Modina

It seems that I was wrong. A loose standing MPPT would use a buck regulator (taking high PV voltage to low battery voltage), but the circuit topology used in Axpert inverters is very different to what one would normally expect. Coulomb knows the hardware better than anyone, so they actually use a boost inverter. Well, sooner or later this energy must then be converted down with an even larger step-down buck regulator. At every conversion stage one has power losses... it all adds up. Have you downloaded the service manual from this forum's resources page? Don't expect full schematics or that your particular inverter is covered. But it does give you a general idea of the basic functional blocks and a very basic schematic of each block. Coulomb is right about SMD caps. The challenge is that these components have no markings. So without service manual one sits with a guessing game. Then one needs to analyse the immediate circuit around it, look at reference designs, data sheets, application notes etc.

There you go. The magic of wireless tech.

I am surprised at all the replies, suggesting/discussing rewiring. Look, even a small geyser is a heavy consumer and plugs normally sit on a 20A circuit breaker. So you don't have much spare to power some very hungry kitchen appliances. Your current setup is not ideal and adding an extra pair of cables is the best solution. Granted. Now I don't like WiFi ONE BIT. I keep mine OFF for 99% of the time, but I am a crazy old man. Ordinary folks have no problem at all with it. So why didn't anyone suggest an IOT Sonoff relay that is controlled wirelessly from a Rasberry PI or whatever other smart devices you guys play with. All you want to do is make sure the geyser is off during L/S. If you do it via the cloud, then at least Bill Gates can turn it off as well, if he feels that you have been a very naughty boy and used up all your carbon credits. I am sure Klaus Schwab would approve noddingly as well.

A MPPT charger is basically a SMPS Buck (step-down) converter that has input and output voltage and currents monitored and some processor then implements a max power protocol, by way of altering the PWM duty cycle. If you say the solar voltage is clamped down to battery voltage, it sounds as if your MOSFET in the MPPT buck regulator is fully switched on or has blown into a short circuit. If finger pressure releases the bus voltage it sounds not like a blown MOSFET but rather that it is driven permanently on by the driver circuit. Check for a copper track crack, or a component failure in the MOSFET gate driver circuit. Passive SMD devices such as caps and resistors can sometimes crack and go open circuit. Those cracks will be hard to see with the naked eye. Measure for approximate resistance over the gate current limiting resistor and pull-up resistor. Also check around the npn/pnp driver transistors. These should be small TO92 or SOT23 devices. If you can't find the fault with simple multimeter measurements and/or visual inspection, I would suggest maybe resoldering the above mentioned components. I would not recommend modifying by fitting extra components, except if you really understand the particular section of the circuit and the consequences of such a component. Note that although I am talking about the MOSFET in the singular, the actual circuit is likely to use 2 or more devices in parallel.

Fat fingers. 😆 For the record, my 2.5KW MCE is specified for <35W idle power. I measured self-consumption by removing all loads, then switching off inverter, switching off the grid, switching of Hubble, connecting my Fluke in line with the battery cable and then switching back on. Once the LCD backlight switched off and no fans running I then measured the actual DC current and multiplied this by the battery voltage. This gave 40W. So I got higher consumption than manufacturer specification. Must be the GSR/SCL (Skin Conductance Level) of my fat fingers in an excited state.😁 I would not trust the inverter's own internal measurements for low power levels. There is likely to be quantization error, measurement noise and DC offsets that could lead to measurement error.

