[xolpy]

This concerns a new 2019 Accord Hybrid (~7000 miles). Under normal operation under 75 mph, the battery cycles between 20-50% charge on the highway, with the engine coming on intermittently to charge the battery.

However after driving for more than 60-90 minutes, the engine no longer cycles and stays on. During this time, the battery will charge but not discharge, causing the charge level to climb to 100%. Trying to force EV mode results in a message "Battery Temperature at Limit", despite the outdoor temperature being anywhere from 20-70 deg F and cabin temperature being 66-70 deg F. This condition resets after about 10-15 minutes of driving, with the engine cycling on and off again. After a further 10-15 minutes, the engine will be stuck on again. This cycle results in a loss of about 5 MPG over the standard behavior.

My read on this is that the battery is used for both regenerative braking as well as adding or removing the torque load from the engine to keep it at the optimal brake specific fuel consumption for the given road speed (and RPM at lockup). When the battery goes into an overheat condition (i.e. battery temperature at limit) the ECU no longer uses the torque loading/unloading trick, resulting in the engine operating at suboptimal BSFC and causing the MPG loss. The battery is only used for regenerative braking, which means that the charge rises until the battery eventually cools and the condition resets.

My question is, is this something that any of you have noted? Does your MPG drop after > 1 hour of driving? Is there something wrong with my battery/cooling system? I'm quite skeptical that Honda would design the system to not be able to cool the battery down at steady state highway driving.

 

[jackpine]

I share your skepticism. My Hybrid does NOT do what yours does. Possibly, there's a problem with the cooling system for the battery. And this is just a check, are the cooling vents under the front of the back seat clear?

However, I don't believe this is the way your car should function. It's under warranty, so I'd take it in to the dealership if I were you.

- Jack

 

[themacfreak1]

This is NOT normal behavior. Even in hot Florida temperatures at highway speeds, my HAH continues to cycle between EV and hybrid modes. I have never had a battery overheating alert.

 

[xolpy]

Thanks for the feedback. I’ll take it in to the dealer as soon once we get out of lockdown here. Stay safe everyone.

I share your skepticism. My Hybrid does NOT do what yours does. Possibly, there's a problem with the cooling system for the battery. And this is just a check, are the cooling vents under the front of the back seat clear?

However, I don't believe this is the way your car should function. It's under warranty, so I'd take it in to the dealership if I were you.

- Jack

Thanks, I checked the vents, and found them clear. I can hear the fan inside running too. Its a pity that the car does not expose any of the battery temperatures or the cooling fan speed over OBD2.

 

[Zx12-iowa]

Ditto. Mine does not do this...

 

[JeffJo]

This concerns a new 2019 Accord Hybrid (~7000 miles). Under normal operation under 75 mph, the battery cycles between 20-50% charge on the highway, with the engine coming on intermittently to charge the battery.

However after driving for more than 60-90 minutes, the engine no longer cycles and stays on. During this time, the battery will charge but not discharge, causing the charge level to climb to 100%. Trying to force EV mode results in a message "Battery Temperature at Limit", despite the outdoor temperature being anywhere from 20-70 deg F and cabin temperature being 66-70 deg F. This condition resets after about 10-15 minutes of driving, with the engine cycling on and off again. After a further 10-15 minutes, the engine will be stuck on again. This cycle results in a loss of about 5 MPG over the standard behavior.

It seems your cooling system is not working properly. Both charging and discharging the HV battery will generate heat in the battery. The outside temperature doesn't really matter, except for how it affects the cooling system. In normal operation, the system should allow the battery to reach a steady temperature where either mode's heat generation is balanced by heat dissipated by the cooling system.

Yours doesn't. So after 60-90 minutes, it has overheated and won't let you discharge anymore. What concerns me, is that it does let you charge. It continues to accept charge until the battery reaches its 100% level (which probably isn't all the battery can hold). Then the engine should stop generating as much power, the battery will cool, and the normal function will be restored for a while.

