Clarity life after 50596 miles

MrFixit said:
I would say that I was getting the 14.5 kWh charges initially, but it has been gradually declining until now. Roughly, I now get 13.5 kWh charges, This decline represents about 94% of original capacity. It would be very interesting to get a readout of this to see if the capacity as measured by the vehicle has declined accordingly... I might expect to see 94% of 55.1 Ah, or 51.8 ?
Click to expand...
Seems like a failed battery would be down to about 9 KWh for full charge: A reasonable assessment would be the battery is still good if it takes more than 10 KWh for a charge. I am wondering if anyone is seeing significant degradation in KWh for a full charge...

Does anybody have a battery that is taking less than 12 KWh for a full charge?
 
Last edited:
The amount of charge is a good easy way to check, but one thing to keep in mind is that 0 EV can be a widely different starting point in terms of SoC. So although you may think it goes from empty to full, "empty" needs to be defined. (I'm not talking about the bottom end buffer.)

I have personally seen my 0 EV state of charge at the beginning of recharging from 6% SoC all the way up to 15% SoC. It averages ~10%. I have seen SoC numbers quoted here as low as 3% SoC and I suppose 0% SoC is a possibility. Again, these take into account the BMS buffer as well.

Just note that a "full charge" will vary depending on the starting point, along with other factors noted earlier.
 
Ray B said:
Just note that a "full charge" will vary depending on the starting point, along with other factors noted earlier.
Click to expand...

Yes, certainly a factor. This just means that you should never over-interpret a single data point. If you examine a relatively large number of 'full' charge levels, your eyeball will give you a number that is pretty close to the 10% average SOC. You could do some math for this, but I think you can get surprisingly close without bothering.

And I still contend that using input charge energy as a figure-of-merit for battery health will be way more accurate than any attempt to use EV range.
 
MrFixit said:
Yes, certainly a factor. This just means that you should never over-interpret a single data point. If you examine a relatively large number of 'full' charge levels, your eyeball will give you a number that is pretty close to the 10% average SOC. You could do some math for this, but I think you can get surprisingly close without bothering.

And I still contend that using input charge energy as a figure-of-merit for battery health will be way more accurate than any attempt to use EV range.
Click to expand...

Yes very true. And I think this technique has merit. But one other caveat to keep in mind is that we don't know if/how Honda may adjust the BMS buffers at the top and bottom of the battery SoC to maintain range for the user:

ev-driving-range-web.jpg

Link to image: https://batteryuniversity.com/_img/content/ev-driving-range-web.jpg
(source: https://batteryuniversity.com/learn/article/electric_vehicle_ev )

Such periodic adjustments to the buffers may obscure some of the true EV battery degradation which is occurring. So we may be measuring x% loss by checking the kWh to accomplish a full charge, but it may actually be x% + y% buffer reduction.
 
Last edited:
insightman said:
However, I'd hate to try balancing a Clarity PHEV battery in my lap like I did this 0.9 kWh, 70-lb replacement battery for my gen-1 Insight. Replacing a Clarity's 17kWh battery would be much more difficult.
Click to expand...

It wouldn’t surprise many folks that replacing a 17kWh battery would be more difficult than replacing a 0.9kWh battery. A similar comparison might be replacing the Diesel engine in a motorhome to a gas engine in a wood chipper.

The Clarity PHEV battery is comprised of 14, 1.2kWh modules. I have not seen a posted weight from Honda for a single module, however, LiFePo batteries with similar capacities from other manufacturers weigh between 25-30lbs. So, it may actually be a more pleasant experience to balance in your lap, a battery with 33% more capacity and a weight less than half that of the Insight replacement battery.
 
Using an "educated guess" for the SOC at start of charge and then monitoring via input charge current what is put back in is really a great idea and there is a lot of data here. For me, since "warranty" is defined by Honda as the pack ampere hour readout on their i-HDS interrogation of the car's BMS that was the only point I cared about. However, knowing the trend and when it makes sense to go get the "official" report from the dealer (or checking yourself if you have the requisite OBDII probe and software) is very valuable. My narrow minded focus missed that... The vehicle has a coulomb counter it uses to track capacity but it also uses voltage curves for our gauges since the Li-ion cells are somewhat predictable unlike LiFEPo4 that voltage is really a complete guess without coulomb.

I am beginning work getting the PID query and response for the pack Ah readout but that may take a while. I have an idea of how to get it but have to finish the install of RP before that is an option and that (with the covid crisis) will have to be when the kids can go back to school... :confused::(

My cheap OBDII (it's a knockoff ELM327, only supports 1.4 codes) and Android "Car Scanner" software will read the pack in/out amps and HV pack voltage. The top of charge and bottom of charge (within the BMS limits) is "knowable" info. Since the voltage will always swing between these points regardless of actual capacity (its the Ah's that change, not the voltage) it makes the guess a little less of a guess. Admittedly this voltage should be gleaned as an OCV (no load on the battery at all, open circuit) and that is not possible, it is still better than using the percentage gauge on the car app.

