# Demonstration of Generator Support

• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
So what they're telling you there is that if you use a higher gen input than what the inverter's transformer is capable of balancing, the input current will override what the transformer can do for leg balancing and it won't work due to the excess current circulating in the center tapped transformer.

I'm not an expert in transformers, much less in generators. So, I may be wrong. That's how I understand this:

Say there's a load on L1, and no load on L2. Say the load is big enough to decrease voltage between N and L1. Then N-L2 winding will have higher voltage than N-L1. As a result, the N-L2 winding will be acting as a primary winding of a transformer, and N-L1 will be acting as a secondary, so that current through N-L2 will induce current in N-L1, which is then is consumed by the load on L1. This will continue until voltage on N-L2 increases to the same value as N-L1 (assuming windings are the same). As a result, the load on L1 will be supported by both the generator and the transformer, and N will be exactly between L1 amd L2. It does not automatically mean that there will be equal current in both L1 and L2 coming from generator, so there could be some help with balancing but far from 100%.

When the generator is weak, it is not capable of keeping N between L1 and L2. There may be a good amount of balancing in the process of moving N to mid-point between L1 and L2.

When the generator produces output with N being already in the middle between L1 and L2 (as Generac does), there will be no effect at all. That's exactly what I observed with Generac.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support

I think that's right, with the exception of your idea about the Generac's neutral "being in the middle". The neutral on a split phase generator will always be a winding center tap, providing a return path to the winding for unbalanced L1 or L2 loads, which then causes the neutral to carry current. With that neutral hooked to the transformer center tap you are powering one side of the transformer primary more than the other side.

So in the case of the XW nothing will basically happen until you attempt to exceed the max amps setting for the AC2 L1 or L2 input. At that point the inverter starts limiting current on the maxed out leg, forcing the transformer to begin balancing it to prevent leg overload on the generator. It's really a quite ingenious method.

I got intrigued with this morning and have been studying the generator support methods used in the XW vs the Radian. In all honesty, as impressive as the Radian is, the XW is just as impressive. And the XW has been around longer and is a pretty much proven platform with a very efficient charger.
--
Chris
• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
So in the case of the XW nothing will basically happen until you attempt to exceed the max amps setting for the AC2 L1 or L2 input. At that point the inverter starts limiting current on the maxed out leg, forcing the transformer to begin balancing it to prevent leg overload on the generator. It's really a quite ingenious method.

This doesn't work that way, although it would be nice if it did.

The charger acts as a 240V load. Say, for example, I have a charer going on at 23A. And I have a limit set to 32A. I can see 24-25A load on each leg (23A from charger and 1-2A my loads on each leg). Then my water pumps kicks in (6A 240V). Load on each leg goes to 30-31A. Everything's working just fine. However, if, instead of water pump, some 120V load comes up, such as cofeemaker (10A 120V). One leg would go to 34-35A and the other would stay at 24-25A. That would be fine for generator, I guess, because this is just a small imbalance and overall load is less than for the water pump. But XW dials down the charging to 19A to keep the high leg below 32A. The decrease affects both legs! So, now you get 30-31A and 20-21A. Still unbalanced, and also undersized.

But that's not all. Batteries react very slowly. So, after XW wants to do quick change to charging, it causes overshoot, and the charging current falls further than the entended level, say to 15A. XW tries to move it back. This causes stable oscillations between 15 and 19A with frequency of 0.2-0.3Hz. As Generac tries to adjust to that you can hear its motor frequency going back and forth. Bad for genrator and bad for batteries. Everything stabilizes when 120V load goes away. The only way to prevent this, is to avoid using large 120V loads when generator is charging.
ChrisOlson wrote: »
I got intrigued with this morning and have been studying the generator support methods used in the XW vs the Radian. In all honesty, as impressive as the Radian is, the XW is just as impressive. And the XW has been around longer and is a pretty much proven platform with a very efficient charger.

I think the XW hardware is good, but firmware is very buggy. For example, motor starting. When it's charging and I open a garage door, it's a short current spike. No doubs, if left alone, it wouldn't do much harm to the generator. However, XW sees the spike and dramatically dials down the charging. Once XW is done with that, the spike is gone. What Generac sees is that a huge load is suddenly removed. It cannot adjust quickly enough and XW simply drops it. Bad for generator. Very annoying. Very easy to fix in firmware.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
NorthGuy wrote: »
The charger acts as a 240V load. Say, for example, I have a charer going on at 23A. And I have a limit set to 32A. I can see 24-25A load on each leg (23A from charger and 1-2A my loads on each leg). Then my water pumps kicks in (6A 240V). Load on each leg goes to 30-31A. Everything's working just fine. However, if, instead of water pump, some 120V load comes up, such as cofeemaker (10A 120V). One leg would go to 34-35A and the other would stay at 24-25A. That would be fine for generator, I guess, because this is just a small imbalance and overall load is less than for the water pump. But XW dials down the charging to 19A to keep the high leg below 32A. The decrease affects both legs! So, now you get 30-31A and 20-21A. Still unbalanced, and also undersized.

There is no way for it to balance the loads, that I can see. Only an attempt to balance generator input during generator support mode (where the inverter would otherwise be running in overload).

