Modina

There have been various suggestions of what you could do. I will concentrate on reducing your heating element radiated power. You could replace your 3KW element with a lower power element. Reading this forum, it seems 2KW elements are common. The disadvantage with a new element is that the geyser(s) need to be drained to fit the new element. Although I have spoken a lot of using capacitors to reduce the brightness of LEDs, I never thought of using a capacitor for high power applications. I have not checked @frivan suggestion of using 2x 70uF capacitors, meaning I have not made any calculations to see how much power your element would see. The calculation is not difficult, we know that a 3KW element has a resistance of about 17.75ohms and it is simple to calculate the impedance of a 140uF capacitor @ 50Hz. Then it is a simple voltage divider calculation. Another way to reduce the power going into a 3KW element is to reduce the input voltage with a mains transformer. You would connect the primary as usual, across the 230VAC and the secondary winding would be connected in SERIES with the element, in such a way that the transformer secondary is 180deg out of phase with the mains. A 40V transformer would reduce 230VAC to 190VAC and with this lower supply voltage the 3KW element would dissipate 2KW. The current would be reduced from 13A to about 10.7A. This would mean your transformer should be rated at about 450VA. By varying the output voltage of the transformer, you could set your power dissipation in your 3KW element to whatever value you choose. Now, a 450VA transformer would be a large lump of iron and it would be pretty pricey. Most likely over R1500. The alternative (for a DIYer) would be to get an old microwave transformer and remove the high voltage secondary, and then rewind with much thicker wire, but only a few turns. (They often use these transformers to make spot-welding machines). Power could also be reduced with phase control using a triac, but this would produce a lot of electric noise and a very bad power factor that your inverter would handle, but be a bit unhappy about. Another way to reduce the power would be a simple electronic circuit that removes a certain number of full sinewave periods, switched on/off at the zero-crossing point. The average power would go down, no electric noise and the power factor would be maintained at 1. However, you still have 3KW impulse loads. For you @Arzy there is one other solution but it also has certain disadvantages... Because you have TWO geysers, with (I presume) two identical 3KW elements, you could connect both in series. Each element would then dissipate 750W and your inverter would see a resistive load of 1.5KW. This would be an elegant solution, but it would treat both geysers the same. One would be able to refine this idea by adding relays that could switch, and thereby revert the one, or other, or even both geysers back to full 3KW operation. Then there is another cheap and easy way of reducing your power by 50%. Simply connect a suitable 20+ Amps 400V diode (mounted on a heatsink) in series with your element. This would result in half wave rectification and would reduce your 3KW element to dissipate 1.5KW. One could then also connect a relay across the diode, as @frivan did with his capacitor, to revert to full 3KW dissipation. Again, a diode would not cause electric (switching) noise and the load would remain resistive with a PF=1. I am not sure if pulsed DC through your element has negative consequences. I am not sure if galvanic corrosion could occur with current only flowing in one direction.

Inverters like all electronic gadgets use 100s of components and any one could fail. The symptoms can be extremely diverse and there are many failures that can go undetected for years. For instance, there are many safety and protection circuits that can fail and the end-user would be blissfully unaware. MOVs are an example of this. As they are called upon to perform their duty (absorb over-voltage spikes) so they deteriorate over time and finally go open circuit. One would expect that a certain model has a few weak spots and typical failures. This is correct, but there are often out of the ordinary failures that surprise repair technicians who know the model inside-out. Even firmware bugs can go unnoticed for years and then get ignored or the end-users accept to live with them. Manufactures too keen to convince the user to pay for the new "improved" model with even more bugs and suspect design. Equipment forever chases more bling, specmanship and questionable functionality that often is not needed and just over-complicates the design. There must be more manhours being placed in producing WiFi or Bluetooth enabled devices, with a seamless user interface, to allow an occasional voltage to be changed from the smart phone, than hours put into designing and qualifying the real hardware circuits.

