Battery Power Supply

[rickmcinnis]

I had 3 of these for the 3.3 and the 5 volts rails: http://www.batteryspace.com/lifepo4-pri ... d-dgr.aspx

Only SLA was the OPTIMA for P4. And I used an AUDIOWIND for it. I have a BATTERY TENDER brand device but it would charge it much too high.

Four of the smaller A123s were used for the P24 12 volts rail.

My chargers were four of the recommended AUDIOWIND regulators. They are working fine.

The batteries show voltage but cannot do any work.

I am mystified!

All of this was working fine before I started the SOEKRIS project. I figure if I am away after replacing this stuff I will disconnect the chargers from the batteries and let them SIT.

Any tricks for re-awakening a LiFePo battery?

[jkeny]

I had 3 of these for the 3.3 and the 5 volts rails: http://www.batteryspace.com/lifepo4-pri ... d-dgr.aspx

Only SLA was the OPTIMA for P4. And I used an AUDIOWIND for it. I have a BATTERY TENDER brand device but it would charge it much too high.

Four of the smaller A123s were used for the P24 12 volts rail.

My chargers were four of the recommended AUDIOWIND regulators. They are working fine.

The batteries show voltage but cannot do any work.

I am mystified!

All of this was working fine before I started the SOEKRIS project. I figure if I am away after replacing this stuff I will disconnect the chargers from the batteries and let them SIT.

Any tricks for re-awakening a LiFePo battery?

I've never seen a LiFePo4 battery with 3.2V but no current output which is what you seem to have - it sounds like some internal fault. AFAIK, the type of chemical reactions in LiFePo4 that cause damage are not reversible

Keeping them at a constant regulated 3.3 or 3.4V is absolutely no problem - my devices have been doing this for 5 years or so. Only two things will damage these batteries over-charging or the more probable discharging below 2V down to 0V

Are all your LiFePo4 batteries like this, including the A123s?

[nige2000]

I had 3 of these for the 3.3 and the 5 volts rails: http://www.batteryspace.com/lifepo4-pri ... d-dgr.aspx

Only SLA was the OPTIMA for P4. And I used an AUDIOWIND for it. I have a BATTERY TENDER brand device but it would charge it much too high.

Four of the smaller A123s were used for the P24 12 volts rail.

My chargers were four of the recommended AUDIOWIND regulators. They are working fine.

The batteries show voltage but cannot do any work.

I am mystified!

All of this was working fine before I started the SOEKRIS project. I figure if I am away after replacing this stuff I will disconnect the chargers from the batteries and let them SIT.

Any tricks for re-awakening a LiFePo battery?

never seen anything like that
not even from the cells i abused

[rickmcinnis]

It is very strange.

The small A123s would show no voltage at all (12 v rail).

Live and learn, I guess.

I have left them off/disconnected from the supplies with the hope they might need a rest! I know I am dreaming.

I originally was using the A123 26650 for 3.3 and 5. Learning how to use them resulted in all kinds of abuse which was only made worse by the abuse they suffered with my tab attachment. Those had an early death but that was not a surprise. I thought they might not be up to the task so I got the bigger ones and I did think the sound was better, more effortless and all of that kind of thing. I will get those again.

From here on in I only get 26650s with the welded tabs. Have some on the way for the 12 v P24.

[jkeny]

One thing to be careful with, Rick is when connecting 4 batteries in series to get 12V (4*3.2 = 12.8V), is that the current draw from each battery will differ when outputting 12V to your device & so they will become unbalanced i.e some falling below 2V while others stay at 3.2V. You will end up with a dead battery (current drawn from a battery below 2V can damage it & kill it) - if you just attach 12V PS to this battery string in the hope of trickle charging them & preventing this unbalance, it won't work as the batteries will also be charged up in an unbalanced way (the 12V charging isn't split evenly between batteries in the string) - so some will over-voltage & also be damaged.

Was this your charging setup that maybe caused your battery failures?

Nigel had a scheme of using a trafo with 3 independent windings rectified & regulated down to provide independent 3.3V supply to each battery in the series which I think may work but I haven't tried it & I'm not sure if he has?