In normal mode where the inverter is charging, plus running loads with pass-thru from the generator all it can do is limit charging amps to prevent leg overload. That's obvious. BUT - once you get to the other side of the curve where the inverter is supplementing the generator from batteries (generator support) then the charger is working in reverse and it can use GunSupp Plus to help prevent gen leg overload. But that's only up to the limit of the transformer capacity, which once beyond that you're going to get a circulating current between one leg of the generator and neutral, and one side of the transformer primary and neutral. That's why the 5 kW limitation - that's the transformer's capacity that it can balance at.
I think the XW hardware is good, but firmware is very buggy. For example, motor starting. When it's charging and I open a garage door, it's a short current spike. No doubs, if left alone, it wouldn't do much harm to the generator. However, XW sees the spike and dramatically dials down the charging. Once XW is done with that, the spike is gone. What Generac sees is that a huge load is suddenly removed. It cannot adjust quickly enough and XW simply drops it. Bad for generator. Very annoying. Very easy to fix in firmware.

This is also normal and it's not buggy firmware. The old SW Plus will do the same thing with our Champion generator on it. With the Honda it don't. It's due to governor over-shoot on varying loads - one thing the Generac Guardian is not designed to handle.

I cannot stress enough to people that there is no substitute for a real generator for off-grid power. And the Generac Guardian is not a real generator. It is a cheaper built unit, designed to compete at a price point in box stores. It is not a premium piece of equipment. And with an off-grid system the generator is probably one of the more important pieces of equipment you will buy. It is the only thing that works when wind and solar don't. And both wind and solar power are fickle - they work at Mother Nature's whim. A good generator works at your whim - every time.
--
Chris
• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
In normal mode where the inverter is charging, plus running loads with pass-thru from the generator all it can do is limit charging amps to prevent leg overload.

They can do something.

The simpliest thing thay could do is to limit sum of amps on L1 and L2, instead of L1 and L2 separately.

Second thing they could do is to recognize that generator can handle overloads and do not stop charging immediately. This would take care of current spikes.

Another thing they could do is to keep charging at steady level for longer periods to give time for batteries to adjust.

Finally, they could analyze input from the generator to actually see how it's doing. It's quite simple to see if the generator is sugging and the load needs to be decreased to keep it running.

These are simple things. Implementring all these is one day of work for a programmer. And would make for much better generator support.
ChrisOlson wrote: »
BUT - once you get to the other side of the curve where the inverter is supplementing the generator from batteries (generator support) then the charger is working in reverse and it can use GunSupp Plus to help prevent gen leg overload. But that's only up to the limit of the transformer capacity, which once beyond that you're going to get a circulating current between one leg of the generator and neutral, and one side of the transformer primary and neutral. That's why the 5 kW limitation - that's the transformer's capacity that it can balance at.

I agree that this is important. However, even if I turn on all my loads, it still will not be enough to overload an inverter or generator working alone. So, I didn't have an opportunity to test how it's working with XW.
ChrisOlson wrote: »
I cannot stress enough to people that there is no substitute for a real generator for off-grid power. And the Generac Guardian is not a real generator.

I agree with that too. Honda is robust because it work reliably with any loads or any inverters that you throw at it. It wouldn't be a problem for Honda to work with really bad inverters. That is what making it so good.

Similarly, XW would be robust if it could work with anything that you throw at it, with any generator, with any batteries, with any loads ...
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
NorthGuy wrote: »

The simpliest thing thay could do is to limit sum of amps on L1 and L2, instead of L1 and L2 separately.

Second thing they could do is to recognize that generator can handle overloads and do not stop charging immediately. This would take care of current spikes.

Another thing they could do is to keep charging at steady level for longer periods to give time for batteries to adjust.

Finally, they could analyze input from the generator to actually see how it's doing. It's quite simple to see if the generator is sugging and the load needs to be decreased to keep it running.

Well, yes, yes, yes and yes. To an extent. But I see some things that need to be considered too:

limit sum of amps on L1 and L2, instead of L1 and L2 separately
This is a good idea when you're dealing with grid power, which is virtually unlimited in the case of most home loads with a 200 amp service. It is not such a good idea when dealing with power sources that have limitations, such as inverters and standby generators. Leg overload is a real problem with limited power sources on split phase systems, and it either has to be monitored and dealt with or your power will go out due to power source overload.

Second thing they could do is to recognize that generator can handle overloads and do not stop charging immediately.
They already do this in both the XW and the older SW Plus. The problem is your Generac, not the inverter. While your Generac can indeed handle a surge, it cannot do it without severe voltage and frequency sag. It's inherent in that generator design. I wished you lived next door - I'd bring my Honda over and hook it up to your XW and show you the difference. The Honda can go from a 3.5 kVA constant load to 5 kVA overload, instantly, and drop less than .2 Hz and keep voltage within 1 volt of nominal for up to 15 seconds. With that kind of a generator on your inverter it has no need to cut charging amps to save the day and the generator is allowed to handle the surge.