A microwave uses a 50Hz conventional step-up transformer that takes your 230V to approx 2000VAC and then uses a diode and capacitor to double that voltage. Note that a magnetron cannot be power regulated. It is either on or off. So if you adjust your microwave to 60% power, an electromechanical timer or an electronic circuit will switch the magnetron on and off at a 60% duty cycle. Every few seconds you can hear this as the humming increases or decreases. On microwaves with electronic displays, the switching is done by a relay. So there should be no problem adding another relay, as long as it is rated for 15A. A microwave draws a lot of power, but doesn't really have substantial inrush current. (Note that EVERYTHING, including a small mains operated radio, has a certain inrush current, as capacitors charge, elements heat, or whatever. This is the reason why a lot of failures happen at switch-on.) So you can disconnect the microwave abruptly, if you don't run it at 100% it "disconnects" every minute or so anyway.

Yes, I had a look, this one is affordable and has a maximum input voltage of 138V. This would be perfect for those 2x 450W panels. https://shop.thesunpays.co.za/products/epever-xtra-solar-charge-controller-30a-mppt

@Scorp007 makes it sound easy. If funds are unlimited you can buy a Victron but you will be paying more than what you did for your inverter. You need to select based on max voltage, max current and max power. 1. Voltage. There are many MPPTs that are rated for 100V. Two 450W panels OCV (open circuit voltage) could come to a whisker of that rating, cutting it very thin. So either try to source panels with an OCV not more than about 45V or you will need to go to a 150V device as Scorp007 suggested. 2. Current. Your panels will produce 900W peak. To push 900W into a 25V battery would require 36A. So selecting a 30A charging current will be a good fit. 3. The MPPT max power should be able to handle the input power of your solar panels. I think other people would agree that you could use panels that are rated for slightly more power than the MPPT can handle, maybe 10%. The MPPT power rating is what it will pull from the panels. If the panels have a little bit excess power it shouldn't be a problem. Based on the above and assuming you use panels with not too high voltage, the minimum speced MPPT would be a unit like this: https://www.communica.co.za/products/sr-mc2430n10-mppt-controller?variant=31078585401417 Please note that I cannot say if this make is any good or not. Best would be to go to youtube and search for the product and listen to the reviews. Here is a page that lists mainly Victron MPPTs. As you can see, the prices of higher current/voltage units can get eye-watering expensive. I don't see a 30A unit such as Scorp007 suggested, but the 150V/35A Victron is about 5.5K https://www.sustainable.co.za/collections/mppt-charge-controllers

You either have the wrong hardware lurking under a false label or the guys loaded the wrong firmware. I suspect the latter. Whichever it is, this inverter needs to go back to the supplier.

That's crazy. Sounds as if someone possibly put the wrong sticker on a 1000/1200VA 12V model. What does the sticker on the side of the actual inverter say? All Axpert inverters have these stickers with the a summary of the main specifications. The 12V models are normally in a smaller enclosure, so maybe measure your Axpert size and then compare it with your 12V manual and a 24V manual. User manuals are normally reasonably easy to find and download. You need to contact your supplier. Seems this is a Takealot purchase. You can request a tax invoice and the supplier details should hopefully be on there. Keep the battery disconnected for now.

Build a thermal sand battery. You use solar panels to electrically heat sand. At night, the hot sand then releases the heat.

🤫🤫🤫 @Scorp007 don't give the game away.... the 8KVA guys will laugh us out of here if they read you have 5 "mickey-mouse" inverters, some running on 12V. We are certainly not part of the "in" crowd. 😆 The absolute minimum entry ticked is a 48V system. For me, having say, 2 inverters for different loads makes sense. One inverter can run 24/7 and the other(s) used only where and when needed. That is why I am investigating some form of load-sensing with automatic start. Anyway, I think one should learn to drive in a Uno first, before driving big trucks. Tomorrow's trucks will be better.

@Scorp007 Huh? You have a 24V inverter, you will need a very, very special battery balancer to use this one. Those still need to be invented. 😅😂

That would be a good fit for my 12VDC bus for lights & auxiliary circuits. 😀

Running in UPS mode will have no detrimental effects on inverter or battery. I am not sure why there is even this setting in the first place. There must be a reason for giving an APLiance mode option, but I can't think of one now. Noise on the power lines can look like a possible power loss event if the detection window is narrow because the instantaneous voltage of the sinewave is monitored and checked against the expected/theoretical voltage of the AC phase. In APL mode the inverter looks over a slightly wider window to decide if a possible power loss event has occurred. This wastes a few milliseconds of time. Perhaps someone else can give a reason for when one would rather want to operate in APL mode.

