Help with inverter battery sizing

That is the point. Quite a few I looked at had NO BMS at all yet they are on sale in SA

So then how are you recommending multiple strings? Do you think everyone's got the time to build their own BMS? Do you think that everyone's got the money to buy a BMS? OP isn't designing a datacentre.
 
Sorry to butt in - its all so harmonious here I couldn't resist :)

Just a quick point. Don't forget the inverter and batteries themselves are a load. So when your discharging particularly at small loads, you will likely not get what you hope for - because say the inverter and batteries will be using 50W too. (Not sure on the numbers I know from experience that cheap UPSs are pretty bad). All that heat has to come from somewhere!

It depends on your batteries and inverter, but may be shown in the spec sheet. The 'efficiency' shown will be at near full use, so hopefully they have one at 10% and 50% use too for you to estimate.

Indeed.

Some numbers on the 1100VA SSPro I spoke of above. Using a Cisco 3560 as load, which draws 43.68VA. I'm not sure what the powerfactor on this switch is, I have not measured it. Lets go with 1 for ease of calculation, thus 43.68W.

While inverting, on the battery side it drew 73.75W with only the 3560. Thus at least 30W is dissipated by the UPS. Probably more, taking PF into account.
 
On my cheap UPS's I have also seen them use around 30W themselves when powering a 5W LED. I will factor in an extra 50W for the inverter.
 
I become a pain when I detect bull****. And since when is calling people out on bull**** a personal attack?

Maybe if you could back up any of the bull**** you spew, I'd take you seriously.

So then how are you recommending multiple strings? Do you think everyone's got the time to build their own BMS? Do you think that everyone's got the money to buy a BMS? OP isn't designing a datacentre.
I have made NO recommendations.
This started off as a technical discussion.
I have commented from a technical point of view on what others' have said. YOU are the one that brought up many of these things trying to show us all that you think you are the only expert on the forum..

I have stuck to the facts and the basic theory.

So now you stick to the facts as well please and just stop with trying to continue your personal attacks and make positive contribution or shutup.
 
On my cheap UPS's I have also seen them use around 30W themselves when powering a 5W LED. I will factor in an extra 50W for the inverter.


Will continue with my relatively simple technical post dealing with considerations relating to batteries, inverters and loads that affect what actually happens the day you need that backup power.
 
In my post https://mybroadband.co.za/vb/showth...ery-sizing?p=19223646&viewfull=1#post19223646

I stopped at the point where it becomes necessary to refine the calculations in any attempt to work out what size inverter and what size battery to get tp provide the power back up. There is plenty on the Internet about this subject much of it taking terrible shortcuts and applying formulas without understanding how and when they should be applied. And then we have the sales persons and the online sales sites selling "stuff" they have no idea about or how to sell it correctly.

Why is it that persons are so ready to believe everything they find on the Internet? Especially when buying complicated stuff like chargers, inverters, batteries and UPS systems?

One of the crucial elements in a standby power setup (forget for the moment the rather spurious argument over a UPS vs an inverter plus separate battery bank). BTW the correct way to talk about batteries is to refer to "Battery Banks" and not "strings". String is used to tie up parcels.

So this post will highlight without going into a deep explanation of anything. You can all go and read the reference I quote for yourself. Be warned though that the maths in the reference gets more and more complicated and will test even the so-called "experts" who think they know what is going on.

Batteries:

This thread has talked a lot about a few things such as:

1. "Discharge the battery to 50% of its capacity": This is not just a value, it is backed up with very sound reasoning as will be seen in the reference I use.
2. Isheed in his post suggested a "battery efficiency factor": Again he did not pluck that out of thin air, it is a pretty good estimation that can be used in a sizing calculation for ALL battery technologies.
3. Battery Banks: Plenty has been said about battery banks and whether it is a good idea to have split banks and batteries in parallel or not. The fact is that without some or other way to monitor the health of the battery banks, it is rather tricky to ensure that all the batteries will discharge at the same rate and recharge at the same rate. An essential fact when building larger systems, but I am suggesting that quite a few systems being sold "on line" to ordinary people that suggest you can do this are doing the consumer in.
4. Estimate the load properly: don't ignore anything and take all the fixed loads into account. One has to know what the ultimate loads are and have to take them ALL into account when estimating the size of the system you need.
5. How deep should one discharge a battery bank?: Well here the urban myths really kick in. What is fact is that batteries last the longest if they are either never discharged or if the discharged by more than 5% , but also never by more than 50% (at least for optimal life and usefulness over the life of the battery. The final value is a compromise between performance and battery life, and cost.
6. The last issue I have found quite strange is that no one trotted out an application of "Peukert's Formula" yet. Again there is a right way to use this formula and most definitely a wrong way and the Internet is crawling with examples and articles written on the subject applying the formula incorrectly.

