Author Topic: Sanity Check (Read 14342 times)

Madscientist267 · « **on:** May 21, 2011, 10:08:35 AM »

Seriously.

Thanks to Peukert factor, full capacity can't realistically be expected. This is understood.

But I have a 'brand new' SLA here that is (by my calculations) under-performing.

WTF, I ask you?

Thanks for the insight ahead of time, because I'm apparently off on this, or SLAs are just clearly pieces of $#!+ from the git go.

My FLAs don't seem to behave this way.

Steve

Madscientist267 · « **Reply #1 on:** May 21, 2011, 12:04:57 PM »

According to 'average' performance (generic numbers given to a Peukert calculator), the time to 100% discharge SHOULD be 4.5 hours.

So 50% would be 2.25 hours...

Right?

Steve

dnix71 · « **Reply #2 on:** May 21, 2011, 07:30:06 PM »

C/6 is harsh, unless it's a battery that was designed for that. The forklift batteries I replaced last week had two ratings on them. c/5 and c/20 and those weren't much different. But these batteries were GNB Absolytes. The non-GNB brand replacements, 6v 220AH, were $250 each, wholesale in Miami, so there is a price to pay for quality.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=16838&start=0 Kind of a long winded thread on this. If you don't go below 80% DOD, at c/5 or c/6 there can be a LOT less usable AH than what's printed on the label. But this only applies to SLA's, not lithium or nicads.

Madscientist267 · « **Reply #3 on:** May 22, 2011, 01:28:16 PM »

What I don't understand is why the SLAs behave this way, and FLA, while still affected, seems to be a lot more robust.

Been reading up on it, and from what I can tell, FLA has a lower effective Peukert number than SLA does.

Damn shame, because I'd rather use them, with that whole liquid acid thing being a factor, but their performance always seems to disappoint.

I've come to the conclusion too that they should not be brought as high up as FLA in terms of charging voltage, and especially not floated as high...

I've recently trimmed the dump controller on the new battery back to 13.7V, and although I realize I'm not hitting 100%, the pros of doing so seem to outweigh the cons.

I learned that watering an SLA doesn't get me very far, and so my next strategy is to avoid the gassing as much as possible instead.

My conclusion about SLA at this point is that they are much more finnicky in many ways than FLA, and so as (I believe) DamonHD put it, "they need to be babied". Pushing for that final 10% of charge seems to do more harm than good most of the time, and maybe should be considered as the 'equalization' for SLA, only going above 13.8V occasionally.

I'm also finding that the constant current mode should be extended somewhat, providing gassing protection toward the top of the charge. Once the threshold is reached, it seems as though they like to be let go and returned to a lower constant voltage charge than FLA does.

Then if that wasn't enough anal-retentiveness for one chemistry, resting appears to be key to obtaining maximum capacity from them. I think this is probably due to the bubbles being trapped in the matting, unable to escape as easily, resulting in an insulating effect on the plates. Once they have rested, I've noticed that the internal impedance is much lower, and they handle loading better.

The part about that which I can't seem to figure out, is this appears to be an issue on discharge as well. They 'grab' the charger better and show a lower impedance if they are allowed to sit after being discharged. This is most apparent after a heavy load. Light loading (<C/20) doesn't seem to bother them as much in this area.

I want a portable solar 'plant', but with SLA I can't get the capacity I need, and with FLA there are other problems to consider, namely the liquid spill potential. I've considered Lithium, but my preliminary encounters with smaller lithium phosphate didn't exactly leave me creaming. $:-\$

NiCd, maybe. If I could get my hands on some really high capacity cells that are all in good shape. Problem there is, there's almost twice as many cells for the same nominal system voltage, and that brings it's own set of problems to the table.

NiMH is out of the question. They are crap.

So my options here are limited... Damn this is frustrating!

Steve

dnix71 · « **Reply #4 on:** May 22, 2011, 03:03:38 PM »

Being starved for water is what messes up SLA's. Allowing them to sit a short time before recharge allows the electrolyte to even itself out. I noticed that when recharging my ebike.

I would love to have them, too, but they are a lot more expensive and less tolerant of abuse. They are 2 or 3 times the price of FLA's of the same capacity here.

I would love to be able to put a big FLA on my bike because of the extra range it would get me, but the water weight and sloshing acid isn't something I want to risk. If I had 3 or 4 wheels on the ground, golf cart style, FLA would be the way to go. Sam's club has $80 6v 200ah FLA's

Madscientist267 · « **Reply #5 on:** May 22, 2011, 04:32:26 PM »

Is AGM susceptible to the same problems as SLA?

Water, Peukert, etc?

