Using the Generator AC input of the OutBack Radian GS8048

Options
swmspam
swmspam Solar Expert Posts: 57 ✭✭
in Grid Tie and Grid Interactive Systems #1
The OutBack Radian GS8048 has a Generator AC input (240 split-phase only).

I assume the generator AC input is rectified and down-converted to the 48VDC bus voltage. Then the DC power can be directed to either battery charging or feeding the inverters.

This leaves the important question: is the Generator AC input power factor corrected (PFC) for the rectification and down-conversion to the 48VDC bus voltage?

This strongly impacts the usage of the Generator AC input. A non-PFC input would require a expensive heavy-duty generator head (4 pole preferred) because battery charging and conversion to DC typically has lousy power factor, which can overheat or damage lighter generator alternators. A PFC input would allow light-duty (cheaper) generator heads.
· Share on Twitter

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #2
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    Where did this rumour about Outback chargers having poor PF get started? It isn't so in my experience (and the Radian is basically four FX's stacked internally).

    When charging, the inverter runs as charger and the loads are passed-through to the gen. When load-sharing with the gen, there is no charging going on and the inverter syncs its output to the generator, essentially 'stacking to the gen'. If anything PF would be even less of an issue on a Radian than an FX because they do not do load sharing.
    · Share on Twitter
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    #3
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    Cariboocoot wrote: »
    Where did this rumour about Outback chargers having poor PF get started? It isn't so in my experience (and the Radian is basically four FX's stacked internally).

    My understanding is that the OB chargers have about .96 power factor when they receive a true sine wave. The pf can be much lower if the AC is from a generator with a poor waveform.

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #4
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    Well 0.96 certainly isn't poor. :D
    I wonder how the PF corrected Magnums do if fed 'bad power' from a gen with high THD, low Voltage, drifting frequency, et cetera. My guess is they don't fair that well under those conditions either: garbage in, garbage out.

    Bill has, in another thread, quoted a 0.67 PF for chargers but I don't know what ones those are. Possibly standard battery chargers or power-converter types.

    Since I run the small gen with the FX and can put full load on it (13-14 Amps AC) without failure (high elevation too) I'd say the Outbacks are not bad in PF on the charger at all. But then the inverter-gen puts out a very stable and clean waveform.
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,439 admin
    #5
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    The 0.67 is for Iota (as I remember) and is very common for non-PFC type supplies and chargers...

    It is possible for the Outback to have good PF near 100% of rated output, and very poor PF at light charging current (are or below 50% of rated output).

    Anything that is >= 0.95 is considered to be Power Factor Corrected.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #6
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    I'm focused on the Generator-to-Battery path. Batteries are reactive loads, which can lead to generator overheating if uncompensated. That's one reason why generators designed for battery charging are overbuilt (expensive): they need to cope with poor power factors from reactive loading. If the battery charging circuit in the Radian corrects for power factor, then the generator will see an "easier" load when battery charging. This substantially reduces the specifications (cost) of the generator.
    · Share on Twitter
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    #7
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    The Outback inverter poor PF myth was recently discussed in THIS thread..


    While the Outback GVFX and VFX inverters may not be power factor corrected per se, in real world tests their power factor is very good - often in the high 90s. At least that has been my experience and others.

    See THIS thread in the Outback forum to see one persons measurements of VFX charger power factor under different loads.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #8
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    I'm focused on the Generator-to-Battery path. Batteries are reactive loads, which can lead to generator overheating if uncompensated. That's one reason why generators designed for battery charging are overbuilt (expensive): they need to cope with poor power factors from reactive loading. If the battery charging circuit in the Radian corrects for power factor, then the generator will see an "easier" load when battery charging. This substantially reduces the specifications (cost) of the generator.

    The only way that would happen is if the charging demand greatly exceeded the generator's capacity, which is extremely unlikely if you size the gen correctly for the application.
    Whereas it is true that running a generator very lightly loaded isn't good, it is also true that running one heavily or over loaded is not good.

