Mackay's Marvelous DIY Powerwall

ShaneE

New member
Joined
Oct 8, 2016
Messages
5
Greetings,

Thought I would start a thread for my own build progress.

As anyone who has done this before, or has started doing this knows, this is a long process. I started on 23rd September dismantling my first laptop battery. Now on 8th October I have 14 laptop batteries left to dismantle (plenty more to come) and have salvaged 204 cells from 40 laptop batteries. Out of those cells 39 have been capacity tested and recharged and are labelled ready for service (RFS). I have started using some of the lower capacity cells to put some charge in to the cells that read below 3v so the charger will charge them.

Still waiting on my TP4056's to arrive (25 of them), 25 x 4 cell battery holders, the plastic holding thingies hat I ordered on 17th September and was only told yesterday that there has been industrial action in the postal service and as such have been delayed significantly...awesome, 3d printing it is. The PIP4048MSE will arrive next week so I am very much looking forward to that!

In order to get this project off the ground relatively quickly, I acquired some 6 volt 130Ah AGM batteries from work. They are reclaimed but still tested to about 125Ah capacity. The idea is to hook them up to the PIP4048MSE so I can run my PC's and pool pump off the batteries during the day and recharge at night off peak. The pool pump uses about 8Kwh per day so the more I can use off peak for that the better, plus I can add solar panels when I get them to run it off grid as much as possible.


image_fqklpq.jpg

image_zswujy.jpg

image_qsivlf.jpg
 
You are probably not saving any money charging from off peak. Agm batteries retain about 70% of the energy put into them. And if your load current is high you will only get 35-50% as high discharge current depletes the battery faster
 
looking good so far I'm going to try to talk on here some times will this site work on a phone?
 
It seems to work on a phone. I am currently using an iPhone to check all of this goodness out.
 
phillip20 said:
looking good so far I'm going to try to talk on here some times will this site work on a phone?

Seems to be in my tests.
 
DIY TESLA POWER WALL said:
You are probably not saving any money charging from off peak. Agm batteries retain about 70% of the energy put into them. And if your load current is high you will only get 35-50% as high discharge current depletes the battery faster

Not sure where you got your statistics from? If I put 100Ah in to a battery I can take 100Ah back out, assuming the batteries are still decent. A poor quality or near end of life battery may be different. It may also happen if you try and charge the batteries at more than 10% of their rated Ah. EG if you have a 100Ah battery and you charge it at any more than 10A.

In my case I have 6v 130Ah batteries x 8, but they are actually 2v 130Ah cells but each battery has 3 2v cells in them. So if I take my worse case scenario load of 4kw from my inverter, at 48v that will equal 83A draw from the batteries (assuming 100% efficiency). That means only 10.4A per 6v battery or 3.5A per 2v cell. That is only 2.7% of their rated capacity.

Obviously nothing is 100% efficient but at least that gives plenty of wiggle room, and I won't be using the inverter at 4kw anyway, will probably max out at about 2.5kw.

I may not save any money recharging in off peak over the long run but at least my PC will have clean power all the time, and won't rely on a UPS to switch over in the event of a power failure. And this setup will allow the connection of solar panels as I get them.
 
Simple logic. If charge voltage is ALWAYS HIGHER than resting voltage, and even if amps in = amps out (which it doesn't) then how would it be possible to get 100%?

Input power, consists of volts x amps. Let's do volts first.

Charge voltage = 14.4
Resting voltage = 12.85

12.85/14.4 = 89.2%

But it doesn't stay at 12.85 does it? Even at 100% full, put a large load on it and the voltage falls to say 11.2v. Now it's 11.2/14.4 = 77.7%. Or, go down in state of charge.

But remember, I am still assuming amps in = amps out. But that doesn't happen. Typically you get 80 out for every 100 in. Its just fact.

So let's adjust our results:

89.2 x 0.8 = 71.3%
77.7 x 0.8 = 62.1%

Here's a scientific paper showing 71%
http://www.battcon.com/PapersFinal2...lyte Mixing VLA Cells in Renewable Energy.pdf
 
14.4 is not the charge voltage of AGM, 13.8v is. 14.4v is used if you are trying to clear it of sulphation. In some cases the charge voltage can go as high as 14v in certain temperature situations but 13.8v is the nominal charge voltage.

