Hi, this is my first post. Thanks for reading and I hope some of you can help.

Sometimes someone turns on the microwave, or kettle, or oven, or hair dryer, or some other appliance briefly for a short while.

When this happens the load, previously say 2000W, will increase to 3000W or 4000W vs solar power of maybe 2500W.

The inverter, Axpert MKS II 5kW, will then use grid electricity to provide ALL this power, even if the batteries are full and there is 2500W solar power, which is just shut down and wasted (I can't sell to the grid).

This is causing unnecessary climate change and pollution, not to mention throwing money away.

Once it switches over to grid power it will then persist with using grid power for everything for some minutes (I timed it a couple of times and it looks like it's on a clock for ten minutes) even if the load has fallen down to what can easily be handled by the solar panels. E.g. if someone turns on the microwave for 30 seconds grid power is used for everything for the next ten minutes.

At first I ignored it since the losses were small and it only happened about once a day but lately we are using more electricity in the day and it now happens constantly (5-10 times a day) and is starting to get annoying as well as probably pointlessly wasting probably between about 1KWH and 3kWH/day.

When the solar is 2500W and the load is 4000W I want the grid to provide only the difference of 1500W not the total of 4000W. I know it's possible to combine solar and grid because I've seen the inverter do it, but for some reason only when the batteries are disconnected. Is there a setting in the program that I can use to make it do this?

The setting 23, overload bypass, is set to BYD (bypass disable) but I also tried changing it to BYE (bypass enable) once and it made no difference. I couldn't figure out for sure whether this setting was relevant or not but I thought I'd try it. In fact, it was finding some threads on this topic that brought me to this forum.

Is there a good reason why the inverter won't combine solar and grid? Is there any downside to trying to make this happen?

I am a novice pretty much so please keep it simple. Thanks for your help.

Edited December 28, 20214 yr by Green Power

AFAIK the MKS 2 cannot mix grid and PV to power the load. The MKS 3 can. 

Do you have batteries connected? If so then the batteries should add the 1500W for the brief period up to the point that the battery back to grid level is used. 

With batteries connected when setting 16 is CSO and setting 1 is SOL or SBU the PV will be used and not wasted. 

I quick way is to see if you have a setting SUB under setting 1. This is needed in order to blend PV with grid for the 1500W needed for a short period. 

This is how I understand it. 

Thanks for helping out. Yes, the batteries are connected. The batteries will step in and add a few hundred watts or maybe a thousand watts if needed. For example, with a load of 2300W might see 2000W from panel and 300W from battery. But when the difference is larger such as a load of 4000W vs 2000W from panel the batteries don´t provide the difference. It just uses grid for everything.

The battery bank is connected. However it is fairly small (four 100Ah batteries). In some cases, that may explain why the batteries don't provide the difference because they may not be capable.

Setting 1 in SBU, setting 16 is OSO (only solar). I can try setting 16 to CSO later and see if that helps.

1 hour ago, Green Power said:

Thanks for helping out. Yes, the batteries are connected. The batteries will step in and add a few hundred watts or maybe a thousand watts if needed. For example, with a load of 2300W might see 2000W from panel and 300W from battery. But when the difference is larger such as a load of 4000W vs 2000W from panel the batteries don´t provide the difference. It just uses grid for everything.

The battery bank is connected. However it is fairly small (four 100Ah batteries). In some cases, that may explain why the batteries don't provide the difference because they may not be capable.

Setting 1 in SBU, setting 16 is OSO (only solar). I can try setting 16 to CSO later and see if that helps.

Now you can check if setting 12 back to grid is not too high. You need to check on the display left side what the battery volts are when you add say 1500W load and PV is contributing 2500W. This can be set lower at times of no LS. 

1500W for a short period is only 30A from the 48V batteries and not high provided the batteries are in good nick and charged. 

You can compare the battery voltage when you add say 300W to the PV and when you add a 1500W load to the PV. 

Setting 16 on CSO will help to keep the batteries full when no PV just for the test. 

