Impact of SOC on performance / drivability

[anko]

Maby and Titan (and probably some others) have reported performance / drivability being impacted by low SOC. Both report excessive high revving by the engine. I have always claimed that there should be no impact even at a very low SOC (unless you totally trash the car), as the battery would still give you plenty of power. Well, both Maby and Titan don't seem to be trashing their cars, so I have always a hard time understanding why / accepting that there would be a real impact.

Without going back to what was exactly written, I must say I have always have made following interpretations when reading their stories:

- SOC made the difference between the engine running at reasonably low RPM versus running at excessive RPM.
- They were accelerating uphill, so power demand must have been pretty high (>60 kW).

Now I think, maybe the situation may have been slightly different from how I interpreted it:

- They were mildly accelerating uphill, so power demand was above average but not extremely high (maybe 40 to 60 kW or such).
- SOC made the difference between the engine not running at all versus running at excessive RPM.

Now, this could be explained. Once the engine is running* the car will try to fulfil the total power demand by the engine alone and potentially even recharge the battery a bit. Only when the engine cannot deliver what is needed for driving (because it is maxed out at 4100 RPM in serial mode or because it is maxed out at the RPM associated with current speed in parallel mode), the battery will assist.

So, when you would be able to end your hill climb with more then 26 ~ 30 % SOC and power demand stays below 60 kW, you could do it all pure EV. No excessive revving.
If you end up wit less then 26 ~ 30 % SOC, at some point the engine will kick in (start of the hysteresis cycle), and it will try to do all the work until the cycle ends, result in excessive revving for a longer period of time. This part would be dependent on SOC at the bottom of the hill.
If you end up using more than 60 kW of power, then engine will kick in and will try to do all the work, which again may result in excessive revving. This part would be independent of SOC at the bottom of the hill.

So, starting a moderate hill climb with enough SOC so that you can make top side with > 26 ~ 30% SOC left could indeed prevent excessive revving, enhance drivability and give a sensation of increased performance.

Maby, Titan, does this make sense?

* Because of low SOC, Save or Charge mode, that is. I am not sure about heating.

 

[Titan]

That's pretty much it for me. If I come across a hill while running in Save/Charge and I'm at low speed or starting from stationary, I will switch Save/Charge off and try and use just battery as, if the engine is already running, it tries to do it in Serial mode with maxed out revs. Once I can transition to parallel mode the issue naturally disappears.

 

[maby]

Yes, this makes a lot of sense. I have reported that the car seems to trip over into a different mode where engine revs go high and stay high much longer than seems necessary. I guess that your explanation makes engineering sense too - once you've started the engine and got the revs up, you may as well keep them there for a while.

I think the actual situation may be slightly more complex than you are suggesting - since I am driving on Save when doing any significant distance, the engine will have been running anyway. The car will be in its standard pattern of power control in either serial or parallel hybrid depending on speed and operating the relatively small hysteresis cycle, oscillating the SOC around the Save point. Power is coming from both the battery and the generator with engine revs kept relatively low. I generally drive at around 60mph on long journeys and the car is cycling actively between pure EV, serial hybrid and parallel hybrid.

I get the impression that when you demand a lot of power on a low SOC, the power control algorithm trips over into a different mode where it tries to source all the power demand from the engine - presumably to avoid the risk of turtling it. Having got into this state, it then seems to continue to run the engine at high revs longer than is really necessary, sourcing most of the power to move the vehicle direct from the generator.

Obviously, once you get up to parallel mode speeds things change - engine speed is dictated by road speed. With a low SOC, it does not have the option of dropping down into Serial mode for additional acceleration, so you can find yourself without that electric turbo kick if you floor it at 50 or 60mph. I don't particularly want to be driving in Save on a full battery all the time - but I don't want to be driving on empty for long distances either. My strategy is driven out of laziness.

 

[maby]

P.S. I think the same can happen accelerating hard on the flat too - even with a moderate SOC, it will kick the engine in when you floor it and it then can hold the revs up high for a lot longer than is necessary.

 

[Titan]

P.S. I think the same can happen accelerating hard on the flat too - even with a moderate SOC, it will kick the engine in when you floor it and it then can hold the revs up high for a lot longer than is necessary.

Yes, had this as well.

 

[jaapv]

On a sidenote, in an overtaking situation it is wise to activate "charge" beforehand to avoid the hesitation caused by the ICE having to start when sudden acceleration is called for.

 

[Grigou]

...

I think the actual situation may be slightly more complex than you are suggesting - since I am driving on Save when doing any significant distance, the engine will have been running anyway. The car will be in its standard pattern of power control in either serial or parallel hybrid depending on speed and operating the relatively small hysteresis cycle, oscillating the SOC around the Save point. Power is coming from both the battery and the generator with engine revs kept relatively low. I generally drive at around 60mph on long journeys and the car is cycling actively between pure EV, serial hybrid and parallel hybrid.

