MG S5 ev: using the API #3

Does my S5 stop maintaining the 12V battery when ev battery SoC drops to 20%?

There seems to be a lot of 'chatter' about electric vehicles based upon supposition.

Opinions posted on the net often get repeated, then eventually get peddled as fact.

Of course this is not a problem unique to the ev.

But I've heard people say that ev functionality changes when the traction battery SoC falls to 20%. And it seems reasonable that protecting the main battery might be more important than maintaining the charge of the system 12V battery; after all, its easier and quicker to re-charge the system battery.

However, its worth remembering that the ev battery on our Long Range S5 has almost 2000 times the capacity of the system 12V battery. So maybe any safety limit (if such a thing exists) would kick-in much less than 20%.

When the SoC on our car dropped below 20% recently, it seemed like an ideal time to check whether periodic 12V battery charging was, or was not maintained.

So here is a rather 'busy' plot, taken from api data...

To clarify:-

• ChargingEV (red) is not a voltage, but a 'state'; red blocks indicate the times when the ev was being charged.
• systemBattery (blue) is the reported system (12V) battery voltage.
• Charging12V (green) is a marker set at an arbitrary level, above which the system battery is definitely being charged from the main ev battery.
• surfaceCharge (amber) is another threshold; battery voltage between amber & green lines represents unstable 'surface charge' following the end of charging.

Unfortunately the x-axis is not linear. This is because not all requests for data via the api are successful, so some data points are missing.

However, the point of the test is that it shows the 12V battery is still being topped-up from the main (traction) battery, even when SoC is less than 20%. The lowest SoC during this test was 15.3%.

api comms reliabilty

Its a bit of a puzzler why comms via the api can appear very stable for some quite long periods (days), while at other times, its pretty dreadful.

When it works, it works well!

 

During testing, there was a period when all requests for data returned "error 4: The remote control instruction failed...".

Plugging in the ev cable (but with charger off) appeared to wake comms and it ran happily for quite a while. However, this could have been a chance thing! I need a lot more testing before suggesting that was, in some way, by design.

using our car as a test-bed

It would be nice to carry out lots of tests on our S5; maybe allowing the SoC to drop right down to 2 or 3% and see if 12V battery charging is maintained.

However, I have to keep reminding myself that this is our family car; even if I'm not planning to drive anywhere, maybe my wife is!