I need a robust battery that doesn't have physical dangers associated with it, and can handle the occasional abuse from extended floating and over discharge. Its not something I regularly allow, but does happen.

I know its all one big compromise, and so I am trading weight for capacity and such already. Cost is af factor too, but I'd rather pay a little more up front and know it will be there when I need it than cheap out and get 'caught in the dark'...

Where to from here? I could do a bank of SLA, I feel that this isn't really escaping the problem. And knowing what I know, I'd still have trust issues with them.

Large capacity lithium phosphate? $$$

And definitely not until I get a good understanding of them from someone with a lot of experience with them. Like I said, my first encounter with them, while only a set of 4 "AA" lithium phosphate (yard light variety) wasn't exactly impressive...

Steve

Simen · « **Reply #6 on:** May 22, 2011, 06:49:23 PM »

I think your first meeting with LiPo wasn't the best one could have...

All Lithium rechargeable batteries needs the most 'babysitting' of all batteries, and that's why Li* battery packs should have a BMS, so the user don't have to babysit.

Keeping on topic; i have a 36V, 750W electric motor on my bike, and the first year i used three 12V, 18Ah SLA's. In theory, i should have around 50 minutes of runtime at full speed, but in reality i had more like 15 min before the LVD in the motor controller cut the power.
Now, for the last two years i have used a 36V, 16Ah LiFePo pack (with BMS), and i get around 45 minutes at full speed, and there's no degradation of the performance yet.

And the weight! 1/4 of the SLA's!

So, SLA's really doesn't like harder load than their C/20 rating...

Edit:
Found the spec for the SLA's i used on my bike (12V, 18Ah), and the C/20 rating are down to 10.5V, which are much lower than both me and any solar charger controller would be comfortable with. According to the discharge chart for my battery, discharging with the C/20 value down to 11.7V (which i consider 100% discharged) would take 15hr (not 20hr)...

Madscientist267 · « **Reply #7 on:** May 23, 2011, 12:25:23 AM »

Don't confuse LiPo with lithium phosphate...

LiPo (3.7V, prismatic, like in a cell phone, or cylindrical like in a laptop) is an awesome technology except for the whole setting itself on fire issue when they get irritated. I won't use that particular chemistry for very much because of that very thing... Shame, too - they are efficient and can deliver the goods when called upon to do so. I currently only employ a single cell in one application for which it was not designed, and even though the BMS is in place and the power levels are low, I still reserve a small amount of paranoia in its use. $:-\$

Lithium phosphate (3.2V, like in some yard lights) is exponentially safer, and doesn't necessarily require BMS (as a safety mechanism), but doesn't have as high of a power density. This in and of itself would not be a problem for me, but in the AA version I played with, they acted funny. BMS is probably a very good idea still, since even though they tolerate severe over charge/discharge (in terms of safety), it apparently still has devastating consequences to the cycle pattern, and most likely to overall life span as well.

I'm not against using them, as I have heard good things, some of which sound very promising, but my personal introduction to them was dismal. As I eluded to, I probably brought that on myself.

That's why I want to hear more from people like you that have used them extensively. They are a bit on the pricey side to be acquiring expensive high capacity cells, only to kill them with inexperience.

With that said, cost effective wise, what am I looking at in terms of differences between a 36AH (ish) set of phosphates, vs SLA with equivalent effective capacity when my discharge currents might be as much as 5A continuous?

Steve

Simen · « **Reply #8 on:** May 23, 2011, 02:41:31 AM »

Quote

Don't confuse LiPo with lithium phosphate...

I'm not, but i think you are...

One are a bit generic when one say Lithium Phosphate, since there's several LiIon (common name) cells that have those elements in them. What mainly makes them different are the material used in the positive electrode. The ones that sometimes catch fire, are the early ones, LiCoO2; most newer LiIon cells uses a different cathode material that makes them much safer.

One correction to my previous post; the cells i use are NOT LiFePo, but LiMnNi (Manganese spindel), like these: http://www.batteryspace.com/limnnirechargeable26650cell37v4000mah10arated148wh.aspx

You say around 36Ah battery and a 5A load, which means you want around 6hr capacity?
A 36Ah LiIon battery will deliver 5A steadily for 6hr while a SLA would need at least 72Ah capacity to deliver the same 36Ah down to what i consider empty (11.7V). An SLA might drop 0.5-0.7V when loaded with 5A, but a LiIon drops maybe 0.1-0.2V

Simen · « **Reply #9 on:** May 23, 2011, 05:05:13 AM »

I re-read your last post, and i see that it's easy to get confused with the abbrevations (i do.