My read on this is that the battery is used for both regenerative braking as well as adding or removing the torque load from the engine to keep it at the optimal brake specific fuel consumption for the given road speed (and RPM at lockup). When the battery goes into an overheat condition (i.e. battery temperature at limit) the ECU no longer uses the torque loading/unloading trick, resulting in the engine operating at suboptimal BSFC and causing the MPG loss.

Close enough. To be picky, the traction motor is used for regenerative braking, and the power it generates is sent to the battery. As for the engine, here is Honda's strategy in graphical form:

Note that the curved lines through the plot represent constant power. And yes, for my critics, I'm going to explain it again. Because it is appropriate for this question.

1. Below about 45 mph and at low power requirements, the car is in Hybrid Drive. The engine drives only the generator. Without the battery, it would have to pick a blue dot that matches the power requirements of the car. With the battery, it will move to the center purple dot (2000 rpm, 88.5 lb-ft of torque, BSFC of 208.1 g/kWh) and adjust the flow of power, from the generator to the motor, by using the battery.
2. Between about 45 and 80 mph and at low power requirements, the car is in Engine drive (the clutch engages). Without the battery, it would pick a yellow dot that matches the speed (2000 rpm = 54 mph) and power requirement. With the battery, it will move to a purple dot that matches speed, and now use the traction motor/generator to adjust the power sent to the wheels the same way.
3. At high power requirements, the car will go into Hybrid Drive, and the engine will move up along the blue line (called the "Hybrid Drive Line") in what I call "blender mode." All the power is sent to the motor, and maybe a little more is added from the battery
4. And when the battery has sufficient charge, the engine will turn off and use EV drive.

The battery is only used for regenerative braking, which means that the charge rises until the battery eventually cools and the condition resets.

This is a common misunderstanding, and the main reason I'm explaining things. Most of the charging done is from "charge" mode in #1 or #2 above. Regen braking is used, well, only when you are braking or coasting. What concerns me is that your car is doing this even when the battery is too hot for EV drive. The power levels are lower, but in my opinion it should stop all battery usage.

My question is, is this something that any of you have noted? Does your MPG drop after > 1 hour of driving? Is there something wrong with my battery/cooling system? I'm quite skeptical that Honda would design the system to not be able to cool the battery down at steady state highway driving.

No, it is not normal. The steady temperature that the cooling system achieves is the most efficient temperature for the battery. Your loss of mpg isn't due to temperature, it is due to running the car at the yellow or blue dots after the battery reaches 100%.

 

[InTotalAccord]

One other thing to check at home is whether the level of the coolant for the inverter is correct. There is a separate reservoir for the inverter coolant (separate from the regular radiator and its reservoir). Off the top of my head I don't recall how easy or hard it is to check the level of inverter coolant. The owner's manual shows where it's located under the hood.

 

[icspots]

One more person chiming in to say mine doesn’t do this either.

 

[xolpy]

Yours doesn't. So after 60-90 minutes, it has overheated and won't let you discharge anymore. What concerns me, is that it does let you charge. It continues to accept charge until the battery reaches its 100% level (which probably isn't all the battery can hold). Then the engine should stop generating as much power, the battery will cool, and the normal function will be restored for a while.

That was my supposition earlier. I think there are two phases of overheating.

At lower temperatures, it disables EV mode, disables battery use for traction motors, and only permits regen braking. Only indicator is that pressing the EV button shows the message "battery temperature at limit".

At higher temperatures (i.e. Power Reduced warning, power system light), it disables all use of the battery. I have not experienced this yet.

I suspect it is an MPG impact issue. Honda probably is trying to reduce the MPG impact for vehicles that are regularly operated at high temperatures.

edit: One more possibility here. Honda has been very cautious to optimize for consistent brake feel here. I also suspect that the retention of regen braking under mild overheat conditions is an optimization to ensure that brake feel remains consistent when the system is operated in hot climates. Also similar to how the car uses the generator coupled to engine braking to dissipate energy instead of the service brakes when the battery is full.

