The Wayback Machine - https://web.archive.org/web/20130731070056/http://windpower.org.za/batteries/batteries.html
     Lead acid batteries   Back to Wind Power
 
Web windpower.org.za
Logo

Have you ever wanted to make your own homemade lead acid battery? Lead acid batteries were invented in 1912 and in essence haven't changed much since. Two plates of lead immersed in 30% sulphuric acid produce about 2V of electric potential after being charged at 2.15V or more. The greater the surface area, the greater the maximum current that can be taking in during charging, and given out by the battery cell. A lead acid cell is typically about 70% efficient - to get 100 AmpHours out you need to put 143 AmpHours in. At room temperature, charging a cell below 2.25V is 'safe' in that it will not cause gassing - at higher voltages water is split into hyrdogen and oxygen gasses. Deep cycle batteries are ideally charged at about 2.5V per cell. Forming the plates is the process of using crystal formation to roughen the surface of the plates, greatly increasing the surface area exposed to the electrolyte.

Thin plates very close together give the best maximum current, but are subject to distortion under heavy current, and corrode away more easily. Thick plates give a long life

I am not sure what effect the distance between the plates has - I suppose it introduces resistance, and so ideally is kept to the minimum without danger of the plates touching or being bridged by sediment.

So the only way to know is to try! I cut out about 80cm2 of 1mm thick lead sheet, plus a tab to connect the wire to. A large jam jar with a plastic lid, melt two slits in the lid, and presto a battery cell.

I covered the plates with battery acid (33% H2SO4), so had 160cm2 area per plate. There was about 130mV generated by this for some reason!

I then used a small DC power supply to push 3 volts into it, and it drew 0.25A. I cranked up the supply voltage to "4.5V", which increased the current to about 0.64A and the voltage actually stayed at about 3 to 3.5V. The plates bubble away merrily, with the positive going rust coloured (lots of small bubbles) and the other looking the same silver/grey (larger bubbles).I think what is happening is the positive plate is getting lead peroxide, and the negative plate is just reuniting electrons with H+ ions.

Be warned the bubbles are hydrogen, and explosive when mixed with air. Keep well ventilated, and no flames or sparks!

The analogue ammeter is connected via a 1K Ohm 1% resistor, to make a simple voltmeter (showing 3 "volts", or 30mA). The digital meter is showing amps.

The DC power supply is marked as rated for 500mA. It seems to survive 650mA! Another one I tried to run above its puny 9W rating worked for a while, and then made a horribling fizzling noise. May it rest in peace. It seems the cell is happy to absorb more current, but the DC power supply is not able to. With too much current, at some point the heating and rapid bubbling may cause the peroxide to flake off the positive plate as it forms.

Once that has charged, I will draw a current to discharge the cell. This should result in PbSO4 forming on both plates. Then charging it again should convert the PbSO4 to spongy lead on the negative plate, and PbO2 on the positive.

Will the peroxide (PbS04) powder and fall off the plate? If so some sort of retaining mat will be needed. Or what about having the positive plate horizontal at the bottom of the jar?

More tests are needed...

27 July 2006

For the next attempt, I wanted to get 6V and more surface area, so I cut out six strips of lead sheeting, 70mm x 400mm, which when rolled up fitted neatly into a medium sized jam jar. To keep the plates from touching, two sheets of woven glass fibre mat were put in between the sheets of lead. So each cell has two sheets of lead and four sheets of glass mat.

This time I left it to soak in the acid overnight, as I wanted lead sulphate (PbSO4) to form on the plates. Once a thin layer of PbSO4 forms, it covers the remaining lead and stops the reaction. So in the morning I put it on charge using a small DC power pack. I set it to 3V, which actually develops 7.1V open circuit. I kept track of the current flowing into the cells.

This time there was no bubbling.. perhaps the bubbling was a symptom of over charging the first model?

I noticed a salt like sediment forming in the bottom of the jars. Perhaps these are the sulphation crystals that are mentioned in warnings about overcharging batteries. To get rid of them I put the jars in a 1-2cm bath of boiling water, and stirred the jars once they had heated. This seems to get the sediment back into solution.

So back to charging.... I am not getting very much charging current... the positive should go rust coloured from PbO2 formation. Perhaps this is due to the lack of spongy lead which will hopefully form after repeated charge/discharge cycles (thereby hugely increasing the surface area).