Keep in mind that the job of a good inverter is to supply clean, stable power to your loads, no matter what. If you have a generator that cannot stay within spec a good inverter will react instantly to prevent the "dirty" power from being passed on to sensitive equipment that it may be powering. A cheaper, less capable inverter will simply pass it on and risk damage to your equipment. While it may be annoying, the inverter is only doing what it was designed to do - protect your equipment from a generator that can't do the job.

Another thing they could do is to keep charging at steady level for longer periods to give time for batteries to adjust.
I assume you refer to the de-rate factor when the internals of the charger start heating up. This is an issue, but I think kind of normal if you're pushing the charger right to its limits. Outback inverters do the same thing - I have a neighbor that has a VFX3524 and he pushes it right to the limit with a 7 kW generator and gets just irate at it because after a half hour the thing gets hot and starts to de-rate. He's even tried blowing a fan on it and that caused it overheat internally and shut down. No matter how big the charger is, it has limitations imposed by ohm's law and to date, nobody that I know of has found a way around ohm's law.

Finally, they could analyze input from the generator to actually see how it's doing.
Actually, they already do this too. That's why the inverter will spit your generator off within milliseconds when it can't recover from a surge and the governor over-shoots. Rather than damage your equipment with a power surge the inverter saves the day and tells a "dirty" generator to take a hike and come back when it can do the job.

I hope you're beginning to see why a stable, good generator is so important on an inverter/charger for off-grid duty. The normal residential standby units are not up to the job. BTDT. And it's not pretty when you're pushing it right to its limits. The inverter units that everybody talks about here are good and stable, but they're not that efficient at the full loads and surge loads that we're talking about here. Diesel generators are generally excellent because they are typically very high torque units with fast governors, inherent in diesel engines. When it comes to spark ignition, they are few and far between to find the good ones.
--
Chris
• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
Keep in mind that the job of a good inverter is to supply clean, stable power to your loads, no matter what. If you have a generator that cannot stay within spec a good inverter will react instantly to prevent the "dirty" power from being passed on to sensitive equipment that it may be powering. A cheaper, less capable inverter will simply pass it on and risk damage to your equipment. While it may be annoying, the inverter is only doing what it was designed to do - protect your equipment from a generator that can't do the job.

That's a very good point. Dropping generator is certainly better than killing computer.

The problem is that when the surge comes, Inverter tries to dial down charging current too quickly, which is too big a change for Generac. For some reason, Generac reacts much better to load increases than to drops. Generac copes somehow with the surge itself, but when Inverter quickly decreases charging current, Generac can't keep the frequency low enough.
ChrisOlson wrote: »
Finally, they could analyze input from the generator to actually see how it's doing.
Actually, they already do this too. That's why the inverter will spit your generator off within milliseconds when it can't recover from a surge and the governor over-shoots. Rather than damage your equipment with a power surge the inverter saves the day and tells a "dirty" generator to take a hike and come back when it can do the job.

This "do or die" Inverter's mentality is too harsh for the generators.

It may be a long time since generator starts dying under the load until it drops frequency/voltage out of useable range. May be few seconds, may be even 10 seconds. Inverter can see that right away. Why not to use this information to dial down the charging load a bit? Then keep it at this lower level until situation improves? That would be much better than relying on max amp setting supplied by the user. At least as an option. This way it would run freely most of the time, but then if some extra load comes up, it would be saved.
ChrisOlson wrote: »
I hope you're beginning to see why a stable, good generator is so important on an inverter/charger for off-grid duty. The normal residential standby units are not up to the job. BTDT. And it's not pretty when you're pushing it right to its limits. The inverter units that everybody talks about here are good and stable, but they're not that efficient at the full loads and surge loads that we're talking about here.

Actually, I'm now thinking that DC generator would be more suitable for my needs. It definitely wouldn't work for you because you have huge loads that inverter and generator can only run together. But, all my loads can be carried by inverter only. So, the DC generator would charge the batteries and inverter would be supplying power. The worst that can happen is that due to high loads charging current gets less than I wanted, but there will be no other inefficiencies, no changes in generator load level, no generator drops etc.

I even think that, perhaps, even now I can put a rectifier on each of Generac's legs followed by MPPT charger. It may make for better charging than using XW. This completely eliminates power quality issues, provide very stable load on the generator and let XW hadle varying loads. Could be more efficient too (95% for MPPT + 1% rectifier vs 89% for XW). Bad power factor may be an issue though.
• Solar Expert Posts: 3,029 ✭✭✭✭
Re: Demonstration of Generator Support

The following comments may not apply to any of the last few posts, BUT ...

Several years ago, was whining here about my pair of SW+ 5548 yanking the little Honda EU6500isa Inverter generator out of voltage regulation when in Absorb, usually at loads above about 80% of available rated output of the genset. Member -- RCinFLA chimed in to mention that as a voltage regulation strategy, the SW and SW+ inverters would dump current into the AC Grid terminals (where this genset was connected), and quite probably this would also be the same on the generator input terminals of the 5548s, and could quite probably cause this behavior.

Using a single Inverter as the charger helped, and dialing back the Charger AC Current setting would elenimate the problem, but would not make great utilization of the genset.