The best case is someone like @MrFixit that has detailed information of charging parameters. Combine that with the start and end HV pack voltage and the dealer reported battery Ah capacity and you have the makings of a real home guestimeter. Probably more work than it is worth, admittedly - but if you know the bottom and top and can get that with a $10 ODBII and free software, then one charge from empty to full becomes that "picture worth a thousand words." :):cool:

@Landshark , Battery replacement guide (simplified and bad PDF conversion, sorry) attached.

I need get those numbers.... as soon as I have some place to drive again.... (due my work... - my wife is the one working, and I am scrambling to home school the kids... thank God for Khan academy so I can keep up with my 6th grader!)

Cheers,

Cash
 

Attachments

Last edited:
I guess the OP has abandoned us...

Seems like a good response to people questioning battery health would be to suggest they do a full charge (Get battery to EV range zero, and let it charge till it stops). They will have to do this at a charging station that measures energy. If it takes less than 10 KWh to charge, they may have a battery problem. Otherwise, their problem is elsewhere.
 
Let me offer another thought to this discussion...

Even if you don't have an EVSE that measures energy, I think you can still do a meaningful measurement.
This will not be perfect, but it will be pretty close and particularly useful if you use it for a comparison over time (once or twice a year or similar).

Run your car until the battery reaches an EV range of zero. Plug in and measure the time it takes until charging stops (this is the hard part because you aren't going to sit there gazing at the light to know exactly when it stops). You likely have a feel for this from experience so that can narrow the window so you sort of know when to be watching.

For the two main cases - the OEM charger (Level 1, 120V 10 amp), and for a (Level 2, 240V 30 amp - Clarity maximum) just use 1200 Watts, or 7200 Watts respectively. Multiply the charge time in hours times either 7.2, or 1.2 respectively. This gives you the total charge energy in kWh. For example, if it took 2 hours with a Level 2 charger, the charge was 7.2 x 2 = 14.4 kWh. This is equivalent to a Level 1 charge for 12 hours (1.2 x 12 = 14.4 kWh).

OK, I know there are some important 'assumptions'. Your line voltage will not always be the nominal 120 or 240. I know the BMS tapers the current at the end of the charge, but my observation is that this 'tail' includes very little energy so it is mostly negligible.

I think this has significant utility and it would be worthwhile for a non-energy reading EVSE user to conduct upon occasion.
 
If you have a "fancy" home charger you can get the data there as well. Even a "killowatt" meter on your included Level 1 charger would give you an idea. The charge efficiency will go down with level 1 versus level 2, however the battery will accept a more balanced "full" charge from level 1 usually. I would assume 85% efficiency on the charger/system is conservative. You have losses in the AC-DC charger (likely ~88-93% eff) built into the car, transfer losses in the cabling at higher currents (level 2 vs 1) and BMS losses in the distribution (unknown), then finally the chemical efficiency of the storage cells converting watts in to watts available/stored (very high so maybe 1-2% loss).

WAG is good enough when it is all you have!

@Ray B , I highly doubt Honda is adjusting the HV pack charge/discharge buffer limits to maintain driver expected "range." As we are already seeing a reduction in range with use and age, and that would not serve Honda bean-counters, as it would cause a greatly accelerated rate of cell decay over time. This would only serve to increase their risk of warranty claims since they have already stated in their warranty documents that it is battery amp-hour capacity that is used to qualify for replacements. Although from an "owner" standpoint such system parameter adjustments would be highly preferred as the "greatest range" is what we really want most. The packs will die eventually either way. And as one who really enjoys getting the most out of their warranty as possible, I would prefer it too!

As for changing the battery or servicing it ourselves. No, not easy like the insight - however it is possible. Rich Rebuilds has swapped several Tesla packs using pretty "home gamer" equipment successfully. Probably doable with the car at a local shop with side post lift and a palette hoist. Then use a shop hoist to put it on a trailer - take it home and do the work, then reinstall days later. No, not fun unless you are an enthusiast but Rich and others proved it is possible. With the cancellation of the Clarity BEV, there are lots of packs about to hit the market on salvage/aftermarket and a module is a module!

Cheers,
Cash
 
Last edited:
This 'efficiency' thing keeps coming up.

It is my contention that for this purpose, that has no bearing. Experience of myself and others has shown that with a 'full' charge, you are inputting 14.4 kWh into the vehicle as a black-box system. It doesn't matter if the car has efficiency losses of 10%, 15%, or 25%. The 14.4 kWh at the 'input' is the baseline full charge energy when you have a 'good' battery. Same for the other nuances you mentioned like losses in the AC-DC, Chemical losses, BMS losses and the like. Transfer losses in cabling are likely trivial, but that cancels out anyway if you repeat this test with the same wires, etc.

