Charging LTO at Max Speed?

[Addy]

I get that you don't want to go "inside the black box", but you're trying to explain things that you're not understanding.

With a CC-CV charger, it is in CC mode as long as the current is higher than the CC limit. As soon as the current is below the CC limit, the charger is now in CV mode. There is no in-between state.

 

[Username1]

I actually am trying to understand what happens inside the black box. So from that perspective...

It seems like with cc/cv charging, you start by setting max Amp and Volt limits. The charger will then monitor it's Amp output and adjust (raise) output Voltage as needed to maintain that max Amp limit. CC stage.

Eventually you end up hitting max outut voltage. Since you can't raise the output voltage anymore, the difference in voltage between the charger and battery begins to shrink, causing the amps to also shrink. CV stage.


Do i have this right? If i'm still getting this wrong don't feel obligated to keep explaining it. The main purpose anyway of this thread was about what it would take to fast charge at 6c since i'm interested in doing so. But i certainly don't mind people trying to help me understand how the charging works.

 

[Sunder]

The answer to that, I don't know for sure, as I'm not an electrical engineer. Just have a high school level of physics.

However, I believe that the easiest way would be to have a variable resistor in series with the load, and a shunt or other current measuring device would be the input to some kind of microprocessor to adjust the variable resistor to get the correct voltage and thus the correct current. You'd have to start from zero and move up instead of start at max and move down however, because if you had a short across the terminals, that could quickly blow the charger.

The above would not be the most efficient, as the variable resistor would consume a fair bit of the power. So... Possibly the same effect with a switch mode instead?

 

[Username1]

Oh i didn't mean like what's going on in detail with what parts it's using and all that. I just wanted to understand how a cc/cv charger works and how it interacted with the battery (basic principles). Seems like i'v got the gist of it finally.

 

[Addy]

It seems like with cc/cv charging, you start by setting max Amp and Volt limits. The charger will then monitor it's Amp output and adjust (raise) output Voltage as needed to maintain that max Amp limit. CC stage.

Eventually you end up hitting max outut voltage. Since you can't raise the output voltage anymore, the difference in voltage between the charger and battery begins to shrink, causing the amps to also shrink. CV stage.


Do i have this right?

Yes, that is correct.

However, I believe that the easiest way would be to have a variable resistor in series with the load, and a shunt or other current measuring device would be the input to some kind of microprocessor to adjust the variable resistor to get the correct voltage and thus the correct current. You'd have to start from zero and move up instead of start at max and move down however, because if you had a short across the terminals, that could quickly blow the charger.

The above would not be the most efficient, as the variable resistor would consume a fair bit of the power. So... Possibly the same effect with a switch mode instead?

What you're describing with the variable resistor is what's known as a linear charger (like a linear regulator). Linear chargers are used for small battery powered devices, where the power lost is small enough that it's not an issue. For bigger batteries the charging is done with a switchmode power supply, but the CC-CV regulation works the same.

 

[Username1]

So i know used Ex telecom and server room type power supplies were recommended for their reliability compared to lower quality chinese stuff...

But what about car audio battery chargers, like the one i linked? These are made for powerful (i'm assuming high end) car speaker systems. Are they really considered "cheap chinese"? I believe they're from Brazilian companies whatever that means. They seem to have various protections like overheat, short circuit, low voltage cutoff, internal fuses etc. Some examples of companies i'v found selling these are Taramps, Stetsom, Spark Usina.

These chargers are quite affordable compared to even most used power supplies i'v found so far, and offer very high amps (less parallel needed). The way my battery would be setup, it's easy to charge in two 12v sections. After looking at my panel situation, i'm pretty much limited to 3c charging currently. So i could use a single 180a charger per bank, with the option to add a parallel 180a in the future for 6c.

They appear quite suitable to me, but i could easily be missing things here. What do you all think of these? If these really aren't suitable i'll drop the idea and scour ebay some more.

