Charging Curve...hacking?

[Ravensfan1996]

That video was showing the new charge curve. When Mach E first came out, after 80% it slowed to a crawl of about 12kwh, but the new update allowed to go up to about 45kw after 80%. (This about a year an half ago i think)

Sponsored

 

[lightningandmache]

That video was showing the new charge curve. When Mach E first came out, after 80% it slowed to a crawl of about 12kwh, but the new update allowed to go up to about 45kw after 80%. (This about a year an half ago i think)

In the video the sessions are restarted and the power changes quite a bit each time well before 80% SOC.

 

[SWO]

Also, has anyone tried unplugging and replugging in the same session to see if you can get higher rates? Doesn't seem like it should be possible, but maybe?

Yes I have (unintentionally) and yes it does....kind of, with limitations.

On a trip I stopped at an EVGO charger which would stop charging every ~8 minutes. First 3 times I plugged in it would start at 158-163kw before tapering to 120-something. The last plug-in SOC was higher and it only peaked to low 140s?

I wouldn't say it was faster but I did leave 10min before the car said I would be done charging (admittedly the car is usually pessimistic). My EVGO account has P&C.

 

[lightningandmache]

Yes I have (unintentionally) and yes it does....kind of, with limitations.

On a trip I stopped at an EVGO charger which would stop charging every ~8 minutes. First 3 times I plugged in it would start at 158-163kw before tapering to 120-something. The last plug-in SOC was higher and it only peaked to low 140s?

I wouldn't say it was faster but I did leave 10min before the car said I would be done charging (admittedly the car is usually pessimistic). My EVGO account has P&C.

VERY interesting! Do you remember what rough SOC this worked up to?

 

[DustyShades]

I think it is partially time based. I remember reading about this when the car first came out. Also having just returned from a road trip, I started charging sessions from anywhere between 4% and 25% I would always get an initial burst around 120-130kW that would then slowly taper off. I also noticed that even though the beginning speed and the end speed approaching 80% was pretty consistent 50-60% range charge speeds varied a lot and seemed to correspond with the starting charge state.

Anyway, this is all anecdotal, I didn’t record any data and there could have been other factors that affected it. But I think charge time is at least one of several factors affecting the charge curve

 

[Ravensfan1996]

In the video the sessions are restarted and the power changes quite a bit each time well before 80% SOC.

Looks like it was charger glitch why they restarted, only once was it alot more when restarted, but only for a bit. As the battery warms you get better speeds, but unplugging and restarting definitely wont be faster IMO.

 

[lightningandmache]

Looks like it was charger glitch why they restarted, only once was it alot more when restarted, but only for a bit. As the battery warms you get better speeds, but unplugging and restarting definitely wont be faster IMO.

It does seem to keep power higher in the middle SOC range than if a single session. But...I am now noticing the battery temps may explain all of this. Because the battery was colder it may have reached these higher power levels on a single session too.

 

[AZBill]

I think that's not correct. While I'm not going to sit there testing it, it stays at 150kW for such a short time that you won't be gaining anything, considering the time it takes to end a session, then start a new session, and have the charging kick in.

In other words, let's say you gain 90 seconds of 150kW. You lose that in ending and restarting. And, when it drops, it drops to the right point on the curve, i.e. you're at 72% and that's where it's going to fall.

This car's been out for 2 years. If this could be gamed, it would have been done dozens of times.

What I have seen, is if I start a charge at a low SoC, like 20-30%, by the time I get to 80% it has ramped down to 65-70kw. But if I start a charge at higher SoC, it can be as high as 90kw at 80%, before the drop off. So the curve does shift, based on where the charge starts.

 

[Mach-Lee]

Yes, I've heard of people unplugging and repluging every 10 minutes to get a faster rate. I suppose you could call that a Mach-E charge curve hack. It's probably not the best for the battery, but the charging curve is very conservative right now so it can probably handle a little abuse.

The idea with the initial boost is the DC wires should be cold, so they can handle more current than their max continuous rating of about 100 kW for a few minutes until they heat up. Ideally the charge curve would then settle into the max continuous current of around 85-105 kW for the reminder of the curve up to about 70-80% before tapering exponentially above that point.

Repeated restarting and operating at 120+ kW for too long could overheat the DC cables, so be careful.