According to the Spec sheet your unit uses a PWM charger. This means the solar panel is connected via a MOSFET transistor directly to your battery. So you could damage the battery if you where to use a too big solar panel. My recommendation is to use a panel with: - nominal output voltage between about 14 to 18V - a maximum short-circuit current under 3A - a panel rated at no more than 50W Here are two panels that would work: https://www.takealot.com/30w-solar-panel-gd-30w/PLID92860657 https://www.takealot.com/30w-monocrystalline-solar-panel/PLID46870689 The battery is rated at 280Wh. A 30W panel will likely charge at the rate of about 11.5Vx1.65A, about 19Wh /hour. So expect over 12 hours to recharge an empty battery. The biggest panel you can (safely) use is a 50W like this one: https://www.takealot.com/solar-panel-50w-sosolar/PLID46868257 Charging rate will be approx 11.5Vx2.8A or 32W(h) per hour. This would give a recharge time of slightly over 8 hours.

Glad to hear that you can now listen to your music without stutter. Sad that your inverter whistles during the interludes.🙉

Here is another idea, not sure how practical: Your 20m cable I presume is an underground cable. Those cables are required to have an armoured steel sleeve. It might be possible to use the steel sleeve as an extra conductor to carry a low voltage/power (say 12V AC or DC) between it and N or GND. For instance, one could use a small 10VA 12V 50Hz transformer that always sits on the grid and powers a contactor remotely. Your supply cable might not have an extra unused conductor? Correction: Keep it AC. We know what the sparkies have to say about DC. DC could also lead to possible galvanic corrosion.

Ah, now I get it. One always assumes the main DB to be the origin of your grid power. Many moons ago people used the mains wiring as a communication bus. For instance intercoms and even computer networks up to about 5 mbps. Basically one adds a high frequency carrier onto the mains. That's how the geyser controllers worked where Eskom/Municipality would remotely switch off your geyser. You could use one of those old geyser contactors, I just don't know where you find the "sender" that would need to be mounted at the SunSynk. I am sure your SunSynk would block any Eskom/Muni control signal that might still be used.

I never just answer the question posed but try to understand the actual requirement that lead to the question. An old varsity lecturer taught me that and referred to it as " 'n bose ontwerp". From your problem statement I am confused why you need a wireless solution. You could create a non-essential circuit fed from a high-current contactor mounted in, or close to your DB. If Eskom goes on strike, the grid powered contactor would drop out, disconnecting the non-essentials. But that begs the question, why run the non-essentials through the SunSynk in the first place? Perhaps I misunderstand you. I doubt that you would find wireless relays that could handle the full 40+ amps that your non-essentials might need. Normally these devices have a 15 or 20A limit. You would then need to add yet another contactor capable of the required current.

Low frequency hum is always 50 and 100Hz and yes that has a habit to increase during high loads. But if the noise is described as a buzzing noise that would be much higher in the frequency spectrum. There are many circuits that could interact. I have seen many hi-res pictures of Axpert inverters and it doesn't seem as if any toroids (those things that look like mini donuts) are covered in any resin. Normally such inductors are at least covered with some tape, but all I see is raw enamelled copper wire. I feel that 10 or 14K is a lot of money for an inverter until I realize that even an entry level AV receiver will cost that and top models sell for 100K. Then we don't even talk about real HiFi which has become eye-watering expensive. All things considered, Axpert inverters are very good value for money. No wonder that some corners are cut in QA.