...

I agree with you.

And I'am afraid that I don't agree with anko writing "Once the engine is running* the car will try to fulfil the total power demand by the engine alone and potentially even recharge the battery a bit. Only when the engine cannot deliver what is needed for driving (because it is maxed out at 4100 RPM in serial mode or because it is maxed out at the RPM associated with current speed in parallel mode), the battery will assist."

In long uphill driving in Save (Savoy, 2 people on board, slope 5% to 8%) , the motors help the engine. And of course the SOC decreases constantly.

And I'am not quite sure, but the car seems more "comfortable" to drive with 60 % SOC than with 35 % SOC.

Good new : I've just received my OBD2 dongle

 

[maby]

...

In long uphill driving in Save (Savoy, 2 people on board, slope 5% to 8%) , the motors help the engine. And of course the SOC decreases constantly.

And I'am not quite sure, but the car seems more "comfortable" to drive with 60 % SOC than with 35 % SOC.

...

I'm glad it's not only me! Several here such as jaapv are insistent that the SOC has no impact on driveability, but it always seems a bit underpowered to me when running with a flat battery. I suspect it is a question of how hard you drive it - I like the kick up the backside that goes with serial hybrid on a full battery!

 

[spellinn]

I've always thought that performance does drop off with low SOC.

I never "floor it" - even with regular driving the engine revs seem abnormally high (and relate almost directly to the amount of throttle) and the car appears sluggish. My "power" dial is always in the green but the engine is revving like crazy with power demand at just "10o'clock" and one-bar shown on the battery.

I've no idea why this would be the case, as I understood from other postings that when the battery meter is "empty" there is still a significant amount of usable SOC left (20%?) to cope with power demand whilst the ICE kicks in to keep it charged at around that 20% "empty" level.

Certainly when I'm doing urban driving not motorway driving the effects are more noticeable as you hear the engine revving up and down as your throttle demand changes, and the car appears to require more throttle to give the same performance (although this could possible be subjective as you listen to engine roar!)

Regards

Neil

 

[greendwarf]

But at lower SoC the faster the batteries charge so presumably this will tend rev the ICE higher to meet this demand?

 

[Grigou]

...
I've no idea why this would be the case, as I understood from other postings that when the battery meter is "empty" there is still a significant amount of usable SOC left (20%?) to cope with power demand whilst the ICE kicks in to keep it charged at around that 20% "empty" level.
...

Neil

A plausible explanation would be that there are situations (moutain, towing) demanding battery's energy during a long time, so the system prevents from this full discharge a long time before reaching 25% SOC ?

It seems (just feeling, no measurements) that as much the SOC is low, as much the engine is trying to charge the battery, or simpy compensate the SOC.
Maybe it simply compensates at 30 % SOC, try to charge at < 30 %, and let discharge at > 30 ? If so, it let discharge the battery differently at 35 % than at 60 % IMHO ...

(all this in Save mode)

 

[maby]

...
I've no idea why this would be the case, as I understood from other postings that when the battery meter is "empty" there is still a significant amount of usable SOC left (20%?) to cope with power demand whilst the ICE kicks in to keep it charged at around that 20% "empty" level.
...

Neil

A plausible explanation would be that there are situations (moutain, towing) demanding battery's energy during a long time, so the system prevents from this full discharge a long time before reaching 25% SOC ?

It seems (just feeling, no measurements) that as much the SOC is low, as much the engine is trying to charge the battery, or simpy compensate the SOC.
Maybe it simply compensates at 30 % SOC, try to charge at < 30 %, and let discharge at > 30 ? If so, it let discharge the battery differently at 35 % than at 60 % IMHO ...

(all this in Save mode)

I don't think this 25 to 30% floor on the battery charge is there to somehow leave us with something in reserve - I think it is more a case of protecting the battery. Just think for a moment about your mobile phone or laptop - both of which will be using some kind of Lithium battery - in each case you will think you are doing well if it lasts three or four years without losing a significant amount of capacity. Car manufacturers putting Lithium batteries into their products will be hoping that they will last for at least ten years - that is the minimum life we expect out of a car before it needs an expensive repair. That's a very long time for a rechargeable battery - they address it by putting in a much larger battery than the car needs and only using part of its capacity - as it ages, it should be able to maintain the specified range for longer even if the battery has started to lose a significant fraction of its capacity.

If that is the case, then it would make sense that the power control algorithms should start to reduce the drain on the battery as it approaches the floor level. We already know that it will do this if you manage to drag the battery down close to the 20% level - dropping into turtle mode. Anko and others have pointed out that as the battery approaches fully charged, the charging current drops off - it seems to me that the corollary of this is that as the battery approaches empty, the available discharge current should start to drop off. Is Mitsubishi just protecting the battery by demanding less of it when it is getting low?

