in the transformer...but if said losses were reduced, we'd basically have an inverter that could do a much higher power output...catch-22 I suppose 😉.

2 hours ago, BlackWaterPark said:

2) I'm slightly confused about the AC input side of things... This unit has an AC charger onboard, right?  If so, can you go into the minutia here?  (input voltage range, output range, 3 stage? etc) My interest, being far far away from the nearest utility pole, is 100% offgrid functioning, including genset charging of the battery bank.

Yes.  AC charging requires AC pass-thru first (AC output connected via relay to the AC input terminals), then the inverter can use the FETs and the transformer to make an impromptu "boost converter" to reach the desired battery voltage.

Input voltage range doesn't have a particular nailed down "min-max" at least at this point (currently the firmware will disconnect if it's over 280v for a 240v input).  The mechanical limits of input voltage range have to do with the transformer core saturation voltage (reduced efficiency)...and/or the fixed voltage divider ratio of the transformer.  For example, the GS6 48v inverter has a 240v-32v transformer (no secret, you can see that number from the Diagnostic Info screen, haha!)...but why that's important is that if you put 280v into it, the fixed ratio of (240/32 = 7.5) results in 37.3vAC out * sqrt(2) will result in a minimum battery charge voltage of roughly 50v.  At 240v in, minimum charge voltage will be roughly 43v.

Maximum battery output charge voltage...not sure just yet 😉.  I might have a generator, but it's still NIB after a full year, so...admittedly haven't tested the limits of this.  Should be able to reach 60v though.

Charge efficiency is currently ~60%, which is pretty poor.  (I do find it very curious that I have NOT found this characterized on any other LF inverter...probably because it's about the same!)  I do have plans to convert the charging to more of a PFC waveform, which hopefully will significantly improve the efficiency.

2-stage charging as of right now (though admittedly without an auto-switch-between "bulk" and "float" right now--just being up front and honest), no temp compensation (yet!), though the software can easily be updated as I learn of what is needed/desired.

Charge current is fully customizable.  I have yet to run experimental maximum charge loads to see what the max sustained current is--though in most cases, the AC input current will be the limitation.

Genset charging, care for a (seemingly industry-first) 2-channel genstart output?  Yup, you read that right: 2 completely separately controllable contacts (across 3 leads).  Every other inverter I've seen always gives you a single switch output "NC C NO"...which you can't use to run a "Start - Run" sequence, or a remote control pushbutton "on" or "off" trigger...haha, I like features 😉

2 hours ago, BlackWaterPark said:

3) Did I seriously read that right somewhere on here? A scant 20 watt self consumption?

Yup.  20-25W is expected, and that's with the WiFi connected, LCD backlight on, inverter producing 240vAC out, no load.  Here's a photo of a GS 6kw inverter on my bench, running no load, full backlight, connected to WiFi (and the remote MQTT server).  Admittedly, this power consumption is a TAD lower than a regular production unit, seeing as this prototype control board is using the smaller 4.7uF output filter caps instead of the spec 10uF filter caps.  But it shouldn't be much over 500mA @ 48vDC.  (Worth noting: the transformer is almost dead silent at no load.  There is NO very obvious 60Hz hum from the inverter at no load--as seems to be the case with all other LF inverters I'm aware of.)  FETs and transformer will get barely warm after about half an hour of no load operation.

<img class="ipsImage ipsImage_thumbnailed" data-fileid="907" data-ratio="75.00" width="676" alt="image.png.a9a12530d8b679e9e3bd01decf484490.png" data-src="/monthly_2021_11/image.png.a9a12530d8b679e9e3bd01decf484490.png" src="/applications/core/interface/js/spacer.png" />

Power Save mode goes even lower, closer to 4W (yes, with WiFi connected and the LCD backlight on full brilliance.)

Yes, these numbers are extremely unusual in the LF inverter market.  Hey, it's possible 😉

2 hours ago, BlackWaterPark said:

4) How field serviceable are the internals, on a scale of Outback to one of those $129 40kw split-phase alixpress inverter/charger/mppt/improvised vertical flat-top grill units?

