Hyundai IONIQ Forum banner

181 - 200 of 228 Posts

·
Registered
Hyundai Ioniq premium -19, 28kWh ~46k km
Joined
·
53 Posts
@giora I think we can safely assume the buffers are not used until proven otherwise, it makes no sense at all having extra capacity just to mask degradation(Hyundai would have been better off selling a 30kWh model with 300km NDEC range). Using the buffers would only expedite the time until the battery reaches end of life.

Degradation is happening in our batteries and is measurable either through how much charge you put in or how much you take out (the latter is harder to do in a controlled and comparable fashion) compared with how much you used to be able to put in or take out.
We want to measure the current usable capacity and compare to an old reading, preferably one from right after delivery, to determine the loss of useable capacity. :)
 

·
Registered
Joined
·
28 Posts
@TNSe yeah, then it's just down to low temperatures. Low temps are as bad as high temps, but usually unavoidable where we live. Here is mine at -10°c. Cell voltage is 4.06v at 100% (and has recovered to about 4.12v at 100% at current temps).
View attachment 33820
This is very interesting. You also have SoC BMS much lower than the 95% everyone assumes the car has at 100% SoC. Your SoC BMS is at 92.5%, which means the BMS does something with SoC scaling on low temperatures. But does it remove energy, or does it eat into buffers? It's so hard to test because the battery heats up during testing. :p
 

·
Registered
Joined
·
162 Posts
I think we can safely assume the buffers are not used until proven otherwise...
How do you explain then SOH 100% reported all over even for cars over 100000 km or 4 years old, and no report yet of lower SOH?
Agree all batteries age and degrade, the 100% SOH reported can only be explained by flexible buffers or do you have an alternative explanation?
 

·
Registered
Joined
·
28 Posts
How do you explain then SOH 100% reported all over even for cars over 100000 km or 4 years old, and no report yet of lower SOH?
Agree all batteries age and degrade, the 100% SOH reported can only be explained by flexible buffers or do you have an alternative explanation?
Well, this post on this forum claims he is seeing 96% SoH. I'd like to see some more information but he never got back to us. Battery degradation
 

·
Registered
2018 Hyundai Ioniq Electric Premium SE Black
Joined
·
171 Posts
This is very interesting. You also have SoC BMS much lower than the 95% everyone assumes the car has at 100% SoC. Your SoC BMS is at 92.5%, which means the BMS does something with SoC scaling on low temperatures. But does it remove energy, or does it eat into buffers? It's so hard to test because the battery heats up during testing. :p
I believe it's to protect the battery at low temperatures because of higher internal resistance, it can degrade the battery more at high SoC's.
 

·
Registered
Joined
·
237 Posts
You should remember that the classic Ioniqs battery is air cooled. When using the whole capacity of the cells the top buffer is there to keep it from being damaged when hot. Liquid cooled packs are kept cooler when compared to air cooled ones so there is a very good reason to limit the maximum voltage of a air cooled pack.
 

·
Registered
Hyundai Ioniq premium -19, 28kWh ~46k km
Joined
·
53 Posts
@TNSe I agree with @MDave the BMS limits the amount of charge allowed at low temps, and displays it as 100%, no more charge is allowed to occur, its easier for the end user to see that the car won't charge more and this is the range available today. If it displayed 97.5% a lot of questions would be asked and the service centers would be filled. (now with ambient temps above freezing again its back to the standard 95%BMS=100%Displayed)
It does this by limiting max voltage as @PetteriSeppänen says the high voltages are bad for the battery health, both at high and low temps. Essentially it is increasing the top buffer size to ensure battery longevity, this is very good.

@giora I have also wondered about this SOH a lot, but I still don't have a clue what it measures, I've read reports of it going both down and recovering back to 100%. Maybe it's just measuring the cell balance or maybe 100% means that all cells are equally degraded. This is a mystery.
What I do know is that to access the buffers, the BMS would have to raise the max cell voltage over 4.14v or min voltage below 3v, something I've never heard of on the Ioniq.
There is a mod for the Nissan Leaf that does just this, increasing max cell voltage allowed to give a higher capacity(to be used only when really needed ofc), essentially overcharging the battery.
Most li-ions go to atleast 4.2v or even 4.4v in some cases, but this shortens their lifetime, which might be okay in competitive RC-racing but not in a production car expected to roll atleast 200-300 K km on the same pack.
 

