Full state of charge vs 80%

I didn't say I was an expert on this or any subject.

Briefly:

The person who started this thread has made the claim that "coulomb counting" is the method—indeed, the "gold standard"—used to calculate state of charge on the Prius. As I have said, this is incorrect.

Coulomb counting is, not surprising, the counting of coulombs: the amount of free electrons that pass a given point in a given amount of time. The coulomb is the SI unit for electric charge, and one coulomb equals the amount of electrons that pass a given point in one second when current equals one ampere.

So when we talk about coulomb counting, we are talking about current over time. We can use this method to observe the amount of charge that enters or exits a battery cell. However, we cannot use it to judge how much charge was already in that cell, and this is one of the reasons why coulomb counting alone cannot be used to calculate a cell's state of charge. For a rough idea of SOC, we will also need to know cell voltage.

With some battery chemistries, changes in voltage roughly follow changes in state of charge (lead-acid comes to mind). For those chemistries, voltage estimation may be sufficient. Unfortunately, this doesn't work so well with nickel-metal hydride cells. So while we can start with a rough SOC estimation from voltage alone (as, for example, at start-up), we will need to then combine voltage and coulomb counting to get a more accurate idea of state of charge, especially as the battery starts to charge and discharge.

However, voltage and coulomb counting together are still not enough for us to calculate SOC as accurately as we would like. Why? Because there is no free lunch: any time you push charge into or out of a cell, you lose "charge". There are losses, and those losses must be counted. Monitoring cell temperature helps, because chemical reactions have different efficiencies at different temperatures.

So now we have voltage, current, temperature, and time. Each measurement is partly a reflection of the last measurement. Over time, inaccuracies accrue, and the calculation is susceptible to "SOC drift". With a plug-in vehicle, this can be corrected to some extent whenever the vehicle's battery pack is charged to 100% capacity. However, the Prius is not a plug-in vehicle, and its battery thus does not benefit from a charge to 100%. State of charge can also change slightly between the time the vehicle is shut off for the night and the time it is started up the next day.

To allow for such issues, algorithms are written in to the vehicle's battery management system (BMS). Such algorithms are proprietary intellectual property, and for obvious reasons, are not discussed in detail, but those who are curious can search the literature and learn some basics about them.

Finally, the BMS must allow for the fact that it is managing a multi-celled pack. For single-cell systems such as cell phones, the BMS can be relatively simple. For systems with more than a hundred cells, decisions must be made as to how much detail the BMS will need. The Prius pack could have been designed, for example, with voltage and temperature monitoring for each cell. Cost, however, is a factor, and we have ended up with a single voltage monitor for every twelve cells (two modules, six cells to a module) and a handful of temperature sensors.

There is much more to it, of course, but this is a start.

Actually, I did more than that. I gave the person who started this thread several hints as to how he could find out more on the subject and correct his misconceptions with very little effort. However, he showed no interest in either explaining his various theories or finding out more about the subject. Hence, I saw no reason to push the issue. Not everyone is open to learning.

Again, there was no reason to do so until you asked. The person who started this thread doesn't take kindly to folks who point out the many errors that he posts on this forum. To the contrary, he usually responds with some sort of deprecating remark about their technical knowledge. Unfortunately, he usually declines to provide explanations for his theories. Nor does he own up to his own errors, even when faced with multiple forum members patiently pointing out that his assumptions are wrong.

I'm sure he means well, and it appears that he is eager to help folks, but I get the sense that he has, to put it kindly, a rather high opinion of his own technical knowledge, and will not tolerate any posts that reveal any flaw in his logic, regardless of their accuracy, relevance or usefulness. Given, however, that such conduct actively stifles the free exchange of useful information, it's hard to feel sorry for the guy.

 

You showed that picture before. Lots of folks have swapped a battery out of a Gen 1.

Can you give an example? Please be as specific as possible, with links to the original posts.

I have never said whether or not I have hands-on experience. You're just fishing.

Do you still think coulomb counting alone can be used to calculate state of charge?

 

jk450,

Thanks for your explanation of SOC measurements. It always helps those of us that are not as knowledgeable as many others on this site.

As for Bob Wilson, I have only been on this site for about 7 months but I have seen a lot of Bob's work on here that has benefitted the rest of us greatly. It is much appreciated by the PC community.

In my opinion, you would better serve the PC members by sharing your knowledge with more data and facts and less pointing out the errors of others posts. Just saying....

Thanks again for the excellent post on SOC.

Dwight

 

My SOC meter never gets to the top. Im a total rookie prius owner, with an 08. Its mostly blue and about 3 bars down or at the halfway level. After coming off the highway or sown a hill it recharges to green, and only go as high as 7 bars. but NEVER to the top.