)...
When you say LiPo, you mean Lithium Polymer, and Lithium Phosphate = LiPO4.

You said your Lithium Phosphate was 3.2V, which would probably make them Lithium Iron Phosphate cells (LiFePO4). (Other Lithium cells with different chemistry have different nominal voltages, ranging from 3.2V to 3.7v...)

There's a jungle of Lithium cells out there, and it's a bit hard to get a handle on the technology.

One common thing though; a good BMS makes most of the newer Lithium based batteries a very good, and a very safe power carrier. (although, a bit expensive, as you mentioned...

)

DamonHD · « **Reply #10 on:** May 23, 2011, 12:34:12 PM »

I'm very happy so far with my (small) LiPO4 prismatic cell battery taking most of the daily cycling load off my (much bigger) SLA bank. (Small = ~240Wh, 100% usable; big = 4800Wh, about 50% usable).

Rgds

Damon

Madscientist267 · « **Reply #11 on:** May 23, 2011, 01:39:10 PM »

Simen -

I agree with the confusion part, hence the clarification with voltages and typical uses etc... There are definitely a myriad to choose from.

Yes, the ones I mentioned are iron phosphate, and had been abused by overcharge/discharge before I really got to testing them. It is unclear to me at this point if they had already been significantly compromised in the process; my loose research suggests that it may have. They don't have any BMS integral to the cells, based on their behavior during testing. They will continue to pull current well after hitting 5 VPC, and don't disconnect when taken down to less than 1.5 VPC. Given that their recommended range is between about 2.5 and 3.8V, I'd say there's nothing there protecting them. Since its not a safety concern, they leave it out and rely on a shunt controller in the solar light to limit voltage.

Is BMS on the large batteries typically integral to a multi cell design, or done externally?

I got my attention now that I'm fairly convinced that the lack of capacity was my own doing...

Steve

Simen · « **Reply #12 on:** May 23, 2011, 03:43:17 PM »

To my knowledge, no single LiIon cells come with BMS integrated. (and still looks like a single cell.)
('Battery Managment System' - Doesn't it really lie in the name?!

Usually, a BMS monitors every cells in a serie, for example 4pc of your AA cells @3.2V each in series, making a 12.8V battery. usually, to get a higher Ah, there's several cells in parallel for each 3.2V group. The BMS monitors then the whole group of paralleled 3.2V cells as one cell. Simpler BMS monitors only any cell (or group) for low voltage and overvoltage. Better BMS also monitors the same, but for all cell(group) individually, and provides individual cell equalization etc...

When you're talking about abuse of your cells, LiFePO4 Can't handle more than 4.2V charge, and 2.0V discharge, which actually are rather good for LiIon batteries; many other chemistrys cannot handle much more than +/- 0.3-0.4V i think...

Madscientist267 · « **Reply #13 on:** May 23, 2011, 07:43:27 PM »

Quote

To my knowledge, no single LiIon cells come with BMS integrated. (and still looks like a single cell.)

Many prismatic cobalts do, simply because of the dangers involved. When they're pre-assembled for end-use (such as a cell phone), it's usually a tiny board at the end of the cell, and is where the contacts to the 'battery' are made. It's left out of the others as far as I can tell, even for end use.

I looked over that site, and it looks like they have a few that I would potentially be willing to use (meaning that would fit my needs), but the battery management puts the cost up over the top. Sets of cells are cheaper, but don't contain the management. And if I'm going to put that kind of money into a pack, I'm not going to run it without management... The rest of the system is designed to be idiot proof, I'm not going to make the most expensive part the weak link.

These things really perform to "100%" (cutoff) without ill effect? When I'm around to mitigate, they would get cut off earlier, but as I said, idiot proof. They need to be able to protect themselves from abuse as much as possible.

Part of my goal here is to make a 'portable plant' that I can hand off to my parents (for example, since they go camping regularly) for them to use and test out for me in real world conditions. I mimic as much as possible, but the real world always ends up stressing the components a little further than I seem to be willing to go in a safety-netted environment where the grid is right there to save the day. It's a whole psychological thing, but the result is that my testing doesn't give me 100% accurate data to design with.

I notice they really start to shoot up in price too when there's more than about 20AH (@14.8V) of capacity. The packs that I looked at had BMS built into them, but were upwards of 300 bucks.

I saw a nice barebones kit that are basically just modules made of 4 of the cells you mentioned, and due to the spacer design, can provide a virtual infinite possibility of configuration for a customized pack. They're like $60 each for 16AH @ 3.7V, but of course don't come with management. I could put together a 32AH @ 14.8V pack for ~$436, but the CMS (in this case) is another 50 bucks.