@Bobster. is 100% correct with all his comments, but everyone overlooks inverter no-load consumption and overall efficiency. These are normally very closely linked, but you can get exceptions. @Scorp007 will roll his eyes skyward at my comment.... perhaps we should ask him to share the secret of his 20Watt no-load inverter consumption. A number of newbies on this very forum have reported their electricity costs to notably rise after having an inverter installed. The losses are substantial. For low power consumers like yourself, the no-load power consumption will be the important figure. For high power uses, those wanting to run kettles, microwaves and washing machines, the overall efficiency is the more important criteria. I tried to find technical specifications on your old Schneider inverter but couldn't find anything. You say you got 8 hours when keeping your loads under 300W. So I read that as 300W peak. Your nominal consumption might have been only about half of this. Your new inverter will likely have 60W no-load consumption. Yes, @Scorp007 . Possibly more. As long as there is utility power or you have large solar panels with sunshine, you don't need to worry too much about efficiency. But the moment the batteries kick in, I am hyper vigilant on any power wastage. So much so that I recommend people to rather replace certain consumers such as older TVs then purchasing expensive lithium batteries to be needlessly drained by inefficient old-tech. BTW, inverters are not the only inefficient device. Nothing is 100% efficient. Lead acid batteries are known to also waste a huge amount of power. Also, please do not look at your inverter's maximum efficiency of, say 93%, and then assume that a 100W load will draw 100/0.93 from the batteries. That figure is the MAXIMUM efficiency, the sweet spot, normally attained at about 80% of rated power. Running a 30W load on a 5.5KVA the efficiency is near zero. Think of it like a car idling at a red robot, it's fuel efficiency at that stage is infinite liters /100km. I have spend many hours thinking about the ideal system. I personally like the idea of using low voltage DC as the always on supply and then have an inverter only kick-in on demand. This is problematic for many people as this requires DIY and customized solutions. However, there is hope. The high power USB-C standard provides up to 48V 5A - that is 240W. I expect more and more devices to make use of this standard in future. The alternative would be a low power always-on inverter and a high power on-demand only, inverter.

I hate it if products come with these wishy-washy descriptions. I searched quite a while for HA-01 balancers and although there are many sites and even a lot of youtube videos nobody explicitly says that the HA is an active balancer. However, judging by the small size and the completely potted nature, it seems that this is indeed an active unit that transfers power as you indicated. Unlike you, @Scorp007 I am very concerned with how things work, to the point of opening up the box BEFORE even plugging the thing in. 😃 Sometimes I see such a mess, that I refuse to even try using the product. There is simply no limit as to how low some Chinese manufacturers can stoop. Youtuber BigClive never fails to amaze me with his findings of dangerous, mains operated junk. The price is NOT always a good indicator. You get very expensive junk and sometimes you get surprising cheap gold nuggets.

I think that most balancers are of the passive type, meaning they have a transistor that switches a resistor into the circuit to bleed away the extra charge. There are active battery balancers that actually transfer power in the way you mention. I would suspect that active battery chargers would be quite complex due to the bi-directionality, and thus expensive.

As long as your inverter is on, it will waste about 40W. On a 5KVA this will typically be 60W. This is enough reason to want to switch the inverter off during the night. Also, the fan noise that a lot of inverters produce can be a nuisance. Note that the inverter will still consume a few watts even when "off". Your previous 720W inverter was very basic and will likely just have had a relay contact drop out and switch your output to bypass. If you want to retain that function you could implement this with a separate switch-over relay.

How did you determine your balancing current? Did you connect a multimeter in series? If you do that, you should connect your meter in series with the common lead. The meter will have a finite resistance that will add voltage drop and can effect the voltage difference the balancer sees. Up to 10A? That's crazy. At least for a passive balancer that wastes the excess energy as heat. My "gut" tells me one would want to balance with a smallish current (say <500mA) over a longer time (for typical inverter sized batteries). I am no expert on batteries but I think the battery chemistry needs some time to settle. For instance, thermal deltas will not be the same throughout the structure.

Is that 2KVA inverter a 12V or 24V unit? I presume it is only an inverter and doesn't have a build-in charger. So you charge the batteries with that 12V or 24V SMPS power supply? Does that PSU have a settable current limit, or just a so-called over-current protection? And the charge voltage? These SMPSs have a voltage adjust preset pot, but they normally cannot be adjusted for a voltage high enough to charge a 12/24 volt battery. Did you change one of the resistors to allow for a higher output voltage? This sort of battery charger is not ideal, to say the least. It's no good for lead acids and worse for Lithiums. It's OK to use this to manually charge a battery once in a while, but not as a set-and-forget, automatic charger. You really need a microprocessor controlled charger to properly take care of whatever battery you use. Perhaps you added one of these little voltage window comparator PCBs that activate a relay to switch the SMPS on and off, but that is still far from ideal. It is a huge problem finding reasonable priced Lithium chargers. A Victron 12V10A charger is about 2.5K. When adding the cost of the inverter one has basically reached the price-point of an entry level Axpert.