    I doubt Honda designed mine for "battery charging" and yet it works fine. This is because the full demand from the charger is only 5 Amps AC, and the load current is limited to the balance available.
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,439 admin
    #9
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    I'm focused on the Generator-to-Battery path. Batteries are reactive loads,

    Just to be accurate with terms (not that I am any paragon of virtue in that).

    Yes, batteries are "reactive loads" in the sense that Capacitors are reactive--(had to look it up--I am so used to talking about inductors and an reactive current--I forgot/or never knew that capacitors were also reactive :blush:).

    However, we do not connect AC Generators directly to batteries--But through some sort of device (aka, battery charger).

    It is the battery charger that determines what the AC loading looks like.

    Typically, a simple battery charger uses a full wave rectifier to convert the AC into DC for battery charging (through a transformer, or the front end of a switching power supply).

    The diodes are charging a local capacitor (or the battery itself)--The diode does not pass any current until the AC voltage exceeds the DC storage element's voltage--Then the current changes from zero to very high peak current... Then as the sine wave drops back down, the current flow cuts off again to zero for the middle of the AC sine wave (through zero crossing, back to negative peak).

    Because voltage or current wave form energy can be looked as the square of the current/voltage (P=I^2*R=V^2*R) the "average" is done using Root Mean Square as opposed to a simple average... So a high current peak (say 3x "average") really is 3^2=9x wasted heat... As well as 3x higher peak magnetic field (saturation of transformers/magnetics in generator).

    Here is a beautiful picture showing the relationship between voltage and current for a rectifier/capacitor front end (from here):


    Attachment not found.

    Note, from my comment in another thread, it is possible for a Non-PFC battery charger to have good PF at "full/rated" output--Because the current is drawn for more of the voltage sine wave cycle.

    As the output current falls, the current "turns off earlier" in the AC cycle--Which reduces the PF to poor numbers.

    Now--That low PF at low charging current is not near as bad as bad PF at full load (bad PF at very low power, is still a relatively small VA loading of the genset).
    which can lead to generator overheating if uncompensated.

    Inductive loads can be compensated by adding capacitors to the circuit (usually near the inductive load--such as a motor, etc.). You can add the currents from the inductor plus the capacitor and get near 1.0 PF (current in phase with the voltage--Note, for motors you usually do not want to compensate to exactly 1.0 PF--There is something where the motor can turn into a generator with the capacitors and cause strange current/power flow during power failures/turning motor off).
    That's one reason why generators designed for battery charging are overbuilt (expensive): they need to cope with poor power factors from reactive loading.

    In general, the "cheap generators" are designed to run only relatively few hours per year and at much less than 100% rated power...

    Prime Mover rated gensets are designed to run for 2,00-8,000 hours a year at >50% to near 100% of rated output during that whole time.

    Charging a Battery bank that was well discharged--allows a person to size a battery charger to use near 100% of the generator's output for hours at a time--The inexpensive generators are just not designed for this.

    And yes, poor PF will add to the I2R heating of the windings/magnetics/etc.--But not doing any "useful" work. So--as part of "conservation", if you can find a PFC supply--it will help reduce load on generator's motor and electrical at the generator head.
    If the battery charging circuit in the Radian corrects for power factor, then the generator will see an "easier" load when battery charging. This substantially reduces the specifications (cost) of the generator.

    Yes and no--The "motor" part of the generator (gasoline, diesel, propane, etc.) does not really "see" the poor Power Factor. It is the Alternator Head that does that. So, good PF does reduce circulating currents and peak I2R heating losses, which is good for the alternator head.

    The other factor is the generator "loading"... With a typical home, the power usage may range from under 100 watts to over 2-4 kW over the day (standby loads, refrigerator, microwave for a few minutes, perhaps an AC system running at 50% duty cycle, etc.)--So the average load on the generator's motor is relatively low.