This behavior was ONLY in Absorb (and probably in EQ), where the voltage needs to be regulated. The Honda's output voltage would be dragged UP, until it exceeded the Max AC voltage out setting in the inverter, and would be dropped. Then the cycle would begin all over again.

I do not see this behavior when using the larger Diesel genset, connected to the Generator terminals. The Diesel is a much stiffer current source (and sink).

Just FWIW. NorthGuy, hope that you have good luck with the Generac. The common wisdon is that these and many other of the Home Stand-By gensets are not up to the long grind of charging batteries, AND that the Warranty on most gensets in this class have no coverage for Off-Grid use.

Looked at the Warranty statement of one of the large genset mfgs, and noticed their definition of the "Stand-By" classification as for those sites served by reliable Grid Power. Some of these warranties also seem to have a limitation of the number of hours/year of use. Realize that you, at this point, DO have grid power, so you may be OK if a Grid power service limitation is in the Generac Warranty.

EDIT: Regarding DC charging, you could do this ... would seem safer is there was a transformer twix the genset and the rectifiers. Also, you probably should have a capacitor bank at the output of the rectifier to reduce the effects of ripple current on the input caps on the MPPT CC. Without a transformer to reduce the resulting voltage out of the rectifiers, you would exceed the Vin rating of the XW SCC (if that was what you had in mind).

YMMV, Good Luck, Vic
Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH [email protected], 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
NorthGuy wrote: »
This "do or die" Inverter's mentality is too harsh for the generators.

Yessir. Absolutely. But I think you have to look at it from the inverter's standpoint too. It's dealing with stuff on a basis where sometimes things have to be corrected within 2 AC cycles (1/30th of a second). Most generators can't react that fast, but inverter electronics can. So here we got a good thing going under a surge load, then suddenly the generator is unloaded and the governor over-shoots. The inverter sees one cycle where the peak of the sine wave has gone from 120 to 125, the second cycle it sees at 130 - and it has approximately 1/100th of second to do something about it before the third cycle hits at who knows what and blows your computer right off the desktop. And it's going to "fix" it with a relay that takes probably another 15-16 ms to open the points and get rid of this problem.

It takes some impressive stuff to do all that.
It may be a long time since generator starts dying under the load until it drops frequency/voltage out of useable range. May be few seconds, may be even 10 seconds. Inverter can see that right away. Why not to use this information to dial down the charging load a bit?

Because in order to deal with very bad generators, the inverter would go into a yo-yo situation the way some generator's AVR's and governor control them. You have to look at this from the inverter's standpoint too. It expects a power source with a nominal voltage and freq. You give it a range that you say is "OK", you specify what you want for charger amps, and that's what the inverter has to work with. How is it to know what, within that range, is "good" or bad". It can't.

If it dialed back charger amps every time the voltage drops some (which it already does anyway due to Power Factor), then you're going to get mad at it and say, "Another thing they could do is to keep charging at steady level for longer periods to give time for batteries to adjust." When the real problem is that your generator has dropped from 240 nominal to 230 (or whatever). So the 30 amps @ 240 to get 7200 watts now turns into 31.3 amps @ 230 to get the same 7200 watts, and the extra 1.3 amps to maintain power kills you in I^2R losses in the generator winding, wiring and inverter transformer.

There just is no free lunch here. You have to reach the realization, sooner or later, is that what that inverter needs is a good stable generator feeding it. The reason residential standby units are the absolute worst choice is because an inverter/charger on off-grid duty is going to subject that generator to continuous loading, and wildly varying and surge loads that it will never see in normal residential standby service. I just haven't seen one yet used off-grid 24/7 that will last past 1,000-1,500 hours. And that's why Generac will void the warranty on it for off-grid use. It ain't designed for it.
Actually, I'm now thinking that DC generator would be more suitable for my needs.

Man, Kohler builds a really, really, really, REALLY nice DC genset for off-grid duty. And it runs on natural gas. View on that baby and drool 8)
http://www.kohlerpower.com/residential/detail.htm?sectionNumber=13561&categoryNumber=13061&prodnum=23364002
--
Chris
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
Vic wrote: »
the SW and SW+ inverters would dump current into the AC Grid terminals (where this genset was connected), and quite probably this would also be the same on the generator input terminals of the 5548s, and could quite probably cause this behavior.

Vic, I can verify that the SW Plus also dumps power into the AC2 terminals for the generator input as a voltage regulation method when the inverter is in absorb or float charge stages. Our Honda EM4000SX can act as a "sink" with no problem when it does this. Our little Champion 46538 backup generator has a problem with it and it kicks the breaker out on the generator sometimes.
--
Chris
• Solar Expert Posts: 3,123 ✭✭✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
Man, Kohler builds a really, really, really, REALLY nice DC genset for off-grid duty. And it runs on natural gas. View on that baby and drool 8)
http://www.kohlerpower.com/residential/detail.htm?sectionNumber=13561&categoryNumber=13061&prodnum=23364002
--
Chris

But it looks like somebody in their marketing group is a little unclear on some concepts:
```[B]Standby[/B] Fuel Consumption at
100% load:		 1.2 m³/hr (42 cfh)
75% load:		 1.0 m³/hr (35 cfh)
50% load:		 0.9 m³/hr (32 cfh)
25% load:		 0.7 m³/hr (27 cfh)
```

Or am I just not understanding what that means?
SMA SB 3000, old BP panels.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support

inetdog - I don't know what standby fuel consumption is. LOL!

