Hybrid Battery Survey-Gen2 Prius 2004-2009

Electricity use in Ready mode can vary, with lights on, fan on , radio on, it can be as high as 400 W. Add rear defrost, front defrost, or AC and you have even more.

The test is interesting, you could speed it up with additional constant (not variable like AC) electric drain and monitor battery discharge current and SOC in % on OBD port.

From full to empty usually means from 60% to 40% SOC, this is just 0.26 kWh, from all full bars (80%) to 2 bars (40%) would be 0.52 kWh.

In cold weather engine will start sooner, but you can partly avoid that with fan off button. Coolant can then drop bellow 40° C without starting up the engine, but I don't really know how low it will go before starting up again.

 

The test sounds good in my weather all year round except mid winter. I will recommend it to a friend with a Gen2. Can do it with headlights on and from 6 bars to engine triggering and monitor in time.
Could also be good for first impression for a potential buyer of used car if the weather is right.

 

I don't believe you are suppose to turn on all accessories, headlight and AC to do this simple test. As a car gets older, the battery will naturally be weaker and won't show performance numbers the same as new. As I was told, if the battery can maintain a charge on "ready" mode for 30 minutes without any accessories on, just gives you an idea the battery is good and doesn't have any bad/too weak of cells. I don't think you should make it too scientific as it probably would not measure up correctly

 

Well, if that is the case, then I guess I buck the trend. :-/

I have a 2006 Canadian model Prius which was purchased as a demo car in April 2006 (~4K miles), and my HV battery failed in May 2013, at about 170K miles (both converted from KM, of course). The good news is that true to form, the car never stranded me, and reverts to "turtle mode" to still get my from point A to point B until I figured out what to do about it.

As far as driving pattern, I am fairly heavy, driving virtually every day, average 70-100 miles daily in the city (long commute, work on the road...), and Canadian summer and winters. Not many short trips, regular use, and somewhat high/low temperatures in the peak of the seasons, but not so much that I feel that would lead to premature failure.

With regards to repairs, I shopped three different Toyota dealers who all quoted me a full battery pack replacement in the $3K neighbourhood as the only option (they claimed Toyota did not have a part number for individual cells).

Anyway, to make a long story short, I located a "Prius rebuilder" north of Toronto (Autodex in Concord, ON) who makes a business of savaging wrecked Prius and selling them (full disclosure, of course), and very reputable guy. I found him through reinvolt.com. Anyway, he was able to rebuild my battery for me, along with a 1-year warranty, for considerably less than half of the price Toyota wanted. 7 months later, so far so good. If anyone near Southern Ontario wants his info, let me know.

 

You're just 10k km ahead of me Tideland, though I'm probably catching you as I'm doing about 30,000+ km pr year these days.

Like yourself I'm also out of warranty, at 198000 km now but still going well. Personally I'll consider it a success if I make 241,000 km (150,000 miles), as that's currently my benchmark for getting a decent run out of the original battery. Seeing as that's going to take under 18 months, I'm pretty confident of making it to 150k miles.

 

Just thought I'd add my info to your query: 2005 Prius, 257,000 km. No issues with the hybrid battery or the 12 volt either for that matter so far. I live in the Pacific Northwest and average 600-700 km/week.

 

In your next survey also ask about climate (hot, warm, moderate, cold - heat is the enemy) and terrain (flat, hills, mountains - Mountains cause stress on the HV battery) as these are major factors determining HV battery life.

JeffD

 

Do you know the one thing which would have the greatest impact on the reliability of the data. This question

* Did you come to these forums specifically because you were having traction battery problems, or problems that ultimately turned out to be due to a failing traction battery.*

Keep those people in a separate category and then we might have some meaningful data.

 

Ok, we understand the reasons why our data could be skewed, but what exactly do we know about the consumer reports data?

Does anyone have any information on exactly how the consumer reports data was collected, just to be sure that it isn't also skewed. I mean, they didn't just ask Toyota how many batteries they've replaced did they?

 

Also many of the batteries replaced are not warranty replacements. There are many that install Dorman batteries and other remanufactured options. Not sure those are part of the survey 2% figure.

 

Yep, that was exactly the point I was trying to make JC. Without knowing exactly how the CR figures were obtained, how do we know that it's not just some limited subset. Who knows, it might only represent out of warranty replacements at dealer service departments or something very limited like that.

The other thing that sits uneasy with me about those really low numbers, like 1 to 2 percent total, is that more than that many probably get written off every year (not total like the CR battery replacement figures, but EVERY year). So why the heck are battery rebuilders struggling to find salvageable batteries? Why are salvage batteries being snapped up as quickly as they become available. Why is supply and demand forcing the price of salvage batteries ever higher? They should be piling high to the roof in junk yards if those are the true TOTAL figures!

 

Consumer reports reliability data comes directly from vehicle owners through a survey sent out every year. It has nothing to do with manufacturer reporting, warranty status etc.

I believe salvageable batteries are still fairly hard to find as all those that go to Toyota are supposed to be recycled rather than ending up in a junk yard. My understanding is that Toyota is still offering a $200 "bounty" on old battery packs to encourage recycling, so even those that end up at junkyards by other means have a good chance of finding their way back to Toyota.

Rob

 

Thanks for the info Rob. That gives me a bit more confidence in their data now.

 

According to CR survey responses were received for 1.1 million vehicles in the last survey, don't know how many Priuses are in there but seems like it should be a pretty decent sample size.

More info on survey:
Car Reliability FAQ | Answers to Reliability Questions - Consumer Reports

 

CR claims the survey dataset tends to track very well with total sales volumes by year / model. Assuming that's true, we could do a very rough estimate of the sample size by estimating that the 1.1 million responses represents an equal sampling (~0.45%) of the ~246 million vehicles in the US "fleet." Applying that to the sales data above:

Year: Estimated Sample Size
2002: 90
2003: 110
2004: 241
2005: 482
2006: 478
2007: 810
2008: 709
2009:624
2010: 630
2011: 610
2012: 1058
2013: 1047
Total Sample: 6,889

Right off the bat the older models are probably exaggerated slightly, as I haven't made any effort to correct for the number of cars going out of service each year. I doubt its worth trying to do so, as this is a pretty rough estimate anyway. It should be sufficient to give a reasonable idea of the order of magnitude of the sample set.

Rob

 

You could also combine the two and come up with an estimated number of reported failures:

Year: Estimated Failures / Responses
2002: 4/90
2003: 4/110
2004: 2/241
2005: 10/482
2006: 14/478
2007: 8/810
2008: 4/709
2009: 2/624
2010: 1/630
2011: 1/610
2012: 1/1058
2013: 1/1047
Total Sample: 52/6,889

​

Again this is a very rough estimate based on certain assumptions about the data set as above.​

 

Thanks again Rob.

Based on those numbers then the overall results should at least give us a good indication, but I'd like to warn that the numbers (and amount of variation) for individual year models are too small to make conclusions like, "2006 are less reliable than 2005", as it appears some people have been tempted to do.

For example, based on your numbers I just did a quick statistical comparison (Chi-squared) and it showed that there is definitely no statistical difference between 2005 and 2006 models.