Does actual capacity of traction battery decrease linearly with age?

Basically, ymmv

 

No. Failure is related to loosing the lowest capacity cells due to "reverse charging" and/or loss of electrolyte. Some of the cells in the battery will be "almost as good as new" when the battery "fails". So some cells of the battery will still hold almost as much charge as when new, maybe even more. Other cells will hold less and still other cells won't function when the battery "fails".

Unless you can measure each module of 6 cells it is very difficult to predict failure. For example, if you loose one cell you loose 1.25V across the entire battery. That small a loss could be caused by anything, so you would have to "hunt down" the defective cell to be sure that was the problem. When the Toyota tech "checks" the battery his equipment checks each module, looking for problem ones.

 

Moreover, as far as I know the standard Prius does not use any sort of adaptive capacity calculation. While the total battery capacity may degrade over time, the amount of Ah used by the HV controller to get from one SOC level to another remains (I believe) constant. Its not until the capacity is degraded to the point where it starts hitting the hi and low voltage limits within the normal SOC operating space that you start to see odd behavior due to SOC recalibration triggered at those limits. This would be problematic in a BEV where a significant fraction of the total capacity is regularly used, but has proven reasonable in the Prius where a rather small portion of the total capacity is typically used.

I would think a reasonable way to gauge health of a battery in a pre-owned Prius would be to look at the Voltage balance between cell-blocks, particularly at the very high and very low end of the allowable SOC range. IE, see what it is at typical mid-range. Then force charge to see what it is high, and then sit with the AC blowing until it gets low. There should be threads on here that give you can idea what is typical / reasonable for the max imbalance, and what could be indicative of a weakened battery.

Rob

 

A great deal of research is going on regarding battery longevity. I've read some interesting articles:

How to Prolong Lithium-based Batteries - Battery University

This one is about phones and tablets, but still interesting:
Battery Life in Gadgets - The Science of Battery Life in Electronics - Popular Mechanics

We could get more than double the power out of our hybrid batteries if we charged them up to full capacity and discharged them to the limit, but longevity would be sacrificed. We don't mind replacing our phone battery every year, but if that happened with a hybrid, we'd probably swear-off the brand for life.

 

The ordinary Prius use Nickel Metal Hydride (NiMH) chemistry for the main traction battery. The Plug-In Prius uses a massive LiON pack shown in the Toyota Dismantling Manual. My interest has been with the NiMH batteries.

The NiMH battery works by converting between two nickel hydroxides and storing hydrogen atoms in the metal hydride:

• Ni(OH){2} <=> NiO(OH) + H{2}O + e-
• H{2}O + M + e- <=> OH- + MH

What ages our NiMH batteries is the loss of H{2} and O{2} from the electrolyte.

During normal charging, a small amount of water electrolyzes into H{2} and O{2} gas. Sealed, the metal hydride can recombine the H{2} and O{2} into water, a good thing. But the weak seals of the earliest 2001-03 battery modules could leak a little of the gas as they got hot and the gas pressure built up. Charging is the risk because it generates heat and gas pressure against the terminal seals:

The left photo shows rubber "O" ring is partially melted on the right side, the black smear. You can also see the nickel compound has changed color from the normal silvery color of the other electrode.

Now as the H{2} and O{2} leak out, the electrolyte, a saturated solution of KOH, dries. This gradually reduces the conductive area in the plastic mesh separator, increasing the resistive heating to a smaller and smaller area. When the heat becomes hot enough to melt the separator, it forms an internal short and that cell is failed:

But the melting doesn't happen instantly. First the reduced electrolyte increases the internal resistance of the cell. This reduces the capacity, the Ahr, and charge-discharge efficiency. So the obvious solution is to add water.

There is an excellent Toyota-Panasonic patent on refreshing the electrolyte and some of us have tried different approaches:

I have successfully returned full capacity to worn, battery modules by adding water and cycle charge-discharge . . . replicating the Toyota-Panasonic patent. But I sealing the modules turned out to be a problem.

I tested stainless steel inserts and epoxy:

The stainless steel nut plates even with heat did not form a gas-tight seal. Epoxy would not adhere to the plastic and the different thermal expansion rates would not be reliable either. I tried plastic welding like the Toyota-Panasonic patent only to realize we would have to use salvage case material and a mold fitting and still it would be marginal. But I could never figure out a solution for melted or weak "O" rings. So how does this relate to Prius efficiency?

Thanks to Patrick Wong, the early traction battery capacity loss occurs in the first 5-6 months:

So unless someone is buying a new Prius or replaced the traction battery with a new one, this short interval really doesn't come into play. Most of the time, the Prius battery is only doing small, fractional charge-discharge cycles UNLESS the driver chooses to put a heavier load on the battery.