 

Will changing to CSO stop it flicking over to grid/utility, and if so will this only be because the battery is fuller/voltage is higher? What I really want is to use less grid/utility electricity overall. Should CSO (rather than (OSO) achieve that? I don't want the grid/utility charging the battery after sunset so I'll have to check if it does that now I made that change.

I currently have setting 12 on 46V on setting 13 on 48V. Would it make sense to set 12 to 44V during the day time when I have had high loads - would that help it stop depleting the battery. And then change it back to 46V in the late afternoon to protect battery life? 

What do you mean by LS?

I'll probably need a few days or a few weeks to try out changes and see the result, there is no huge hurry here...

9 hours ago, Green Power said:

Will changing to CSO stop it flicking over to grid/utility, and if so will this only be because the battery is fuller/voltage is higher? What I really want is to use less grid/utility electricity overall. Should CSO (rather than (OSO) achieve that? I don't want the grid/utility charging the battery after sunset so I'll have to check if it does that now I made that change.

I currently have setting 12 on 46V on setting 13 on 48V. Would it make sense to set 12 to 44V during the day time when I have had high loads - would that help it stop depleting the battery. And then change it back to 46V in the late afternoon to protect battery life? 

What do you mean by LS?

I'll probably need a few days or a few weeks to try out changes and see the result, there is no huge hurry here...

LS=loadshedding. 

Why I mentioned CSO as you then just have to disconnect PV to charge the batteries once of before adding a 1500W load on top of the 2500W from PV at peak time during the day to see that it adds battery power to PV without going to grid. 

I would once of use setting 12 at 44V but not on an ongoing basis as it is not good for the battery. It is again just to see if you can prevent going to grid. It is just to confirm your inverter goes to grid due to low voltage. 

I am afraid things just got more complicated and confusing than ever. Yesterday (after my last post) I had someone in our gardening doing some soldering work and they kept causing power cuts but when this happened they just kept trying over and over again. I went out and suggested he stop but he insisted it worked fine in other houses and it was nearly finished so he kept trying and the power cut out several more times - either that circuit or the entire house. I then had a look at his tool and had an amperage dial that ranged between 30 amps and 200 amps and he had it set to about 50 or 60 amps I think. The circuits of my house are 10,16 and 25 I believe amps and the whole house is I think 32 amps, and the inverter can only deal with I think 5000W or 5500W. So I then insisted he stopped. However, after he left, we have had about 5 or 10 more power cuts that evening and today, all when we try to use something that uses a lot of electricity like a kettle. It wonder if the soldering tool may have damaged the inverter, the battery, or some other part of the system. It seems likely to me the use of that soldering tool did something as we have had as many power cuts in the last 20 hours as we had in the previous several months. At the moment we seem to have stable power as long as we don't plug in anything that uses 1000W+, if you do we sometimes get a power cut, sometimes it works but only on grid, and sometimes it works normally.

On the other hand we already were having power cuts occasionally before then, perhaps once a month or once a week, due to what I thought was an error on the inverter as they happened randomly at night. I found this old thread about that months ago which might be relevant: 

Most of the power cuts either give the error code 58 "Output voltage is too low", the code 8 "Bus voltage is too high" or, just now, BP "Battery is not connected".

That last one didn't make sense at first when I saw it last night as the pictures on the inverter show the battery being charged. However, now I think about it, it may make sense. That could explain why the grid electricity is being used too often. It would also explain why in the last few days I happened to notice on one day the battery had a low charge just after sunset when it would normally be high, and other day this week I noticed the house running off grid electricity as early as 9pm at night, when it would usually be battery. At the time I just assumed we had used more electricity in the day that usual, or due to clouds.

At this point it may make sense to fix the battery connection issue first, and then see what issues remain after that?

Re your last post, I just set it to 44V and plugged in the car and it went over to grid. It should have been possible to provide the electricity with solar + battery, even just solar alone without battery should almost certainly have covered it, and it still went over to grid. Before going over to grid there was a few seconds when it was trying to provide the energy without grid and showing error 4 "low battery" before giving up and going over to grid.