The pack modules:

http://www.batteryspace.com/limnni26650battery37v16ah592wh40aratewopcb.aspx

The controller I would need:

http://www.batteryspace.com/CMBfor14.8VLi-IonBatteryPack16Alimited.aspx

Is it really worth it?

Almost $500 for 32AH?

Steve

DamonHD · « **Reply #14 on:** May 24, 2011, 01:38:03 AM »

Consider this sort of "I'm really an SLA" design with integrated BMS:

http://lifepo4.info/products.html

When I have some spare $$$ to rub together I'm keen to try some out. All they really need is a constant charge voltage applied, so I am told, and they will do the rest internally, if need be.

Zero maint.

Rgds

Damon

Simen · « **Reply #15 on:** May 24, 2011, 02:41:25 AM »

Quote

Is it really worth it?

Considering that LiIon can be safely discharged down to 80%, and discharge/charge cycles are at least 4x to 5x that of SLA, the price does even out. (To get the same useable Ah and lifespan with sla's, one have to buy approx. 2x Ah of sla's, 5 times during the same lifespan of similar LiIon.)

But it is a lot of money at once...

The controller you linked to have max 16A discharge, and the ability to charge and discharge at the same time. If you are ok with 10A max discharge and charge/discharge separate; a $15 controller would be fine:
http://www.batteryspace.com/Customize-Polymer-Li-Ion-Battery-Module-14.8V-9Ah-133-Wh-7A-rate.aspx
Both of these have cell balancing, which i consider important for LiIon's lifespan.

Edit:
And as Damon says; zero maint.

Madscientist267 · « **Reply #16 on:** May 24, 2011, 04:16:01 AM »

I'm fine with 16A limited discharge; the meter box is fused at 8A-SB to protect the 10A shunt.

My maximum average discharge continuous would be in the range of about 5A. That's running my (currently) most efficient laptop, a hacked all to hell Peltier cooler, my 'tent ceiling fan' (more on that in the diary thread later), and charging a cell phone all at the same time.

I'm avoiding anything above 36-40AH as this would be difficult to recover in any reasonable time with the panels I have (60W, 45 typical). There is DEFINITELY one BIG plus to running Lithium - I don't have to worry about partial discharges sitting any length of time like I do with lead acid.

I want the higher capacity however so that when there is a drain, and no real usable light (cloudy), I can still have reserve left over to do some odds and ends. The cell phone comes to mind.

Ok so now that I think you guys have helped me decide that Lithium may very well be the way to go, where's the cheapest place to get them (with the controllers) to get max bang for me buck?

Is 'batteryspace' the way to go here?

Steve

Simen · « **Reply #17 on:** May 24, 2011, 05:01:46 AM »

Cheapest?
There, i cannot help you any more than Google can...

But i think there's cheaper places than Batteryspace when it comes to the cells, though their controllers does look nice...

(My 36V, 16Ah e-bike battery from Goldenmotor consist of 40pc 26650 cells and a 36V (10cell) pcm controller (30A cont, 60A peak) in an alu-case, costs $395 - but; they offer only air freight @ $135, so...)

Madscientist267 · « **Reply #18 on:** June 20, 2011, 01:27:50 AM »

Little update here...

I haven't been able to do the lithiums yet, but have learned a little more about the SLA issue.

It's definitely the discharge rate. At ~C/10 discharge, I get about between 7 and 8 AH out of the battery that prompted this poll.

SLA just isn't as robust as FLA when it comes to higher discharge rates.

It's within reason however knowing that the battery sat for several months before I got it, and factoring peukert into it, I won't complain.

Charge rates and finishing/floating voltages are also very key, not just charge to charge, but overall lifespan!

Oh well, it will have to do for now, but I appreciate the info and responses. Lithium is coming as soon as I have the cash to throw at it!

Steve

Norm · « **Reply #19 on:** August 12, 2011, 11:25:44 AM »

One of these days I'll have to do a study on some of my
NiCads....right now I'm just pedaling and charging my DoubleStuff
2.6 Amp/hr 14.4 volt batteries only have two right now that I put
20 watt/hrs. in each. after I have let them rest for about a half hour
they are reading 16.6 I'm anxious to see how they hold up, maybe see
how many 1/2 holes I can drill in wood.....
In the meantime also making a charging outlet that I can plug 5 of my
dual twelve 2.6 amp/hr NiCads in at a time.
Just having a terriffic time kiitchen door open and keeping cool with a ( was 68,
now 72 degree partly cloudy day)
Have fun !
Norm.

News:

Poll

What run time would you realistically expect from a typical SLA at C/6 rate?

Author Topic: Sanity Check (Read 14342 times)