It seems that a way of monitoring the sulphation and discharge process is needed. Or perhaps I should push more current into the cells on charge? And should discharging be done by just shorting out the cells (individually or all together?) or by controlling the discharge current?

Is there anyone out there with more knowledge of the chemistry involved who can suggest anything? Please post a comment here, and I will post it here.

Currently (ha ha), the cells are drawing about 5 - 10mA on charge at 6.9V. A new test is an electric motor which spun for 7 seconds from the batteries - I will now alternate charge and discharge and monitor this to see if it increases or decreases. More to follow soon...

29 - 30 July 2006

So here is a graph of what should happen during charging. It makes it clear I was using too low a voltage, so I increased the charge voltage to a nominal 6V, which delivers 7.9-8.2V during charging. After charging at this voltage, this time the motor ran for 35 seconds. Thereafter (with the load removed) the voltage climbs slowly back to around 5.4V after a few minutes. It then went back on charge, which started out drawing 78mA, dropping to 33mA after 20mins.

So following a sort-of pattern of allowing the charge voltage to rise to over 8V, and then running the motor, then charging again gave the following results. Sometimes I shorted the battery out after the motor stopped. In all cases, the positive plate stays much darker than the negative, meaning PbO2 remains on the positive plate. To get maximum surface area, should all the PbO2 be converted to PbSO4 after discharge? post a comment hereAny ideas or suggestions?

Time charged Final charge voltage Seconds motor ran  
2.5hrs 7.17 7 Charged at "4.5V" setting
14hrs 8.55V 35 Switched to "6V" setting
1.5hrs 8.29 40  
75min 8.10 52  
12min 8.05 55 Then short out for a few minutes
15min 8.01 69 Then slow discharge via LED
18min 8.02 80  
8min 8.1 99 Switched to 6V battery charger
2min 8.56 96  
8min 8.58 103  
  Here I reversed charged, but the PbO2 started to flake off alarmingly
25min 8.60 141 Back to normal charge polarity
  Here I filtered the electrolyte using a coffee filter paper (after heating a bit)
?+12+11min 8.32 147  
11min 8.44 174  
2min 8.1 140  
10min 8.35 203  
10min 8.44 209  
24min 8.31 220  
4+9+2+24+16 8.38 203 Tried a bit of reverse charging again
34+10 8.44 221  
30 8.44 290  5V:239   5.5V:200 Now measuring secs to 5V & 5.5V
5 8.46 270  5V  5V:243  5.5V:210  
21+3   330  5V:276  5.5V:232  
7 8.38 341  5V:298  5.5V:254  

I think the motor draws 750mA at first, and working with this, the battery is about 0.062 Amp hours! Not so great - to get 10 Amp hours would need plates half a square meter each! (plates are currently 70x400mm each, two per cell)

If left for 15min after charging, the battery settles at about 5.91V, 1.97V per cell. This is too low! Will repeated charging improve this? Suggestions and ideas welcome!

I will now continue the charge/discharge cycling (down to 5.5V), and record and graph the time the motor runs. Results to follow...

2 August 2006

One of the cells started to behave strangely - it didn't bubble when charged, and didn't maintain its voltage under load. So I drained it, rinsed, unrolled it, and cut new glass fibre mat to size, and rolled it up again. The electrolyte was filtered (coffee filter paper again), and topped up with a bit of water and acid. Part of the rerolling process was banging it flat, which would have destroyed the high surface area pitting, so this cell is likely to take a while to catch up to the others. The glass fibre mat had changed consistency - it was stuck to the lead in places, and was very weak. It 'came apart in me hands' as I pulled it off. Perhaps its not the best material to use as a separator; perhaps some sort of rubber mesh would be better. Any ideas? If so please let me know.

I cut a sliver of lead off the negative plate and looked at it under a microscope - the surface was very rough, and I could see it had been pressed smoother even from the light pressure of handling. Unfortunately I can't take photographs through the microscope.

So now the charge/discharge cycle continues, and here is the graph of the motor time test so far. I keep trying things, like varying the charge time, putting the cells in a hot bath, shorting out after discharge (or not), etc., so the variation is to expected.

After this (point 19 on the graph), the new cell took a few charges/discharges to start catching up to the other two more mature cells. A feature of a lagging cells is the battery voltage drops rapidly to around 5V where it is maintained for quite a while.

More... (page 2)

Back to Wind Power

Please bear in mind acid is dangerous, and safety and disposal precautions should always be observed.