I have been disconnecting the battery for days at a time and I noticed that when I did this grid use in the day is reduced. I think when there are high loads the inverter goes to the battery and when the battery doesn´t respond it goes to grid.

I also noticed that when the battery is disconnected things work smoothly and there have been no power cuts so far.

I am fairly confident that grid and solar is being mixed and simultaneously used when (and ONLY when) the battery is turned off, as shown by the picture on the inverter. I also took a bunch of meter readings and compared to load and PV input and this seems to confirm this.

It looks like the battery may be in need of replacement. I thought it would last longer than 1.5 years but after using daily to provide power in the evening and some hundreds of cycles I suppose it´s plausible that the battery is pretty much dead. That seems to be the best explanation for now.

I had my friend who installed the system come over, and that is what he thinks as well.

I still have a small doubt...it could be the inverter after all that is the problem, but that looks less likely.

Edited January 15, 20224 yr by Green Power

Alway feel sorry for someone getting such a short life out of batteries as it is the most expensive item of a PV system. This being the main reason why I only cycled my batteries during LS and every few days to prevent them dying of old age.

Perhaps a good time to have the inverter checked for proper operation in different modes as well before new batteries are connected. You don't want to buy another set after 1.5 years.

I still wonder why your mode SUB vs SBU is not working as it should by using the full PV available and only mixing with grid when you exceed the PV available.

Edited January 16, 20224 yr by Scorp007

my Axpert MKS II 5kW doesnt mix grid/pv, but goes to the batteries directly if the pv cant keep up, highly noticeable with our rainy cloudy weather these days.

it appears to be a one or the other scenario, leading to unnecessary battery cycling.

the sunsynk 5kw is a good replacement inverter, you can sell yours to someone just starting out (making sure its working ok first), offsetting the cost of the upgrade.

I seem to be getting around a year or so out of my good used sla's, maybe its worth considering, if 1.5 years to 2 years (seems to be the ave with actual use) is all one gets out of new ones.

 

Edited January 16, 20224 yr by Nitrious

Sorry for the late reply.

On 2022/01/01 at 2:46 AM, Green Power said:

I had someone in our gardening doing some soldering work

I think you mean welding. There's a big difference in power (50 W versus possibly over 2000 W).

On 2022/01/01 at 2:46 AM, Green Power said:

I then had a look at his tool and had an amperage dial that ranged between 30 amps and 200 amps and he had it set to about 50 or 60 amps I think. The circuits of my house are 10,16 and 25 I believe amps and the whole house is I think 32 amps,

You can't directly compare those currents. Welders operate at around 28 VDC, after a transformer and rectifier. So 60 welding amps translates to less than 6 A at the utility (stepping down the voltage steps up the current). Call it 6 A with losses. That's about 220×6 = 1320 W, a moderate load. Note that 200 A welding current might translate to about 28 A at 220 V, or some 6 kW (depending on the voltage setting of the welder, efficiency, and many other factors). That's why welders often require a stronger than normal outlet. In Australia, they usually come with a 15 A plug, when most appliances have a 10 A plug. It will be different in South Africa. Although 15 A would not be nearly enough for 200 A @ 28 V.

Thanks for that explanation, in which case it could mean that the power cuts were due to my system´s problems rather than the cause of them.

I experienced similar grid priority problems with my system of similar configuration.

An electrician friend assessed the matter and came to the conclusion that the inverter is not a true hybrid (not able to supplement solar with grid power), and was cycling the batteries regularly to draw the difference, which led to premature battery failure. The inverter then eventually switched to grid more frequently, as the batteries had effectively been worn out.

That inverter is best used for off-grid application. The solution which was recommended for true hybrid operation was to replace the inverter first, since it was the cause of short battery life. Not an easy message to swallow, but rather than throwing money at the problem, I had to choose to invest in an effective solution.

LoodPyp, Did your system have lead acid batteries?