This is a common misunderstanding, and the main reason I'm explaining things. Most of the charging done is from "charge" mode in #1 or #2 above. Regen braking is used, well, only when you are braking or coasting. What concerns me is that your car is doing this even when the battery is too hot for EV drive. The power levels are lower, but in my opinion it should stop all battery usage.

I don't think I have misunderstood the system here. I meant that under low overheat condition ("battery temperature at limit") that the BSFC optimization is switched off and regen is left on. I don't know the threshold temperatures for the states, since I don't have access to the sensors under OBD2. I'm under no impression that the battery is only used for regen braking under normal operating conditions.

One other thing to check at home is whether the level of the coolant for the inverter is correct. There is a separate reservoir for the inverter coolant (separate from the regular radiator and its reservoir). Off the top of my head I don't recall how easy or hard it is to check the level of inverter coolant. The owner's manual shows where it's located under the hood.

I'll check it, but I doubt so. If the inverter overheats, I suspect that it would show the "Power reduced" message, since engine power still has to be routed through the inverter when the clutch is not at lockup.

 

[JeffJo]

I don't think I have misunderstood the system here. I meant that under low overheat condition ("battery temperature at limit") that the BSFC optimization is switched off and regen is left on. I don't know the threshold temperatures for the states, since I don't have access to the sensors under OBD2. I'm under no impression that the battery is only used for regen braking under normal operating conditions.

You expressed more knowledge than most have, which is why I directed that comment at the general population. But what I was trying to say was that it is "BSFC optimization" that is causing your State of Charge to continue to rise when you can't go into EV drive. It really is just semantics that distinguishes it from "regen braking." All "BSFC optimization" means that you are running the engine harder than it needs, and applying "brakes: to absorb the difference. The point is that you can't turn either off as you described

 

[Zx12-iowa]

xolpy.

I just took a 700 mile drive and watched the battery charge bc I was curious am based on this thread. What I noticed is that unlike driving in town where my battery charge runs the gamut, the interstate drive had my battery charged right about mid level most of the drive and it only fluctuated up/down by one bar. I drove from Alabama to South Florida with temps approaching 90 and full sunshine-weather report to show even with heat I had no overheating/overcharging.

 

[xolpy]

xolpy.

I just took a 700 mile drive and watched the battery charge bc I was curious am based on this thread. What I noticed is that unlike driving in town where my battery charge runs the gamut, the interstate drive had my battery charged right about mid level most of the drive and it only fluctuated up/down by one bar. I drove from Alabama to South Florida with temps approaching 90 and full sunshine-weather report to show even with heat I had no overheating/overcharging.

Thanks for the interest. I saw in the other thread that you were driving 75-80, and I think the car behaves differently in that range. I refer to the figure posted by JeffJo in post #6, also linked in the Honda R&D paper: Development of SPORT HYBRID i-MMD Control System for 2014 Model Year Accord . The below discussion will apply for normal vehicles, and not one with the overheating batteries.

At 45-70 mph speed, the steady state torque required (dotted black line) is less than the BSFC-optimal line (solid blue line). That means that increasing torque loading on the engine by way of the generator increases efficiency. Therefore, the engine cycling is an optimal strategy.

At >=75 mph speed, the steady state torque required is greater than the BSFC-optimal line. At this point, nothing you can do can increase the steady state engine efficiency. The car thus operates in engine only mode, and you won't see the engine cycling too much unless you reduce torque by allowing the car to slow down or going downhill.

If you drive at 70 or less, I suspect you will see the battery charge fluctuate from 20-50%.

You expressed more knowledge than most have, which is why I directed that comment at the general population. But what I was trying to say was that it is "BSFC optimization" that is causing your State of Charge to continue to rise when you can't go into EV drive. It really is just semantics that distinguishes it from "regen braking." All "BSFC optimization" means that you are running the engine harder than it needs, and applying "brakes: to absorb the difference. The point is that you can't turn either off as you described

No worries. I get what you are saying. I don't think so, the charge doesn't accumulate appreciably when under overheat conditions when driving steady state torque on the highway. It does accumulate when I start going downhill or letting off the gas, when I see the "tach" needle drop into the green range, which indicates some regen is going on. I can also see some differences in the instantaneous MPG figures as well, with it being higher when the battery is in overheat mode, since the engine isn't being run harder for BSFC optimization.