Then you simply look to see how you are doing with respect to the 66% warranty. When you start to approach 66% of 14.4 kWh, then you have a chance at a warranty repair.

Maybe the argument is that the 'losses' are fixed, while the battery capacity isn't (it is degrading). OK, so maybe that means the 66% criteria must be tweaked a little to compensate for this? If these fixed losses are let's say 15%, then your threshold for warranty might be 85% of the 66% warranty threshold, or 56%?

Am I missing something?
 
Last edited:
No, I agree, but putting 14.4 kWh in the black box is not 14.4kWh into the battery, that was all I was saying. If you have a fancy charger than can track data such as you, then you have a good starting point. I just meant that 9.5 kWh would likely be much less than the 66%. Sorry to sound so technical, just figured with the inclusion of graphs it was a fun discussion.
 
Wait.. doesn't this come out in the wash.

Let's use the 85% 'efficiency'.
Out of the 14.4 kWh input, only 12.2 kWh makes it to the "battery" (this is a 'good' battery).

If the "battery" degrades to 66%, then it is now storing only 8.1 kWh. At the input side, that would be 9.5 kWh (warranty threshold).

9.5 at the input vs. the good value of 14.4 is still a ratio of 66%.
 
MrFixit said:
This is equivalent to a Level 1 charge for 12 hours (1.2 x 12 = 14.4 kWh).
Click to expand...
Seems like this could be the 'standard' test. Run your battery to zero EV. Plug in the Honda charger. If the charger is still running at 10 hours, your battery is pretty good. A warranty problem shows up with a charger that stopped about 9 hours or less. New battery stops about 12 hours.

I like the idea that you get a go / no go at 10 hours... Nobody wants to sit and watch a charger.

The test could include checking the charger at hourly intervals, for more information. ...or you could enjoy social distancing, and watch the charger...
 
ClarityBill said:
I like the idea that you get a go / no go at 10 hours... Nobody wants to sit and watch a charger
Click to expand...
66% of the 12 hour nominal case is more like 8 hours, but I do like your thought.
Check the charger at 8 hours, 9 hours and 10 hours... Some criteria like this:

Off_Time Status
< 8 hrs BAD < 66% capacity
< 9 hrs POOR < 75% capacity
< 10 hrs FAIR < 83% capacity
> 11 hrs GOOD > 90% capacity
 
Changing our topic a little, but back to the original question:

What would likely be the problem with his EV range, if the battery has not failed?
For me, the heater drops EV range, but his ambient temperature shows 70F.
I know a bad wheel bearing, or stuck brake pad will drop the range (my experience is gasmobile MPG).
I was driving my Clarity with a busted up bumper without range loss, so aerodynamics don't seem that big.

Any other obvious choices?
 
New Clarity HV Battery = 55AH (from documents available) X 348Vmax (I measure it at 343.46v on mine but that is not OCV, Honda's safety guide says 352v max) = 19.14 kWh x .66 = 12.63 kWh < "battery failure point", you are absolutely correct that we don't access the full battery capacity due to the very discussed buffers of roughly 2.14 kWh (by Gov Spec to 17kWh). 12.63 kWh - 2.14 kWh = 10.49 kWh. 10.49 kWh / 348 = 30.1 Ah. Which is below the warranty threshold Honda capacity of 66% at <36.6 Ah...

*edit: forgot to mention that the lower top voltage is due to it not being OCV and I was charging level 1 which likely tops lower due to the controller seeing the lack of inrush current.*

The point is that Honda bases the "warranty" on the battery system total amp hours, not the usable amp hours we can charge or discharge. You are absolutely correct that it is a black box and charge delta from "new" to "failure" is just a ratio and doesn't matter. Charging efficiency is just additional lost headroom that will only make your run to the dealer for warranty more a sure thing. However with range now around 50% of "new" on average (for this hypothetical owner), I am betting no one will be surprised, and that goes back to the original post.

I may be completely wrong about the math. I am currently trying to teach my 6th grader her math class and my 2nd grader her language arts and probably will be unable to speak English by the end of the day....

Also everyone comes at this question from a different perspective and philosophy, for me - (I know no one is surprised) it is to get a new "free" battery at 7 years and 11.5 months. So that's the data point I watch.

Cheers,

Cash
 
Last edited:
ClarityBill said:
Any other obvious choices?
Click to expand...

I don't think a mechanical condition causing drag (as you postulate, dragging brakes or wheel bearing) seems plausible. With that much of a 'hit', I think you would have severe symptoms like smoke pouring out of a red-hot caliper, or extreme grinding from a bearing.

If I were the OP, I would start with the battery disconnect just to see if a cold reboot would clear out some kind of a software glitch. If that didn't work, it would be off to the dealer for a battery test.

I hope we hear back... I hate loose ends like this.
 
Last edited:
You must log in or register to reply here.
Back
Top