Would it be fair to say that with lithium batteries the better cycle life, and perhaps better ability to withstand heavier demands, might mean that this inverter might work Ok with lithium batteries?

 

35 minutes ago, Green Power said:

LoodPyp, Did your system have lead acid batteries?

Would it be fair to say that with lithium batteries the better cycle life, and perhaps better ability to withstand heavier demands, might mean that this inverter might work Ok with lithium batteries?

 

We are getting back to my post in December where I had it only the MKS iii does blend PV and grid. This is normally the Axpert king, max and some Kodak models.

Although the Axpert are good I myself would not load it at maximum power. Lithium would give you more cycles but to reduce grid to save a would also go the Sunsynk route.

I thought that the 450V input units could blend as per some other topics but may be only the MKS iii.

@LoodPyp

I think years ago hybrids did mean mixing grid and PV. I think sales PPL started using the term for inverters that could use PV to charge and back up combined with a AC charger. Thus we must rather look at specs to know how the unit works.

We are finding the same with EV. Hybrid did initially charge the battery while driving and combined with the fuel engine. Lately the newer hybrids also allows batteries to be charged via a plug at home or charging stations. Nothing stays the same and hybrid as a word can mean different things.

 

 

Edited January 16, 20224 yr by Scorp007

6 hours ago, Scorp007 said:

Alway feel sorry for someone getting such a short life out of batteries as it is the most expensive item of a PV system. This being the main reason why I only cycled my batteries during LS and every few days to prevent them dying of old age.

Perhaps a good time to have the inverter checked for proper operation in different modes as well before new batteries are connected. You don't want to buy another set after 1.5 years.

I still wonder why your mode SUB vs SBU is not working as it should by using the full PV available and only mixing with grid when you exceed the PV available.

Now replying to an earlier post in the thread that I hadn´t gotten round to replying to yet.

The trouble is that making your batteries last longer by just using them as a power backup or to a lower depth of discharge is also kind of a false economy because you aren´t getting much value out of them.

I am thinking that the bottom line is that batteries don´t work as a truly good financial investment. If you care about climate change, or need power backup, they make sense.

I actually prefer that the inverter uses the battery to supplement high loads because this is very likely a better thing to do from a climate perspective (35% coal and >50% fossil fuel here in Chile) and I´d rather avoid more grid electricity use, even at the expense of a shorter battery life, as long as the sudden heavy zapping when things are turned on isn´t actually damaging any part of the system. If it just means the lithium battery lasts say 5 years instead of 8 I´d settle for that as long as it functions fairly well for that time and doesn´t have issues.

As long as the lithium battery lasts 5 years+ we will get enough value out of it to justify the cost. I may only recoup half the cost of the battery in that time in electricity savings, but the other half I am happy to spend to have a good backup as we have regular power cuts, and to reduce our family´s carbon emissions.

At the moment I´m still leaning towards replacing the battery first and hoping for the best but I am dithering and you might all be able to convince me. The inverter is expensive but it´s also the heart of the system. I don´t even know how to take out the inverter and connect the house back to grid. So I either have to pay a professional to come to the house and evaluate it here or pay another professional to take out the inverter and connect to grid and then someone else to evaluate it in their store and factory.

58 minutes ago, Green Power said:

LoodPyp, Did your system have lead acid batteries?

Would it be fair to say that with lithium batteries the better cycle life, and perhaps better ability to withstand heavier demands, might mean that this inverter might work Ok with lithium batteries?

 

Yes, but lead acid soon became dead acid batteries.

I did not confirm whether the older inverter's co-operation with a lithium battery may have been acceptable, as I upgraded to an inverter with greater capacity, grid supplementing and more user friendly controls.

@Green Power1.) Fist of all, lead acid batteries, gel batteries, deep discharge batteries, whatever you want to call them is not suitable for solar applications in my view. If you discharge them below 70% SOC, you damage them and they will not last. Yes, they are cheaper than lithium iron batteries, but they are useless. Don't waste your money, get a lithium iron or LiFePO4 battery. You can probably discharge them to 20% SOC without damaging them and get about 6000 cycles out of them. 