As long as we're on the topic of gen support, inverters that work, and what it takes for a generator to make it work, I made a little video of how rock stable our Honda generator is, running at full rated load with the inverter supplying support to run the microwave (about 14 amps), one induction range surface "element" (about 12 amps), the clothes dryer (about 22 amps), the washing machine (ave ~8 amps), and other normal loads (~6 amps). So the total load on the system is around 62 amps in this video, or about 7.5 kW.

In one part of the video the microwave gets done cooking something and unloads the system by 14 amps, reducing the total load to around 48 amps. The load on the generator is 33 amps, the inverter is supplying the balance of the total load amps at all times, the batteries are in float and the wind turbines are all running supplying DC power to the bank at about 1.1 kW.

I would like to note here that we have never had 100% success with generator support with SW Plus inverter until we bought the Honda generator. We used to use the EcoGen with it and it worked. But we would get an occasional spit-off (for instance like when the microwave unloads the system by 14 amps in this video used to cause the EcoGen to get spit off). And sometimes we would get a spit-off due to low voltage from the generator. In the end I had to reduce the continuous load on the EcoGen to no more than 20 amps, or ~4.8 kW, to keep the voltage above 230 after the generator head got hot. The EcoGen's rated output was 25 amps, but it would not maintain voltage or frequency at that output.

I have never seen yet - not a single time - that the Honda EM4000SX has ever gotten spit off during gen support mode.
--
Chris
• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
Yessir. Absolutely. But I think you have to look at it from the inverter's standpoint too. It's dealing with stuff on a basis where sometimes things have to be corrected within 2 AC cycles (1/30th of a second). Most generators can't react that fast, but inverter electronics can. So here we got a good thing going under a surge load, then suddenly the generator is unloaded and the governor over-shoots. The inverter sees one cycle where the peak of the sine wave has gone from 120 to 125, the second cycle it sees at 130 - and it has approximately 1/100th of second to do something about it before the third cycle hits at who knows what and blows your computer right off the desktop. And it's going to "fix" it with a relay that takes probably another 15-16 ms to open the points and get rid of this problem.

Generators have flywheels. They cannot suddenly go to halt or speed up. Not with my loads, anyway. Inverter has plenty of time to react.
ChrisOlson wrote: »
Because in order to deal with very bad generators, the inverter would go into a yo-yo situation the way some generator's AVR's and governor control them. You have to look at this from the inverter's standpoint too. It expects a power source with a nominal voltage and freq. You give it a range that you say is "OK", you specify what you want for charger amps, and that's what the inverter has to work with. How is it to know what, within that range, is "good" or bad". It can't.

You tell it.

It goes something like that:

Every second (or half a second) do the following:

- monitor frequency. If it falls below 50, drop the generator
- if it is below 57, calculate overall load on the generator. If it is more than (X)kW, decrease the charging load, so that the overall load is less than (X-5%)kW
- if at least 60 seconds has passed since the load has been decreased on previous step, frequency is higher than 57, and load is less than (X)kW, increase charging load by 5-10%, but not to exceed (X)kW.

You let user specify advanced settings - frequencies, delays, speeds of increase, figure out the default, perhaps recommended settings for widespread generators, and this could be used to tune up to about anything.

You spent 2 hours of programming time, 10 days on testing with various generators, and you have a product that works fine with any generator. Radian, perhaps?
ChrisOlson wrote: »
There just is no free lunch here. You have to reach the realization, sooner or later, is that what that inverter needs is a good stable generator feeding it. The reason residential standby units are the absolute worst choice is because an inverter/charger on off-grid duty is going to subject that generator to continuous loading, and wildly varying and surge loads that it will never see in normal residential standby service. I just haven't seen one yet used off-grid 24/7 that will last past 1,000-1,500 hours. And that's why Generac will void the warranty on it for off-grid use. It ain't designed for it.

It is certainly not the best generator and may not last long (although I hope for 10 years at about 150hrs/yr, may be too optimistic). However, when I bought it, I put it through good tests. I run huge loads, such as electric stove, which switches 3kW up and down several times a minute, different motors. It took everything very well. After what people say about Generac, I was really surprised.

My "off-grid" loads are much less demanding that this - they're generally small and mostly soft starting, except one - XW, which also works very well.

Problems start when XW thinks that there are too much loads and starts dialing back charging current to "help" generator. It does this so clumsily that it causes problems. I cannot tell that it doesn't work. It does, just not as good as I would want it to.

Would Honda do better with the clamsy XW treatment? Possibly. However, why develop firmware to work with the best generators only while you can easily support nearly everything?
ChrisOlson wrote: »
Man, Kohler builds a really, really, really, REALLY nice DC genset for off-grid duty. And it runs on natural gas. View on that baby and drool 8)
http://www.kohlerpower.com/residential/detail.htm?sectionNumber=13561&categoryNumber=13061&prodnum=23364002

Thank you Chris. That looks really good, although a little bit small. I looked at the Kohlers two years ago when I bought Generac, and this thing wasn't there
Vic wrote: »
would seem safer is there was a transformer twix the genset and the rectifiers

Yes, you do need transformers. Or, you could go without transformer with full 240V, but it's hard to find an MPPT that would handle 350V input and 100A output.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
NorthGuy wrote: »
Problems start when XW thinks that there are too much loads and starts dialing back charging current to "help" generator. It does this so clumsily that it causes problems. I cannot tell that it doesn't work. It does, just not as good as I would want it to.

