Batteries

The cells themselves are fairly standard 6.5Ah 1.2V NiMH units, but in a special modular form-factor. As far as I know, there's not that much special about them, apart from much more rigorous quality control: you can't afford to have any faulty cells when you wire 168 cells in series.

The main trick for longevity is that the car only uses around 40% of the capacity, at most. The battery never gets charged above approximately 80% and it's never discharged below about 40%.

The control computer carefully monitors and controls the temperature, voltage and current of the pack as a whole, as well as individually monitoring the voltage of each of 14 pairs of battery modules.

And if necessary, the control system will subject the battery to a very occasional conditioning cycle. I've never seen this happen myself, but others have noticed it.

 

I have 2 questions: 1) Are the batteries "no memory"? 2) It then is normal for thr battery graph to read on the lower side & never reach the top of the graph? Thanks.

 

1) Not sure so I'll let other chime in

2) The battery graph really shows 40% charge as "empty" and 80% charge as "full" since the car won't allow you to gow below/higher then that. This is a good thing. I've been reading about the insight stick shift on greenhybrid.com and the batteries seem to fail on it because people drain/recharge them to much. Having a computer take care of it seems a lot better for longevity and the Prius has been shown to handle >150.000 miles with no detoriation of battery power.

 

i think i heard once that NiMH don't have to be calibrated by charging up and all the way down and all the way up (aka no memory). Laptop batteries do but they aren't NiMH

 

The batteries are different due to the form factor. This does two things: It give the batteries more current capability and lower losses than regular cylindrical cells. It also allows for air flow around each cell, so the batteries can be kept cool and more uniform in temperature. Being quite thin makes them much easier to cool, and the sides have bumps to keep them spaced a bit apart so air can flow.

They do have memory, and that is the purpose of the conditioning cycle that changes the charge level set point sometimes. I have seen this at least 3 times over tha last year on my car but may not notice all of these.

The battery memory conditioning method is patented, and the patent numbers have been posted here before so you may want to do a search.

Besides the mentioned battery control, cell measurements etc, it is easy to see with a can view that the batteries current usage is very tightly controlled with temperature. The batteries almost shut down when they are either too cold or too hot, and this is also probably a big factor in extending their life.
They like to operate around room temperature. This is why the EV button some of us have added will not function when the weather gets cold enough.

 

NiMH cells don't have the problems associated with NiCads. Yes, they're nickle-based cells and the have some of the same characteristics; but there's no perceptible memory issues with NiMHs. That's the reason they don't need to be deep cycled prior to charging. The Prius' controller makes adjustments to it's predicted state of charge (SOC) according to the battery management system.

The prismatic form factor, battery pack construction, and cooling allow for only a 1-2 degree © differential across the entire pack. Ambient temperature is a primary factor in battery life. Along those lines, I've heard the 40% SOC swing number (from 80% to 40%) on here a lot; but I know the NiMH cells can't swing that much and produce the cycles they do. I suspect the SOC swing is 10% max (maybe 90% to 80% SOC).

 

-

 

Do read the owner's manual when you get it. Note that you *can* drain the main battery in various ways:

1) leaving the car on ("Ready") but in Neutral (it doesn't re-charge the main battery while in N)
2) leaving the car in IG-ON (the engine will not start while in IG-ON and so cannot recharge the main battery)
3) leaving the car on and in Park until it runs out of gas
4) driving the car until it runs out of gas, then turning it on again and driving on battery alone.

Any of these will eventually drain the main battery, and you will be hosed. "Hosed" here means that the main battery may not be damaged, but the car will have to be towed to a dealer to get recharged using a special charger.

 

Here is a copy of a post by Wayne that describes the battery conditoning sequence. I can't find the post but I had this copy because I sent it to one of my non-Prius engineer friends. Everything that follows was written by Wayne about a month and a half ago:

QUOTE
Yes, I can speculate, I just hope I can make it comprehensible when
I do. I think that it is very possible that she was in the midst of
clearing or cleaning up what Toyota calls an HV Battery `Discharge
Memory Effect.' There are two kinds of `memory effects;' `Charge
Memory Effect & Discharge Memory Effect. Both of these memory
effects are relatively rare but, they do happen and Toyota has done
some really nice work first, to detect them & then, to clear
or `clean' them up. Let's take a look at what causes them & how
they are `cleaned' up.