·
Registered
Joined
·
28 Posts
@giora I have also wondered about this SOH a lot, but I still don't have a clue what it measures, I've read reports of it going both down and recovering back to 100%. Maybe it's just measuring the cell balance or maybe 100% means that all cells are equally degraded. This is a mystery.
What I do know is that to access the buffers, the BMS would have to raise the max cell voltage over 4.14v or min voltage below 3v, something I've never heard of on the Ioniq.
So this is the mystery, because I've seen a youtube video of a hypermiler pulling about 26kWh out of a 10k km Ioniq, and TeslaBjørn pulling about 26Wh out of a 100k km Ioniq. (He also pulled about 26kWh out of an ioniq at earlier point).

So how is Ioniq doing the buffers? Is it tricking by showing a calculated kWh/100km drained instead of actual drained?

Edit: Added relevant youtube videos at relevant time stamps.
365.3km @ 7.2kWh/100km = 26.3kWh (10 546km)
142.6km @ 18.5kWh/100km = 27.8kWh (5 807km)
227km @ 10.9kWh/100km = 25.8kWh (92 095km)

First and third car seem to be 2017 models. The second car is a very early model from 2016, probably a press car. Did Hyundai deliver some Ioniq28 with more buffers available (that were actual 28kWh available) and later they limited it to ~26kWh?
 

·
Registered
Hyundai Ioniq premium -19, 28kWh ~46k km
Joined
·
53 Posts
The buffers are formed by limiting min and max voltage from the values recommended by the cell manufacturers(lg) min/max limits(bms 0-100%), nothing special there.

It could be that the later models measure power draw at a later stage and don't fully account for some losses? I feel the hypermiling attempt should have gotten close to max extraction with low C-rates and warm temperatures. Regen is also accounted for in the displayed numbers somehow (and probably a significant amount in the low speed hypermiling test compared to high speed winter tests).

I prefer measuring charge added (and use this method track my monthly efficency) to charge used(displayed by the car). This is what I pay for in the end, just like I measured diesel pumped into my old car, not the reported amount used. :)
Gom is good enough to know how far I can go and ABRPtransmitter(or abrp no-map mode) helps with pacing when I'm unsure, the cars reported consumption is practically irrelevant, even if fun to compare.

I have never tried calculating energy used by average consumption and comparing it to what I'm actually able to put back into the pack. I'll try to remember to do that next time I'm on low soc, it's a good test.
 

·
Registered
Joined
·
28 Posts
It could be that the later models measure power draw at a later stage and don't fully account for some losses? I feel the hypermiling attempt should have gotten close to max extraction with low C-rates and warm temperatures. Regen is also accounted for in the displayed numbers somehow (and probably a significant amount in the low speed hypermiling test compared to high speed winter tests).
Yeah, because my car follows the hypermiling car closer than the winter-test car. (I think I can dig up some 70kW+ charger tests of that winter-test car. The charging curve was better than mine (but similar shape).

I prefer measuring charge added (and use this method track my monthly efficency) to charge used(displayed by the car). This is what I pay for in the end, just like I measured diesel pumped into my old car, not the reported amount used. :)
Gom is good enough to know how far I can go and ABRPtransmitter(or abrp no-map mode) helps with pacing when I'm unsure, the cars reported consumption is practically irrelevant, even if fun to compare.
I agree that what you put in is important when it comes to what you pay. But what is in store and how you pull it out is more relevant for how far you go, which is the critical moment for me.
 

