Extender battery not fully charging

My current project is a 2015 leaf with 24kWh battery @ 58% SoH

I have added an extender battery into the boot, from a 2015 30kWh leaf @ 62% SoH. This is wired directly into the PDM, using the same bolts that connect busbars inside the PDM. This avoids possible water ingress into spliced HV wiring. Balancing is done weekly with a separate LBC. It all works well. No fault codes. Range is extended but I haven't run it below 30% SoC yet

Combined battery capacity should be around 32kWh

I can fast charge it to 100% SoC

But it doesn't slow charge (L2) beyond about 75% SoC

Any ideas what is causing this behaviour and what needs to be done to fix it?

TIA

Hi xafalcon :

I'm interested in this, where is your expansion battery

As far as I know, this is normal behaviour. 75% of 32kWh is 24 kWh.

The onboard charger will only charge the amount that it thinks the battery should be able to handle. You will need a CAN-bridge to fix this behaviour.

FinnishCoffee said:

As far as I know, this is normal behaviour. 75% of 32kWh is 24 kWh.

The onboard charger will only charge the amount that it thinks the battery should be able to handle. You will need a CAN-bridge to fix this behaviour.

Click to expand...

Thanks fir your reply

I'm not sure how the leaf "knows" how much power has gone into the battery???

Only half the power is going through the current sensor inside the battery pack. The only other current sensor is in the inverter, measuring motor power

And why does fast charging allow a full charge? Maybe a charging algorithm based on cell voltage would answer this question. But it seems odd that there would be different charging algorithms for fast and slow charging

At the moment I'm ok with fast charging it to get it full, as local chargers at shopping malls are only 25kW = 12.5kW per battery. But in NZ there is a rapid increase in EVs and its getting harder to find a vacant charger. So the home charging (to full) is needed soon

My understanding of the can bridge was to install larger factory batteries in place of the standard factory battery. All electrical power going through the current sensor

bonnechiang said:

Hi xafalcon :

I'm interested in this, where is your expansion battery

Click to expand...

It's in the boot (trunk) floor. It raises the floor by the width of the modules, plus the new lining on top.

I used 10mm hardboard to make the boot floor flat. I have used the factory mounting hardware to mount the batteries into the boot, with several 50mm x 700mm steel reinforcing straps under the floor to spread the load in case of collision. The factory bus bars, cut out fuse, contractor box were also used

The 30kWh modules were from a car that I replaced the battery with a better one. It was perfect for this extender battery experiment

Send me a pm with your email and I'll send you a picture

email: [email protected]

I approved your post, Bonnie, but you should be aware that 'bots and spammers can 'harvest' any personal information you post here. You can mask email addresses by writing, say, "dot com" instead of the literal address. You can change or remove that info just by editing your post.

xafalcon said:

FinnishCoffee said:

As far as I know, this is normal behaviour. 75% of 32kWh is 24 kWh.

The onboard charger will only charge the amount that it thinks the battery should be able to handle. You will need a CAN-bridge to fix this behaviour.

Click to expand...

Thanks fir your reply

I'm not sure how the leaf "knows" how much power has gone into the battery???

Only half the power is going through the current sensor inside the battery pack. The only other current sensor is in the inverter, measuring motor power

And why does fast charging allow a full charge? Maybe a charging algorithm based on cell voltage would answer this question. But it seems odd that there would be different charging algorithms for fast and slow charging

At the moment I'm ok with fast charging it to get it full, as local chargers at shopping malls are only 25kW = 12.5kW per battery. But in NZ there is a rapid increase in EVs and its getting harder to find a vacant charger. So the home charging (to full) is needed soon

My understanding of the can bridge was to install larger factory batteries in place of the standard factory battery. All electrical power going through the current sensor

Click to expand...

The onboard charger keeps track of energy fed into the battery for some unknown reason. This was an issue in the early days of bruteforce Leaf battery swaps.

DC fast charging bypasses all of the car's circuitry and connect direcly to the battery DC bus, and all charge controls are done by the external charger. Thus a DCFC can charge the extended battery fully.

The onboard charger keeps track of energy fed into the battery for some unknown reason.

Click to expand...

Presumably as a fail safe way to prevent overcharging.

FinnishCoffee said:

xafalcon said:

FinnishCoffee said:

As far as I know, this is normal behaviour. 75% of 32kWh is 24 kWh.

The onboard charger will only charge the amount that it thinks the battery should be able to handle. You will need a CAN-bridge to fix this behaviour.

Click to expand...

Thanks fir your reply

I'm not sure how the leaf "knows" how much power has gone into the battery???

