Re-celling EN-EL4 also (a)

[Seapy]

I have mentioned in unrelated threads that I have been engaged in re-celling one of my Nikon EL-EN4 D3 batteries.  That has caused some interest in the forum...

I am somewhat cautious about relating my adventures on this project, the consequences of getting it wrong are serious.  The EN-EL4 case is very tightly fitting assembly containing three 18650 cells, there is not spare room for wires or even excess solder!  Unless you are skilled at working with these type of electronics and have relevant experience, I say you should leave well alone.  The risks of working with these powerful and potentially dangerous cells is well publicised,  this involves working with fully charged, live batteries which if shorted by accident or overheated by soldering can spontaneously ignite or explode.

Not sure I can do this in one go, I will first introduce the preliminaries then go on to the actual task later.

This was instigated by my visit to Kielder on a cold but clear night during which I intended to take Milky-Way photographs.  I set up by the lake in a designated 'dark space' to try to get some photographs of the moon setting.  That was about 2am, I intended trying for the Milky-Way at about 3am once the sky had darkened sufficiently, however after about 120 exposures with the D3, on time lapse, the first of my three EN-EL4 batteries gave out, this was a year old, non original battery from a recommended German source, in normal conditions these batteries have generally been giving me about 1000 exposures per charge.  I fitted my only genuine battery, which was on life 4, but that only took a further 60 exposures.  Somewhat concerned, since it was getting even colder, I stopped using the D3.  My  D300S was plodding away, quite happily also on time-lapse duty with no issues.

We went up to the higher viewpoint in readiness for a truly dark sky, I got about 90 exposures of the Milky-Way with the D3 before the third battery gave out.  The batteries were freshly charged.  It was about -6C, very little wind.  On our way home (about 120 miles) we stopped for the sunrise, I managed another 20 exposures or so with the D3.

This prompted me to rethink my battery strategy, I had been contemplating making an external power source, either a small generator or a portable large power pack of some sort.  The investment in time and travelling expense, fuel, meals, etc. of these trips is quite an item, the failure of the equipment is very unwelcome.  New Nikon batteries are not an option @ £150 apiece, realistically I would probably need at least six, that's £900.  More than I paid for my D3, and a lens...

I have started to gather components to make an external power pack, but that's another story.

I usually get my electronic bits and pieces from CPC, Farnell who's warehouse is relatively local for me but on this occasion I decided to cast around and see what the alternatives are.  There are all sorts of claims when it comes to batteries, the 18650 cells are available in a wide range of capacities up to about 3500mAh.  In my opinion 2600mAh is a realistic capacity, taking into account lifespan, reliability, quality and reliability.  I finally decided on Samsung, I already have several Samsung 18650 cells which I used to build some batteries for my D1/D1x cameras ten years ago, they are still going strong, so I felt that given the likely improvements in capacity etc. the current Samsung cells should be at least as good.  I also have four Samsung 18650 cells in my Nichia LED - 365nm UV light source, they are also 100% good after about 8 years.

So rather than experiment with 'pie in the sky' unlikely high capacity cells from unknown sources, I would stick with known quality, Samsung.  NO! I don't have shares in Samsung.

Then came the dilemma, buy ready tagged and solder or spot weld myself...  I found a very good 'You-Tube' item about a DIY spot welder:

https://www.youtube.com/watch?v=Ceos88VO6p4

The principle of measuring the actual power used to make the weld is a winner for consistent, reliable welds but the kits seem to be hard to obtain and I wasn't sure I really wanted the investment of well over £100 for a specialised tool, plus suitable battery or power source, which I only hope to use a few times, so it did not seem the right path for me. Your situation may be different, in which case a spot welder may be the best route, using soldered copper tags isn't easy either.

Given the extremely restricted space in the EN-EL4 casing linking the cells with  wire is not an option, in my opinion, apart from the two positive leads which Nikon use to link from the bottom of  the pack to the PCB, which I retained.  One of the things which took me some time to source was some suitable copper foil or sheet about 0.18mm, 0.007" or seven thousandths of an inch thin.  That is about the thickness of the original link straps which Nikon use in these batteries.  The original links are nickel plated steel, so copper will provide a slightly lower resistance.  I have measured the strands of wire and links within the EN-EL4 battery and they are at least 1.2mm² cross sectional area.  I have made sure any links I have introduced are at least 1.2mm² cross sectional area, because when I re-constructed my D1/D1x cells initially they did not work.  I had used wire which was slightly too thin, which did not allow the shutters capacitor to recharge quickly enough.  Be sure to use sufficient capacity links, at least as good as the original.