CHARGE MEMORY EFFECT
Charge memory effects are caused by repetitive discharges before the
HV battery receives a `full' charge in respect to its SOC upper
limit. When this happens the HV Battery is not able to take on much
of a charge & in fact, will report a full charge via voltage level
long before the normal or appropriate quantity of amps has been
admitted. Toyota has a system that very effectively follows the
voltage vs. amperage I/O & can discern quite effectively when
a `Charge Memory Effect' is presenting. In fact, their literature
indicates that they can isolate such symptoms & begin remedial
processing up to three months before it becomes lastingly
detrimental to the battery.

DISCHARGE MEMORY EFFECT
Discharge memory effects are caused by repetitive full charges
before the HV Battery experiences a full discharge in respect to its
SOC lower limit. When this happens the HV Battery is not able to
discharge for very long before indicating, via an accelerated drop
in voltage, that it can no longer be discharged, when in fact, the
amperage quantity delivered is far below what the HV Battery should
have delivered before manifesting such low voltages.

TYPICAL NiMH RECOVERY
Generally the NiMH industry handles such problems by admitting
excessive (higher voltages than normal) charges & then lower than
normal voltage discharges; especially when there are several cells
or modules in series. This often & effectively brings all the cells
or modules closer to one another in SOC range holding capability &
cleans up any memory effects pretty nicely. What this also does
however, is generates, through electrolysis, an abnormally high
amount of hydrogen molecules H2 & oxygen molecules O2, which cause
harmful internal pressures to the cell or module. These gases can
be vented but, then you loose a portion of your electrolyte &
consequently some capacity so, most manufacturers try to avoid
repetitive conditioning of NiMH batteries this way.

ANOTHER DOWNSIDE
Another negative side effect of this type of conditioning is that if
one were able to retain the gases contained in the cell or module
then, during the chemical recombination reaction of the molecules, a
fairly healthy temperature increase transpires within the cells or
modules which heat causes accelerated deterioration unless a very
stable & low rate charge is applied. This low, stable rate charge
can be applied to batteries outside the Prius but, attempting to
apply this kind of charge to a moving Prius is almost impossible
without severely affecting both the performance & efficiency of the
Prius. Similarly, asking the customer, via DTC or MIL warning
lamps, to bring the car in for battery conditioning was something
Toyota felt would be negatively received in many ways and spent a
lot of time finding a better solution.

TOYOTA'S REMEDY – MOVE TARGET SOC
Toyota has a unique & patented way to handle `memory effect' by
causing the HSD to administer a `Target SOC' that has been moved
into a higher or lower point than normal. Normally, the Target SOC
is held somewhere around 60% and the system attempts to hold it
below 70% & above 50%. I say somewhere around 60% because ambient
temperatures & HV Battery temperatures cause the Target SOC
to `crawl' around a little.

TARGET SOC FOR CHARGE MEMORY EFFECT
Because the `Charge Memory Effect' has resulted from repetitive
discharges before reaching a full charge, the HSD moves the Target
SOC up the scale. If the normal Target SOC were say 60%, it might
be moved up as high as 70% and the HV Battery SOC would be held for
a relatively long but, specific amount of time at a Target SOC of
70%.

This has the same effect as the application of excessive voltage but
has none of the side effects of generating excess H2 & O2 gases &
pressures or the heat of the recombination reaction. While
operating in this new Target SOC range, the Prius performs almost as
efficiently as it did while in the `normal' 60% Target SOC range and
the HV Battery is put under significantly less stress; a very nice
remedy indeed.

TARGET SOC FOR DISCHARGE MEMORY EFFECT
Discharge Memory Effects come from repetitive charges without
reaching the full SOC discharge state. To remedy this memory
effect, Toyota designed the HSD to move the Target SOC down from say
a normal 60% Target SOC to a 50% Target SOC. Then the Target SOC
is held here for an extended but specific amount of time. This too
is much easier on the HV Battery than the typical NiMH conditioning
method & allows the HV Battery to continue delivering a very
efficient service while it is going through its remedial process &
without having to take the Prius in for HV Battery conditioning.

TIGHTER SOC TARGET TOLERANCE
During these cleansing or clearing events, the ±10% limit above &
below the Target SOC is held much more tightly than when things are
operating in the normal 60% Target SOC. During the `normal' Target
SOC which is generally around 60%, the system will allow us to go
almost all the way to 80% and almost all the way down to 43% both of
which greatly exceed the ±10% band that they system tries to keep us
in most the time. The system however, seldom allows the ±10%
limits to be exceeded when going through a `memory effect' removal
process.