2.) You need an inverter that can run on solar and if a cloud moves before the sun, to use grid power to make up the deficit. Your Axpert MKS II 5kW is making use of battery power to make up the deficit. You are continuously cycling your batteries, they are so confused they don't know if they are charging or discharging. The batteries will never last. 

3.) Batteries are expensive and you don't want to hammer them each day with high loads and reduce their life expectancy. Yes, we all want to get our money back from what we paid for them. Rather switch over to batteries at say 21h00 or 22h00 at night when the load is relatively low and constant to run until the sun rises the next morning to charge them again. To run of batteries during the day when the wife switches on the kettle every 30 minutes while doing the ironing and heating up yesterdays leftovers in the microwave while the tumble dryer is running, is a battery killer.

48 minutes ago, Green Power said:

I´d rather avoid more grid electricity use, even at the expense of a shorter battery life

I admire you for trying to save the world and paying for it out of your own pocket, but for now, first get the right system in place that works for you , you can save the world later.  

Edited January 16, 20224 yr by Don

1 hour ago, Green Power said:

LoodPyp, Did your system have lead acid batteries?

Would it be fair to say that with lithium batteries the better cycle life, and perhaps better ability to withstand heavier demands, might mean that this inverter might work Ok with lithium batteries?

 

Just a question. Is your grid power measured by a smart post paid, prepaid meter or do you have a old type turning disk meter? 

If you were a DIY guy with basic elec knowledge it would be easy to explain how to bypass the inverter for removing it so that you have all loads on the grid again to get the unit tested. 

It is an old type turning disk meter. Nothing smart, and the rate is always the exact the same at every hour of the day.

24 minutes ago, Green Power said:

It is an old type turning disk meter. Nothing smart, and the rate is always the exact the same at every hour of the day.

If one goes back to your 1st post and the fact that you batteries are shot I would still think that once you switch on a 3000W+ load and PV(2500W) is not enough it switches to batteries but as the batteries cannot supply the load within a few seconds or even within 50ms it switches from battery to grid to power the whole load. Your timing of 10min before it switches back to PV is based on the setting RETURN to batteries and it takes 10min to recharge the batteries enough to reach this point.

1 hour ago, Don said:

You are continuously cycling your batteries, they are so confused they don't know if they are charging or discharging. The batteries will never last. 

 

Thanks again for your help Don. I see the point of switching to batteries at 9pm or 10pm but from a climate change and pollution perspective I suspect it makes more sense to run the battery from sunset onwards because that is a high demand period for electricity grids. If you can avoid using grid electricity until 10pm, that probably helps more to reduce gas and coal, and even get the plants shut down.

I suppose if I replace the inverter now at this early stage we´ll have to factor in 3 inverters during the life of the system rather than 2. I´ll likely end up spending US$1000 (??) on the inverter. We`ll also lose some money in spending more on grid electricity during the day - say $100 a year. Maybe $1500 spent after 5 years. (Add on installation cost and a bit of grid spend while the inverter is out of action, that´s another $200, but say I get back $200 from selling the inverter. So let´s say $1500.)

Would I save US$1500 in batteries during the say 10 years lifetime of the new inverter? Even financially, without concern about climate change, it´s not clear to me if buying an inverter works at this point.

What do you mean by "never last"? Might the daytime and tea time cycling reduce the lifetime of the battery by half? More than half? Less than half? Some data and studies might be out that to help on this. Haven´t found anything so far. Maybe will have another look.

 

22 minutes ago, Scorp007 said:

If one goes back to your 1st post and the fact that you batteries are shot I would still think that once you switch on a 3000W+ load and PV(2500W) is not enough it switches to batteries but as the batteries cannot supply the load within a few seconds or even within 50ms it switches from battery to grid to power the whole load. Your timing of 10min before it switches back to PV is based on the setting RETURN to batteries and it takes 10min to recharge the batteries enough to reach this point.