Well, the MOSFETs are on boards that just require a dozen screws each to remove, and then can be replaced--see above photo.  (Current GS inverters actually are using a Power Jack mainboard, though the next production run will use our in-house design.)  Removable FET boards make the inverters very serviceable.

If the FETs have the misfortune to go, the LF driver usually gets damaged as well--but it is also socketed, and easily replaceable.  Due to the driver circuitry design, damage cannot (and has never!) gone further upstream in the inverter.

At least in my book, the GS inverter is highly serviceable.  Heck, if you want to change the system voltage between the available options (12 / 24 / 36 / 48), that can be done in the field without replacing any parts except the cooling fans (you'll just need new fans of the appropriate voltage and specification).

2 hours ago, BlackWaterPark said:

5) What's the typical lead time for one (or two) of these things? 

Up to Sean to do final assembly on them and ship 'em out the door: they are currently in stock.  Hopefully less than a week, depending on how much work he has to do--or life's happenings.  We're real people, individuals...not a large scale company with lots of money 😉

2 hours ago, BlackWaterPark said:

6) What would be the difference between paralleling a pair of 120v units to create 240v (assuming that can be done in the first place), or Master/slave setup of a pair of 240v units?  

Using a pair of 120v single-phase inverters to generate split-phase 240v is actually the preferred method; we call it "split-sync", where one inverter is the master, and the "slave" generates an AC waveform 180 degrees apart.  Surge amperage is doubled with this method.

Please note that 240v AC input is not possible with split-sync inverters, at least with the current design (as the AC input on the "slave" inverter is used for the sync signal.)  This also will limit AC input to 120v, and charge to a single inverter.

Full-on parallel operation is quite a bit of a challenge due to the fact I tried to design the inverters to not require a parallel cable.  As of right now, the function is not useable in the Rev. C boards (which are currently shipping), though this is a simple matter of getting the firmware up to speed.  Full parallel operation will not double the surge rating of the inverter, unfortunately...so there's pros/cons to either method 😉.

2 hours ago, BlackWaterPark said:

7) what is the range of the low voltage disconnect?

Down to 40v for a 48v inverter, though you'll start to lose the pure sine output below 45v (at no load).  (32vAC * 1.414[=sqrt{2}] = 45.238v mathematical min).

I've probably got you more questions now; I'll answer as I can.  Off to bed for now though 😉.

Here's one for the solar powered only crowd to think about for future firmware.  On low load draw (user definable) / no load detection (user option to enable / disable the detection), modify the drive waveform to reduce the inverter's self-consumption.  This is something Victron has as one of the power reduction modes on the multiplus (and phoenix?).  In the older Victron Windows configuration tool this used to be referred to as modified sine wave, but it's not the horrible stepped arrangement instead it results in a ... narrower .... waveform.  More on the standard sinusoidal closer to zero, but steeper towards the peaks.  Old CRT TV hacks might think 'that sounds like velocity modulation'.

lSo much to digest...

couple of things that stick out in my mind though,

This place is great!

So, considering a 60% ac charging efficiency, as I'm going with a lifepo4 bank this time around, I'm going to guess that using an external lithium optimized charger would be preferable here?  Quite honestly, I was really looking for a quality LF inverter with a low self consumption anyway, so this is a non issue for me.  I currently run a Solinba 3kw LF unit, and I've never used it's charger function, preferring the Iota.  So nothing new there, haha. You already had me at ~25w self consumption and the the 120v "split-sync" dual inverter design. This is *exactly* what I was looking for, and anything cool on top of that is just gravy.

But that begs further inquiry into your balancing modules... at those related at all to the on board charger? And are they more for the DIY battery banks, or are they applicable to prebuilt ones as well? I ask because I'm entertaining several options right now, including (but not limited to) the EG4 Lifepower 48v 100ah rack units, a bank of battle borns, and 4 or 5 Simpliphi phi 3.8 48v modules. I see that the EG4s have specific coding to communicate with the growatt crap... is that something on the horizon for you with the GS inverters in any way? If you have any recommendations here, I'm all ears  I'm a DIY type of guy, but I'm looking for reliability/longevity over all other concerns.