So here is a very good reference by no means the only one, BUT one of the references that is thorough and explains all you ever wanted to know about sizing of battery banks whether they are for UPS designs, inverter + battery standby systems or even alternators and a deeper understanding of what goes on a typical car electric power system.

The reference is: http://www.smartgauge.co.uk/technical1.html
http://www.smartgauge.co.uk/technical2.html#depth

and a pretty good read on Peukert's Formula if you do not want to read the whole thing. http://www.smartgauge.co.uk/peukert_depth.html
 
Last edited:
When you say battery, you should point out you're talking about Lead Acid.

eg Peukerts Law, discharging to no more than 50% etc.

Other chemistries don't have the same issues.
Lead Acid quirks are well known - its a chemistry we've been using more than 150 years after all.

Lithium Ion mass use only really dates back to the 90's (although it was first developed in the 1960's)

Some history here -
https://qz.com/338767/the-man-who-b...attery-at-57-has-an-idea-for-a-new-one-at-92/
 
Yes correct. But the reference does show how the theory can be applied to other battery technologies.

Peukert (1897) came up with his empirical formula to try an answer a question still around "Why does a battery never deliver its rated capacity?"
 
I have made NO recommendations.
This started off as a technical discussion.
I have commented from a technical point of view on what others' have said. YOU are the one that brought up many of these things trying to show us all that you think you are the only expert on the forum..

I have stuck to the facts and the basic theory.

So now you stick to the facts as well please and just stop with trying to continue your personal attacks and make positive contribution or shutup.

You're right, you've contributed nothing useful at all. Lots of theory, but nothing that matters in the real world. Even your monologue on battery management amounted to nothing more than you enjoying the sound of your own voice. You should follow your own advice and "stop with trying to continue your personal attacks and make positive contribution or shutup".

In my post https://mybroadband.co.za/vb/showth...ery-sizing?p=19223646&viewfull=1#post19223646

I stopped at the point where it becomes necessary to refine the calculations in any attempt to work out what size inverter and what size battery to get tp provide the power back up. There is plenty on the Internet about this subject much of it taking terrible shortcuts and applying formulas without understanding how and when they should be applied. And then we have the sales persons and the online sales sites selling "stuff" they have no idea about or how to sell it correctly.



One of the crucial elements in a standby power setup (forget for the moment the rather spurious argument over a UPS vs an inverter plus separate battery bank). BTW the correct way to talk about batteries is to refer to "Battery Banks" and not "strings". String is used to tie up parcels.

So this post will highlight without going into a deep explanation of anything. You can all go and read the reference I quote for yourself. Be warned though that the maths in the reference gets more and more complicated and will test even the so-called "experts" who think they know what is going on.

Batteries:

This thread has talked a lot about a few things such as:

1. "Discharge the battery to 50% of its capacity": This is not just a value, it is backed up with very sound reasoning as will be seen in the reference I use.
2. Isheed in his post suggested a "battery efficiency factor": Again he did not pluck that out of thin air, it is a pretty good estimation that can be used in a sizing calculation for ALL battery technologies.
3. Battery Banks: Plenty has been said about battery banks and whether it is a good idea to have split banks and batteries in parallel or not. The fact is that without some or other way to monitor the health of the battery banks, it is rather tricky to ensure that all the batteries will discharge at the same rate and recharge at the same rate. An essential fact when building larger systems, but I am suggesting that quite a few systems being sold "on line" to ordinary people that suggest you can do this are doing the consumer in.
4. Estimate the load properly: don't ignore anything and take all the fixed loads into account. One has to know what the ultimate loads are and have to take them ALL into account when estimating the size of the system you need.
5. How deep should one discharge a battery bank?: Well here the urban myths really kick in. What is fact is that batteries last the longest if they are either never discharged or if the discharged by more than 5% , but also never by more than 50% (at least for optimal life and usefulness over the life of the battery. The final value is a compromise between performance and battery life, and cost.
6. The last issue I have found quite strange is that no one trotted out an application of "Peukert's Formula" yet. Again there is a right way to use this formula and most definitely a wrong way and the Internet is crawling with examples and articles written on the subject applying the formula incorrectly.

So here is a very good reference by no means the only one, BUT one of the references that is thorough and explains all you ever wanted to know about sizing of battery banks whether they are for UPS designs, inverter + battery standby systems or even alternators and a deeper understanding of what goes on a typical car electric power system.

The reference is: http://www.smartgauge.co.uk/technical1.html
http://www.smartgauge.co.uk/technical2.html#depth

and a pretty good read on Peukert's Formula if you do not want to read the whole thing. http://www.smartgauge.co.uk/peukert_depth.html

post-truth
adjective

Relating to or denoting circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief.