@sweet rolls Why do you have a 8KVA inverter if your house load is so small? 😄 I don't have an off-the-shelf solution for you, just an idea for an entrepreneur.... Years ago cars normally used 65A alternators but I would expect that the alternators these days are likely to be 100A for medium sized cars due to all the extra electric loads. Assuming a car idles at 850rpm and uses a 1:3 alternator pully, the alternator will spin at over 2500rpm and this should allow for 80A at about 14V. That's 1120W. Yes, we need to allow some power for the car's electrics like the fuel-injection system, but 1KW auxiliary draw should be possible (at engine idle speed). One would need to design a boost converter to take the nominal 12-14V to 48-56V with a microcontroller implement circuit similar to a MPPT lithium charger. The unit would connect directly to your battery. A sensible, easy to reach, high current power connector would need to be fitted on the car somewhere. Not necessarily to be used every day, but as an emergency backup measure. It would be silent and fuel efficient. I doubt that it would use more fuel than the normally very crude, carburetor fed genie. I would not be surprised if the fuel consumption would be a fair bit better. Just an idea.

I am still learning.... The way I understand it, only the Axperts with removable LCD and the new round LCD displays can communicate with a battery's BMS. Is that correct? So for anyone running an older Axpert the above is a given. I initially presumed that the Hubbles are LiFePO4, but even that seems incorrect. The high bulk charge voltage is a hint to that. Further irritation is the requirement for the cloud interface that costs 3K just to know how the battery feels. So at the moment I am operating in the dark iro battery SOC, cycle counter, etc.

@P1000 You are 100% correct. Any 50Hz transformer would immediately saturate and blow any fuse. Any device with a SMPS would not be effected AT ALL. Just about all switch mode power supplies could be run from DC indefinitely (albeit at a slightly higher DC voltage to that of AC). A firmware bug could hold the inverter's H-bridge momentarily still, causing the output to go DC for a moment. The only way to confirm this would be to check the AC output with an Oscilloscope and this would require a lot of caution (the scope would need to be run from an isolation transformer or be battery operated and AC would need to be reduced with a suitable resistor voltage divider.) Over- and under-voltage will NOT cause a fuse to blow, at least not if the 230V varies by 30 or 40V.

Only Victron seems mad enough to make 3 and 5KVA inverters that run from 12V. The MUST 3KVA inverter runs from 24V. @skaiThe 30A is the recommended charging rate for the Hubble AM4. The maximum continues charging rate is given as 115A, so you could increase it to 50A without concern. Note that <17> Bulk Charging Voltage (C.V Voltage) = 29.2 is a bit high. According to Hubble this should be 29V. See the summary here: https://www.hubblelithium.co.za/axpert

Install a Geyserwise dual AC/DC element PV powered (standard) geyser. Once the house is signed off, disconnect the solar panels and connect them to your inverter. 🤣😂

1. I hope you at least tried to contact your bank's fraud centre and reported the scam. 2. It might be a good idea to name and shame the person or company on this forum to warn other people.

DC miniUPS boxes have lithium battery management system that is always on. However, the output of these devices can be turned on/off with a switch on their front panel. The UPS has an internal boost converter to step up the Lithium battery voltage to a well regulated 12VDC. This boost inverter will have a certain quiescent current when on. So it is better to turn it off when your lights are off. I always turn off my router over night with UPS switch. My guess is that you intent using a "proper" light switch? There would be no problem having the UPS on 24/7 and just placing an ordinary on/off switch in series. I do not quite know why you want to use strip lights for L/S and not ordinary MR16 12V LEDs. Here is the problem: Strip lights are normally cut to length and thus do not have a voltage control chip. They use simple voltage dropping resistors. These lights thus require a precise 12V. Under voltage will result in dimmed lights. Overvoltage will burn out the LEDS. MR16 LEDs always have an internal SMPS and can run from AC and DC. They should be more than happy with any DC voltage between 10... 16V. This means that you can use a lead acid battery with a corresponding charger. Doesn't matter how much your battery is discharged or charged, the fluctuating battery voltage will have no effect. You could also use a lithium battery but lithium chargers are expensive and hard to come by. With a little bit of DIY, the cheapest way would be to get a suitable 230AC 12VDC power supply, modify an internal resistor, to take the output voltage to about 13.8V (these supplies are adjustable, but the voltage is always shy of the 13.8V), then add a small high wattage, series resistor (a 5W 0.33 or 0.47 ohm might do it) and use this as a trickle charger on a suitable dimensioned SLA battery. Ideally you want the PSU to have a settable current limit, but such PSUs are as scares as hens teeth. So this is why one would place this small resistor on the output. It is a crude way of voltage/current regulation that fit lead acids characteristics reasonably well.