Regarding the Meanwells, well, they are like the Honda of the driver world (don't know about their inverters, per se)... pretty efficient and dead reliable, but loathe to afford you any real features to speak of. I've wired lots of them into horticultural light rigs, and for my own, that's all I'll use. Much like Hondas actually, which are all I'll own for vehicles. Your probably starting to see a theme with me by now.

So, here is a question then... you say you've got a new upcoming proprietary board to replace the PJ boards currently in use now... is that something I should wait for, or is that something that I could get later to field replace the PJ board that would otherwise be in my inverters now, If I were to order a couple?

regarding the lvd, so what's the upper threshold? I should have elucidated more with my question actually, because I was more looking for the upper range (for instance, the Simpliphis require something like 52.6 or some such).

regarding the genset charging reply, I'm not exactly sure what you said, but it sure sounds tasty!  I'm not exactly an EE over here, hehe...

Well, 60% being what it is...I would say to wait on purchasing a dedicated charger if the inverter can do what you need (albeit at a slightly lower efficiency for a little bit here).  I need to revamp the charge code to a PFC-style charging methodology, which hopefully will gain another 15-20% charging efficiency...it's a simple firmware update. 

Though it is worth noting that the inverter "charge" efficiency will always be lower than the "inverter" efficiency.  Transformer loss will remain the same--but most of the charge amperage goes through the FET body diodes (i.e. more lost energy in heat).  I would be happy to see charge efficiency around 75-80%.

But like I said...I have not seen ANY other LF inverter manufacturer publicize their charge efficiency number...probably because most people don't check it 😉.  They only characterize "transfer efficiency", which as a straight relay, is basically a useless spec.

Not at the current time...though they technically COULD be at some point.  There's an extra serial port on the inverter control board that COULD be used to interface to various other system elements if desired/needed.

GS balancers are for DIY banks, not prebuilt ones.  They're a very simple and basic design--intended to be a better choice than the Chinese "BMS units" that are only good for 0.03A balance current in the first place.  (That simply isn't enough for the large battery banks required for off-grid solar systems.)  The GS balancers are definitely not "polished showroom ready products"--they're more like the components inside the fusebox of your car.  Very functional, but not picturesque.

Yes, this is something we are considering.  I would have to make a breakout for the serial port on the inverter for communicating with products like this, and it would be a hardware addon for the inverter (i.e. more than just a firmware update).  To support a particular system, I would have to have comm protocol info to implement it...and a test unit (or test customer) to debug the functionality...but it is very possible to implement.

I'm more of a DIY-over-prebuilt batteries...simply because of the outrageous costs of the prebuilt systems.  They like to sell LFP batteries at $600-800/kwh...

...but I built my entire 12kw LFP bank for roughly $1,200.  Yeah, that was due to a very good price on NOS 32650 cells on eBay, but it's currently possible to get cells (Lishen/Eve) from China for under $200/kwh.  Whether or not they have the same longevity is yet to be determined, but...when it's 1/3 the price 😉.

Yes, we do have an upcoming "mainboard" that will replace the PJ boards currently in use.  However, for 6kw inverters, the PJ board DOES function just fine.  We will be replacing it with our design "just because"...and also because we would control the specification, and not have their changes affecting our design.  From a functional perspective however, there would not be any significant improvement/change--as the boards are quite capable of doing 6kw continuous.

We only had to design our own board (which is VERY similar in actual design...just with a few improvements/modifications) in order to make a 12kw inverter possible at all...and are just taking that design down in size for the 6kw.  More of a "product line" thing than a necessity.

In summary: no, you do not need to wait for the new boards.  While technically you could replace the PJ mainboard at a later date...the mounting holes won't line up, and it's pretty nigh impossible to drill new holes in the stainless steel chassis.  (Believe me, I've tried!)  And really you wouldn't gain much at all. 

Upper battery threshold on the inverter design?  We have a customer running a GS inverter up to 66v...and they've put that inverter through the wringer, promising to blow it up (but it is still running just fine, haha!)  I wouldn't recommend running a 48v GS inverter much past that--but that's far beyond the normal range of a 48v battery system anyway.