I agree that this is a bit odd, but note that an Axpert inverter has at least 2 high power converters and the high power inverter. Additionally there will be a handful of low power SMPSs to supply power to all internal circuits. Further, there will be a number of high power filters such as common-mode input filter, output chokes and likely another common-mode filter on the output. It could be any of these wound components. I mentioned that other AV receiver that blows fuses... Yours doesn't, but it drops out. So this is not a DC problem. Please first ensure that your setting <3> is set to UPS and not to APL. In UPS mode the transfer takes about 10mS while in appliance mode the transfer takes twice as long. My own Axpert uses APL as the default mode, chances are yours might too. My guess is that your AV receiver doesn't like the short blackout period when 230VAC drops away and the inverter is enabled. Even a 500mS blackout should be OK for the individual amplifiers because of large bulk-storage capacitors in the power supply. What I suspect is happening is this: Many AV and HiFi amplifiers have elaborate speaker protection circuitry. This consists of a power-on de-thumb protection, a switch-off de-thumb and a DC offset protection. At power-on and power-off state changes, the amplifier goes through an unstable phase which can cause loud thuds in the speakers. Not only unpleasant but no good for the speakers either. So in a stereo amp the speakers are switched through a relay. Not sure if relays are used in AV amps, because there are so many extra channels in AVs. Anyway, this protection circuit is rather quick as it wants to switch the speakers off before the bulk-storage capacitors in the PSU discharge. In other words, on the first indicator of a power blackout this circuit might trigger. At switch-on, or in this case when power is re-established, the circuit is designed to keep the speakers off for a number of seconds (normally 2 or 3s) , to allow the amplifier to stabilize it's various DC operating points. In a HiFi amp you would hear the relays click. As I said, not sure if AV amps use relays, they might use other circuitry such as FETs at their inputs or some other technology. If the above hunch is correct, this would not be damaging in any way. But it is very obstructive. There is no easy solution for the end-user. If it was my amp I would open it up and modify the protection circuit to not drop out that fast. Adding a small capacitor in the base drive of the transistor that drives the protection relay might do it. The problem is that AV amps are packed. If one is unlucky, it could be a huge mission to strip various boards to reach the protection circuitry. On HiFi amps this would likely be much easier because they generally have a lot of room and most use only a single main PCB. What is the model number of your Marantz? I think I might have some service manuals for AV amps. From the circuit diagram one would get a good idea what type of protection is implemented and if it would be easy to do a mod. The problems that people have with various systems misbehaving in some or other fashion when on inverters, never ceases to amaze me.

@BritishRacingGreen Only certain Axperts have the capability to run in parallel? Do all those that can, have a feature where they can be synchronised 120/240 deg-out-of-phase with the master, for 3-phase operation?

This is a rather complex subject. But in summary: 1. A 150W inverter can kill as easily as your 230V wall-socket. 2. A current flowing through your body is what kills, to create the current, you need a high enough voltage. 3. Touching across 12V battery terminals will also create a current (through your body), but it is super, super low so that you don't feel it and it is completely harmless. 4. To prevent electric shock, you want to have an earth-leakage unit and an earth. 5. Earthing is not ALWAYS used for protection. Certain equipment will work poorly when not earthed. For instance a "crystal" AM radio. 😄 6. Even if the equipment is FULLY protected (fitted with earth-leakage and grounded) you could still kill yourself if you where to touch L and N simultaneously, especially if done with left and right hands so that the current flows through your heart. In this case, the earth leakage unit would not even trip. 7. Here is a shocker (excuse the pun). if you have no earth leakage unit fitted, it might be a safer NOT to ground a trolley inverter or other small system such as you have. The reason for this has to do with the fact that a bird doesn't get shocked when sitting on a HV power line. To be specific, if a trolley inverter is standing on rubber wheels, it could be considered electrically floating - as if suspended in air by a bunch of halogen balloons 😄. Touching the L output terminal would then be as good as touching the Neutral output terminal. You would not feel a thing. AS LONG AS YOU DO NOT TOUCH L & N SIMULTANIOUSLY. However, if you plug in multiple devices, there is a danger that the loads could somehow find a way to electric ground.... and then touching L is a no-no. 8. An earth-leakage unit will only work in detecting a fault current. This means for the EL to work, the equipment must be earthed. 9. @cbrunsdonOver current protection has nothing to do with earth-leakage and user safety. 10. If you only want to watch TV and power a light, use a 12V LED and purchase a 12V powered TV (I have a 24" Sinotec, purchase many years ago, specifically for L/S) or buy a TV that runs from a laptop-like 19VDC power supply such as many 32" Samsungs. Then you need to get a small boost inverter (Communica sells them for about R 120). This will take your 12V battery to 19VDC. You will then have a low voltage system that is 100% electrically safe. Your cut out an inefficient inverter. Your battery will thank you and live much longer. Many inverters also have noisy fans, that too would fall away, you could then watch 7de Laan in silence. 😂😅

