FalconFour
Well-known member
Figure I'd pop in here with some info. Mine started exhibiting exactly the same symptoms, but long after the update. Nothing prompted the change except for one figure I'll discuss in a second. Now, I'm lucky to get 4 bubbles any time above 2 bars remaining on the display - at 80%, I get nothing at all for quite a while. It's completely dependent on the amount of time regen is active - that is, the longer I slow down, the more regen power it pulls in before I come to a stop. It will keep regenning more and more power until I apply the accelerator and get back up to speed - short bits of power, not fully using all the regen'd energy - don't seem to "reset" it. It'll sway from 2 bubbles (including center dot - effectively "1 dot") all the way to 4 active bubbles (out of 5 - which I never get anymore except at LBW/VLBW) and then back to 1 active bubble (center dot only) within one stop light. HUGE variation, and clearly being limited by the battery controller's current-request reporting, same way it limits regen in typical Leafs over 80% SOC.
I have a theory to this. There's a figure that's reported by the Leaf Spy app (previously the Leaf Battery app for Android by Turbo3), that was previously called "Health" ("Hlth", but you can figure...), but is now just called "Hx" due to the unknown nature of the figure. For me in my Leaf, it started at 68.67% when the app first included Hlth reporting, with that figure on August 19th, '13. I thought that was rather low. Everything here is "post-update", too. I had P3227 done pretty early on. As time wore on, that figure never really trended upwards, but only downwards with a few upward burps within a range of error. That's led me to believe that "Health" or "Hx" is related to internal resistance of the battery cells, perhaps an average, or maybe a figure of "lowest health" of the cells in the pack. I have two cells that perform noticeably worse than the others in the graph - always at a lower voltage than the others, at all times and all SoCs.
Suddenly, around 63% "Hx", I noticed a severe degradation in regen ability - it was very pronounced and I can definitely point my finger to a specific time it started behaving differently. I've been driving with reduced regen now for the past few weeks, maybe around a month, ever since fall started turning into winter. We've had our warmer days, though, and the warmer weather has not changed it in the slightest. And it's been taking a pretty big hit on my efficiency - not to mention the winter. I'm now lucky to get 50 miles out of a full charge with just 1 bar gone (lost around the middle of this year).
However, the notable point is that the Hx degradation has come to a complete stop, and started reversing! On 10-01-2013, I have a screenshot showing 65.44% "Hlth" with the old version. On 10-11, it was 64.49%. 10-14 saw 64.68% for a second, and on 10-23, it was 63.25%. But on 10-26, it showed 63.54% - it had started improving, and nothing but the regen behavior had changed! Today? 63.68% again, just like a month ago. The first time it's ever actually started trending upwards. The data's a bit incomplete since I don't seem to have any screenshots stored for November - I was only looking back at past screenshots auto-saved to Dropbox for other reasons. I do seem to be missing quite a few, though... I swear I took some in November. :?
The stresses that regenerative braking puts on the battery are similar to those put on it when fast-charging. Charge activity enters from the edges of the cells and propagates through the material in each cell with actual, chemical state changes taking place - so I'd imagine that regeneration combined with acceleration levels each other out quicker, chemically, than the forces put on the cells by quick-charging, but the stresses are there none the less. I do no quick-charging *whatsoever* because no QCs are available in Fresno. Only 3 are on my car's counter, from my 4 visits to the SF Bay and one trip had no QCs I could use without a Blink card.
It seems that the regen "throttling" is being done to protect what I could now pretty accurately guess *is* a "health" figure for the battery, at least, as far as the car is concerned. It's trying to protect itself. Which I think is pretty cool, except that it totally limits my range.
The voltage rise/fall for acceleration/regen demand seems in line with this concept, too. Accelerating has a noticeable voltage drop on the cells, and decelerating (regen) puts the voltages up about 0.1v at most with one glance test I did, since I don't have a mount for the tablet to watch it more often. I imagine it watches these voltage peaks/drops during driving to come up with this Hx figure, and to determine the physical state of the battery. There's a WHOLE lot of lithium-battery research that almost certainly went into the thing.
Amusingly, not having yet come to this conclusion (hadn't yet looked at the data above), I took my Leaf to the dealer today after calling the 1-877-NO-GAS-EV help line (I have rather limited confidence in the dealer... "5-star" battery review a month ago? Really?), and the help line said it was a valid cause for concern since they hadn't been aware of this issue. The dealer, after sitting there for 4 hours, determined there was nothing malfunctioning that they could figure out, even having called the tech line to talk about it. They saw the issue too, but didn't know what to make of it.