    With a Battery Charger--Discharge your battery to 20% SOC and through it on the genset with a "large" battery charger (near generator's max continuous power rating)--It is very easy to run that genset for 8 hours or more at >80% of rated output (battery is a fixed load able to absorb power for many hours). That is what causes (in my humble opinion) the non-prime mover rated gensets heart aches--And why some of the major genset mfg. do not warranty there generators for off grid/solar backup applications).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    #10
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    Cariboocoot wrote: »
    I wonder how the PF corrected Magnums do if fed 'bad power' from a gen with high THD, low Voltage, drifting frequency, et cetera. My guess is they don't fair that well under those conditions either: garbage in, garbage out.

    I haven't a clue about what type of pf correction Magnum has. It is possible that they have some sort of 'active power factor correction' where circuitry tunes the pf correction to the conditions.
    mtdoc wrote:
    The Outback inverter poor PF myth was recently discussed in THIS thread
    Thanks, mtdoc, for reminding me. I was too lazy to spend the time searching:blush:

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
    · Share on Twitter
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    #11
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    Batteries are reactive loads, which can lead to generator overheating if uncompensated.

    Wow, :roll::roll::roll:

    Abstractly, if subjected to a bidirectional low impedance bipolar AC source around its terminal voltage. But it does not follow the ELI the ICE man rules of reactance so within that definitions they cannot be called a reactive load.
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #12
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    Wow

    Hurumph. As previously pointed out, the battery's reactive impedance is seen at the AC source through the step-down transformer and rectifier bridge. Even worse, the diodes only conduct part of the time, resulting in nonlinear current draw in respect to voltage. PFC converters bypass this technicality by using switch-mode high frequency PWM that force current conduction during the entire AC sine wave. This isolates the battery's DC characteristics from the AC source.

    OK, most of that was unnecessary. But it is still important because the PF of the aforementioned process is about 0.6, which means the alternator current is (1/0.6) = 1.7x. If the PFC is employed, the PF as seen by the alternator is approximately 1. This has nothing to do with engine size.

    The question still remains. How does the Radian work?

    When using the Generator AC Input of the Radian, does the generator power go directly to the 48VDC bus, which then powers the output inverters? I've looked through the Radian manual, and there is no clear schematic, mech drawing, flow chart, or other truth table to illustrate how power is routed in various situations. Ideally, I'd like the generator power to go directly to the 48VDC bus so it can feed the output inverters. The manual suggests (although I haven't decoded all the logic) that the generator AC input is routed to the AC output in pass-through mode. Ugh! This seems silly. Why have this nice expensive split-phase inverter if all it does is "pass-through" the generator power? If this is true, how are the batteries charged? What performs the AC-to-DC step down to the batteries? Using the Radian's inverters backwards? Seems like a waste of good inverters.

    It also seems unsophisticated to "pass through" the generator AC. Consider this sequence of events:

    1. Generator off. Inverter powers loads from 48VDC bus from battery.
    2. Battery approaches low SOC.
    3. Generator starts.
    4. Radian switches from battery-to-inverter mode to generator-pass-through mode.
    5. Switchover causes glitch when the first sine wave (from battery-inverter mode) is exchanged for the new sine wave (from generator)
    6. The balance of generator power not going to loads is taken by the batteries through [how?].

    Great, now my computers crashed, along with whatever else connected that are sensitive to glitches. Kindof defeats the purpose, eh?

    Here's my desired scenario:

    1. Generator off. Inverter powers loads from 48VDC bus from battery.
    2. Battery approaches low SOC.
    3. Generator starts.
    4. All generator power is routed to 48VDC bus through a PFC downconverter.
    5. Inverter continues to power the loads from the 48VDC bus without interruption.
    6. The balance of generator power not going to the inverter is taken by the batteries straight from the 48VDC bus.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #13
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    Thought I explained this but it might have been in a different thread.
    When an inverter-charger has an AC In and is charge mode the inverter does not operate: the loads are connected to the AC IN and the inverter operates as a charger. The transfer is utterly unnoticeable and this method of operation is far more efficient than running the unit as a continuous conversion UPS
    As has been discuss in other threads and proven by those of us who have them these inverter-chargers do not have a low PF at all.
    If the generator can handle the initial charge stage the rest is easy, because as the battery charges its current demand decreases.

