Drawing was reverse engineered by Ukrainian guy from what looks like Chinese clone of Voltronic design (EPEver, Must, etc.)
Last edited:
Multi-String High voltage MPPT with communication for large / multiple battery banks.
- Thread starter ksmithaz1
- Start date
zanydroid
Solar Wizard
Picture #1 - addressed in the system diagram posted above
Picture #2/#3 - quite likely a UL requirement.
IOW I don't think this is unique Blue Eurotrash magic.
zanydroid
Solar Wizard
Sure, I think that's fine. There are a few details that would be different, like maybe different HV rail voltage, and maybe needing two different pictures depending on how the split-phase is achieved. One for 2 stacked inverters, and another for single inverter into a neutral forming transformer.
Zwy
Emperor Of Solar
I'm not certain it needs remote trip.
This of course brings up the question of how Victron does a disconnect? FET based?
fmeili1
Solar Enthusiast
I was a bit unprecise with my question. You've mentioned "Battery is inherently isolated due to HF step up transformer unless batt- and PV- are tied together internally". Do you know if the batt- and PV- are tied together inside the EG4-6500EX?
Oh yes, forgot that Solark is split phase. Could be achieved by combining two 120V inverters side by side fed from center tapped HF transformer producing +/-200Vdc.
No idea, never seen inside of one before.
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 20,795
zanydroid
Solar Wizard
Another variable is how air gapped the AC in and AC out can be, and whether there are more internal CTs to help with higher end features. Like generator/grid assist and AC coupling. Some of those don’t matter for this thread but would matter for other discussions
Not sure if this Voltronic is something primitive like SUB/SBU architecture with no blending
fmeili1
Solar Enthusiast
Very interesting schema to better understand what's going on!
Is my assumption right, that if both relays RY1 and RY2 are closed (bypass mode) that this schema would be able to work both as an off-grid inverter (like the EG4-6500EX) and it could also be used as a grid-tied inverter and feed energy back to the grid (if the firmware/controller of the AIO allows this working mode)? From the schema it looks like that this may be possible?!
I know that one of these Voltronic clones (called Phocos PSW-H-6.5KW-120/48V and looks nearly identical to the EG4-6500EX) does have a setting to allow to feed energy back to the grid. It looks like all of these AIO's using the same circuit schema independent if off-grid or grid-tied or both.
Zwy
Emperor Of Solar
Then how would Victron trip for an event to shut down PV?
fmeili1
Solar Enthusiast
OK, I will open and disconnect all wires on one of my EG4-6500EX and will do a resistance measurement between batt- and PV1/2- to be really sure if they are isolated or not. I would sleep better if I know that a failed FED would not be able to destroy all BMS's and battery cells!
Maybe someone have a not yet installed EG4-6500EX on hand to do this resistance measurement to verify it?
Yes. The inverter has to synchronize to grid waveform in order to inject solar into AC loads with grid assist. The hardware to grid export is there but this feature is not enabled probably due to lack of software support or extra hardware required to comply with anti-islanding, I am not sure.
Last edited:
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 20,795
I don't know that victron would. SIC-40 has an output to do that.
To retrofit, you may want an over-voltage detector if LiFePO4 battery goes above a voltage representing 4V per cell or something like that. But watching whole battery may be too slow to prevent cooking off one cell. BMS control would see it, but its gate drive to FETs may get clobbered when FETs fail.
Oh, if voltage between BMS and SCC goes to something stupid high, that is what should trip it. BMS is supposed to see regulated charging voltage and it watches cells. Maybe 65V or 70V applied to BMS would be a good threshold. Possibly just a 5V coil relay in series with 65V Zener. When it closes, it signals remote-trip, or completes a circuit from battery through 1A breaker into 5 ohm resistor.
fmeili1
Solar Enthusiast
From the theory point of view an off-grid inverter is a voltage source which generates it's AC output frequency by itself (via PWM) and a grid-tied inverter is a current source which must follow the grid waveform (by pushing current into the grid by trying to raise the voltage a bit over the grid voltage). It's interesting, that just with a different PWM modulation and different controlling of the PWM sinewave H-bridge, both scenarios could be achieved with just one circuit.
zanydroid
Solar Wizard
I've heard this many times, but I don't think it's a healthy abstraction for all kinds of thinking.
If it was truly an ideal current source, it would immediately blow out YOLO attempts at AC coupling with incompatible hardware. But I don't think it has a 100% instant kill rate.
More likely, there is a lag in the PWM control loop when adjusting to maintain the target output power.
ksmithaz1
Solar / EV Junkie
Not to be a d*ck, but watts an milliamps are two different but related things.and noise
There's a lot of differences people glance over.. Self consumption of the EG4 is 25W compared to 15mA on the Victron. That's quite a bit of heat/inefficiency difference. Victron is also fully isolated.. along with other differences.. so you do in fact get a higher quality product so it isn't exactly a 1 to 1 comparison solely on price... whether those differences matter to you is up to you.
RCinFLA
Solar Wizard
- Joined
- Jun 21, 2020
- Messages
- 3,565
Actually it is my created diagram. It is a generic HF AIO inverter. It is cheapest models with no isolation to PV charge controller.
The PV inputs ride on top of the H-bridge PWM high voltage DC polarity switching causing AC being present on both PV lines relative to neutral of inverter, which is normally ground bonded. This is why you need to use dual pole DC breakers cutting both PV pos and neg lines.
Split phase HF inverters like Deye and SolArk HF inverters have a DC to DC converter between PV SCC and internal HV DC bus. It serves to allow up to 500 vdc PV input and supplies two isolated 250 vdc outputs with PV power sharing between the two 120vac series connected HF PWM H-bridge circuits in the unit to provide 240/120vac output.
Here is updated drawing showing optional DC to DC converter. EG6500 has extra DC to DC converter to allow 500 vdc PV inputs with a 250 vdc HVDC for 120vac inverter, but it does not provide isolation to PV inputs.
Just about all the low cost HF AIO inverter have PCB layout similar to this:
Diagram below is a low freq inverter like AIMS. LF AIO inverters PV charge controllers feed battery directly with their SCC output going directly to inverter battery terminals.
Last edited:
zanydroid
Solar Wizard
These all seem like they can have momentary or continuous parallel operation (if there is a firmware bug). A lot of POCO require interconnect agreement even for momentary.
Can the PWM transistors be used to disconnect high voltage bus from the grid AC to allow charging when on grid?
Can the PWM transistors be used to disconnect high voltage bus from the grid AC to allow charging when on grid?
RCinFLA
Solar Wizard
- Joined
- Jun 21, 2020
- Messages
- 3,565
The inverter AC side PWM transistors are the AC input synchronous rectifiers from AC input to supply HV DC to be used by battery to HVDC converter to charge batteries.
PV SCC also dumps its power output to HV DC bus.
On HF hybrid inverters, all power, with exception of AC input to AC output pass-through, flows through HV DC bus.
Similar threads