I know what you're talking about, and frankly it doesn't sound any different than our SW Plus. I have to pre-start the generator if I want to use my welder in the shop because the inverter by itself can't supply the surge needed to run a Lincoln 225 amp arc welder at high welding amps.

When the generator is manually started, the inverter goes right to battery charging. So I go to the shop and strike an arc with that welder. If the shop door is open I can hear the generator go to full iAVR overload power, and at the same time the inverter "panics" and drops the charger like a hot potato. I guess that's one way to test the integrity of your system to see what it's made of.
--
Chris
• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
I know what you're talking about, and frankly it doesn't sound any different than our SW Plus.

Yes, looks like they're almost the same. Mine XW is actually working very well except some small glitches. I put it through lots of testing and I found very little wrong. I'm glad I didn't buy Magnum PAE, which was my first idea.

Do you need to use a separate AGS unit to start the generator, or everything is built-in in SW+?
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
NorthGuy wrote: »
Do you need to use a separate AGS unit to start the generator, or everything is built-in in SW+?

The SW+ has an external GSM:

It's a little different than the AGS on the XW though. The AGS has internal programming and logic in it, interfaced and programmed with the SCP. The SW+ has a built-in Inverter Control Module (plus we have a remote one inside the house) and all the programming is saved in non-volatile memory in the inverter itself. The GSM is just a "dumb" box with three relays in it and it hooks to the SW+'s GSM port with an ethernet cable.
--
Chris
• Solar Expert Posts: 3,029 ✭✭✭✭
Re: Demonstration of Generator Support

Chris,

My main reference to a transformer was for safety. One could use a transformer with individual secondary windings on each half of the genset's split phase 240 VAC output, FW rectify each of these -- about 160-ish VDC and run each side of this 160 VDC into its own Classic 200, or similar.

FWIW, Vic
Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH [email protected], 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
• Solar Expert Posts: 1,925 ✭✭
Re: Demonstration of Generator Support
Vic wrote: »
My main reference to a transformer was for safety. One could use a transformer with individual secondary windings on each half of the genset's split phase 240 VAC output, FW rectify each of these -- about 160-ish VDC and run each side of this 160 VDC into its own Classic 200, or similar.

If you rectify each leg separately with common bridge rectifiers, you will receive two DC sources. It'll be 120V AC between the negatives of these two DC sources. If I'm not mistaken, MPPTs have a common ground and the negative of input is connected to negative of output. When you connect each of the DC sources to MPPTs, you will short their negatives. But you cannot do that because there's 120AC between them. I figured that all out when you told me about transformers.:D
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
NorthGuy wrote: »
If you rectify each leg separately with common bridge rectifiers, you will receive two DC sources. It'll be 120V AC between the negatives of these two DC sources.

No, it won't. The DC side of a full-wave bridge is completely isolated from the AC side. If you rectify the legs of a 240V split phase generator separately you'll get 165V DC power from each leg.

The problem with this, however, is that it will have way way too much ripple to feed the DC into a Classic controller without causing damage to the input caps in the controller. Rectified three-phase power is very clean with little ripple. Rectified single phase is too rough.
--
Chris
• Solar Expert Posts: 600 ✭✭
Re: Demonstration of Generator Support

FWIW, when I got home this evening, I did some more testing of my GVFX inverters gen support functioning, including Chris's test of overloading the inverter then turning on the generator to see if it kicks in and takes over some of the load. It passed all tests with flying colors. Video below. Apologies for the shakes and poor lighting - it was done with one hand holding my iphone and one hand pushing buttons, etc. Notes on the video further below.

A couple of points. I show the settings on the Mate and the mate status meter but mostly I show the Wattplot software status bar on my netbook because this is has more detail and is clearer than the Mate screen IMO. The top numbers under FX-1 are from left to right: the battery voltage, inverting amps, "buy" amps which is amps coming from the AC source (in this case the generator), the AC voltage in and out, and the load amps. The second line under DC-1 is the battery voltage, state of charge, battery temp, shunt measured amps to or from battery - red means battery is discharging.

I start out with a load of about 28 amps at 120VAC. The generator is supplying about 9 amps and the inverter the rest. Then i turn on a space heater and the amps increase to 39 amps at 120 VAC. Then I turn off the generator at the transfer switch. The inverter carries the entire 39 amp load for a while then I turn the transfer switch back and after a pause of 20 seconds or so the generator input kicks back in to help out the inverter. I'm guessing the pause is to give the generator a chance to warm up if I was using an auto start system.

From there I go outside to generator and my utility AC disconnect which is turned off. Back in to show the display again and turn off the generator input again. I repeated all this several times and it worked every time.