Only half the power is going through the current sensor inside the battery pack. The only other current sensor is in the inverter, measuring motor power

And why does fast charging allow a full charge? Maybe a charging algorithm based on cell voltage would answer this question. But it seems odd that there would be different charging algorithms for fast and slow charging

At the moment I'm ok with fast charging it to get it full, as local chargers at shopping malls are only 25kW = 12.5kW per battery. But in NZ there is a rapid increase in EVs and its getting harder to find a vacant charger. So the home charging (to full) is needed soon

My understanding of the can bridge was to install larger factory batteries in place of the standard factory battery. All electrical power going through the current sensor

Click to expand...

The onboard charger keeps track of energy fed into the battery for some unknown reason. This was an issue in the early days of bruteforce Leaf battery swaps.

DC fast charging bypasses all of the car's circuitry and connect direcly to the battery DC bus, and all charge controls are done by the external charger. Thus a DCFC can charge the extended battery fully.

Click to expand...

Thanks for the explanation

Is there a way to overcome this?

How does the OBC keep track of energy fed into the battery? I can't see any current sensor in the charger section

https://imgur.com/HTFAvxY

In the diagram in this picture, within the on-board charger there is drawn a resistor in series with the positive line. This must be the mystery/offending current shunt/sensor. This is logical - almost all switch mode power supplies have a current sense resistor/shunt in order to prevent damage from overloading.

tripper1000 said:

https://imgur.com/HTFAvxY

In the diagram in this picture, within the on-board charger there is drawn a resistor in series with the positive line. This must be the mystery/offending current shunt/sensor. This is logical - almost all switch mode power supplies have a current sense resistor/shunt in order to prevent damage from overloading.

Click to expand...

Thanks. Unfortunately I can't expand the picture enough to decypher it.

A CT (current transformer) sensor is located on the battery box. There is another one on the DC side of the inverter. The former reads half the actual current (and half goes to/comes from the extender battery), and the latter reads correctly. I was thinking of adding a bypass around the inverter CT, to halve the current this sensor reads. If I use the same cross section copper, the resistance will be the same down both paths, and the CT will read exactly half, mirroring the battery box CT sensor situation

FinnishCoffee said:

xafalcon said:

FinnishCoffee said:

As far as I know, this is normal behaviour. 75% of 32kWh is 24 kWh.

The onboard charger will only charge the amount that it thinks the battery should be able to handle. You will need a CAN-bridge to fix this behaviour.

Click to expand...

Thanks fir your reply

I'm not sure how the leaf "knows" how much power has gone into the battery???

Only half the power is going through the current sensor inside the battery pack. The only other current sensor is in the inverter, measuring motor power

And why does fast charging allow a full charge? Maybe a charging algorithm based on cell voltage would answer this question. But it seems odd that there would be different charging algorithms for fast and slow charging

At the moment I'm ok with fast charging it to get it full, as local chargers at shopping malls are only 25kW = 12.5kW per battery. But in NZ there is a rapid increase in EVs and its getting harder to find a vacant charger. So the home charging (to full) is needed soon

My understanding of the can bridge was to install larger factory batteries in place of the standard factory battery. All electrical power going through the current sensor

Click to expand...

The onboard charger keeps track of energy fed into the battery for some unknown reason. This was an issue in the early days of bruteforce Leaf battery swaps.

DC fast charging bypasses all of the car's circuitry and connect direcly to the battery DC bus, and all charge controls are done by the external charger. Thus a DCFC can charge the extended battery fully.

Click to expand...

I have thought about this and am now of the opinion it is not relevant to my situation. I'll explain why I have come to this conclusion

The bruteforce battery swap used higher capacity modules and the original LBC and other battery box components. The battery architecture remained factory, but with larger capacity modules. The capacity limitation is built into the LBC, and measured through the CT sensor inside the battery pack. The battery CT sensor reads correctly in this instance (unlike mine, which reads half the actual current), and once 284 GIDS are present in the battery, it stops accepting charge

In my situation, the LBC records about 160 GIDS when the battery is fully charged by a fast charger. Cell voltages are about 4.1V, confirming a full charge. At a fast charger I can see the current being delivered by the charger eg 50 amps, and through leafspy I can see current being recorded by the battery CT sensor = 25 amps in this example

The only other CT sensor is in the DC side of the inverter, which records the true current feeding from both batteries into the inverter. I'm thinking of adding a bypass around that inverter CT, to effectively halve this reading. As I'm thinking this mismatch may be part of the problem

I would welcome any counter-comments if anyone thinks I have this wrong