@RustyGate Did you have any luck yet with a repair? If a repair fails, I would say definitely replace with another Axpert. The many complaints regarding all sorts of issues on SunSynk inverters shows that expensive products do not buy peace of mind. What you could consider is maybe paying a bit more and buying a Kodak for the simple reason that spare parts seem to be generally available from a number of SA suppliers. I see Xenononline keeps a large selection of replacement PCBs and even Geewiz lists some. @BritishRacingGreen have you purchased replacement PCBs from any of these? Equipment should be repaired on the component level. However, with high power electronics, you sometimes have catastrophic failures that burn PCB tracks or burn holes into boards. Any repair will then forever be a butcher-job at best, making for unreliable system operation. In such instances PCB replacement is the only way to go. I am not sure how close the Axpert clones follow Voltronic designs and if there is any possibility to say, fit a Kodak replacement board in a Synapse,

I had a quick look. All I see is sale, sale, sale. Who in his right mind: 1. Keeps stock of so many, many different makes of inverters. A legit company will normally have resale agreements with only 3 or 4 suppliers. 2. Puts everything on sale during times where more and more people are panicking over ever worsening L/S. The reality is that real world distributors will have websites and pricelists listing sold-out, sold-out, out-of-stock, out-of-stock and here and there is still an item in stock, at a price about 20% higher than at the beginning of the year. I am just very suspicious of these super cheap, we-have-it-all sites.

You get a few different types of "ware" 1. Hardware 2. Software 3. Firmware 4. Vaporware Vaporware come in many flavours. I spend 2 weeks solid, trying to source reasonable priced hardware and that was in February. Since then things have only gotten worse. I am reasonably happy with my system but am keeping my eyes open for a well priced 3KW 24V or possibly 1KW 12V inverter for another site. I would even consider buying a blown-up unit if the price is right. So I am following the prices and offers. An Axpert 3KW for 6K is Vaporware. It is click-bate and most likely a fraud. Just this morning I had such a laugh at the energy-solarcity.co.za website which we discussed in "Has anyone used energy-solarcity.co.za?" Most online retailers seem to sell vaporware.

@cbrunsdon Who is the supplier that sells these inverters for 6K? What model Synapse Inverter is it? Is it the 3.0+? Battery trolleys can be build much cheaper. Very short battery cables, no trunking, often no proper circuit breakers. Very seldom do they fit earth leakage units. The units can be assembled in a central location. No CoC required. It is only the trolley itself that is an extra expense.

I can't figure out how he can do it for that price, but in that case, go for it.

I hate to blow your bubble, but those prices sound too good to be true. If you are really lucky you will be able to buy an inverter and a cheap 2.6KWh Lithium battery for 25K. Then it's likely that the inverter is only a PWM unit. No installation, no panels, no circuit breakers, fuses, cables, etc. No CoC. I brought an Axpert 2.5KW high power MPPT clone and a Hubble AM4 in February for R 7200 and R 20000 respectively for a diy installation. Since then the prices have increased SUBSTANCIALLY, and as indicated, you need quite a few extra bits & pieces.

No it is not even that. If you go and scroll to the bottom of the home-page and click on "payment policy" it takes you to a page called Terms & Conditions and then starts off at: Our Contract Lorem ipsum dolor sit amet, consectetur adipiscing elit. Morbi feugiat molestie ex quis finibus. Quisque tincidunt dolor ut convallis lobortis. Proin ullamcorper eros ac ultrices imperdiet. and it goes on and on in a language Google translate detects as Latin. And the google translate leads to more laughter. At least these guys are only trying to earn a few dishonest bucks. Most Western Governments publish equally pathetic dribble but with consequences much worse than "only" loosing 35K, or whatever.