 

[jaapv]

The battery protection kicks in by turtle mode.

 

[Grigou]

I don't think this 25 to 30% floor on the battery charge is there to somehow leave us with something in reserve - I think it is more a case of protecting the battery.

I agree with you, and didn't spoke about "reserve".

The choice of ~25% and 30 % are for the preservation of the battery's life indoubtedly. But we can also consider that the ~15 % to 30 % range is a sort of "reserve", which the system is always trying to fill ...

 

[anko]

If I come across a hill while running in Save/Charge and I'm at low speed or starting from stationary, I will switch Save/Charge off and try and use just battery as, if the engine is already running, it tries to do it in Serial mode with maxed out revs. Once I can transition to parallel mode the issue naturally disappears.

Haha, indeed this is what I forgot to mention: I thought engaging Charge mode during the climb or being in the "engine on" phase of the Save mode cycle would have the same effect as having a low SOC.

 

[anko]

And I'am afraid that I don't agree with anko writing "Once the engine is running* the car will try to fulfil the total power demand by the engine alone and potentially even recharge the battery a bit. Only when the engine cannot deliver what is needed for driving (because it is maxed out at 4100 RPM in serial mode or because it is maxed out at the RPM associated with current speed in parallel mode), the battery will assist."

In long uphill driving in Save (Savoy, 2 people on board, slope 5% to 8%) , the motors help the engine. And of course the SOC decreases constantly.

I don't see how what you are saying contradicts with what I am saying. I am pretty sure that on this slope the engine is maxed out given the limited RPM associated with your current speed. I mean, you can see see this already happen on an onramp of a bridge. You do not mention what speed this is, but don't forget, at 100 km/h you have no more than roughly 42 - 45 kW available. At lower speeds it is even less. Check your power meter. Is it 2/3 to 3/4 in the green? Then the E-motors will have to support as your engine is maxed out. Torque Pro will show you a relative load of 100%. Wanna bet?

 

[anko]

Yes, this makes a lot of sense. I have reported that the car seems to trip over into a different mode where engine revs go high and stay high much longer than seems necessary. I guess that your explanation makes engineering sense too - once you've started the engine and got the revs up, you may as well keep them there for a while.

My line of thinking was that the engine did not switch from normal to high revving, but from totally off to high revving. I think this happens:

At some point, you reach the SOC low water mark and the hysteresis cycle kicks in. The engine starts running and keeps running until SOC is up by 1.5%. But in order to propel the car uphill AND recharge the battery at the same time, the engine needs to produce a fair share of power. And for this it needs the revs. And it does need them for a longer period of time to complete the cycle.

I think the actual situation may be slightly more complex than you are suggesting - since I am driving on Save when doing any significant distance, the engine will have been running anyway.

Save mode does not mean that the engine is running perse. Well maybe if you hit Save at 90+% SOC, but not normally.

But apart from that, from what I hear you say in other posts, you are not normally driving in Save mode at sub-parallel speeds. So, the above must apply to parallel speeds. And at parallel speeds, high revving will normally only occur when power from the engine in parallel mode + power from the battery together are not enough. Not even 100% SOC can stop that kind of high revving.

IMHO, the only other reason for high revving at parallel speed is that you are looking at a big turtle symbol in your dashboard and actually experiencing the limited driving mode. Which I do not believe is the case.

Power is coming from both the battery and the generator with engine revs kept relatively low.

This may very well happen in parallel mode. But I don't think this will happen in serial mode, not even on the flat. As long as the engine has power reserves, the car will not take power from the battery for driving the car (well, maybe for a very short moment to bridge the power gap of the engine.). Like I said, I believe that once the engine is running, the car will let the engine do all the work. Until the engine is maxed out. Only then, the car will call upon the battery for assistance. If you think differently, please explain how you have come to this conclusion. As the graphics on the dashboard or on MMCS have no way of representing such situation.

(To split one or two hairs: The above is not entirely true, because even in (according to the instruments) 100% pure parallel drive some power is taken from the battery to eliminate E-drag in the front motor and generator and generate minimal torque in the rear engine. This is something I can actually witness on my TP dashboard. But this is very minimal compared to total power demand).

I generally drive at around 60mph on long journeys and the car is cycling actively between pure EV, serial hybrid and parallel hybrid.

Serial mode? When 'driving around' at 60 MPH? I guess you do not mean these short moments of serial drive in between EV drive and parallel drive? The only reason I can see for serial drive at 60 MPH is when you are looking at a turtle or are requesting > 90 kW of power (1/2 into the white area of the scale).