Tesla does amazing things and finds the limit the hardware is actually capable of. Our peak charging rate is limited to only 400 amps, but here's a Model Y pushing 683 amps (255 kW) into the pack for comparison:

Ford found the absolute limit of what the pack was capable of outputting with the GTPE, I wish they would apply that same derate logic to the DCFC curve to get the max the hardware is capable of doing.

 

[Jiji]

I won't comment on hacking and there isn't anything new here on charging but I am proud of my new DCFC support in OBD Record. Someone wanted to see what a DCFC charge curve looked like and while it can't yet be viewed in OBD Play you can peek at it in the database until I can push an update with the detailed charge views.

This was recorded at an Electrify America 150kW charger, with no preconditioning, looking closely you can see that the battery temperature increase was due to charging and that at about 35% SoC the MME started cooling the battery (purple SoC in %, blue HVB temp ℃, orange HVB coolant inlet temp ℃).

The charge power curves are here:

purple	EA advertised power (constant 180kW)
orange	kW requested by MME
purple	kW at output of MME charger module
blue	kW delivered to HVB

Starts off requesting 125kW but within 3 minutes it has dropped to below 100kW where it stays for the next 14 minutes. The big step down occurs at 80% SoC where the requested power drops from 60kW to 37kW.

39 minutes to get from 3% SoC to 80% SoC, 44 more minutes to go from 80% SoC to 100 SoC.

 

[Mach-Lee]

The charge power curves are here:

purple	EA advertised power (constant 180kW)
orange	kW requested by MME
purple	kW at output of MME charger module
blue	kW delivered to HVB

Starts off requesting 125kW but within 3 minutes it has dropped to below 100kW where it stays for the next 14 minutes. The big step down occurs at 80% SoC where the requested power drops from 60kW to 37kW.

Why is there a ~20 kW discrepancy between requested and delivered power that decreases with time? Are you using actual pack voltage or target pack voltage to calculate the kW? The traces should all be very close <6 kW difference or there's some kind of problem. You should graph the amperage curves and make sure they have the same pattern as the power curves.

 

[lightningandmache]

I won't comment on hacking and there isn't anything new here on charging but I am proud of my new DCFC support in OBD Record. Someone wanted to see what a DCFC charge curve looked like and while it can't yet be viewed in OBD Play you can peek at it in the database until I can push an update with the detailed charge views.

This was recorded at an Electrify America 150kW charger, with no preconditioning, looking closely you can see that the battery temperature increase was due to charging and that at about 35% SoC the MME started cooling the battery (purple SoC in %, blue HVB temp ℃, orange HVB coolant inlet temp ℃).

The charge power curves are here:

purple	EA advertised power (constant 180kW)
orange	kW requested by MME
purple	kW at output of MME charger module
blue	kW delivered to HVB

Starts off requesting 125kW but within 3 minutes it has dropped to below 100kW where it stays for the next 14 minutes. The big step down occurs at 80% SoC where the requested power drops from 60kW to 37kW.

39 minutes to get from 3% SoC to 80% SoC, 44 more minutes to go from 80% SoC to 100 SoC.

39 minutes? Based on this curve, this must be a SR pack not your GT?

 

[Jiji]

39 minutes? Based on this curve, this must be a SR pack not your GT?

Yes, vehicle is a Premium SR Mach-E. I should have included that in the write up.

 

[kltye]

So I'm the guy that recorded the video that OP found. And yes, unplugging and replugging does get you faster speeds initially by "restarting" the curve, but as others have said - it's not worth it. That speed tapers off really quickly, and the amount of time spent for the charger to handshake with the car takes so long that any speed gains from the "hack" makes that a losing proposition.

It's not just because of pack temperatures, etc. I've done this several times (I exclusively charge at public chargers because I have no at-home charging, so I have a good amount of experience with DCFC), and the story is always the same. As @Mach-Lee said, the car should be able to stay at around 100kW forever, but for whatever reason Ford has chosen this extremely shitty method of calculating the charging curve. There's no reason we can't charge at 1C deeper into the pack, but here we are, I guess.

 

[SWO]

VERY interesting! Do you remember what rough SOC this worked up to?

I don't exactly recall, but I plugged it pretty low....in the teens percentage wise.

Sponsored