MG5 Supercharger: FAQ 2 - What upgrades does my MG5 need to use a Tesla charger?

FAQ 2 - What upgrades does my MG5 need to use a Tesla charger?

What is the current situation?

This is the MG buelltin posted by Miles Roberts describing the EVCC upgrade. I found it worth while having this with me to show dealers who may not be aware of it.

Note that the EVCC part numbers are NOT the same as on the label. Based on owner information, we have established that the corresponding part numbers on the label are: 11170476, 11182835, 11313393.

Do I need any hardware upgrades?

The answer is a definite maybe. The EVCC fix is only compatible with three versions of the EVCC module as stated in the bulletin at the top of this FAQ. These were only fitted to PFL LR models. This means that the PFL SR and possibly, very early LR models will need the EVCC replaced.

How to identify which EVCC is fitted?

The EVCC is located next to the 12 volt battery as shown. The label is placed facing the battery and there is a small gap between the two that an endoscope can be used to see the numbers on the label. Failing that, the cover needs to be removed and the unit unbolted to expose the EVCC. The key number is the one beginning with three 1's. It should be marked EVCC-02. If it is an EVCC-01 unit, then that needs to be upgraded. If it is an EVCC-02 with a number that is different to 11170476, 11182835, 11313393, it most likely will need replacing as well. The units all look physically the same.

Can I use the eZS app to identify them?

No. While it reports firmware and hardware revisions, these are the manufacturer's (VMAX) use only. They do not change with the upgrades as the before and after screen shots show.

eZS app is android based and was written in Thailand for the MG ZS. It requires a ELM327 compatible OBD2 interface to talk to the car. It does seem to work with the MG5 but note that charging and BMS values are based on the Thai implementation which produces slightly different values.


What software upgrades are necessary?

There are two: SC052 and the EVCC fix. The SC052 EP22£EP21£ZS11EMCE CCU Software Update will not fix the Tesla problem on its own but it does address the HV battery fault alert caused by the 12V battery not providing a high enough voltage to start the system. This is done by raising the DCDC battery voltage to around 14.6V thus allowing the 12V battery to be fully charged. It does improve the CCS charging reliability - presumably because the 12V voltage is a bit higher - so it is worth making sure it is done. It is a warranty FOC repair. The EVCC fix can be carried out at the same time - mine was - but that is according to MG, chargeable.

Even if you do not have the EVCC fix, make sure that the SC052 update has been done.


Do I need to pay for the EVCC fix?

It depends on your car model. FL cars can have it under warranty FOC. For all other cars such as the PFL SR and LR models, it is classed as an enhancement and the dealer is charged, who then passes that onto the customer. This has prompted the great "warranty vs enhancement" debate.

Typical costs are 1 hour labour for a software upgrade and an additional £300 plus labour for a replacement EVCC module. The prices do vary because some dealers raise the prices because they simply do not want to do it. It can pay to shop around.

Dealers do seem to be taking a more pragmatic approach and if the car is in for other work, they are more amenable to do a free upgrade. However bear in mind they have to run a business and they may or may not need to charge you. That is their decision but ask nicely and who knows.

What happens if I don't get the upgrade?

Nothing. The car will still charge at other networks. If you charge mainly at home, there may be no real advantage in getting it, especially if you have to pay for it. Tesla superchargers cost a lot less - 50% or less compared to Ionity or Osprey - and so the decision to do so can be simply based on financial savings that it offers. If you do a lot of public charging and an open supercharger is nearby, then it might be worth while as the payback could take only 3-4 sessions.

I would recommend making sure that you get the SC052 upgrade - it's free and should give more reliable charging.

 

[Coulomb]

This is done by raising the DCDC battery voltage to around 14.6V thus allowing the 12V battery to be fully charged.

Is this change all the time that the DC-DC is running, or only while charging?

Either way, it seems unwise to me to leave the battery voltage at 14.6V, as this is above the gassing voltage for lead acid batteries (14.4V @ 25°C). The proper fix for the auxiliary (12V) battery not being charged enough to start the car would be to have a computer wake up at regular intervals, check the 12V battery voltage, and go back to sleep if it's OK. Otherwise, if the main battery is above about 30% SoC, then turn on the DC-DC for 10 minutes or so, or until a certain voltage is reached, or the charge current reduces below a threshold. That's what my 2012 Nissan Leaf does.

I'm also confused about linking the low auxiliary battery voltage issue with the Tesla supercharging issue, unless there is some requirement for a high auxiliary battery voltage specifically during supercharging. I can't think of a reason why this would be the case.

This is another great article, shpub, thanks for taking the time to write them.

 

[shpub]

Is this change all the time that the DC-DC is running, or only while charging?

It is not permanent and it drops down to between 14.1 and 14.3 volts. This is the display voltage reading BTW so again don't know what error it has and perhaps more importantly what the corresponding current is.

Either way, it seems unwise to me to leave the battery voltage at 14.6V, as this is above the gassing voltage for lead acid batteries (14.4V @ 25°C). The proper fix for the auxiliary (12V) battery not being charged enough to start the car would be to have a computer wake up at regular intervals, check the 12V battery voltage, and go back to sleep if it's OK. Otherwise, if the main battery is above about 30% SoC, then turn on the DC-DC for 10 minutes or so, or until a certain voltage is reached, or the charge current reduces below a threshold. That's what my 2012 Nissan Leaf does.