·
Registered
Hyundai Ioniq premium -19, 28kWh ~46k km
Joined
·
53 Posts
@TNSe yeah, agreed.
But it is really hard to do a controlled drain test without lab equipment(I guess laps on an oval/track/motorway with cruise control would do) so for calculating degradation it's easier to use the ammout charged, we can fit this into our daily schedules on a day with similar weather as last time.
Degradation should affect drain as much as charge, if you can put in 10% less you should also be able to drain 10% less, or 10% less range in effect.

But if we don't know the original chargeable/drainable capacity when the battery was new, it's going to be hard to know how much we have lost.
I did my first readings at 35000km(Sept -20), so I'll only be able to compare to that. Degradation should be worst in the first few years and 250-500 cycles. My car was built in September -18(I don't know the date on the battery, but it's going to predate assembly) and had probably already lost some capacity at delivery in May -19 from calendar aging.

It could be that all the battery cells were manufactured by LG in 2016 as part of a large order, so the later models had pre-degraded batteries only giving 26kWh whilst the press car, presumably one of the first of the assembly line had fresh cells at the time capable of giving the max 28kWh.
What is weird is that 26*108% is ~28, that is if you remove the buffers(3%+5%) from 28, you get approx 26, probably only a fluke though.

Unfortunately we will have to wait and see when we start getting reports of capacity/range loss, but the battery seems resilient so maybe the largest loss is due to calendar aging(slow) and not cycling/high power on these cells.
 

·
Registered
Joined
·
28 Posts
Sadly, I'm still unable to trace down the origin of these pictures, but apparently it is a spanish Ioniq. Over 200k km before degradation kicks in? Pretty okay (1400 cycles)
33919
33920
 

·
Moderator
IONIQ Electric, SE (w/ CCP) 2018.
Joined
·
893 Posts
Good to finally see an EVNotify screen shot with something other than SoH = 100%. Makes me think that perhaps it is reading a PID or something from the vehicle to get the value. Now to determine if the pics are real or fake.
 
  • Like
Reactions: KevinT

·
Registered
Hyundai Ioniq premium -19, 28kWh ~46k km
Joined
·
53 Posts
I was reading an article on bms systems last week and it stated that most automotive bms have two ways of measuring soh.
By estimating available capacity (best for EVs and phev) or by estimating max available power output (best for hybrids, Small battery that needs to accept and deliver high C-rates during regen/boost).

That got me thinking that maybe, since Ioniq is the first of Hyundais EVs and its based on a mixed platform, the BMS might be stating SOH as a function of max power output. A 90% SOH would then be able to provide 90% of max power requirement (88kW engine + 7kW heatpump/ptc), not 90% of original capacity as one would assume (capacity might be lower or higher at that stage). As the battery is large(compared to a hybrid) it might take awhile before the available power drops below the requirement, thus it would stay at high soh for a long time as even with degradation it is able to provide the car with its full power requirement or 100% soh.

Just a theory though, but it does fit the avaliable data.
 

·
Registered
Joined
·
28 Posts
Just a theory though, but it does fit the avaliable data.
IF that is the case, it should be rather easy to test. Just check SoH on very low charge on very low temperatures. The battery can't deliver more than about 50% at 0-10 at 2-5% (adjusting for hidden bottom buffer?), and from there on, it just goes into the cellar. Alternatively, if you wanna destroy your battery for science, heat the battery to 55C.

However, if the spanish ioniq is actually correct, the battery pack temperature is pretty much perfect and should not have any reduction in output due to temperature itself.
33921


Edit: I have been able to pull nearly 94kW out of the battery during tests, without climate pulling a lot (anything?). (Controlled sustained acceleration). Controlled regen during same test gave me 82kW back.

Edit2: We now seem to have 3 competing theories:
1: The SoH is showing remaining capacity, but does not decrease until the hidden buffers are used up.
2: The SoH is showing a ratio of possible battery pack power output (as a base of 94kW?)
3: The SoH is showing the balance of the cells, and the less balance the cells have, the lower SoH becomes. (This was a theory on Facebook in a closed norwegian Ioniq forum by a Kjetil Tønnesen)
33922

Translation: My SoH has been 97% once, a full charge to 100% made SoH show 100% again. The battery had not been at 100% SoC for months before this. I have long suspected that SoH shows how well the battery pack is balanced, and not the health of the battery pack.