AGM's aren't designed for high current loads, unless they have been specifically designed for that. Like I said you shouldn't normally exceed 10% of the Ah rating for both charging and discharging. The battery voltage under normal load conditions will sit around the 12.5v mark and will stay that way until it starts to get low in capacity. To keep your batteries in good condition and to last a lot longer you shouldn't let them get below about 12.2v but at the most 12v. So let's say thats the 70% discharge mark (which is about accurate in my tests of the batteries I have). If the batteries reach 12.2v (or 48.8 in my system) then it will kick the charger in to recharge them.

A full AGM battery will rest at 12.8v with no load, it will drop to 12.5v under 'normal' loads. It will stay at 12.5v for the first ~70% of it's capacity.
A fully discharged AGM under 'normal' load will sit at 11.6v (will bounce back to 12v with no load in most cases, unless the capacity of the battery has been diminished)

Your logic doesn't take in to account the battery resting at 12.5v under load for ~70% of the time.

PS That link doesn't work for me, 404 error.
 
Lots of quoting voltages for each state of charge, but no actual rebuttal nor evidence to back up your statements.

ShaneE said:
If I put 100Ah in to a battery I can take 100Ah back out

Says who? Please provide evidence. Energy is not amps. Amps in is not amps out.
I know this first hand. I had a coulumb meter running on my agm bank. Every day, without fail, the soc (which is amps out vs amps in) would reach 100% long before the charge controllers reduced the current to a trickle. Therefore, when the amps out were replaced by an equal number of amps in, the batteries were still not charged and more amps were required. Thus proving that amps in does not equal amps out. I searched Google high and low, and was unable to support your claim that amps in equals amps out. So you are on your own there.

ShaneE said:
at 48v that will equal 83A draw from the batteries (assuming 100% efficiency). That means only 10.4A per 6v battery

What? If it's 83A, and you have 8 batteries in series, then each battery (and each cell) has a flow of 83A. Show the math giving you 10.4.
 
ShaneE said:
Your logic doesn't take in to account the battery resting at 12.5v under load for ~70% of the time

That's because logic doesn't factor in fairy tales. How about we factor in 90% efficiency on the inverter? Now your 83A is 92A. And I assure you, at 92A you do not have a voltage of 48v. More like 45. So now your current needs to be 98A
 
Let's say you want to use 1kw to charge. Just by having the pip connected, you waste 50w. So the actual consumption is 1050w.

Let's assume the charger is 90% efficient. Your 1000w is now 900w.
Your batteries store 50% of that 900w. It's now 450w.
When you run from the batteries, you again pay a 50w penalty. You now have 400w.
The inverter is about 90% efficient. It's now 360w.

1050w in, 360w out. 34% efficiency.

That's assuming no large loads which will give a much worse result. Also ignored is the life used up in the batteries and inverter, which typically costs between 10c and 15c per kwh.
 
DIY TESLA POWER WALL said:
Let's say you want to use 1kw to charge. Just by having the pip connected, you waste 50w. So the actual consumption is 1050w.

Let's assume the charger is 90% efficient. Your 1000w is now 900w.
Your batteries store 50% of that 900w. It's now 450w.
When you run from the batteries, you again pay a 50w penalty. You now have 400w.
The inverter is about 90% efficient. It's now 360w.

1050w in, 360w out. 34% efficiency.

That's assuming no large loads which will give a much worse result. Also ignored is the life used up in the batteries and inverter, which typically costs between 10c and 15c per kwh.

I think you will get your point across better if you used units of measure that are relative to efficiency. By just using 'w" (watts) you are just representing energy at a single point of time.

You must use energy over time to get any real meaning. That's either in joules, coulombs, Ah or Kwh. Kwh is a probably the most accepted term (in this application) as it represents total energy over time. Ah doesn't take into account variations in voltage so should not be used in this application.

Efficiency (in this application) should be measured by calculating total energy in divided by total energy out, over a specific period of time. For example, total energy input into the system (from solar panels) over say, 16 hours divided by total energy taken from the system (output of inverter) over say, 8 hours.

To be honest, your example above is not a real calculation of efficiency and is very misleading.
 
If it makes you more comfortable, add the letter h after w for each number. Ie, 1hr charge and 1hr discharge. Semantics. None of those figures are time dependant so they stand regardless of whether the letter h follows them or not.