You can see that video here: https://www.youtube.com/watch?v=Pq8yYLY8hy

Yes, the fans aren't the world's quietest (rated 250CFM each), but they aren't close to as loud as this video makes them sound!  Yes, you'll definitely hear the fans if they're running at full blast...but you definitely do NOT need earplugs around the inverter.  (Think a vacuum sweeper; the official spec on the fans is 72dB, which is par for the course of that speed/size rating.)

OK, so simpler...

...most inverters out there with a genstart function give you a single "light switch", i.e. "on/off" output.  They give you both sides of the switch--but it's only one switch.

GS inverters give you TWO switches on the output.  This gives you the ability to do an "ignition key" procedure (i.e. "run...START...back to run"), or "press" one button for "on" and another button for "off" (i.e. remote control), etc.

This all sounds really good. I think I'm highly likely to be onboard here, just trying to figure out the rest of the system and make sure everything works harmoniously.  I'm almost certain to buy prebuilt batteries at this point, so I'm looking at EG4s,Energetechs, and possibly Simpliphis (though they aren't high in the running due to price and warranty requirements, though there's no lead time on them at the moment, unlike seemingly everything else).

So, of i get 2 of the 6kws, I'll be able to run 240v as I'm understanding it.  One stays in standby mode and fires up when a load is then detected on that 120v leg? I'm not quite sure on this part.  does each inverter get weird into the breaker box, or only one and the other is essentially daisy chained, so to speak, to the master inverter?

Are there plans for a Genetry app at all?

and thanks for clearing up the confusion about the genset switch. I get it now!

And, I'll try one more stab at the lvd question here (I apologize if I'm not articulating this correctly)... I'm asking what the highest voltage number can be set to activate the lvd.  For example, my brother got some simpliphi batteries, and was using his VFX3524.  That inverter did not meet the warranty requirements of simpliphi because the highest value of the lvd on the inverter was simply not high enough for these Lifepo4 batteries, and the unit is also not firmware upgradable to address it.

For split-sync operation, both inverters only produce a single 120v phase (120v @ 50A), and are synced together to operate 180 degrees apart.  Both get connected to the breaker box to opposite phases.  (Disadvantage being that you can't use 240v AC input for ATS/charge.)

For parallel operation...both inverters are internally wired for normal 240v split-phase output.  One unit is the "master" that carries loads, then the "slave" unit shares the load (split-phase operation on both).  However, at the current time, the parallel function code has NOT been written/tested for Rev. C inverters...I have so many other things to take care of first!

We currently have a "webapp" (platform-agnostic) on the "wifi.genetrysolar.com" page once you're logged in.  Please note that it's a separate ecosystem (not tied to the "Genetrysolar.com" login just yet); this likely will be changing. 

Nice thing about a webapp is that we don't have to consume resources making a "Android app" and then paying regular costs to have an "Apple app"...though Apple persists in being the last one to the game (they barely support WebApps).  The app shows up as a regular app on your phone homepage--and if properly designed, is almost indistinguishable from a regular app--just you won't find it on the Play Store, and it doesn't get sideloaded either.

Got it this time, now I know what you're asking 😉.

You're asking about a Genetry product: The question is, "what do you need the setting to be?"

Genetry inverters calculate UVP as "points below Nominal".  OVP is calculated as "points above Nominal."  Normally, Nominal = battery nominal voltage (for a 48v inverter, this is 48v.)  However, on a GS inverter, you can adjust said "Nominal" value up to 55v (in the event that you want UVP or other "below" thresholds above 48v.)

In short, you basically have an effective UVP setting range of 40 - 55v on a 48v GS inverter.  High enough for SimpliPhi?

Oh,yea... way high enough, haha. This is not especially relevant to my pending system revision (as voltage doesnt seem to be a great way to ascertain SoC, buuutt... I'm going to eventually get my brother to buy into these inverters as I implement them myself. As I'm the one that has to solve his system issues (scary that I'm the most knowledgeable one over here), it makes it easy to have homogenous gear.

Ok. this is awesome, actually.

This doesn't really seem like a problem in my situation, being totally off the grid. I've been using a Yamaha ef2000is to run my charger, which only outputs 120vac anyway.  