As for the regen and the charger/controller, in reply to an earlier post... lol, the charger has nothing to do with the regen system that I can imagine The motor is only an "AC motor" by description, but its windings are directly controlled by a computer, part of the overall "inverter", that can either apply or pull current to/from the motor depending on which direction it wants to apply torque. Applying "negative torque" is regen - it takes the higher voltage out of the motor and feeds it to the battery. Applying positive torque is acceleration - taking the higher voltage from the battery and applying it to the motor. It could, theoretically, pull in 80kW of regen power if that wouldn't smoke the battery pack. :lol:
I have a theory to this. There's a figure that's reported by the Leaf Spy app (previously the Leaf Battery app for Android by Turbo3), that was previously called "Health" ("Hlth", but you can figure...), but is now just called "Hx" due to the unknown nature of the figure. For me in my Leaf, it started at 68.67% when the app first included Hlth reporting, with that figure on August 19th, '13. I thought that was rather low. Everything here is "post-update", too. I had P3227 done pretty early on. As time wore on, that figure never really trended upwards, but only downwards with a few upward burps within a range of error. That's led me to believe that "Health" or "Hx" is related to internal resistance of the battery cells, perhaps an average, or maybe a figure of "lowest health" of the cells in the pack. I have two cells that perform noticeably worse than the others in the graph - always at a lower voltage than the others, at all times and all SoCs.
Suddenly, around 63% "Hx", I noticed a severe degradation in regen ability - it was very pronounced and I can definitely point my finger to a specific time it started behaving differently. I've been driving with reduced regen now for the past few weeks, maybe around a month, ever since fall started turning into winter. We've had our warmer days, though, and the warmer weather has not changed it in the slightest. And it's been taking a pretty big hit on my efficiency - not to mention the winter. I'm now lucky to get 50 miles out of a full charge with just 1 bar gone (lost around the middle of this year).
However, the notable point is that the Hx degradation has come to a complete stop, and started reversing! On 10-01-2013, I have a screenshot showing 65.44% "Hlth" with the old version. On 10-11, it was 64.49%. 10-14 saw 64.68% for a second, and on 10-23, it was 63.25%. But on 10-26, it showed 63.54% - it had started improving, and nothing but the regen behavior had changed! Today? 63.68% again, just like a month ago. The first time it's ever actually started trending upwards. The data's a bit incomplete since I don't seem to have any screenshots stored for November - I was only looking back at past screenshots auto-saved to Dropbox for other reasons. I do seem to be missing quite a few, though... I swear I took some in November. :?
The stresses that regenerative braking puts on the battery are similar to those put on it when fast-charging. Charge activity enters from the edges of the cells and propagates through the material in each cell with actual, chemical state changes taking place - so I'd imagine that regeneration combined with acceleration levels each other out quicker, chemically, than the forces put on the cells by quick-charging, but the stresses are there none the less. I do no quick-charging *whatsoever* because no QCs are available in Fresno. Only 3 are on my car's counter, from my 4 visits to the SF Bay and one trip had no QCs I could use without a Blink card.
It seems that the regen "throttling" is being done to protect what I could now pretty accurately guess *is* a "health" figure for the battery, at least, as far as the car is concerned. It's trying to protect itself. Which I think is pretty cool, except that it totally limits my range.
The voltage rise/fall for acceleration/regen demand seems in line with this concept, too. Accelerating has a noticeable voltage drop on the cells, and decelerating (regen) puts the voltages up about 0.1v at most with one glance test I did, since I don't have a mount for the tablet to watch it more often. I imagine it watches these voltage peaks/drops during driving to come up with this Hx figure, and to determine the physical state of the battery. There's a WHOLE lot of lithium-battery research that almost certainly went into the thing.
Amusingly, not having yet come to this conclusion (hadn't yet looked at the data above), I took my Leaf to the dealer today after calling the 1-877-NO-GAS-EV help line (I have rather limited confidence in the dealer... "5-star" battery review a month ago? Really?), and the help line said it was a valid cause for concern since they hadn't been aware of this issue. The dealer, after sitting there for 4 hours, determined there was nothing malfunctioning that they could figure out, even having called the tech line to talk about it. They saw the issue too, but didn't know what to make of it.
As for the regen and the charger/controller, in reply to an earlier post... lol, the charger has nothing to do with the regen system that I can imagine The motor is only an "AC motor" by description, but its windings are directly controlled by a computer, part of the overall "inverter", that can either apply or pull current to/from the motor depending on which direction it wants to apply torque. Applying "negative torque" is regen - it takes the higher voltage out of the motor and feeds it to the battery. Applying positive torque is acceleration - taking the higher voltage from the battery and applying it to the motor. It could, theoretically, pull in 80kW of regen power if that wouldn't smoke the battery pack. :lol: