Balance at the cell low voltage limit

Dan - that's a good idea to tap at the cellog cable.

I am a bit leery of closing the lid completely; as the converter get pretty warm, more than 100F, cooking the closest battery pack relative to the cooler pack behind. I have a 2x4 propping the lid open for the time being.

that siren from RShack is quite obnoxiously loud, but for <$6 each; quite cost effective.

 

Dan - no, what 6 in plastic tubing ? what is the tube diameter ? is it part of the kit ?

I am also seeing the pack is very well balanced, around 2-4mv overnight delta; without the DB8 balancers.

What would you say the DOC range between the alarm range of 2.8v and 3.75v ? 20%-80% ?

I also plan to beef up the AC cable, I think it needs to be at least 14 or 12 awg as the cable gets pretty warm during charging.

-Paul

 

Dan - no, my kit does not have this exhaust ducting. I know exactly where the rubber flap is and I was thinking of doing exactly that; but I didn't know that it was part of the kit. Hum, I wonder if Jack would send me the extra parts.

Jack said that the 4" exhaust fan is set to come on at 140F/60C while I notice that the converter's fan come on around 100F. I think 140F is a bit high, especially come summer.

My concern with regard to the DOC is to maximize the pack life. I am finding myself looking for outlet every where I go now.

 

Your concern for battery life certainly has merit. Nobody really knows how long those batteries will last. I don't have the datasheet for the Mottcell batteries, but I can't imagine the Enginer system is configured for anything less than 80% or even 90% DOD (generally both "knees" on the LiFEPO4 curves make up about 10% of the battery so it must be at least 90% since they use voltage cutoffs on both ends of the battery). I think it was Dan that posted earlier he was getting nearly 41Ah out of his pack? They're only rated for 39Ah so...

The Thundersky batteries are under-rated on capacity (that is, they are officially rated at 40Ah nominal capacity though initially they have closer to 45Ah capacity). This is how they are able to claim a larger-than-normal cycle life--they over-shoot on capacity with the expectation that the batteries will degrade quickly. From what I hear on these forums the Mottcell batteries have an initial capacity pretty close to what they claim the nominal (average) capacity is over the life of the battery.

I wouldn't be surprised if the target deep cycle lifespan for the battery was less than 1000 cycles for that high of a DOD% (on par with most LiFEPO4 batteries). There is a reason why the Chevy Volt only uses 50% DOD (granted, they also use much better batteries--and I say better as in higher quality, lower internal resistance, more resilient to cold temperatures, longer life span and able to provide higher sustained power output--but still).

Different balancing techniques can also playing into the cycle life of a battery (remember, lithium batteries are restricted by deep cycles, so you ideally want to keep them topped off at all times).

Andrew

 

I would not trust the Enginer voltage monitoring of the whole pack as a very good cut off. I trust the Cellog monitoring of the individual cell more.

So if we assume that the 2.8v and 3.75v threshold are the knee of the voltage plot; then we are operating between 10%-90%. That would be a bit deep for maximum battery life.

Short of actually integrating the current flow; I think we can change the Cellog alarm LV & HV to more conservative, say 3.0 & 3.6. The Prius pack operates between 40-80% and I believe the EV RAV4 as well and their packs seem to survive quite well.

yes, I agree: charge frequent and often.

 

What's the maximum voltage range for the Mottcell batteries? I was under the impression that the Mottcell chemistry was more similar to other LiFEPO4 batteries like K2 and PSI where the max charging voltage is around 3.65v. Does anyone have a legitimate datasheet on the actual battery used by Enginer? All I found was a certificate by some testing company that certified the battery was safe to transport...

The Prius and Rav4EV of course use Nickle which is a separate beast all together, but the overall concept is the same yes. You generally want to stay off of the voltage "knees" to extend battery life--especially if you aren't doing active balancing (even the best balanced packs will eventually un-balance themselves over time due to temperature differentials and slightly differing capacities).

EDIT:

I found the "datasheet" on the Mottcell batteries, or rather the evaluation of the individual 13Ah cells that make up the larger 39Ah cell used. It says the cutoff voltage range is 2.0v to 3.65v, but that would be 0% to 100%. Additionally, current must be tapered on the high end to get a full charge (that is, the charger has to intelligently back off the current to get a full charge).

I wouldn't get anywhere close to 3.65v if you don't have 100% confidence in your charger shutting off properly and quickly. More likely it should be charging to around 3.55v or less depending on whether or not the charger supports tapering.

I also wouldn't go any lower on the discharge voltage than it is now--the voltage drop on the low end tends to tank quickly and 2.0v is likely the "point of no return" voltage anyway. See the Thunder-sky discharge graphs if curious--note how the voltage effectively falls off the face of the earth as it approaches low SOC.

Andrew

 

Hi Dan,

Typically for Lithium batteries, the 2.0v and 3.65v are the absolute max and min voltages. Damage occurs if the voltage goes above or below them regardless of whether or not current is flowing in/out (unlike other chemistries like Nickle where you can slightly fudge the numbers without too much grumbling from the batteries).

All I was able to find was this:
http://www.enginer.us/img/Mottcell Battery Test Report.pdf

It's the testing report for the individual batteries that make up the larger battery cells. The chemistry I'm sure is the same so the voltage numbers should be correct--obviously capacity can't be trusted since it's not the actual battery in use. Note the taper charge current ratings however.

Here is a link to another LiFEPO4 battery made by PSI:
Phoenix Silicon International Corporation -- 40155F1 Specification

This is NOT the same as the Mottcell battery of course, but it gives some point of comparison. Note how steep the voltage "knee" on the bottom of the battery is. These batteries will have a lower internal resistance than the Mottcell batteries do (more efficient but cost more), so don't draw conclusions from these graphs about how the Enginer batteries work--we'll need an official datasheet from Mottcell to do that. Also the voltage cutoffs aren't exactly the same (every chemistry is slightly different), but you get the idea.

The above datasheet also demonstrates my previous point about battery life cycles--note how the battery comes with about 105% capacity and degrades linearly to yield the nominal (average) capacity. I'd assume this holds true with the Enginer batteries as well so just because you're able to get 40Ah out of them now may not mean they will provide that much capacity over the full life of the batteries (the full life cycle is generally denoted by the ability to retain 70%-80% of advertised nominal capacity).

Andrew

 

did u upgrade to the latest cellog software included on the CD ? I recall Banshee reminding folks to do the upgrade; though I don't recall what problem the upgrade addressed.

 

Dan,

2.17 is the newest and the update I suggested is to deal with the power saving modes on the celllog. If you were looking for the mod to make sure power is also drained from cell 7 & 8. Here is the link to the picture. You just need to put a resistance of less than 100ohm.

RC Groups - View Single Post - CellLog is releasing!

 

Dan,

Use a sharp utility knife to pry the case open. That is what I did. Just do it carefully. Yes I did do the mod already. I just used a piece of loose wires, figuiring that it is at least some ohm value.

The power saving move has couple of options. You can have the device turn off completely which would stop all data collection. What I do is turn the display off after 5 mins. The cellogs on my dash board is set to turn off after 1hr.

Hope that helps,

Chan

 

Yes it was difficult. I had to borrow a nice $300 solder iron from my friends to do the mod. My iron did not product enough heat. When you set the device to turn off the display the data collection continues. It is in the man configuration. I will look at mine tonight when I get home and get back to you on the menus to get to the powersaving modes.