Is there something wrong or is this correct? I have a difficult time understanding all the technical talk in this thread. I just want to know if everyone else meter goes all the way to the top?

 

Then your Prius is doing its job. It tries very hard to maintain your SOC in the blue range, and generally in the middle blue range.

 

This is normal. As your battery ages and loses capacity you'll find that the SOC gauge will show wider swings.

For example, my 2004 with 100K miles will routinely move from 8 green bars to 2 red bars in the course of driving up and down hills.

 

As pointed out above, this is normal. Your Prius uses the battery to store excess energy, and to supply energy when needed. To do this it must have some room at the top for storing energy, and some charge to supply energy. This means it needs to keep the charge somewhere near the middle, which is why you see it blue most of the time.

Tom

 

Came for the complicated discussion that makes no sense to me.

Left satisfied.

 

Thank you for the kind words. I'm glad to have helped.

I'm sure he has helped folks in the past, but I have also seen folks being led down costly paths by advice he has given which was terribly wrong. I really don't care who the source of misinformation is, only that it is pointed out or corrected. It is not my fault that most of that misinformation comes from one person.

I can understand such a sentiment arising from a casual read of this thread. But there is a history here.

Last month, I noticed a thread in which a person with a sick car was being pointed toward some rather expensive parts even though available information did not indicate an issue with those parts. Sensing delicate egos among a few self-styled forum gurus, I asked a few preliminary questions about the logic behind their diagnosis.

Despite being unable to come up with reasonable explanations for their recommendations, they showed no interest in changing course or examining their own thought processes. Instead, they went on the attack. In particular, Wilson took it upon himself to type out deprecating remarks which you can read for yourself in the archives. Yet he could not fix the car.

I gave the person enough information to resolve the issue—which they did—and moved on. So yes, I am happy to provide data and facts when someone wants it. But correcting all of the errors would be a full-time job. Life is short.

That same pattern of discouraging accurate information can be seen in this thread: when Wilson's odd state of charge theories were politely questioned, he labeled the correction a "sad and pathetic answer". But he didn't offer evidence to show that his theories were correct. Later, he referred to the correction as "inadequate", "misleading", and an "obviously weak claim". Again, he didn't offer evidence to show that his theories were correct. He then tried to change the subject of the thread (which he had started), and when that failed, termed the correction "willfully ignorant".

But he's still wrong.

You are more than welcome.

 

In the electronics world there is still a lot of "Black Magic" going on especially on EMI problems. Even if you have done everything on the book and with prior experiences, the solution to EMI problem always requires ingenuity and creativity. This is why NTHSA called on NASA to help on the Toyota electronic EMI problem if there is any.

 

I have seen no efforts to intimidate you or anyone else.

I did, however, notice that you started a thread in which you termed a forum contributor a "sociopath" merely because they had taken the time to correct an error in one of your posts. In that thread, you also encouraged forum members to place that person on their "ignore" list. While I don't consider that intimidation, it's not exactly something to be proud of.

Your posts often encourage folks to send you private messages.

It is speculation to state that. See above.

By chance, this is a short list of some of the technical subjects that you have struggled with recently.

If you believe that, why not simply refrain from encouraging folks to PM you?

Back on topic: you have said several times that state of charge is determined by coulomb counting. However, you haven't yet explained why you believe this.

Why not?

 

The primary problem is accountability.

Folks often search the forum archives looking for answers to their problems. They deserve our best efforts to provide accurate information. They deserve contributors who will take responsibility for their own contributions, and who post from a position of service rather than ego.

Any effort to inhibit that serves no one.

Do you have any comments on state of charge? It is, after all, the subject of the thread.

 

I don't charge for my advice on this forum. That's all I will state relating to this topic.

Tom

 

It *was* stated somewhere up the line there that no, coulomb
counting by itself cannot determine overall state of charge over
the long term. Why is this such a salient point to you, jk?
Back off with your micro-nagging, you're not helping.
.
_H*

 

I tried coulomb counting for SOC, but every time I went out to my Prius, I found a little puddle of coulombs under the car. At first I thought it was condensation from the A/C, then I worried that I had a coolant leak. Eventually I realized it was only coulombs. I was relieved that I didn't have a problem, but it did make me realize that coulomb counting would never work for SOC, unless I put a pan under the car and count the ones that got away.



Tom

 

I didn't say it was a salient point to me. I would have been fine if the conversation had ended after the point was made. I simply thought the conversation had been dragged off-topic long enough.