Again, apologies for the poor quality of the video - it was just done on the fly and I've never tried to do one of these before.

So - based on my experience - the GVFX3648 does generator support just fine with a Honda EU2000i. I'm still confused by why Chris is being told differently. I haven't had a chance to call Outback yet since their phone hours are during my work hours. I sent an email to their director of Engineering, Tim Shirley but I haven't heard back yet - If I do, i'll pass on what he says. I'm mostly interested in hearing about the history of this function in Outback inverters and why the inconsistent documentation and messages from some phone support staff.

I can only speak from my personal experience - in that it works on my system and has every time i've tried it. Obviously other inverters have more tolerance of different generators and may have aspects of gen support they do differently and/or better. But clearly the GVFX inverters can do gen support.
• Solar Expert Posts: 3,029 ✭✭✭✭
Re: Demonstration of Generator Support

Chris, YES, regarding the Ripple Current, and THAT is why in a previous post I mentioned that the rectifiers need their own Cap bank to minimize the ripple current working the CC's input caps too hard. FWIW. Vic
Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH [email protected], 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
• Solar Expert Posts: 53 ✭✭✭✭
Re: Demonstration of Generator Support

mtdoc, is that EU2000i remote/electric start? If so, I'm wondering if I can kit out my EU20i in that way when I eventually upgrade to a larger system.
• Solar Expert Posts: 600 ✭✭
Re: Demonstration of Generator Support
newl wrote: »
mtdoc, is that EU2000i remote/electric start? If so, I'm wondering if I can kit out my EU20i in that way when I eventually upgrade to a larger system.

Nope, I've never needed that or wanted it. The only time I need to run the generator is if there is an extended period of no sun during a power outage. Easy to start the eu2000 by hand in those cases.

Though not common for the eu2000s, I believe you can set it up with a remote start. This guy makes a kit. Pretty expensive though... Probably better to use an eu3000 if you really need to remote start.
• Solar Expert Posts: 2,490 ✭✭✭
Re: Demonstration of Generator Support
mtdoc wrote: »

So - based on my experience - the GVFX3648 does generator support just fine with a Honda EU2000i. I'm still confused by why Chris is being told differently. I haven't had a chance to call Outback yet since their phone hours are during my work hours. I sent an email to their director of Engineering, Tim Shirley but I haven't heard back yet - If I do, i'll pass on what he says. I'm mostly interested in hearing about the history of this function in Outback inverters and why the inconsistent documentation and messages from some phone support staff.

I can only speak from my personal experience - in that it works on my system and has every time i've tried it. Obviously other inverters have more tolerance of different generators and may have aspects of gen support they do differently and/or better. But clearly the GVFX inverters can do gen support.
Thank you for the Video, it does show the same thing I found with the GFX. As with you, I found it to be a seamless transition to and from the Generator Support thats important with the Honda EU2000, if it surges or faults it's pretty much useless.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
mtdoc wrote: »
FWIW, when I got home this evening, I did some more testing of my GVFX inverters gen support functioning, including Chris's test of overloading the inverter then turning on the generator to see if it kicks in and takes over some of the load. It passed all tests with flying colors

Pretty cool - now we just have to find out why Outback keeps insisting it's not recommended

It could be to do with the fact that when a manufacturer markets a product with a feature, and the feature don't work with a wide range of equipment, they end up with support issues because people start calling up, complaining that it don't work. Even the venerable SW/SW Plus, which is pretty much the Gold Standard that everything since has been measured against, can have problems on gen support with some generators. And I believe the SW-series is about as forgiving as they come.

It's interesting that Magnum is coming out with an inverter that does gen support. It seems like a simple thing to do. But it's been in testing for over year and they still haven't released it.

There's one other thing I wonder about in your video? Let's say you have a heavy load on and are using gen support in your GVFX. You have really good incoming RE power so the bank is floating, but despite that the inverter can't carry the load without gen support. Can the inverter try to "sell" power back to the generator if the "sell" feature is enabled? Or does it disable that during gen support?

The SW Plus, if the batteries are floating, will "sink" power into the gen head during gen support to control charge stage voltage. That will blow an inverter generator offline when it does that. A conventional wound field gen head will handle it as long as it doesn't see a "short" on the output breaker of the generator.
--
Chris
• Banned Posts: 17,615 ✭✭
Re: Demonstration of Generator Support

My guess is Outback is not satisfied with the performance of their inverters under gen support operation. As such even though the feature is there and does function they don't give "official" support for it. Which would make you wonder what bogeyman is waiting in the wings. :roll:
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
My guess is Outback is not satisfied with the performance of their inverters under gen support operation. As such even though the feature is there and does function they don't give "official" support for it. Which would make you wonder what bogeyman is waiting in the wings. :roll:

Although they do officially support it on the GFX model. And, of course on their flagship unit - the Radian. Without one of Outback's engineers coming here to take the time to explain it, I think it will remain a mystery.