The problem with the battery voltage is that the DC-DC voltage was never high enough to fully charge the battery. As a result, the battery would occasionally not provide a high enough voltage to get the DC-DC to start. This would then be reported as a HV battery fault. Get arounds included pressing the brake pedal really hard to reduce the amount of electrical assistance that the iBoost would offer, switching everything off to reduce the power consumption and also switching everything off and coming back later. Others carry a small 12V pack to "jump start" the DC-DC.

The voltage does drop after the initial turn on so no need to worry about that. According to the Battery University (BU-403: Charging Lead Acid) charging at 2.40 to 2.45 volt per cell can protect against sulphation and give more consistent capacity readings at the expense of gassing as you say. On this basis, A short burst of higher voltage may be very advantageous given that the battery is only used as a to "start" the car.

I am now seeing high battery voltages - 13.4 - on turn on which is about 1 volt higher than before.

I'm also confused about linking the low auxiliary battery voltage issue with the Tesla supercharging issue, unless there is some requirement for a high auxiliary battery voltage specifically during supercharging. I can't think of a reason why this would be the case.

To be honest this does not surprise me one bit. The comms over the PLC is done by a QCA7000 which is reported to be a little bit fussy about its design e.g. power surges and drop outs. It is used in a lot of ethernet over mains adapters as well. For these there is a compliance test but it was found that while most units passed this and should be interoperable, they were not with very slow transmission and drop outs. While supposed to be the same design, very different performance issues.

It is also very susceptible to electrical noise. Just moving a couple of power cables on a test bench can turn an iffy connection into a good one.

The 12V battery powers the car during the charging negotiation. So if this supply is lower than expected, it is likely that the chip supply voltages may well vary/drop and so on causing comms problems.

So my conclusion is that the lower battery voltages can potentially cause voltage drops/surges which can then reset controllers and interferes with the Qualcomm chip that handles communication causing the charging negotiation to fail. Raising the battery voltage, then addresses this problem and thus improves charging reliability. It also addresses the HV battery fault on start and improves the battery reliability. Probably why MG made this a software recall.

This is another great article, shpub, thanks for taking the time to write them.

Thanks.

 

[Coulomb]

It is not permanent and it drops down to between 14.1 and 14.3 volts.

Glad to hear that.

I am now seeing high battery voltages - 13.4 - on turn on which is about 1 volt higher than before.

Huh. I'm surprised, but pleased.

The 12V battery powers the car during the charging negotiation.

Oh! I assumed that the contactors would be closed, and therefore the DC-DC would be operating. That explains everything, thanks.

Heh. No point closing the contactors, most charge attempts fail anyway. What a pessimistic algorithm!

I wonder if a better work around would be to start the charge in ready mode; presumably the car will switch to charging mode (not able to drive off, i.e. no longer ready), when needed. But possibly the firmware will cleverly point out that you can't charge in ready mode, as if the driver is the one in charge of the contactors.

 

[shpub]

Heh. No point closing the contactors, most charge attempts fail anyway. What a pessimistic algorithm!

Assume nothing and you will not be surprised. Keep it simple stupid! Great attitude to take when writing and debugging software. Gold star as far as I am concerned.

I wonder if a better work around would be to start the charge in ready mode; presumably the car will switch to charging mode (not able to drive off, i.e. no longer ready), when needed. But possibly the firmware will cleverly point out that you can't charge in ready mode, as if the driver is the one in charge of the contactors.

What you have just highlighted is one of the CCS compliance problems. The HV should not be connected until the negotiation is finished and both sides are happy. Well can we not connect the DCDC converter to power the 12v stuff? It doesn't say we can't. Nothing is going to happen... is it? There won't be a spike when we do... Oh I forgot we had the heaters, lights and radio on. 50A startup surge here we come!

If it happens and given the CCS comms fussyness, I would not bet on that staying connected. Just to be clear I have seen this occur but not on an MG so don't assume that this is how MG works. It is an example of a typical compliance problem. Believe me there are plenty of others. My mantra is keep it simple.

When I went through the posts - and there are a lot - it was not clear what the procedure actually was. Switch to ready then plug in and stand on one leg facing south, say "Hi Elon" and so on.

I started with keeping it simple: Park. Switch Off, Plug in and use the app to start and stop the charger. It worked straight away. Result. Happy Bunny.

 

[Gomev]

...

I started with keeping it simple: Park. Switch Off, Plug in and use the app to start and stop the charger. It worked straight away. Result. Happy Bunny.

Method works with every charger I've used. You could substitute 'wave contactless card/RFID' in place of 'use the app' and that works too.

My logic is that the the physical connector is dumb and the electronics/communications don't activate until initiated by whatever means (app/card) so it doesn't matter if it's in the holster or shoved in the car socket. Plugging it in first means no delay once activated.

 

[xblackwolf90]

Just as a heads up; I have a 2023 FL, and all dealers I've spoken to are stating the EVCC update is now chargeable, despite it originally being covered FOC.

 

[kevin ellard]

Just as a heads up; I have a 2023 FL, and all dealers I've spoken to are stating the EVCC update is now chargeable, despite it originally being covered FOC.

I think it is possible to get them to include it with the price of a service. They need some income to cover the time the car is in the workshop but once its there the update doesn't take long and doesn't cost them anything.