 

[john61ct]

Yes I've never seen quality DC electricity infrastructure gear in the audio market, and definitely seen a lot of shoddy stuff lots more lying and scamming than you find even in the ebike arena.

But who knows? there are exceptions I'm sure, maybe you can be the pioneer and let us know what you learn.

But, have you even looked through the hundreds of threads here using the keywords I've given you?

Personally I bet that you'll be surprised how much a quality bank of any size costs.

And that you end up changing your desired voltage maybe more than once, that puts you back to square one.

As I mentioned, start out just spending a couple hundred max, accepting a long charge cycle until you've got everything else sorted.

No reason to prioritize your final ultimate charging setup, learn as you go, follow eBay a while and you'll start seeing crazy cheap high value deals all over the place.

Especially if you happen to settle on nominal 48V.

Buying both high voltage and high amps brand new is just silly unless you're wealthy

 

[Username1]

Alright i'll start checking out real power supplies properly as you suggest then. I didn't get too deep into the searching since i didn't quite understand what i needed. Now i have an idea of what i'm looking for, thanks.

 

[john61ct]

Sorry to keep repeating, but forget the focus on the super fast charging aspect for now, prep all your infrastructure to withstand it, but get everything else solved using a 0.1-0.2C rate first.

 

[serious_sam]

It seems like with cc/cv charging, you start by setting max Amp and Volt limits.

I actually think that's the more accurate way to describe the charger hardware. The charger is effectvely a current limiting and voltage limiting power supply.

It just happens to run at constant current (i.e. at the set limit), and then a constant voltage (again, at the set limit), in that order, due to the specific type of varying load (increasing back EMF) that the charging battery puts on it.

 

[Username1]

Sorry to keep repeating, but forget the focus on the super fast charging aspect for now, prep all your infrastructure to withstand it, but get everything else solved using a 0.1-0.2C rate first.

Will do.

I actually think that's the more accurate way to describe the charger hardware. The charger is effectvely a current limiting and voltage limiting power supply.

It just happens to run at constant current (i.e. at the set limit), and then a constant voltage (again, at the set limit), in that order, due to the specific type of varying load (increasing back EMF) that the charging battery puts on it.

Yeah that's how i understand it now after having read the previous posts.

 

[999zip999]

How about automotive-grade cells from electric car like a Chevy volt or ? There are a lot more better cells out there for your needs.

 

[Username1]

After much research and thinking, i'v decided on 10 x 40ah cylindrical cells (960wh). They will be made into 2 x 5s packs which are charged separately (12v), but discharged in series (24v).

I'v found a proper CC/CV 400a adjustable power supply for a good price. This works well because i'v since realized yinlong themselves advertise these as capable of 10c (6 minute) charging. The plan is to charge the packs one after another at first, with the option to get a second power supply down the road.

Now about the actual charging method... To achieve a real 6 minute charge, does this mean you must set the voltage to the maximum (2.8v per cell), and skip the CV stage? I can't find any information about the method used to achieve this.

 

[john61ct]

Sorry if I forgot, but why are you in **that** much of a hurry?

And remember, just do slow charging to start until you have got everything up and running, you will very likely be changing your mind as you gain experience so if you are smart, you will wait and save finalizing your charger spending decisions for last.

I think you'll find that 10C rate is only OK (maybe) up to 80%SoC

And 400A is twice as fast as that, going into only 20Ah. If your PSU allows you to de-rate current, then set it at 200A or IMO lower would be better.

Now about the actual charging method... To achieve a real 6 minute charge, does this mean you must set the voltage to the maximum (2.8v per cell), and skip the CV stage

Please reread
https://endless-sphere.com/forums/viewtopic.php?p=1548381#p1548381

For the first CC stage he **circuit** voltage will be much lower than your setpoint.

But because your current rate is so high, that voltage will hit your setpoint quickly, thus at a pretty low SoC.

Yes you can use a HVC to terminate the cycle there, but likely you want to use a greater % of your cells' capacity.

So, the PSU holds voltage at that CV setpoint, and amps will be falling.