If I had the money and time I'd love to buy a Radian and a XW6048 and compare them side by side, put them thru their paces and see which one comes out on top and gets the title of Off-Grid Inverter Of The Century
--
Chris
• Solar Expert Posts: 2,490 ✭✭✭
Re: Demonstration of Generator Support
My guess is Outback is not satisfied with the performance of their inverters under gen support operation. As such even though the feature is there and does function they don't give "official" support for it. Which would make you wonder what bogeyman is waiting in the wings. :roll:
Time will tell I guess, by the end of the summer it will have, either worked or given up the ghost. One thing about the GFX 1312 being small it will be using Generator Support all the time when the Generator is hooked to it. I cracked it open last night to do some modifications to the cooling and it didn't look any different from the FX's I have seen. They were 48 V so hard to tell if differences are because of voltage or capacity. Like the 48 V has 6 capacitors, GFX 12 V has 3, transformer looked the same. Anyway, I hope it works.
• Solar Expert Posts: 600 ✭✭
Re: Demonstration of Generator Support
ChrisOlson wrote: »
It could be to do with the fact that when a manufacturer markets a product with a feature, and the feature don't work with a wide range of equipment, they end up with support issues because people start calling up, complaining that it don't work. Even the venerable SW/SW Plus, which is pretty much the Gold Standard that everything since has been measured against, can have problems on gen support with some generators. And I believe the SW-series is about as forgiving as they come.

I'll bet that's it - or a least a big part of the reason. I'll bet the fact that they now sell the \$Radian\$ which also has this function has something to do with it as well. The thing is the Radian is larger than most people need IMHO. Most people who are interested in RE - whether off grid or hybrid system - do not run all electric households. Some like you do (and do it well!) - but most don't have the kind of large RE resource (like you do) needed to create a well balanced system with that inverter.

I'll bet there's a whole lot of people out there who could make good use of gen support on a medium sized system and who don't need or want to spend the money on the size of system needed to balance with a Radian. Yes the smaller XW is an option -it appears to be an excellent (though more expensive) inverter - but then you have to live with the buggy software and that stellar Schneider Electric support.:roll: If Magnum gets it right - that market (at least for those full time off grid) will be theirs..
There's one other thing I wonder about in your video? Let's say you have a heavy load on and are using gen support in your GVFX. You have really good incoming RE power so the bank is floating, but despite that the inverter can't carry the load without gen support. Can the inverter try to "sell" power back to the generator if the "sell" feature is enabled? Or does it disable that during gen support?

That's a good question. I've read speculation that if selling was not turned off it could try to backfeed the generator and cause problems. But I read - I believe an old post by boB - who said the inverters anti-islanding feature should prevent that. Also - of all the searches I've done on this topic on the Outback forum, I've never seen anyone say that has happened to them. And the thing is - if that was a risk - it could easily be eliminated by a firmware fix... Finally if that is risk - it would also be a risk anytime a generator was used and all of their recent brochures clearly give examples and even make reccomendations on using a generator with their grid interactive inverters. For example from their latest "Grid Interactive Solutions" brochure.:

So, I dunno?
My guess is Outback is not satisfied with the performance of their inverters under gen support operation. As such even though the feature is there and does function they don't give "official" support for it. Which would make you wonder what bogeyman is waiting in the wings.

Maybe. I just can't think of what that would be. There is nothing about gen support that is sending larger than normal amounts of voltage or current into the inverter - so damaging it seems unlikely. I don't claim to be an electronics expert, but my experience is that when something doesn't work it shows itself pretty quickly with with performance glitches or wafts of escaping magic smoke and dead equipment. Also, if that's the case why is it an option in the firmware to do it at all?

This is why some detailed info from an Outback engineer in the know would help. (paging boB again...).

Oh, well, if I get no reply to my email, I'll make some more efforts to dig around eventually.

BTW - Blackcherry - I'm sure you realize this but one thing I've found is that if you forget to change the "AC transfer control" to "GEN" then the inverter will try to do gen support but eventually the generator gets overloaded, it's voltage output drops so the inverter drops it out, takes on all the load for a bit then goes back to transferring some back the generator and the cycle starts again. This causes a lot of surging of the generator as it's load is increased, it drops out, back on, etc. Interestingly the inverter continues to ensure the loads are supplied just fine and it never lets the generator overload to the point that it shuts down. If you aren't looking at a display of the amp distribution you might not even realize what was happening - except for the sound of the generator surging. Gotta get those settings right... as long as they are it's been rock solid for me.
• Banned Posts: 1,807 ✭✭
Re: Demonstration of Generator Support
mtdoc wrote: »
That's a good question. I've read speculation that if selling was not turned off it could try to backfeed the generator and cause problems. But I read - I believe an old post by boB - who said the inverters anti-islanding feature should prevent that.

According to the manual for the GFX, what I read, it disables the "sell" feature when the AC input is programmed for gen. That's why I wondered what it does in the GVFX.

I agree that the Radian is pretty big. But if you look at it realistically, for somebody who lives off-grid and wants to live fairly comfortably, they're going to have more money than sense. If they had sense they wouldn't be living off-grid because living off-grid is expensive. So from that standpoint is the GS8048 is a perfect match. It's not for the guy that wants small and cheap, and to get by with as little as possible. It's for the premier off-grid system.

Our host has the Radian for \$4150 bucks. I got one in my cart. I just haven't clicked on "checkout" yet
--
Chris