    Your concerns in this matter may seem logical, but they do not exist in the real world.
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #14
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    Cariboocoot wrote: »
    Your concerns in this matter may seem logical, but they do not exist in the real world.

    Thanks for the concise response.

    That seems like a leap of faith, especially considering the system cost. My estimated cost for the electricals (battery and inverter) is about $10k, and that's without the generator. This accounts for about $4.5k for the inverter, $3-4k for the batteries, and another $2k for electrical work and junction boxes. I'm presently having my electrical panel reconstructed. The contractor who originally installed the panel did so very tightly and sloppily. I also installed an eMonitor to track consumption of every circuit. So I'm already about $2k sunk with just electrical work and eMonitoring. Maybe the generator is a boondoggle; the money may be better spent on a solar installation. I'll be ordering the Radian when I'm happy with my wiring. I'm totally sold on NAW&S at this point. Forums did a great sales job.
    · Share on Twitter
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    #15
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    this is hard for some to wrap their heads around. when you feed ac into a inverter/charger it will charge the batteries, but the external ac source is also going straight to the loads. if you look at the spec sheet for the radian,
    http://lib.store.yahoo.net/lib/wind-sun/Radian-specifications.pdf
    going more than halfway down the second page you will see 2 specs. the first says ac input current max---50amps at 240vac and just below that it says battery charger input max---30amps at 240vac.

    that means 50a x 240vac = 12000w to the inverter. from there up to 30a at 240vac (7200w) can go to charging the batteries through the inverter circuits. no inverting is going on here as that same circuit when reverse fed ac power will charge the batteries and not drain anything from them. there is a difference of 20a at 240vac (4800w) that can go to the loads so this is essentially the ac loads in parallel with a battery charger. if the loads are larger than 4800w then it will reduce the charge current to the batteries.

    you may have been thinking that the ac gets fed to the inverter to charge the batteries and then inverted to go to the loads. this is not the case as the charger circuit is the inverter circuit in reverse and both functions of inverting and charging cannot happen at the same time and they shouldn't as that would be rather inefficient to go from ac to dc and back to ac.
    · Share on Twitter
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    #16
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    Yeah, a bidirectional inverter/charger is pretty much what has been available for many years now, except for maybe some very expensive UPS systems and
    satisfy 95%+ of the customers.

    AFAIK, the small APC and similar UPS's you buy at consumer outlets also have a finite transfer time (a few ms) and those are only designed to run for maybe 10 minutes.

    If these inverter/chargers had a separate PFC charger (or any separate charger) inside, the price would go WAY up and they would be heavier
    and larger than they are already. So, the inverter/chargers are designed to have a very fast transfer time for both transitions.


    Another feature that some of these inverter/chargers have, and I ~think~ the Radian has too is "Generator Support"... This allows
    the inverter and batteries to supplement the generator when the generator isn't large enough to supply the AC loads.
    Believe it or not, that is a much more desired feature than the true UPS function.

    I used to have a desktop computer about 15 years ago that would not stay up no matter how short the transfer time was...
    I'm sure it would have taken a true UPS like we're talking about to stay on. They must have had some sort of reset circuit
    is all I could figure... I haven't run into another computer like that since, BTW. Of course, neither would a laptop have
    that problem, luckily.

    If you want a charger/inverter with PFC, the Magnum Energy inverters are that way, but they are bidirectional too.

    boB
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #17
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    niel wrote: »
    you may have been thinking that the ac gets fed to the inverter to charge the batteries and then inverted to go to the loads. this is not the case as the charger circuit is the inverter circuit in reverse and both functions of inverting and charging cannot happen at the same time and they shouldn't as that would be rather inefficient to go from ac to dc and back to ac.