"very misleading"

That's a statement you can only make when you have equally qualified evidence to the contrary. But you don't. And here's why. All of these steps are less than 100% efficient. Do you disagree?
The input number for each gets progressively lower. Agree?
The overall efficiency is the product of each individual inefficiency (disregarding inverter self consumption) multiplied. Agree?

Overall efficiency = charger efficiency x battery efficiency x inverter efficiency
= 90% * 50% * 90%
= 0.9 * 0.5 * 0.9
= 0.405

But we have not allowed for inverter self consumption, so that will account for the slight difference in the examples. So please provide evidence that it is "very misleading"
 
DIY TESLA POWER WALL said:
If it makes you more comfortable, add the letter h after w for each number. Ie, 1hr charge and 1hr discharge. Semantics. None of those figures are time dependant so they stand regardless of whether the letter h follows them or not.

"very misleading"

That's a statement you can only make when you have equally qualified evidence to the contrary. But you don't. And here's why. All of these steps are less than 100% efficient. Do you disagree?
The input number for each gets progressively lower. Agree?
The overall efficiency is the product of each individual inefficiency (disregarding inverter self consumption) multiplied. Agree?

Overall efficiency = charger efficiency x battery efficiency x inverter efficiency
= 90% * 50% * 90%
= 0.9 * 0.5 * 0.9
= 0.405

But we have not allowed for inverter self consumption, so that will account for the slight difference in the examples. So please provide evidence that it is "very misleading"

OK, first up. Chill out a bit. I was only offering advice, not having a go at you.

Secondly, I didn't disagree with anything you posted - I just said meant that the way you presented it was "very misleading". Perhaps a better way to state your point would be to provide real evidence of efficiency rating of some devices currently on the market. To simply state that a battery is 50% efficient gives the impression is that it's a fact of life. That figure is true for a battery in bad shape, however I have test data that proves good quality Li-ion cells can be around 85%-95% efficient.
 
We are not discussing Li-ion, he is using lead acid (temporarily). So that's where the miscommunication is. All good. Even so, the math still would look terrible using Lithium.

Anyway, hopefully some reading this will attempt to do some math before committing vast amounts of time and money into a flawed power system.
 
DIY TESLA POWER WALL said:
We are not discussing Li-ion, he is using lead acid (temporarily). So that's where the miscommunication is. All good. Even so, the math still would look terrible using Lithium.

Anyway, hopefully some reading this will attempt to do some math before committing vast amounts of time and money into a flawed power system.

OK, DIY TPW you got me there. I missed that he was using AGM until he gets his Li-ion bank up and running. Also have to agree, AGM is not the best type of battery for a solar setup. They are more suited to emergency back-up installations. Battery companies like them because they are cheaper to make, don't last as long as other types of batteries (have to buy more batteries) and they're easier to transport than FLA.

However, you can make them perform better and last longer if you treat them nice. The biggest issue these batteries have is not sulfation, it's heat. Heat from either charging or discharging them too quickly. Anytime you are heating them you are going to be off-gassing and evaporating electrolyte that cannot be replaced. Once they dry out, they're useless!

So looking more at Mackay's proposed setup, I agree with @DIY TESLA POWER WALL here. It's not going to be practical to charge these up with off-peak grid power and then drain them with a pool pump. If you don't want to kill those batteries and you want to improve your efficiency, discharge them at around the C12 rate. So if they are 125Ah, that's 125/12 which is around 10A. Since your running at 48V (50V nominal), that's a 500W load - which is probably a lot less than what your pool pump will draw but probably OK to run a PC for 5-6 hours a night. (Insert aforementioned inefficiencies here :) )

Same goes for charging them, if you charge them too fast you will kill them quickly. Again if you limit the charging current to around C12 (~10A) you'll be topping them up gently which will increase the battery life and improve the efficiency. (C12 charge/discharge works well for a 24 hour cycle as well)

As for your pool pump, I would just run it from off-peak grid power. 8kWh seems to be very high, you must have a massive pool or you're running it longer than needed. Just for comparison, I have a 750W pool pump that only runs for an hour/day over winter (0.75kWh/day) and 4 hours/day (3kWh/day) during the summer swimming season. I use a timer to run it when my small 1.6kWh grid-tie solar system is putting out the most in the middle of the day.

Anyway, hope this helps
 
Back
Top