I have no idea where @frivan comes with his buck converter idea. You say it is the AV. So that would mean something like a Denon / Sony / Yamaha 7-channel or whatever unit? Each channel is most likely rated at a minimum of 50Wrms or so, meaning the AV receiver will have a few hundred watt power supply with plus and minus 40-50VDC supply lines and many other voltages as well. Your AV should have no problem at all. We had a discussion a few days ago about an AV constantly blowing it's fuse. This was on another Axpert inverter, not the Kodak. I can't find the link to that discussion now. Anyway, the speculation is that the inverter creates a DC voltage for a moment during startup. AVs mostly use 50Hz power transformers that would saturate when presented with DC. Your TV, lights, laptops, etc. all use switch-mode power supplies and would not be effected. I don't know at what frequency the various switch-mode power supplies of an Axpert run. It should be >20Khz, but even SunSynk was so dumb to run their switchers at 15Khz (apparently now increase to 20Khz with a firmware update). The buzzing noise you hear is normally from magnetic components such as chokes, inductors and transformers where the enameled copper wire turns are not properly covered with varnish, RTV (silicon) or potting resin and then vibrate at the switching frequency. Even if the switching frequency is >20Khz it might still make noises due to beat frequencies. This would occur if two smps run unsynchronized and at slightly different frequencies. For instance one runs at 20 and another at 21Khz. This could produce a beat frequency of 1 Khz (the difference). All the above problems (momentary DC, under 20Khz switching and beat frequencies) should NOT happen in a properly engineered product. But if a much more expensive SunSynk can't get things right, then there is not much hope for cheaper Axpert systems, especially clones. (Your Kodak is not a clone). The AV having a hickup is worrying. I also have an AV but never switch it on during L/S. These receivers most likely have a minimum consumption of maybe 70W at low volume settings... quite wasteful when you run on batteries. If the DC theory is correct, any electronic and even electro-mechanical device with 50Hz transformers and motors could give problems. Devices with small 50Hz transformers could be damaged. Larger transformers as used in AVs are less likely to get damaged, but are likely to blow fuses.

Perhaps you should introduce that household to an invention called a thermoflask. Boil once and have steaming hot water for 3 hours.

Thanks @Coulomb for that explanation. It's difficult keeping up with all the different models and versions. Given a certain model, say a 5KVA VMIII high voltage MPPT Axpert, how standard or different is the actual hardware between (1) say Kodak / Mecer / RCT and (2) a clone? It seems as if spare parts such as control PCBs for Kodaks seem reasonably easily available. (That's the impression I get). Say I have a clone that dies with a faulty controller card, what are the chances of giving the clone a heart, or is it a head, transplant? My guess is don't go there (on a clone). But possibly OK to do so between a Kodak / Mecer / RCT ? I have seen that at least some models have the MPPT on a separate PCB. So more to the point of this thread, would @Inverter Man be able to use a Kodak MPPT in his Mecer? (I have no idea how easy it is to get hold of Mecer spares.)

@DrewZA is correct. The only thing that might be confusing to people is that he drew the change-over switch as having, (from left to right) connections L1 N1 L2 N2 when the physical device has L and N grouped together. I.e. L1 L2 N1 N2. If someone would blindly follow the diagram he could get a very nasty surprise. @Shockin In your picture of the actual installation everything is correct, except that you didn't follow the SA convention of always feeding the power from the top. Question to people with a wiremans' licence: If one connects the feed from the bottom, would that pass a COC? I fell into the same trap a few years ago, not knowing any better. I think the arc suppression chamber inside the CB is apparently designed to work better with power feeding from a certain direction? Or is that bull? Earth leakage units also seem to be directional sensitive devices and these have an arrow showing the flow. I find it totally hilarious that each country uses separate color codes for L,N, etc. and follows different feed-from-top and feed-from-bottom conventions. Very confusing.