Remember, charging is the risk:

Here I am doing a forced charge recording the pack temperature. It generates:

• heat - that weakens the "O" ring, terminal seal
• gas - increases the pressure to leak it out

The best path to a long battery life is to monitor the energy flow and AVOID battery discharge. So when climbing a hill, back-off when the energy flow shows a discharge. The same is true for acceleration, avoid drawing charge from the battery. If not discharged, the car won't have to charge the traction battery and heat it up. But sometime, we have to descend a large hill, 500 ft or so. In those cases, use "B" so the engine will absorb the excess energy that would over-charge and heat the traction battery. But this is just background information about our NiMH batteries: Prius Battery Photos

The Prius shop manual describes a way to force-charge the traction battery to maximum, 80% SOC. Then shifting the car into "R" and holding a accelerator down, the battery is discharged to the minimum. This technique is used to identify weak modules or a marginal traction battery. However, it can also provide a rough, battery Ahr measurement. It also 'stresses' the traction battery which a seller may object.

Personally, try the car and use the original battery to learn how to drive it. IF the battery fails, either buy a quality rebuild or new Toyota replacement. Then use your 'lessons learned' to let that battery last until 'the wheels fall off.'

Bob Wilson

 

I love Bobs posts, but just wish I wasn't so thick and I could understand them in full.

 

This is an interesting thread. Under the assumption that heat is the battery killer, I have endeavored to limit the battery charge and discharge rate to 20 amps whenever possible. I noted that the controller usually limits the charge/discharge figure to about 22 amps. This is also true when one uses the "B" setting. So I figured that Toyota had determined that this was a good compromise between battery charge/discharge and heat generation. For me, this is a significant issue because I live in a very hilly area. It is not uncommon for me to see near an 80% charge at the bottom of a hill. Sometimes I will have to park the car with the battery in this highly charged condition. On a cold start in such cases, after a near full charge during the first phase of the warm-up period, if you allow it, it will run up to 100amps without engaging the ICE unless you push the power demand to the point where you force the engine to fire. My approach has been that, when the engine is cold, I limit the discharge to 20 amps until the controller starts applying ICE power to the wheels. Only then do I increase my power demand.

Bob, your plot looks quite informative, but I am having trouble interpreting it because the legend is not clear to me. The orange line looks like it is plotting current with respect to time rather than volts as the legend implies, yes? Is the blue line percent charge?. If so, then are the green dots individual data points from multiple experimental runs in amps?

 

Sorry, my graphs are often more for personal consumption but I share when I can:

• dark-blue and pink - these are two temperature points in the battery
• orange line is the voltage/10 - so the 34.6 peak is actually 346 volts
• tiny-green dots - current using the right-size scale

In this test, I wanted to find out high up a hill the car would go in reverse. But the hill was too small and a cop came by. Knowing the weight of the car and the change in altitude, I could calculate the available traction battery energy in kwHr.

Bob Wilson

 

Thank you, bob. Very cool. Do we know an approximation of the efficiency of the battery pack? I.e. We input x number of watts into the pack, what percentage is recovered as energy? Heat is clearly energy loss. This has surely been discussed ad nauseum, but I am curious what your opinion or calculations on the subject are.

 

Efficiency is a hard problem if you'll check the efficiency numbers in the charge/discharge chart in my earlier posting. As Ahr capacity returned, it reached ~95%. It may also be a function of how much charge cycles in and out of the pack. A hard efficiency number is beyond the scope of my studies except for what I've already done.

Bob Wilson

 

Wrt battery health I think the easiest way to gain insight into this is to monitor the parameter Vmax-Vmin using an OBDII tool. I use Enginelink and this is one of the "gauges" I strip chart. From my perspective knowing what this looks like over time will help me recognize a problem without having to look at all those individual module voltages, unless and until one of them starts to diverge. On my lower case v, it stays around 0.1x V, although occasionally it will go lower, into the 0.0x V range. I'd be concerned if that behavior changed.

 

Wow, this is Super Technical stuff, BUT awesome information I hope I can someday use.

I believe my question is then perfect for the contributors to this thread.
I'm looking to purchase a new 2013 P3 @ the dealer in SoFl where the temperatures can get pretty hot in the summer and the dealer does not have a covered parking lot.

My biggest concern is the condition of the batteries, both of them!
I'm sure the batteries were manufactured between 6-12 months or more prior to the vehicle manufactured date.
So, that would make the battery at least up to this month, 2 1/2- 3 years old. And I believe batteries start to lose their charge capacity once they are made. So, the traction battery could have been totally discharged if no one test drove the car for a while!
Sorta like the cliche, once we are born, be begin to die. Anyway, enough morbid thoughts!