I use an old Weller 120W soldering gun which has definitely seen better days, I prefer a powerful soldering gun because you can be quickly, in and out, before things get too hot.  Using a small pencil soldering iron can be slow and the components can actually get hotter, which is not good...

A Stanley knife to open up the casing, care is required here because it's quite thin and the joints are 'half-lapped' and actually sort of clip together.  You also need to bear in mind that the battery could still contain some current potential so the need to  manage the penetration of there knife blade is important.  The PCB at the connector end is also vulnerable to damage during the opening up phase.  Once the top (the raised section with the Nikon Lable) is removed, the cells and PCB will likely remain stuck in the lower half.  There are two strips of double sided sticky compound securing the cells into the casing.

An inevitable Dremel to clean up and trim.

A micrometer or electronic calliper for accurate measuring.

I used Gorilla glue, which is slightly foaming to glue the cells together to ease handling and help avoid the possibility of short circuits.

A two pack resin glue to close the case after the operation is complete.

Any comments or queries please raise, then I will follow up with some photographs and my method and comments.

It's doable but not easy, great care is required.  So far mine is running OK but that's early days.

[Hugh_3170]

Great write up Robert;  and thank you for "taking the bait". 

[Seapy]

Thanks Hugh,  it was in my plans but like I say I am concerned that nobody tries this without understanding completely what they are doing.  It could wreck a camera or much worse if a mistake is made.

[Hugh_3170]

Yes, the dangers of Li-Ion cells cannot be understated as they can be seemingly "flat", yet still retain enough charge to be dangerous.

My only experience in cell replacement so far has been replacing the batteries in a battery powered electric drill and I had the luxury of being able to use replacement cells with connection tags already spot welded onto them - which greatly facilitated the job.  Unfortunately not all applications have enough space or room to allow the use of tagged cells.

[Seapy]

The biggest problem isn't associated with the practicalities and safety, it's the economics...

I just checked, I paid £46.60 for two EN-EL4(a?) batteries from Photo Shop Becker, Breslauer Str, Bad Schwalbach, Deutschland, inc. postage.

That is £23.30 each.

The three cells from RS Components (UK) cost me £21.93 delivered, to re-cell one battery.

There is no indication, without opening them up, what the capacity of the German battery is.  I plan to raise this issue of poor cold performance with them and also ask them what the capacity of their cells is.  Edit  The cells are marked 2600mAh.

I can get 18650 cells for about £3.00 each, £9.00 for three.  But, not @ 2600mAh.  Now if the German batteries are not 2600mAh, then they aren't such a bargain.  I don't really want to open them up just to find this out.

One other factor... Despite having 're-calibrated' my rebuilt battery, it's still showing life 4 in the camera.  I just wonder if at some stage it will switch off. Perhaps after a given number of charge cycles.

[Øivind Tøien]

I have been following this with interest as I have an EN-EL15 (Li-ion 20) that stopped working after less than 1 year of moderate use.  (Of course these use a different type of cell.) As it was out of warranty and there has been no recall on this series it has so far been a complete loss. (I have retained it hoping that some recalls one day would be expanded to the series of this battery; there has been some due to overheating.)

With respect to the 18650 cells, a US supplier I have used for some work related specialty batteries has high discharge rate versions from LG and Samsung at 2500mAh capacity at about $8 each, https://www.powerstream.com/18650-high-discharge-rate.htm, however they specifically warn against using multiple cells in series: "If you need higher voltage by putting these in series they can be very dangerous, and must be assembled in a qualified factory with the appropriate matching and balance charge circuitry. We can do this or have this done for you. We sell these individual cells with an understanding from the customer that they will not be used in series.". I recall from the days of using NiCd cells, when incorporated into a package, one should use matched cells. I wonder if this is also the case with Li-ion cells, or if the internal circuitry in the battery package would take care of any mismatches in capacity?

If you recorded any images during the recelling, I would love to have a peek.

I just upgraded my bike light to a Niterider version, the Lumina 900 (nominal 900 lumen for 1 hour at full blast!) , which appears to also use an 18650 cell at 2600 mAh. What is notable though is that it's big brother's Lumina 1100's capacity is rated at 3400 mAh nominal and uses the same battery size. Wired replacement cells are available from Niterider but at about $25 per cell. After some searching the web, I see non-wired single 18650 cells with claimed capacity up to 3.5Ah but at 2x the price of the Samsung cells above.