Another way of ensuring proper charging (but not trickle charging) is to wire up a multiway switch in such a way that it switches the batteries into parallel connection when off & series connection when on. Then a 3.3V PS can charge the 3 batteries in parallel when not running the audio device.

[rickmcinnis]

jkeny,

Yes, that is exactly what I did for the 12 and the 5. Probably not as critical there?

Nigel's idea sounds like a sound one but what a pain for this rail that hardly does anything - when my computer is as I intend it, there is a video card for the monitor. The video card is powered by the CPU battery (which I hope to get this much of the system back working this evening).

Almost makes me think I might use one of the PANASONIC SLAs I used to use with my Gain Clone amps and not worry with all of that. Just cannot imagine this rail making much of a difference sonically. Of course, I would want to be disabused of this notion if that is needed!

I have four A123 26650s on the way.

I had considered them overworked for the 3.3 and 5 volts rails. Should I consider using these and not bothering with getting three more of those monster things? If 3.3 needs it easy enough to parallel two of them since I will have four. Which rail requires the most current these days?

jkeny and Nigel, please repeat to me what you are using for the 3.3 and 5 volts rails?

Those charging schemes sound good but from memory I think even the 12 volts rail needs trickle charging or the computer fairly quickly turns off.

Thanks and take care,

[nige2000]

Nigel had a scheme of using a trafo with 3 independent windings rectified & regulated down to provide independent 3.3V supply to each battery in the series which I think may work but I haven't tried it & I'm not sure if he has?

ive many pots on the stove

yes it does work
in a kind of sledge to crack a nut type of way as the cells discharge uneven the 3.3v supplies have to be able to work harder than you would think

i have it on a constant charge and drain these last couple of weeks so should have fairly solid info soonish

[jkeny]

jkeny,

Yes, that is exactly what I did for the 12 and the 5. Probably not as critical there?

Well, if I understand you, that charging scheme is your problem & most likely what killed your batteries. How soon this happens probably depends on the current draw - see Nige's note about the fact that some of the batteries in a string of batteries connected in series have surprisingly high current draws

Nigel's idea sounds like a sound one but what a pain for this rail that hardly does anything - when my computer is as I intend it, there is a video card for the monitor. The video card is powered by the CPU battery (which I hope to get this much of the system back working this evening).

Almost makes me think I might use one of the PANASONIC SLAs I used to use with my Gain Clone amps and not worry with all of that. Just cannot imagine this rail making much of a difference sonically. Of course, I would want to be disabused of this notion if that is needed!

I have four A123 26650s on the way.

I had considered them overworked for the 3.3 and 5 volts rails. Should I consider using these and not bothering with getting three more of those monster things? If 3.3 needs it easy enough to parallel two of them since I will have four. Which rail requires the most current these days?

jkeny and Nigel, please repeat to me what you are using for the 3.3 and 5 volts rails?

I'll leave Nige to answer that

Those charging schemes sound good but from memory I think even the 12 volts rail needs trickle charging or the computer fairly quickly turns off.

Thanks and take care,

Yes, so Nige's charging scheme is probably the best available solution, at the moment, for series connected batteries. What his scheme is doing is using a floating regulated 3.3V to charge each battery in the series - so for 4 batteries in series giving 12.8V, a transformer with 4 secondary windings, 4 rectifier bridges & 4 regulated supplies are needed. As you say - a bit of a pain!

There might be an easier way that avoids the transformer & rectifier bridges if an isolated regulated supply is used. Something like this might do the trick?

[randytsuch]

I built a headphone amp with 10 batteries, two sets of 5 each in series, one set for positive and one for negative.

I charge them with 18V across each set. I have two 18V walwarts for this. I know that's a little high, but it works fine.

So I know this is not optimal, but I'm not building 10 little regulators to charge the 10 batteries.

If a pack goes bad in a while because of this "abuse", I'll buy another set of batteries and replace them.

BTW, I normally use clones that cost about $7 each, and maybe that's why I don't worry if I have to replace a battery every now and then.

[jkeny]

Sure, Randy, it's a personal choice although the isolated voltage regulators I linked to, are not expensive
I was just outlining what probably caused Rick's battery failures & the issues associated with charging batteries in series.