    It may have been you that also mentioned the term "dual conversion UPS" ... correct. I calculate my existing generator (EF2000iS Yamaha) operates anywhere between 10%-15% efficiency compared to the theoretical energy content of gasoline. Unless I can capture all that waste heat (cogeneration), it's a huge efficiency hit. Boy, it would be nice to get one of those Honda cogeneration units! I digress ...

    Continuing my thoughts: My average household load is ~2kW during winter evenings and the whole family is busy doing homework and other activities. During a blackout, we can congregate and consolidate power usage. During the day, it drops to ~800W average, mostly due to appliances (refrigerator) and computer servers. I run my business from those servers, they're important. They all run Core i3 CPUs on power-saving motherboards, excellent consumption numbers. So I hook a 6kW generator to my system to bulk the batteries. 2kW goes to the loads (average). 4kW goes to the batteries. I'm still taking a major downconversion efficiency penalty. The generator will be running steady-state 6kW, which makes an inverter generator lose its variable-speed benefit but adds another inverter into the stream?

    Guru master ChrisOlsen weighs in:
    ChrisOlson wrote: »
    Once you have your batteries up to Absorb again and your high loads done, you turn off the gen and go back to batteries.
    So using the above logic, I would want a large generator not a small inverter.

    I don't totally agree with this "old school" method because it's incredibly inefficient and requires you to schedule loads to take advantage of the gen run time.

    ChrisOlson proposes (I think) using a smaller inverter generator running all the time instead of a large generator running intermittently. Then battery charging is no longer the generator's primary goal; load support is. Rather appealing concept, although 24/7 noise control may be an issue (maybe turn off at night). ChrisOlson has also presented data that supports using a Honda inverter generator at a specified power is extremely reliable. He apparently has the hands-on experience to make such claims:
    ChrisOlson wrote: »
    We use our Honda for peak load management on our inverters and it will deliver 3.6 kVA for as long as you need it to - and maintain the voltage dead on 240 and the freq on 60 Hz. That's a prime power rating example on a genset that would otherwise be rated at 4,000 watts (at sea level).

    So, the obvious "best choice" that is emerging is a Honda or perhaps Yamaha inverter generator. Being portable, they can be used for all sorts of other emergency purposes.

    However, try this proposal: use a DC generator straight to the 48VDC bus. This is also where solar panels put their power! With a DC generator, there's no battery charger, the loads see the clean power of the Radian's inverters, and DC generators can operate variable power just like an inverter generator would. The problem is availability of DC generators. Kohler and Alpha both have very nice models but expensive (industrial telecom quality). Or I could have a generator maker configure me one (i.e. call Winco).
    · Share on Twitter
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    #18
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    Boy, it would be nice to get one of those Honda cogeneration units! I digress ...

    I didn't know honda made cogeneration systems. Are they sold in the US?

    Here's a very unsafe (IMO) honda cogeneration unit: http://forum.solar-electric.com/showthread.php?15396

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
    · Share on Twitter
  • BB.
    BB. Super Moderators, Administrators Posts: 33,439 admin
    #19
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    A standard generator (non-inverter/generator) is more efficient at heavy loads vs the inverter/generators--So with a known load like a battery bank to charge--You can do pretty well with a standard genset for recharging (at least up to around 80-90% of capacity, when the battery current starts to taper off).

    Heat Harvester (Freewatt Cogeneration System)

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
    · Share on Twitter
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    #20
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    The spec sheet is a little lean. The 50 amp input AC current limit is based on the AC input connect relay current handling capability. The op manual states the relay is capable of handling 55 amps but since there is realistically only 50 amp or 60 amp AC breakers available it is limited to 50 amp AC panel breaker.