[Seapy]

Interesting,  I wondered about the way the cells are wired to the PCB, There are four connections, two positive and two negative.  I have taken a few photo's of the cells, original and replacement.  I don't think  these cells are special with regard to heavy discharge, the wires are 1.2²mm. That said, effectively they are 2.4²mm, being doubled up.  I  think that's to minimise resistance, rather than carry heavy current.

I will post some pix later, am up to my armpits in my daughter's car's brakes right now...

[Øivind Tøien]

A quick Goggle on "li-ion cell balancing circuit" brought up several interesting documents on this topic from major manufacturers of power management ICs.
Better get those brakes fixed first!

[Erik Lund]

........
One other factor... Despite having 're-calibrated' my rebuilt battery, it's still showing life 4 in the camera. .....

How did you attempt to recalibrate the battery? I would think this was not possible for obvious reasons,,, ie Nikon doesn't offer rebuild battery packs,,,

[Seapy]

How did you attempt to recalibrate the battery? I would think this was not possible for obvious reasons,,, ie Nikon doesn't offer rebuild battery packs,,,

On the charger. Press the calibrate button.

Having read two quite comprehensive PDF's and read the Wikipedia entry on cell balancing I now have a much better understanding of the inner mysteries of Li-Ion batteries.  More stuff to read but getting there...

Within the EN-EL4(a) battery there are four connections between the PCB and the  cells.  It's entirely possible for the PCB to monitor and balance the Voltage of each cell and to balance both the charge and maintain equilibrium between the cells during discharge.  Apparently it's vital that 'lazy' cells and 'hyperactive' cells are equalised, because the lazy ones can fail completely if their Voltage goes below a certain level (even by milli-Volts) and  the hyperactive ones can explode due to gassing pressures within the  cell when they get hot.  This is why taking Li-Ion batteries on aircraft which can have variable cabin pressure can be dangerous. The variation in cabin pressure can cause the cell to rupture and burn, even if the  cells are not being charged.

From what I have read it is easier and more cost effective to micro manage variable cells than it is to manufacture the cells to tolerances which avoid this issue, also the cells change their characteristics over their lifetime, they may all start out matched but end up way out of spec by the time they have been charged and discharged multiple times.

I wish I had photographed the PCB and been able to check the circuits and IC markings so I could better understand precisely how Nikon have constructed the circuits.  I think the 'life' shown in the camera menus of battery condition is a crude counter of how often the battery has been charged.  I have used many power tool Li-Ion batteries to death.  They seem to die suddenly, which leads me to think that there is a charge or current counter in the circuitry which one day says "enough".

I do have another EN-EL4a which is on life 4 so I may repeat this, but it isn't really cost effective compared with buying non original batteries. When adding the safety concerns I seriously wonder if it's worth it.

I am going to contact the German supplier of my replacement non-original EN-EN4a batteries today to see what they say about the apparent failure in only moderately cold conditions.

[Erik Lund]

I thought you meant a factory reset, my misunderstanding.


Is there no electronic circuit in the battery pack?

[Seapy]

Is there no electronic circuit in the battery pack?

Yes a small PCB with about half a dozen chips, mainly I think to achieve the cell levelling during charging and discharging, as I described.

This is the external battery connector side of the PCB with two or three IC visible, there are others on the other side of the PCB. As you can see it's pretty compact.



This is the other end of the cells, these are the Samsung 2600mAh cells.  I wrapped red tape around the positive (+) end to make it clear which was which.  In the end I had to remove the tape because it was preventing the casing from closing properly.  It's that tight for space.

[Øivind Tøien]

Thanks for the images and description above. Even without any immediate plans to venture into recelling, it is nice to have a better understanding of how it works.

[Seapy]

I think that sums up how I feel too.  I'm not rushing to do another, yet not ruling it out completely yet.

I feel reasonably happy about the job I have done, my biggest concern is the close tolerances around the PCB and the connector socket.  Things are a little... crushed in there.

As for the cell balancing I am pretty sure that's why they used four connections to the three cells.  It allows the circuitry to balance the charging and usage over the three cells.  From what I have read, obtaining 'balanced cells' is a misnomer, they may all start out balanced but over time and use, the characteristics of the individual cells will vary.  It's due the the chemistry in the cells being very slightly variable and it seems unavoidable unless the cells were much more precisely made, which would inevitably increase prices exponentially, raising questions about the cost effectiveness.

I have eMails my German supplier and asked their opinion about the effect cold had on their batteries.  Await their reaction...

[Seapy]

Øivind, sorry I didn't thank you for the tip to do a search on Li-Ion cell levelling, that was key to a better understanding of the use of multiple Li-Ion cells in power packs.

Thank you very much for that.