At 20Ah, I would terminate the charge when trailing amps hit 0.05C, or 1A or thereabouts.

Of course you are sitting there watching in order to be able to do so.

 

[Username1]

Sorry if I forgot, but why are you in **that** much of a hurry?

I'm not in a huge hurry, i just have a clear goal in mind. I want max charge speed on LTO and i'v been extensively researching what parts to use. I think i'v pretty much sourced everything i need now.

I think you'll find that 10C rate is only OK (maybe) up to 80%SoC

I pretty much expected this, and i'm ok with it. There will be times i'll charge slower since the power supply i'm looking at is fully adjustable.

And 400A is twice as fast as that

No, these are 40 ah cells. 400a is 10c.

Please reread

I think you misunderstood me. I already know how CC/CV charging works thanks to yourself and others explaining it earlier. I was just talking about about setting the maximum voltage. I'm trying to figure out the exact charging procedure to achieve a real 6 minute charge.

Would you agree this can only be achieved by skipping the CV phase at the end? And i suppose you should set the voltage as high as possible (aka 2.8v per cell) to achieve the best possible state of charge while charging at 10c?

 

[john61ct]

I'm not in a huge hurry, i just have a clear goal in mind.

No I meant what is the practical reason for that goal, IOW why are you so focused on a so-called 6-min charge cycle?

> I was just talking about about setting the maximum voltage

And I've been repeatedly telling you, there only **is** one voltage setpoint. Its purpose is to be the maximum the charge regulation allows, and that capping function only becomes relevant at the transition from CC to CV, that is, if you choose to use any CV stage, aka Absorb Hold Time.

By using endAmps as your spec for stop charge you will get to a precise SoC / capacity utilization every time, no matter the charging C-rate involved.

Using an automatic HVC, then it is just a "charge to and stop" termination voltage, and the C rate will determine the SoC / capacity utilization; the higher the former, the lower the latter.

 

[Username1]

First reason is just because i want to. It's really cool and useful to charge an ebike in 6-10 minutes (and still have it last 10k+ charges). Secondly this ebike will be used heavily, and instead of fitting a massive conventional battery (which could also go empty and need a long recharge), i have the ability to quickly recharge as much as i want.

I still think you're misunderstanding me. This whole time iv only been talking about the "voltage setpoint", or what i called maximum voltage.

I basically have 2 questions... Am i required to skip the CV stage at the end, in order to achieve a real 6 minute charge? And should i set the "voltage setpoint" to the battery's spec limit (2.8v)?

 

[john61ct]

> Am i required to skip the CV stage at the end, in order to achieve a real 6 minute charge?

It is you failing to understand me.

You can charge to whatever definition of Full you like.

Whatever level you get to safely in 6min will not be anywhere near that, you've already said getting to only 80% SoC in that time is fine.

So then the question is, if it takes another 6min or 20min to get to your definition of 100%, do you stop charging and rush off only 80% full, or do you finish your cup of coffee?

> And should i set the "voltage setpoint" to the battery's spec limit (2.8v)?

That will be up to you, presumably related to your experimentation wrt the above variables.

My understanding is, such maximum ratings are best interpreted as stressful, "do not approach" limits if you want good longevity.

But no CV stage, "just charge and stop" CC-only, probably even 2.85Vpc would be fine, so long as you're not sitting long at that definition of Full.

If you were holding CV until an endAmps spec of 0.005C, I would advise a lower setpoint, say 2.65V, compared to endAmps of 0.02C where 2.75V might be perfect.

Apparently 2.55V is a good definition of Full *resting** voltage, at least one hour isolated after charging, 24hrs is better.

Your goal is to calibrate your gear to get to that point, or maybe something else if I'm wrong, @Sunder ?

And note when you are expecting 10000 cycles, you probably aren't too concerned if your needs cause you to make choices that are "suboptimal" wrt longevity.

So if you really want to get to 100% Full in 6min every cycle, and that causes lifespan to be "only" 5000 cycles, is that an acceptable trade-off for you?