    As a side related comment. The Xantrex/Trace SW series inverters spec'd 60 amp pass through. The AC relay was a three pole SPST relay rated at 30 amps per contact current with all three contacts tied in parallel. I am a little (but not overly) surprised that U.L. let them get away with this. Three contacts tied in parallel doesn't really achieve a higher current relay, reliably. I am not sure how the XW series AC relay is done, but a true 60 amp contact relay is quite expensive compared to a 3PST 30 amp contact relay.

    It is hard to draw any conclusions about charging efficiency from any of the specs. The inverter IS operating during battery charging, just in reverse current flow. The efficiency of battery charging is roughly the same as forward AC sourcing efficiency.

    As the battery reaches full charge, and the charging current drops off, there is little AC power consumed in charging the battery. The inverter transformer is still PWM's and magnetizing inductance and losses of the transformer magnetic core becomes dominate. If you measure the PF at this low AC input current it will be dominated by transformer inductance so it will be very inductive with a poor PF. But again it it very low VA and is just about identical to operating the inverter forward AC output sourcing with little AC load. The transformer probably burns 4 or 5 watts (feel the core's warmth) just idling with no loads. The AC connect relay burns about 2 watts in its energizing coil. The rest of the inverter's idle power is running the controller board and in the gate drive to the power MOSFET switches doing the transformer PWM'g.
    · Share on Twitter
  • stephendv
    stephendv Solar Expert Posts: 1,571 ✭✭
    #21
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    However, try this proposal: use a DC generator straight to the 48VDC bus. This is also where solar panels put their power! With a DC generator, there's no battery charger, the loads see the clean power of the Radian's inverters, and DC generators can operate variable power just like an inverter generator would. The problem is availability of DC generators. Kohler and Alpha both have very nice models but expensive (industrial telecom quality). Or I could have a generator maker configure me one (i.e. call Winco).

    Other disadvantages of DC generators:
    - you lose the ability to use the "gen support" feature of the inverter/charger where you can sum the power output of gen + power output of inverter (as bob mentioned earlier)
    - you don't have redundancy of an AC power source: if the inverter/charger fails, you have no backup.
    - if your gen is portable, then carting it around a large piece of land to power the odd power tool is quite useful.
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #22
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    Another feature that some of these inverter/chargers have, and I ~think~ the Radian has too is "Generator Support"... This allows
    the inverter and batteries to supplement the generator when the generator isn't large enough to supply the AC loads.
    Believe it or not, that is a much more desired feature than the true UPS function.

    Please see if I have the modes of operation correct:

    During battery operation, the inverters are supporting loads. When the generator becomes active at the AC generator input, a momentary glitch occurs when the inverters disengage from the battery and the loads are switched over to the generator. The glitch is unsubstantial and generally does not interrupt connected loads, similar to how a computer UPS switches from line to battery.

    During generator operation, the Radian operates in "pass-through" mode to support loads. The battery gets the balance of what the loads aren't using. The inverters are charging the battery by operating in reverse. The batteries will change until reaching a stop condition as programmed on the controller. The maximum generator power is set using the controller. The Radian senses the AC input to ensure the combination of loads and battery charging doesn't exceed the programmed maximum. Am I straight so far?

    During generator-support operation, the loads exceed the programmed maximum for the generator. The inverters switch direction, synchronize waveforms with the generator, and use the battery to supplement the generator. Is there a sense time for the Radian to figure out the overload situation and take corrective action? During this time interval, does the generator burden the surge?

    If using a small generator, toggling between generator-only and generator-support could occur frequently as loads are applied and removed.
    · Share on Twitter
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    #23
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    By George I think he's got it! :D
    Even the part about the essential need to get the right size generator.

    I'll also say that if grid-tied the Radian can supply up to its maximum 8kW output for loads or selling back to the utility if the loads don't require the full output.