FINALLY
I believe it is very possible that your lady was experiencing the
effect of having reached the upper limit of around 60% SOC that was
being imposed upon her car as it had likely experienced a `Discharge
Memory Effect,' had a 50% Target SOC imposed upon it, reached 60%
SOC or above while coming home & coasting down her long driveway,
stopped to wait for the cat to cross the drive & then the HV ECU in
concert with the HV Batt ECU started trying to peel off the extra
amps and keep her car within the narrow lower limit of the SOC.
Again, when the Prius is going through one of these
cleansing `Target SOC' regimens, the upper & lower 10% tolerances
are much more stubbornly held to by the system and consequently the
system was being very persistent in getting her HV Battery back down
where it needed to be.

Best Regards,
Wayne

 

How can you tell when the car is doing a conditioning cycle on the battery?

 

When my car does it I notice my SOC is extra high or low for the conditions. Also if it is trying to charge to bring up the level, no way will the engine ever shut off, even when car is warm and it would normally be in stage IV and able to shut off.

Winter my SOC is normally around 55 to 65%. When it equalizes this climbs to 70 or 80%, or it might go down to the 40's range.

This area is pretty flat, so I don't normally get these changes that could also be caused by hills.

 

The only time I have seen the 80%/40% kind of swing was when the day was very hot and I sat in the car with the A/C on and in Ready mode. SOC dropped to 2 (red) bars and then the ICE started. SOC went to 3 (blue) bars and the ICE shut down. My normal driving keeps the SOC between 5 (blue) and 7 (green).

 

You're looking at a graphical display of colored bars. This correlates to a SOC but the user doesn't know what that is; i.e., zero bars might mean 80% SOC and eight bars might mean 90%. BTW - This is about what I think it is.

The SOC is determined by the battery management system (BMS) and uses many factors including total number of cycles. It resets the max capacity (100% SOC) depending on how the BMS and controller is designed to work (proprietary). There's a whole design team just determining how to predict an accurate SOC so the vehicle can operate in its most efficient mode.

 

bars to charge is known and has been posted here several times.

do a search.

 

I've searched and found nothing from Panasonic or Toyota saying anything to the effect of "Eight bars is 80% SOC and one bar is 40%." There's nothing official anywhere (not just on this site).

Now, Cobasys makes many NiMH pack systems ... complete with cooling and a proprietary BMS. Their high power cell voltage is the same as the PEVE cell and I expect the characteristics are similar. Cobasys has info out there for their NiMH cells used in hybrid apps, and here's a quote re their cells and SOC:

"The Cobasys NiMH batteries have demonstrated cycle life of greater than one thousand 80% Depth of Discharge (DOD) charge discharge cycles. In applications such as hybrid electric vehicles which utilize very shallow charge-discharge cycles 200,000 to 300,000 cycles have been achieved."
(http://www.cobasys.com/pdf/tutorial/InsideNimhBattery/inside_nimh_battery_technology.html)

Granted taking a cell down to 20% SOC is extreme, but so is taking it down to 40%. The number of cycles constituting end-of-life increases in a non-linear fashion with decreased DOD. In the Cobasys cell, they expect only 1,000 or so cycles at 80% DOD and hundreds of thousands of cycles with a "shallow" DOD. So, what is shallow? I say it's about 5-10% of initial capacity - calculated by the BMS and taking cell health into account.

The thought of taking these NiMH cells to 40% SOC and expecting them to last (as they obviously have) contradicts manufacturer data for NiMH life cycles. Applying Cobasys life cycle data to the PEVE cell isn't quite apples-to-apples, but it's pretty close ... given the fact that PEVE life cycle data as a function of DOD isn't readily available (I can't find any from Toyota or PEVE).

Now, and this is apples-to-oranges, Valence claims 10,000,000 cycles for their high power Li-Ion cells (hybrid apps). However, they achieved this at a 1% DOD and with charge rates of either C/10 or C/20. Not a realistic charge rate or DOD for any rigorous hybrid application. It just goes to show how much battery life - for any current chemistry - is a function of DOD ... not to mention a host of other factors.

Lastly, in reference to Tumbleweed's post from Wayne, several times he mentions a max SOC of 60%. What would be the purpose of keeping the cells at such a low capacity? I can understand targeting a max SOC of 85% or 90% (when charged by the engine) and keeping that extra 10% - 15% available for regen braking. The only case to set the target SOC (again, from the engine) at 60% is to rely heavily on regen braking to top off the charge.