Now let me tell you WHY we don't give a **** about what you said: It's all irrelevant. Lots of (factually incorrect) pedantic attacks on my nomenclature, lots of references to "sources", but nothing of any value to anyone designing a system. No worked examples, no real world designs, no nothing. Just a lot of hot air. And I mean A LOT! Like pages that contain about 3 lines of information.

Just go sit quietly in corner, before you embarrass yourself any more.
 
Just go sit quietly in corner, before you embarrass yourself any more.

Anyone else think that Geoff.D and Drunkard would make a great odd-couple sitcom? :)
Either that or they are the same person locked in a Jekyll and Hyde cycle?
 
I am solving the problem by simply ignoring anything he posts.

And I invite him to do the same, he is welcome to ignore my posts.
 
I am solving the problem by simply ignoring anything he posts.

And I invite him to do the same, he is welcome to ignore my posts.

That's about what I'd expect from someone who's been called out on their bull****.
 
That's about what I'd expect from someone who's been called out on their bull****.
No that's what an intelligent person does when he realizes he's arguing with a total imbecile.
:D

Just relax and contribute rather than fighting with everyone.
 
No that's what an intelligent person does when he realizes he's arguing with a total imbecile.
:D

Just relax and contribute rather than fighting with everyone.

You've got any facts to back that up, or you just resort to insults like a three year old?
 
So now how does one choose an inverter?

The first thing to do is decide on the type of inverter you want to get. Again there is plenty on the Internet about the various types of inverters and what they are suitable for and what not.

OP decided on a pure sine wave inverter because that is what is required for the laser printer.

The Size? Well here is where things get interesting. One has to choose an inverter to handle the peak load. In this case, the printer is the peak load item requiring 1033 watts. To this you have to add the total static load plus an allowance for what the inverter draws just to operate. Based on the information supplied by OP, that means 1033 + 70 + 50 = 1153 W. It is not a good idea to run an inverter at maximum power and you need to get one that exceeds the maximum load. OP indicated that he intends to get a 1700 Watt inverter which is okay provided the total load has been calculated correctly.

The next step is to try and find out from the supplier of the inverter what the inverter efficiency curve looks like. BUT very few provide this information in their data sheets! The cheaper the inverter is the less likely you are going to be able to get them to commit to giving you anything about inverter efficiency.

Inverters do not perform very well under light loads. Most have an efficiency curve that approaches about 90% when the minimum load is about 20% of capacity. Isheed suggested that an inverter typically has an efficiency of 92%, which normally is the case provided the minimum load is around 20% of capacity. For this scenario that means the minimum load should exceed 340 Watts. Indirectly that means OP can afford to find a few more items to back up and maybe even include emergency lights in his scenario. It shows where I get the 20% value form as well as how the efficiency tails off at the top end of the load.

This what an inverter efficiency cycle looks like.

energies-08-04853-g004-1024.jpg

The last crucial item is to find out the charging capacity of the inverter OR, to work out how you plan to charge the battery bank, because this will tell you how long it will take to recharge the battery bank after a discharge cycle. You may even decide after you know this that you do not want to discharge the battery to 50% simply because the charge time is too long. Here is an older thread on the subject that is worth looking at.

https://mybroadband.co.za/vb/showthread.php/682748-Inverter-Deep-Cycle-battery-charging-question

So now without more information on the chosen inverter I cannot comment on the choice of a 1700 Watt inverter.

In Summary, make sure the maximum power rating of the inverter exceeds the maximum expected drain on the inverter, then calculate 20% of that value and that will give you the minimum load when on a discharge cycle. The reason for this is to ensure the inverter is running at maximum efficiency which should > 92%. Note that inverter efficiency also tails off as you approach its maximum power rating on a discharge cycle.

Good luck with trying to milk out of inverter suppliers the efficiency curve for the inverter!

Update: If you intend to Grid Tie your power setup to the
incoming mains power, find out from the authorities what specifications you must comply with to be able to do so! That is IF the local authorities in SA have even got around to coming up the specifications.
 
Last edited:
The efficiency of lead acid batteries is bad, usually about 70% when new and they go down to less than 50% after about 500 cycles or 2 yrs. That means you need a lot more solar panels and about double the capacity (Whr) for calculated usage for a new system. Lithium (lifepo4) batteries start at about 95% efficiency and can last 5000 to 6000 cycles so they might be worth the extra cost.

The Americans have sold A123 systems to the Chinese so prices might fall soon.
 
Li-ion is amazing, 18650 cells are impressive, amazing how Tesla uses tens of thousands of them to build the batteries in their cars and powerwalls.
 
Top
Sign up to the MyBroadband newsletter
X