    If GT was feasible here, I'd have one of these in my house. But since there's NO GT incentive here and the electric is really cheap it just isn't practical for me.
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #24
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    Even the part about the essential need to get the right size generator.

    The conclusion is (1) too-small generator will toggle the system into "generator support" too frequently and may not have enough excess power to charge the batteries, and (2) too-large generator will finish charging the battery, throttle down, and operate in a fuel-inefficient reduced-power load-support mode.

    This makes a good argument for inverter generators. When the batteries become charged, the generator throttles down and does fuel-efficient load support. With the Radian, the generator input requires split-phase 240VAC, which mandates the Honda EU6500is or Yamaha EF6300is (the only ones that support 240V split-phase). Since I was looking in the 5kW category, this makes a lot of sense. I guess a smaller version would work with an autoformer (to create split-phase), but who wants to do that?

    The counterpoint for a too-big generator is simply turn if off once the batteries are charged, which may suffice until the next day. Chris paints this intermittent operation as "grossly inefficient", presumably because it accumulates losses as the power goes though the AC-DC downconversion, goes inside of the battery, comes back out of the battery, and through the DC-AC upconversion inverters. The accumulation of these losses may be bigger than the losses of an inverter generator operating continuously feeding loads directly. I'll scratch down some math and DTO (Data Trumps Opinion).
    · Share on Twitter
  • solarvic
    solarvic Solar Expert Posts: 1,071 ✭✭✭✭
    #25
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048

    Are you planing this system or do you already have the the OB radian inverter? The magnum MS-PAE4448 could work for you if it is big enough. It outputs 240 ac but generator imput can be 120 or 240 vac. The Yamaha or honda inverter generators you mentioned should work well with this inverter. :Dsolarvic:D
    · Share on Twitter
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    #26
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    Chris paints this intermittent operation as "grossly inefficient", presumably because it accumulates losses as the power goes though the AC-DC downconversion, goes inside of the battery, comes back out of the battery, and through the DC-AC upconversion inverters. The accumulation of these losses may be bigger than the losses of an inverter generator operating continuously feeding loads directly. I'll scratch down some math and DTO (Data Trumps Opinion).

    Except that there are not two separate paths. It is a bi-directional inverter/charger and uses the same FETs and transformer from input to output.
    (FX stands for Fets and Xformers as you all know :D )

    Also, on gen support, there is no "toggling" per se', but only pushing (for selling or supporting) and pulling (for charging) as it were... (my lingo)
    ...which means the inverter trys to either raise or lower the AC voltage it is making, therefore determining the direction of power flow.

    The inverter/charger when charging will "back off" the amount of charging current it taps from the generator so it doesn't blow the AC input breaker
    or exceed the internal relay contact and wiring ratings.

    If the AC loads exceed the capability of the AC pass-through, the inverter/charger usually does nothing except maybe blow its breaker although
    it ~could~ switch back to inverter operation and just current limit the inverter portion and of course that would turn off the inverter's AC output.

    boB
    · Share on Twitter
  • RCinFLA
    RCinFLA Solar Expert Posts: 1,484 ✭✭✭✭
    #27
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    swmspam wrote: »
    Please see if I have the modes of operation correct:

    When the generator becomes active at the AC generator input, a momentary glitch occurs when the inverters disengage from the battery and the loads are switched over to the generator. The glitch is unsubstantial and generally does not interrupt connected loads, similar to how a computer UPS switches from line to battery.
    .

    Inverter is never disconnected from ACout. It can stop PWM switching of battery side of transformer at which point there is just a large unloaded transformer across the ACout node. The ACin relay just parallels the connection to the inverter/ACout connection. It always goes through a syncronization process first whenever AC input is detected that gets the inverter controller the knowledge of the phase and voltage of the ACin, before the ACin relay closes. If you operate in standby, the sync is maintained ready to fire up the PWM switching with near correct voltage and phase to match AC input.