Managed to drop my car off at the dealer earlier this week, so I'll have more to report when it is back. If I still had my car with me, I could post the OBD2 data since MPG is reported there.

 

[xolpy]

Just talked to the dealer. Took a week for the Honda field rep to take a look at it, and they claimed that it is normal. Not sure what to make of it.

 

[Zx12-iowa]

Well I don’t think it is normal at all. Hmm. Frustrated for you on this. But very much appreciate your update.

 

[xolpy]

Yeah, been pretty busy this week, but I plan to write an email to Honda America to see if I can get any traction there.

Sent American Honda the following email. Fingers crossed. Also attached the redacted service report.

My significant other purchased the first 2019 Honda Accord Hybrid in Aug 2019 (VIN xxx2846). We liked the performance so much that I purchased a second 2019 Honda Accord Hybrid for myself in Dec 2019 (VIN xxx8373). Unfortunately, I noticed significant performance differences between the two vehicles when driven under substantially similar conditions.

The second Accord (VIN *8373), when driven for 60-90 minutes, at 60-70 mph, with outdoor temperatures between 20-70 deg F, and cabin temperatures between 66-70 deg F, goes into a battery thermal protection state. During this condition, the engine, which normally cycles on and off on the highway and shuts off when the car is stopped, does not switch off. The battery is no longer used to power the traction motors, and charges to full due to regenerative braking. During this mode, pressing the “EV mode” button results in a message, “EV mode unavailable, battery temperature at limit”.

This causes a loss of 5 MPG and persists for 8-20 minutes before reverting to normal behavior. After an additional 10-15 minutes of driving the battery goes into the thermal protection state again. This cycle repeats until the vehicle is parked for an extended period of time. This behavior is neither observed on the first Accord Hybrid (VIN *2846), nor observed by other Accord Hybrid drivers (Battery temperature at limit after long drives?).

I took the car into my dealer, xxxx, and I attach the service report. During the 12 days that the vehicle was in the shop, they indicated that they “had field rep take vehicle on extensive road test, and all EV functions are operating normally” and said that there was nothing that could be done. However, mileage records indicate that they only drove the vehicle for 19 miles, which is insufficient to replicate the battery thermal condition listed above, which requires at the minimum 60 – 90 minutes of highway driving.

To date, I have not received either a resolution to this issue, or a technical explanation about why the vehicle is considered to be “operating normally” despite the issues noted above. I’m also disappointed that there was no attempt to reproduce the issue despite the extensive amount of time that it was in the shop. I remain ready and willing to assist in the diagnosis process, by either speaking to an engineer or service technician, and/or driving the vehicle sufficiently to reproduce the problem immediately before taking it into the shop.

Please let me know how you plan to address my concerns above.

 

[OG1]

I'm sorry you're going thru this. The error codes alone should be enough to warrant some action as that's obviously not normal. The fact that they had your car for two weeks and only drove it 19 miles, while claiming they had done extensive road tests, is infuriating. It almost sounds like they didn't take you seriously and a tech never even looked at the car.

Good luck and keep us posted.

 

[xolpy]

I'm sorry you're going thru this. The error codes alone should be enough to warrant some action as that's obviously not normal. The fact that they had your car for two weeks and only drove it 19 miles, while claiming they had done extensive road tests, is infuriating. It almost sounds like they didn't take you seriously and a tech never even looked at the car.

Good luck and keep us posted.

Thanks for your interest. The issue here is that there is no diagnostic trouble code raised in the car during this situation. I suspect that there are two levels of overtemp protection. At the first level, observed in my case, no DTCs are raised, and the car switches off some engine optimization functionality to reduce use of the battery. This is to prevent DTCs when a fault-free car is started in hot weather.

It is frustrating, particularly since I can reproduce the problem easily.


For all who are interested, I collected data during my drive today. Planning to do another drive tomorrow with my SO's car to collect the same type of data. This should hopefully convince the dealership or American Honda that there is a real problem with the vehicle.

