Dendrite/crystal removal

[45/70]

When discharging a nickel battery, this breaks up dendrites [EDIT: actually crystalline formations]. I understand, the slower the rate of discharge, the better it works. My question is, when working with a battery with obvious voltage depression, is it practical to run it down at say 100mAh to 0.9 Volts, charge a bit (say 1 minute @ 400mA), then repeat a few times to speed the process up? Or, is it better to fully charge the battery each time and then discharge? I'm using a Maha C-9000.

Dave

 

[Mr Happy]

Re: Dendrite removal

You should probably forget whatever you read about dendrites. These are a particular problem with NiCd cells and do not happen with NiMH cells.

The instruction manual for the C9000 gives advice as follows:

Battery Rescue Steps

For batteries that do not perform favorably...the following sequence can be applied.

1. Refresh & Analyze for one to three times.
2. If capacity is still low, use Break-In mode.
3. If the step 1 to step 2 shows some capacity improvement (> 10%), repeat Break-In mode for one to three times. If no significant improvement, battery probably at end of useful life.

I'd say you could do no worse than to follow this advice.

There is no requirement to discharge the batteries between each Refresh & Analyze or Break-In cycle, but some people like to do it. It will take longer that way of course.

 

[45/70]

Re: Dendrite removal

I'd say you could do no worse than to follow this advice.

Ya, well, my goal is to do better, not worse. By the way, I should maybe have been more specific, but these are NiCds, and as far as I know, NiMH's have them too, just not as easily formed.

Thanks Mr. Happy. :thumbsup:

Dave

 

[Mr Happy]

Re: Dendrite removal

With NiCd's, some recommend a deep discharge right down to zero volts (short out the cell with a flashlight bulb and leave it for several hours until fully discharged). The C9000 will stop at 0.9 V and will not completely drain a cell.

You could try the deep discharge followed by a break-in cycle.

 

[SilverFox]

Re: Dendrite removal

Hello 45/70,

When trying to recover a battery, you have to ask yourself a few questions.

How much is my time worth?

How much time am I willing to invest in the project?

How much does a replacement battery cost?

Are there advantages to replacing the battery (i.e. going to low self discharge batteries, or batteries of higher capacity), or will I be better off recovering this one?

Sometimes you can recover a battery and it works reasonably well. Other times you are simply "spinning your tires."

Cells fail in a number of ways. As a cell ages, its internal resistance increases. This is due to separator and electrolyte breakdown. There is nothing you can do to reverse this.

If a cell has been abused, crystal growth can form. This can be broken down through slow discharge rates, followed by faster charge rates.

If a cell has been stored for a long period of time, you can get dendrite growth that basically shorts the cell out. You can blow those dendrites out, but you still have a hole in the separator, where new dendrites will begin immediately forming. The basic symptom of this is high self discharge. Separator damage can not be fixed through charge/discharge cycling.

The key to battery recovery is to recognize why the battery is not performing up to its potential, and then realizing if it is something that can be recovered, or if the battery is simply ready to be retired.

For example... I was given a bunch of 12 year old 1200 mAh D NiCd cells. These cells are new in the package, but have been in storage for the last 12 years. I am viewing this as a "project" and am not worried about the cost.

The first 8 cells of the batch have yielded 6 that are performing within 90% of their initial labeled capacity, and they are holding voltage reasonably well. The other 2 came in at less than 50%.

As far as the amount of time I have invested in this "project," it is far more than the $8 per cell that the cells cost 12 years ago. I have limited my efforts to 50 charge/discharge cycles per cell, including forming charges and some slow 40 hour discharges. Mind you that I was not actively involved during the whole process, but I did keep an eye on things all the time the cells were charging, and the discharging tied my equipment up keeping me from being able to do other projects.

I can tell you that, if after 50 charge/discharge cycles you don't see good improvement, the cell is most likely "crap."

Tom

 

[45/70]

Re: Dendrite removal

Thanks Tom.