    You can have the inverter continually PWM switching in parallel or PWM switching deactive in a standby mode. Having it running in parallel allows a quicker pickup of loads if AC input fails. The inactive standby mode takes several cycles to activate and stablize its output. The difference in the two load pickup modes (time to pickup loads) may mean the difference between a computer resetting or a refrig compressor shutting down.

    I have my inverters, which are normally on grid, always operating and ready to pick up load quickly. Have been operating that way for about 15 years straight. It cost about 30 watts of continuous power consumption which is relatively small part compared to the battery float consumption of about 50 watts for my 800 AH battery bank. When grid drops out or glitches the inverter detects the overload as it momentarily tries to power the grid then immediately opens the AC input relay releasing from the grid. Inverter continues on, taking over the AC out load. Switchover process typically takes less then a half cycle of 60 Hz. (8 msecs). I have never had a computer reset or shutdown due to a momentary grid glitch.

    The PWM'g of the inverter transformer is the key to operation. Just a slight change in the PWM duty cycle causes the current to flow forward or backwards through the inverter. It is all a matter of matching the AC input voltage. If inverter voltage is just a little lower, then it is chargng. A little higher and it pushes AC current to ACout node. The software turns this basic operation into several modes. The current sensors are inputs for processor to control which way and how much the current flow. If you don't want to sell and battery is fully charged, with no ACout loads then PWM'g is adjusted to closely follow AC input voltage causing little flow in or out of inverter. If you have load shave enabled then it monitors AC input current and adjust the PWM's of inverter to push just enough to make the ACout load with the contribution limit set for ACin current max. It can do the same for charging, backing off of the charging current if the available AC input current is needed to support ACout load.
    · Share on Twitter
  • swmspam
    swmspam Solar Expert Posts: 57 ✭✭
    #28
    Options
    Re: Using the Generator AC input of the OutBack Radian GS8048
    boB wrote: »
    The inverter/charger when charging will "back off" the amount of charging current it taps from the generator so it doesn't blow the AC input breaker
    or exceed the internal relay contact and wiring ratings.

    I thought in this situation, the Radian knows the output capability of the generator from programming (i.e. 4500 Watts). If the loads exceed the generator capability, it stops charging the battery, synchronizes the inverter with the generator, and the inverter starts supplementing the generator's output until the load drops below 4.5kW. Then the inverter switches direction and resumes battery charging. That way, the generator is never overstressed and assumes a constant 4.5kW load (until the battery reaches its programmed threshold). If my generator is 4.5kW, the AC breaker or relay ratings never come into play. I use 4.5kW figuratively. My candidate generator is the Honda EU6500iS, which is probably good for around 4.5kW prime power (although I haven't read the "fine print" yet).
    Are you planing this system?

    I don't have the inverter yet. I'm still architecting the system. The other candidate inverter is the XW6048, which has a proprietary digital communications protocol to control the system. The primary requirement is the inverter is controllable via my own custom microcontroller with CAN or RS-485 bus. The Outback system is due for publication of an open-source protocol. The XW system requires hacking and reverse-engineering of the protocol. I have grid connectivity and Natural Gas. My primary goal is to understand the operating modes before I drop $10k or more to make sure it will work correctly for my needs. For the Radian, I am clear on most operation modes except for the generator input. This has large ramifications for my architecture. I'm already several thousands committed with electrical work and upgrading my electrical panel. I have deep understanding of the electrical grid, including being a prior chairman at both IEC and NEMA for grid-intertie energy storage that included FERC. I'm accustomed to specifying the functionality for devices. The Radian is an existing device that is already programmed, thereby requiring me to reverse-engineer the functionality and create work-arounds to make the device function as I need. I'm hoping for enough flexibility to do that. But hope is not a business plan. I don't spend $10k hoping a device will work as intended. I may offset my purchase until Outback publishes the communications protocol so I can study the messages before jumping in. Outback implies the comm protocol release soon, as in, a few months.
    · Share on Twitter

Categories