Problem Description.pdf

Shared with Dropbox

www.dropbox.com

 

[xolpy]

Problem Description.pdf

Shared with Dropbox

www.dropbox.com

I've updated the slide deck above with additional data from my drive today with a different (and normal) HAH2019.

 

[InTotalAccord]

Very compelling data, xolpy. It also shows that to experience/recreate the problem, it may be necessary for a Honda tech to ride with you in the car for maybe an hour, to experience the thermal issue..... It also may explain their inability to recreate the problem.

Have you paid attention to the "power flow" car graphic, when the problem occurs? I think you are theorizing that when the problem is happening, the battery is not discharging to the main electric drive motor, and that therefore the drive motor is only getting electric power from the ICE-driven generator. The power-flow graphic MIGHT actually show some difference when the problem is happening, possibly showing electricity flow from the generator to the drive motor, but not from the battery to the drive motor. Just a thought....

I think it is the case that the inverter is part of the path from the battery to the drive motor, as it converts high-voltage DC to an AC voltage for the drive motor. The inverter also has a separate coolant reservoir, that is separate from the ICE radiator. (It also uses long-life antifreeze, like the radiator, but there is a separate reservoir.) You might want to make sure that the inverter's coolant level is correct. Inverter overheating might be the culprit, rather than battery pack overheating. (Although, maybe the messages you are getting implicate the battery pack?).

Just a couple of thoughts, while you are waiting to get the car looked at again by Honda....

 

[xolpy]

Very compelling data, xolpy. It also shows that to experience/recreate the problem, it may be necessary for a Honda tech to ride with you in the car for maybe an hour, to experience the thermal issue..... It also may explain their inability to recreate the problem.

Yeah, I'm hoping that they will allow me to roll the car into the shop after I drive it around for 2 hours. I get that they can't just send a guy driving for 1-2 hours on a whim, so I did offer to drive it. They didn't take me up on my offer. We shall see what happens after American Honda gets back to me.

Have you paid attention to the "power flow" car graphic, when the problem occurs? I think you are theorizing that when the problem is happening, the battery is not discharging to the main electric drive motor, and that therefore the drive motor is only getting electric power from the ICE-driven generator. The power-flow graphic MIGHT actually show some difference when the problem is happening, possibly showing electricity flow from the generator to the drive motor, but not from the battery to the drive motor. Just a thought....

I haven't been monitoring that, but I will. The battery is still used in the city for regenerative braking though, so I don't think there will be a functional difference in graphics there except that the car never goes into EV mode. On the highway, however, I fully expect that the graphic will be engine drive only, just like when you take the car above 75 mph.

I have another plot of the battery current draw here, and there is still some usage of the battery during fault state, but definitely far less than during normal operation.

I think it is the case that the inverter is part of the path from the battery to the drive motor, as it converts high-voltage DC to an AC voltage for the drive motor. The inverter also has a separate coolant reservoir, that is separate from the ICE radiator. (It also uses long-life antifreeze, like the radiator, but there is a separate reservoir.) You might want to make sure that the inverter's coolant level is correct. Inverter overheating might be the culprit, rather than battery pack overheating. (Although, maybe the messages you are getting implicate the battery pack?).

I checked the inverter coolant levels and found them to be normal:

I think that inverter overheating is not the issue here. There are two inverters in the PCU, one that drives the generator and another that drives the traction motors. Any power flow that is not through the clutch would have to go through at least one inverter. Two inverters being used in engine-on city driving (engine>generator>3-phase>inverter>DC>inverter>3-phase>traction motor) because the clutch cannot lockup.

During such an inverter overheat condition, power would be reduced in city driving and acceleration to highway speeds. This is not observed.

Just a couple of thoughts, while you are waiting to get the car looked at again by Honda....

Thanks for the interest and advice. I just hope that the slides that I sent actually get looked at by someone technical enough to see the problem. I also can't imagine how much more difficult it would be to prove my case if I were unable to collect my own data.