Since the car is new and the dealer is in the business of selling their cars, and maybe not much concerned in maintaining the battery condition of the car by starting it and letting it charge for a while, should I be concerned? I'm have some amount of trepidation pulling the trigger on the sale.

Even though there is a 8 yr 100k battery warranty on the car in the state of FL that starts when the car is purchased, I may already 'lose' 2-3 good years of the batteries life.

Thoughts please!

 

bump.

 

There seems to be just one question in your previous post. Should you be concerned? Does it matter what people here say to you? You either will be or won't be. Given that the battery is warranted for 8/100 from point and time of sale, you need to figure this out for yourself. I will offer that if any damage has been done to the battery by its sitting, it would be evident well before that time.

Batteries do not necessarilly lose capacity from sitting starting on day one. (Well, Lithium Ion do, but that doesn't apply here.) As has been stated by others, the biggest cause of capacity loss in the Prius batteries is loss of electrolyte due to hard regen. I don't think that is the case here. So, if you can wrap your brain around the fact that the car is probably just fine see what the best deal you can get for the car and buy it if you like the price. If not, move on. Life is too short to vacillate over things like this.

 

Excellent info Bob, thank you!

I do find it interesting that discharge is the enemy, not recharge... but, that's based on my intuition, I have very limit knowledge of battery technologies.

I'm curious what happened on this graph at the 2,400 second mark. Also, what do the horizontal blue bands represent?

 

Sorry, I sometimes get busy:

• Traction battery - operation is the risk. Even in the Florida summer heat, parked, the traction battery won't get in the temperature ranges that cause the seals to leak. However, drive off the lot and do they type of driving that leads to avoidable, charge cycles, and all bets are off. That is why I always recommend parking 'in the shade' in hot climates even if it means walking across a hot parking lot. Drop off those who need cooler temperatures near the door and then park in the shade. But even if the car has been solar heat-soaked, drive modestly, say no more than 65-70 mph on a Florida highway using cruise control with the AC running and the car will be fine. But if you want to 'tail gate' and spend a lot of time changing from 85 to 60 and then up to 75 by manually operating the accelerator, that would be bad. Speed changes, the faster ones, can 'heat pump' the traction battery and that is bad . . . and avoidable.
• Lead-acid 12V battery - these are like tires, consumables. Unless you are prepared to put in a solar, tickle charger, drive until they get weak in cold weather and get another one . . . like tires.

Charging an NiMH battery is exothermic, it generates heat. It also generates some H{2} and O{2} which further increases the pressure against the terminal seals. Sad to say, discharge does not provide enough 'cooling' because ohmic heating of the current is trying to go the wrong way. But discharging leads the car to having to put a charge back in to maintain 60% optimum level. If you want to drive the car to minimize traction battery heating:

• Monitor energy flow - if climbing a hill and the traction battery arrow is feeding power to the wheels, back-off the accelerator until the engine is doing the heavy lifting. Following at a safe distance, a heavy truck up a grade is a cheap-trick and lets following traffic burn their gas racing to the crest. Acceleration should be 'with traffic'; drive on cruise control which avoids 'fretting' the accelerator, and; speeds over 75 mph can often lead to 'heat pumping' in 200-300 ft or higher hilly areas.

The traction battery has four temperature probes: three on the controller side, 19 of the 38 modules, and one air-inlet probe. My Graham miniscanner can only record six data fields so I used the coolest, pink, and hottest, blue, readings. The steep temperature climbs happened during a 'forced charge' as seen in the high charge current.

The temperature probes are located on the top of the modules, not in the center so some of the heat transfer is probably delayed. This test would have been better had I turned off the car and then done a series of spot checks every 2-3 minutes later.

Bob Wilson

 

A quick response to whether storing a new car "in the hot Florida sun" would damage the traction battery is NO. The battery -might- reach 20F above ambient, with no gas pressure inside. When in operation it can run hotter than that and with gas pressure. Of course driving it with gusto off the lot when it's that hot -can- cause problems. Keep in mind, these cars are operated in very hot places such as Texas, Arizona, and Nevada. How hot do you think the traction battery gets when it's 120F outside and you are hammering up the mountain pass between Arizona and California?

The 12V battery, if not recharged at least once every few days (by the car running or by a "battery tender" of some sort), will sulphate when it discharges too much. Discharge can be from two sources, the car (parasitic losses keeping the computers in standby), and internal self-discharge losses. If not taken care of the 12V battery can be expected to fail after 3-5 years. The main failure mode of the 12V battery is caused by loss of electrolyte, because you cannot add water (it's a "maintenance impossible" battery). IF you peal off the label, add distilled water, and charge it up with a proper charging technique you can get 6-7 years of life from it. I'm cheap, and I do this. Many can't be bothered and replace it after 3-4 years.