 

[Username1]

I know that stopping the charge at the cc/cv transition point is somewhat arbitrary, but that's the point where it begins going progressively slower, and 80% at full speed is enough for me. So i can call that "fully charged" for my purposes.

I'm not sure how many cycles this will last at 6-10c. I'v read bits and pieces but would have to research further. The upper number quoted is 30,000 (using some low speed obviously). So 10,000 at 6-10c was just a number i threw out there. Even getting 5000 like you mentioned would be fine. Overall i'm not worried about it, whatever it lasts would be quite long it seems.

I was trying to figure out what procedure yinlong used to get the "6 minute charging" they claimed. But using simple math and how charging works, i can see now as i suspected, that they simply charged with 10c and skipped the CV step. So not exactly a "true" 6 minute charge since technically they didn't fully fill the battery.

 

[serious_sam]

I basically have 2 questions... Am i required to skip the CV stage at the end, in order to achieve a real 6 minute charge? And should i set the "voltage setpoint" to the battery's spec limit (2.8v)?

In simple terms, yes, skip the CV portion of the charge cycle, and charge at maximum current up to the battery voltage limit, then cut the charge off there. That will give you the majority of the charge in the minimum time.

 

[john61ct]

I know that stopping the charge at the cc/cv transition point is somewhat arbitrary, but that's the point where it begins going progressively slower, and 80% at full speed is enough for me. So i can call that "fully charged" for my purposes.

So do that, charge with an HVC at 10C, either just stop after 6min or when you hit your setpoint.

Note that doing a CC-only charge as above might give 80%, or 70% or 95%

you might hit that in three minutes and only be at 50%

that can only be determined by testing, trying various profiles and seeing what meets your needs with safety and repeatable convenience.

Precision is not at all required for your normal cycling.

But do establish a precise capacity test protocol, set your 100% SoH benchmark when the cells are broken in, so you can later tell when it starts declining, by how much per 100 cycles or per year whatever.

 

[Username1]

Been gathering some more information on these cells...

It appears that the larger version (66260 60ah) may no longer be in production. This size appears for sale much less often, and i haven't found any actual 60ah cells (just 55ah probably implying old/used stock). I was originally considering this larger size, but it appears the smaller 66160 size is definitely the way to go.

Speaking of the smaller size, i first thought 40ah was the maximum, but i'v since found they are now making a 45ah version. This works out to a little over 100wh per cell. Curiously they seem to talk about 2.9v maximum with these cells rather than 2.8v. The few sellers i'v found so far seem to be charging around $70-$80 usd per cell.

On charging... I originally talked about 6c being the limit, and it appears i was right. 10c seems to be for SINGLE CELLS ONLY, while 6c/10 min charge is talked about much more often. I also found some good news about high speed charging, with a test showing 6c results in ≥90% charge capacity (same thing for 6c discharge). So it appears 6c charging is indeed the way to go.

This information is very scattered, conflicting, and not well documented. So if anyone can confirm or correct any of this information please go ahead.

 

[john61ct]

a test showing 6c results in ≥90% charge capacity (same thing for 6c discharge)

Again, that is not how charging works for any chemistry any C-rate can get to any target SoC

It is the termination voltage, **combined with** the C-rate

but only for "charge to and stop" CC-only profiles, the lower the C-rate the higher the SoC%

adding some CV / Absorb Hold Time. it is the voltage **combined with** the endAmps spec that determines SoC%

in which case nothing to do with the early-stage C-rate.

Again, any C-rate can get to any target SoC, period, the two are just not related.

 

[Username1]

I believe this test was talking about charge absorbed without the CV phase, meaning fast charging with a constant current until the maximum voltage was reached then stopping. Should have made that more clear.

I'm just trying to say it looks like it could be ≥90% at 6c, which i would be quite happy with,

 

[john61ct]

Yes without holding CV, then going from 10C down to 6C

for a given voltage, a slower rate gets you to a higher SoC