That sounds about right. Is that "return to batteries" a changeable setting in the inverter menu? I don´t recall seeing that. Or maybe I´m confused again. When the batteries were newer it would usually only switch to grid for everything if the load was more than 1500W higher than PV input or if the batteries were very low in charge. But lately even after 10 minutes of supposed charging it tries to switch back to solar and batteries and doesn´t work and goes back to grid again for another 10 minutes. I actually have the batteries disconnected at the moment.

1 hour ago, Green Power said:

Might the daytime and tea time cycling reduce the lifetime of the battery by half? More than half?

If it was only tea time cycling, it would probably not be a concern. The problem is it happens many times a day, depending on load and cloud conditions. Like the time when the guy did some welding and the power tripped 10 -15 times. These are all unplanned events, but they happen. You mention on some days the sun sets and the batteries are not fully charged. Then you run the batteries until 21h00 when they eventually switches back to utility. You cannot drain those batteries every day and expect them to last. 

You mentioned that once you have a sudden increase in load, the batteries used to carry the load, now the inverter goes to utility for 10 minutes. It is because the batteries are no longer able to handle the load. The inverter then bypasses the batteries and goes to grid. Over a period of 1.5 years you have killed the batteries. Lead acid batteries were designed to handle a sudden high demand for power (high amps) over a short period of time (seconds), i.e. starting a car or engine. They were not designed to draw power from over extended periods for hours at a time. It just kills them. 

Don't worry, I had 8 x BSB 250 Ah Solar Batteries. Massive batteries, 73 kg each. Design life span was 5 years. When I got them, I could run my house for 2 days on them without solar. A year and a half later, I was lucky if I got 4-6 hours out of them. I have got the t-shirt, they don't last and I don't want them for free. 

Edited January 16, 20224 yr by Don

I have been coming around to your way of thinking on lead acid batteries, which is why I´m thinking of going to lithium. The question is whether I can run a lithium battery to support loads during the day AND drain the battery in the evening and still get 5 years. I guess I can´t expect anyone to really know that?  I can´t find anything useful from searching for scientific studies online so I´ve given up searching.

500 cycles a year (i.e. more than one a day including daily cycling) would be 2500 cycles compared to the >4500 it´s supposed to be capable of in ideal conditions according to Pylontech. That sounds plausible. Even if the use of daily cycling cuts the total cycles in half it could still be OK.

The battery has a recommended discharge current of 50A so for 48V battery I think that should be 2400W. Load is going to be below PV input majority of the time, and nothing we use is >2400W, so only rarely will things go above the recommended discharge current. And you would think the recommended discharge current in the manual ought to be consistent with the >4500 cycles in the manual.

However that still leaves your point "You are continuously cycling your batteries, they are so confused they don't know if they are charging or discharging. The batteries will never last." Are you saying that continuous flicking between charging is discharging is bad for a battery?  Is that a gut feeling or let me know if there is any evidence to support that.

I am concerned also that what I am looking to do would cause in practice the battery to be cycling a lot between 100% and 90%. That is a problem because for lithium 10 times from 100% to 90% is far worse than 1 time from 100% to 0% as far as I know. That means my 2500 cycles above may in reality be like 3000+ complete cycles. Not sure if there´s a way to stop the max charge % at 80% to minimize this issue.

My wife bought a gas kettle after the electric kettle caused power cuts several times since the battery got trashed. So that should help.

We do not exceed the solar power input constantly during the day. It isn´t jumping all over the place. The electric oven at tea time is extremely frequent for 20 minutes at about 1500W, almost daily. Apart from that, things can be fairly stable and exceeding the input is pretty occasional.

For 8 months of the year, we don´t have much changeable weather here. It is either sunny all day or cloudy up to a point then sunny. That helps a bit.

The Pylontech UP5000 I am looking at it seems comes with a 5-year guarantee. So there is the argument that either it lasts 5 years which means I´m happy, or it lasts less which means I have a chance to get something on the guarantee.

Thanks again for walking me through this.

Edited January 16, 20224 yr by Green Power