Personally I believe that the SoH measurement is same as the Kia Soul 30kWh. But that car only seems to have a small 3% top buffer. Maybe also similar BMS SoH as the 27kWh, and those batteries seem to be getting destroyed these days.
 

·
Registered
Joined
·
124 Posts
Repeating a comment I made 3 pages back that explains why it is possible to see a fluctuation in the SOH value.
'After monitoring BMS data from Kia Soul EVs for many years, to lose 4% suddenly is not a big deal. The BMS adjusts its estimate occasionally both up and down.
These swings can easily be 5% or more. The actual trend is deduced by plotting a chart of multiple readings over several seasons. '

I too believe that the SoH measurement for the Hyundai Ioniq Electric 28kWh is same as for the Kia Soul 30kWh. I have no data to show what the buffer size is on either of these cars. I have never seen a value other than 100% for the SOH of a Kia Soul 30kWh. It is very unlikely indeed that the Kia Soul 30kWh has a small 3% top buffer. If it was that small some of the 3 year old cars would have lost it already. The only car for which we have plenty of data is the Kia Soul 27kWh. The BMS for these cars shows a value for Minimum and Maximum cell deterioration. This capacity is measured from the top of the 'early years' buffer. SOH will be shown by apps as 110% - the value lost. There is also a 'safety' buffer which is never lost.
The Hyundai Ioniq Electric 28kWh and the Kia Soul 30kWh will only show SOH = 100% to be the full usable capacity.
 

·
Registered
Joined
·
28 Posts
I too believe that the SoH measurement for the Hyundai Ioniq Electric 28kWh is same as for the Kia Soul 30kWh. I have no data to show what the buffer size is on either of these cars. I have never seen a value other than 100% for the SOH of a Kia Soul 30kWh. It is very unlikely indeed that the Kia Soul 30kWh has a small 3% top buffer. If it was that small some of the 3 year old cars would have lost it already.
Reason I said 3%, is that I'm assuming the battery pack of the Soul 30kWh is the same as the Ioniq28kWh, which is close to 31kWh.

Assuming that it only has a small 3% top buffer, and we haven't seen any loss of SoH, it may actually be an indication that the SoH is not based on remaining capacity, but something else?
 

·
Registered
Joined
·
254 Posts
I wrote earlier about my suddenly decreased SoH after 140.000 km. At 100% SoC, all cells had equal voltage values. At 8% SoH there is a slight difference in voltage of the individual cells ranging from 3.36 to 3.40. I don't know whether this is a lot or that this may occur also in cars with 100% SoH.
What may also be a sign is that I noticed lately that, when fast charging (CCS), the charging power drops at around 75% SoC, instead of formerly around 80%.
Hope this supports one of your theories.
 

·
Registered
Joined
·
28 Posts
I wrote earlier about my suddenly decreased SoH after 140.000 km. At 100% SoC, all cells had equal voltage values. At 8% SoH there is a slight difference in voltage of the individual cells ranging from 3.36 to 3.40. I don't know whether this is a lot or that this may occur also in cars with 100% SoH.
What may also be a sign is that I noticed lately that, when fast charging (CCS), the charging power drops at around 75% SoC, instead of formerly around 80%.
Hope this supports one of your theories.
Excellent! Thanks for coming back with this update. Also keep on updating if the car drops further in SoH.

My car starts dropping on 50kW chargers around 78% and always has. The 150kW charges I see a drop around 75%.

If at all possible, if you could, using CanIoniq105, make a charging curve on 50kW/150kW chargers?

My car looks like this (with warm battery, cells 15C+, 150kW charger) 13 minute mark is ~75% 9 minute mark is 85% I believe? 3 minute 94%? Don't remember exactly where they are.
33940

33941


Edit; I believe the values for max regen and max power might have been swapped. Also, you may want to get your 12v battery checked. It is very low.
 
181 - 200 of 228 Posts
Top