I get the impression that when you demand a lot of power on a low SOC, the power control algorithm trips over into a different mode where it tries to source all the power demand from the engine - presumably to avoid the risk of turtling it. Having got into this state, it then seems to continue to run the engine at high revs longer than is really necessary, sourcing most of the power to move the vehicle direct from the generator.

I believe it always tries to source all power from the engine when the engine runs. Regardless of SOC and speed. With the restriction that it will not drop out of parallel mode unless it has to. This is what I meant when I wrote: "Only when the engine cannot deliver what is needed for driving (because it is maxed out at 4100 RPM in serial mode or because it is maxed out at the RPM associated with current speed in parallel mode), the battery will assist.".

At 35 MPH the engine will be maxed out at 100% load @ 4100 RPM, which means 60 kW output from the generator. So, only when 60 kW is not enough, the battery will assist. In reality, most of there time, the engine will be able to produce surplus power and recharge the battery in serial mode. But at 40 MPH, the engine will max out at 100% load at 1700 RPM, resulting in as little as 30 kW mechanical at 40 MPH. Only in very optimal driving conditions, there will be some spare power to charge the battery.

Obviously, once you get up to parallel mode speeds things change - engine speed is dictated by road speed. With a low SOC, it does not have the option of dropping down into Serial mode for additional acceleration, so you can find yourself without that electric turbo kick if you floor it at 50 or 60mph.

Why would the engine not be able to drop down into Serial mode, even if the battery was totally rock bottom empty? Upton speeds of 120 km/h, the engine can produce more power in serial mode than it can in parallel mode. But apart from that, the pictures I shared the other day show a battery boost of > 60 kW @ 23.5% SOC. Believe me, if it hadn't been for the caravan on my tail, that would have felt like a turbo kick.

 

[anko]

I'm glad it's not only me! Several here such as jaapv are insistent that the SOC has no impact on driveability, but it always seems a bit underpowered to me when running with a flat battery. I suspect it is a question of how hard you drive it - I like the kick up the backside that goes with serial hybrid on a full battery!

And me too. To bad Jaap hasn't commented on my opening post (yet). I was pretty much in his camp (or he in mine) until this morning. We both believed the drivability as well as performance of the car was not impacted by SOC (until you hit 20% SOC).

Now I still believe performance is not impacted (until you hit same 20% SOC) but I think that (perceived) drivability at low speeds is impacted by having the engine run, regardless of whether this is a result of Charge or Save mode or because of low SOC.

Typical scenario to explain:

While towing the caravan, I firmly pull away from the first roundabout in EV mode. This is quiet and seems totally effortless. Big smiles on all faces in the car. At the next roundabout, I pull out in Charge mode. This results in high revving of the engine. And a nasty smell as a bonus. Even with significant SOC. No more smiles on the faces in the car. Except for mine. So, from a comfort perspective, it would be best to disengage Charge mode. But from a "I am towing and therefor I want to have as much SOC as possible" perspective, it is much better to keep Charge mode engaged. As most likely, the battery is being recharged, even while pulling out of the roundabout. This is why there was still a smile on my face. As I was the driver

And yes, because of the high revving in Charge mode and the hard work the engine is delivering, the car may seem underpowered. But don't forget, the same battery power that was available when pulling away from the first roundabout is still there as a reserve just in case the engine couldn't cope.

 

[anko]

I've no idea why this would be the case, as I understood from other postings that when the battery meter is "empty" there is still a significant amount of usable SOC left (20%?) to cope with power demand whilst the ICE kicks in to keep it charged at around that 20% "empty" level.

At urban speeds, the low water mark is at about 26%. At that point the engine will kick in, in an attempt to quickly restore SOC to 27.5%. In order to do so, the engine must produce all the power needed for driving plus some extra power needed for upping the SOC.

Guestimate: With your power meter in the 10 o'clock position you use 20 kW for driving. Then I would expect the engine to produce no more than 40 kW, as the charge current is limited to protect the battery from blowing up. Also, I believe the engine tries to stay in its sweet spot of 75% load as long as possible, for optimal efficiency. It would require (ballpark figure) roughly 3600 RPM @ 75% load to produce the needed power. Maybe a little bit more if you also run A/C or an electric heater.

 

[anko]

But at lower SoC the faster the batteries charge so presumably this will tend rev the ICE higher to meet this demand?

Interesting thought... In the extreme, at 100% SOC, there is no charging so the engine only needs to produce what is needed for driving. At the cost of 0% EV driving, of course.

BTEW: I don't know if it is truly a "demand" or more a "willingness to accept". I have a feeling the engine will not go above 75% load to meet than demand. I can imagine it will only do this when driving conditions dictate. So, higher revving, perhaps yes. Higher load, perhaps no.