Thanks for managing to answer all my questions Sid. You sound like a pretty busy guy in general, so I appreciate your commitment to timely responses on this forum.

What is the weight and dimensions of the GS6? I looked high and low, but couldn't find those stats...

approx 22 x 12 x9 
55 lbs

Thanks Joe 🤘

Getting close to getting my batteries, so I'm trying to finalize what inverter setup to go ahead with here in preparation.  I'm looking at my situation at it stands now with the fact that I'm running my whole house from a single 3kw 120v inverter, and I'll be adding at least an electric standard size fridge as an additional load, ditching my current LP unit in the process.  I'm wondering if a single 240v 6kw unit is sufficient to start out with, or even a single 120v unit for that matter.  Because I'm ramping up the storage considerably, and because there's not much of there in the way of external 48v chargers that go much beyond 20a output, charging power might be the name of the game for me. My plan is to aquire a pair of Kilovault hab 7.5kw 48v battery systems in parallel, and to get the extended warranty, the charge rate is limited to 50a per, so 100a total, and to get this I'm going to need 240v generator output, so far as I can see.  In the event of multiple inverters, this pigeonholes me into the 240s obviously, but in the case that a single 120v unit proves sufficient to power the house, I'm assuming here that you only loose the 240v input option when you actually put the 120s into spilt sync, but only running one will operate with a 240v input just fine, right?

...honestly, a 'fridge is NOT a bad load 😉.  I have a 26 cubic-foot Kenmore Elite French door fridge...it's been "repowered" with a regular recip compressor, and runs ~1.2A at ~30-70% duty cycle based on the ambient temp.  Defrost runs 4.0A when both elements are running.  ('Fridge originally came with the biodegradeable [but more efficient] linear compressor, which failed on the clock at ~3 years.)

I guess we offer too many options, it can get confusing 😉. 

With the current setup, you would only have the 240v input option if the inverter is internally wired for 240v split-phase output.  This has to do with the transformer mechanics...if the transformer secondaries are paralleled for full output @ 120v, there is no 240v tap or ability to accept 240v input.

In other words, if both inverters are internally configured for 120v single-phase output, they can only accept 120vAC input due to hardware limitations.

TECHNICALLY it should be possible to handle 240v input across 2 single-phase inverters, BUT this requires some hardware modifications and an external sync cable (instead of the current method using the AC Input terminals as a sync signal.)

It sounds like your usecase might better be handled with 2 regular split-phase inverters...with the parallel mode on the slave inverter (after I get that rewritten for Rev. C and hopefully a lot more stable!)

Remember, the GS inverters are rated for 6kw regardless of whether they're configured for 240v (=25A) or 120v single-phase (=50A).  If 6kw will run your house, you just need to balance out the 120v loads across a 240v split-phase output, and you'll have the same powering capability + 240v AC input.

LF inverters are not super picky about how balanced the output load is--yes, it's good not to run much over 1/2 the output rating off a single phase for a very long time, but they'll do it without complaining.  (HF "split-phase" inverters will really complain and/or shut down if the loads are too imbalanced.)

ok, I had it a little wrong about the 120v units in that case, so I get it now.  So in regards to the revC boards, are those now the standard in the new GS inverters (as in if I were to order them now that's what I would get in them) and you just need to do some software updates when you rewrite it, or are they something still down the pipeline?  

And, as they can't currently run in parallel due to the software issues, how else would I use two 240v units together until they can run as you described above?  

In any event, it honestly looks like a single 240v can actualy handle my needs by itself as it stands. Two would be nicer for future proofing of course.

Yes, all GS inverters shipping now are with Rev. C boards.

And yes, it would be a firmware update to add the functionality.  Yes, I know I said that about Rev A.1 / B on parallel mode...it's still somewhat possible, but extremely difficult due to the very limited power monitoring available on said boards.  However, I did prove the hardware ability with the Rev A.1 / B boards...just not the fully controlled ideal end result possible with a Rev. C board setup.

Lack of software is more like it, I haven't rewritten the function for the new board yet 😉

As long as the outputs are separate, (i.e. separate load panels), you'd be just fine running 2 units.