Yes, I realize this isn't a practical endeavor. It's just something I'm playing around with. I have a few NiCd AA's that are about as old as your D's, I suppose. Some are 850mAh and the others are unknown, but purchased before the 850's.

I just thought I'd see what can be done with them. They've sat for long periods, been charged up and used occasionally, but I'm sure they have dendrites/crystals, whatever. The voltage under load drops big time, a classic example of voltage depression. They all hold their charge very well, better than a standard NiMH, for sure.

One of the unknown capacity cells I'm working with, was charged up @400mA and then checked with the C-9000's discharge function @200mA. Result was 230 mAh. After the discharge (then shorted out for an hour or so) it was formed at 50mA for 16 hrs on a NiCd charger. Ran the discharge function with the C-9000 again, 310mAh.

I got to thinking (Uh oh, that's usually a bad sign ), "this is taking forever". That's when I came up with the idea of the short charge, then discharge idea. After repeating the procedure about 10 times, I charged up the cell @400mA, waited a couple hours and then did another discharge test on the C-9000. The result was 590mAh. Hummm.

One thing that led me too this procedure was, as I watched the cell discharge on the C-9000, when the voltage readout dropped below 1.00 Volt, I could see the voltage vary up and down, eg. it would go down to 0.91 volt and then bounce back up to, say, 0.97 Volt, then slowly go back down again. This occurred several times before 0.90 Volt was reached. It may not be the case, but I theorized that maybe, this could be the dendrites dissolving, causing the voltage fluctuation to occur.

I realize that a 1200mAh AA NiCd can be had for under a buck so, again, this is just to see what happens. I'm not too sure that this procedure is good for the cell, hence my starting this thread. It does however, seem to work, so far anyway. On the other hand, what's the point? Ya, well, anyway.

Thanks all, :thumbsup:

Dave

 

[SilverFox]

Re: Dendrite removal

Hello Dave,

OK, now that we have established that you are just "playing around..."

Instead of shorting the cell out, hook it up to a low wattage lamp. You can watch the lamps output fluctuate as the crystals are broken up. With NiCd cells, you can discharge all the way down.

When the lamp gets real dim, unhook it and let the cell sit for around an hour. You should find that the voltage will recover some. Now, hook up your lamp again and let it discharge again. It should take less time to discharge this time.

When you have completed the second discharge, unhook the lamp and once again let the battery sit for an hour or so. Once again hook up your lamp, and this time pay attention to the glow. If it smoothly dims down, you have done well. If it is still fluctuating, you still have a ways to go.

Now you have performed 3 discharges in a row. It is time to "exercise" the cell and fully charge it. You should charge at 1C. Once again you need to go through the series of discharges paying attention to the fluctuations in the glow on the last discharge. When you get a smooth dimming you can then do a 0.1C charge for 16 hours followed by a 0.2C discharge to see where you are at.

Have fun.

Tom

I forgot to add that all of this charging and repeated discharging only breaks up crystalline formations. To deal with dendrites you need to apply high voltage and/or high current in bursts. This can be dangerous, so be careful.

 

[45/70]

Re: Dendrite removal

Ah, dendrites vs. crystals. Mr Happy was correct, NiMH's don't suffer from dendrites, only NiCd's. Due to the symptom (voltage depression as opposed to self discharge), I'm sure I'm dealing with crystal formation with these NiCd's. Their history of being stored in various states of charge for lengthly times supports this, as do my observations. I have done the high voltage trick for dendrites in the past, but not too sure whether the cure wasn't worse than the problem! I'll change the thread title to reflect both.

Tom, thanks for the idea as to how to properly remove the crystal formation. I'll give it a shot. It does appear that there is no way to save time (which is what I was trying to do) "rescuing" battery cells. It seems to always come down to the hard way, or no way! :sigh:

Thanks again for your input, Tom. :thumbsup:

Dave