How an electrician evaluates a DC charger & battery capacity

friendly1uk · Feb 8, 2015

Your balancing time this week won't be the same as next week. It will increase as the cells drift further apart. Normal aging.

Many of us have seen no balance time needed sometimes. Or more depending on how deeply discharged they became. It is variable.

What is 100% charged? For me it is when the green light comes on, because balancing happens before peak voltage. My bms don't wait till something goes to high.

Maybe you could add some leds to each balance channel so you can see when it is done.

Your testing here is a bit futile, because you won't repeat the same duty ever again. Though it might be educational to see it at least works.

bs2391 man.

cwah · Feb 8, 2015

ah you are still discovering the world of the batteries, although you are electrician.

Not sure the time to charge or balance is of any value to us.

What is interesting, and what interest us in term of battery... is other variables such as:
- Calendar life: How it ages over time
- Cell to cell consistency: This one is easier to measure, and closer to what you are doing
- Capacity loss between each cycle.
- Resistance increase between each cycle.

I suppose after 20 cycles or so you should be able to see some patterns, although temperature and cells resistance are an important factor.

friendly1uk · Feb 8, 2015

Kiriakos GR said:
friendly1uk said:

Your balancing time this week won't be the same as next week. It will increase as the cells drift further apart. Normal aging.

Click to expand...

This is a fact but with out having any data collection when the battery is fresh, you may not be aware of the changes over time.
After cycling my battery five times by making 50Km from charge to next charge, now it is ready so me to start recording data.
Today this is my perfect working example of a battery which is amost one year old in age.

friendly1uk said:

Many of us have seen no balance time needed sometimes. Or more depending on how deeply discharged they became. It is variable.

Click to expand...

Apples to apples the specific Lifepo4 BMS cut out when voltage drop down to 31.5V, and therefore is not variable here.

friendly1uk said:

What is 100% charged? For me it is when the green light comes on, because balancing happens before peak voltage. My bms don't wait till something goes to high.

Click to expand...

I would be interested to see your charge current graph if you are able to have one.

friendly1uk said:

Your testing here is a bit futile, because you won't repeat the same duty ever again. Though it might be educational to see it at least works.

Click to expand...

I will promise three reliable charge graphs in the next six weeks, the best that I can do for you. :lol:

Can I number them?
1, You won't know it's all going wrong, till it all goes wrong. Then you will get a new pack, and it won't perform the same. Data from previous packs can't tell you when a cell in your next pack will fail. That is why we use a battery management system that involves constant monitoring. Though if you enjoy such data-logging exercises you might enjoy the challenges of rc lipo chargers. I'm sure you have seen they have PC monitoring software. They know each cell by name.

2, I'm not sure if 31.5 is a factory lvc. If it is, you don't want to be hitting it every time. You should be avoiding it. It would be a pain to have to run out of battery on every charge anyway.

3, I made a mistake before. My green light has two levels of illumination. It will start on level 2, then start switching back in forth between level 2 and level 1. With increasing frequency until it settles on level 1. Though never for hours on end. I think the self discharge or perhaps bms drain is evident. If I don't switch off the controller, It will stay on level 2. I don't graph it though. It seems to have no purpose when you have a bms and some sort of watt/hour meter. Allowing you to check capacity and watch seasonal trends.

Teh Stork · Feb 8, 2015

cwah said:
ah you are still discovering the world of the batteries, although you are electrician.

Not sure the time to charge or balance is of any value to us.

+1.

Kiriakos, there really are only two stages to battery charging: 1. Constant Current. 2. Constant Voltage. Thus the acronym CCCV is often used when speaking of li-ion charging. The act of balancing normally takes place at top-charge, but this is not a rule. Note that the time spent in CV increases when internal resistance grows, and also becomes significant while fast charging. (In fact, my old li-ion battery spends almost 80% of fast-charge time in CV mode). Your 4 out of 5 rule is useless.

CCCV charging and Battery Management System protection operation
Texas Instruments on CCCV charging, buck converter hardware and software

Outroduction: Currently writing my master thesis for M.Sc in EE, specializing in lithium ion battery management.

riba2233 · Feb 9, 2015

Kiriakos GR said:
What my measuring equipment are unable to calculate is the portion of energy in Ampere-Hours which is received after the Green light and three hours later that that.
For such a measurement I would need a 3000$ USD equipment which is not for my wallet, especially when such evaluations does not belong to my major applications list.

Or maybe just an 150$ cycle analyst?

friendly1uk · Feb 9, 2015

Do what math? All the numbers are not present

You could measure the amount of charge in Ah using a $10 Ah meter. It need not be calibrated, because the exact result won't be acted upon. Your just getting an idea of whats going on. Something you could just read from one of the many charge graphs on the internet.

I'm not sure that you are an electrician. I think a maintenance fitter. I test the work of electricians in my role as test and measurement engineer (bs2391) It is for me to decide who does what. I seriously question your competency. When will you be evaluating battery capacity? I'm finding this quite amusing.

teklektik · Feb 9, 2015

meter-masturbation

(mē′tər) (ˌmæs tərˈbeɪ ʃən)

Noun.
The act of playing with meters for self gratification.

origin: Greek

dnmun · Feb 9, 2015

guys spend a lotta time measuring it so that is why they call it a meter.

won't matter much as greece falls into the coming decade of firebombing of all the businesses still left.

$62 billion just left greece headed to german banks last month. anybody with any brains and money has already skipped out.

but they got the guvment they wanted.

dogman dan · Feb 9, 2015

Guys, you are going to let him trick you into saying shit that gets you banned. Be careful guys.

Edit see page two for explanation of this comment.

dogman dan · Feb 10, 2015

True enough, the stakes are different if you are trying to ram a product back down the throat of the seller. Most of us aren't doing that, we just want to know on the fly how things are doing during a ride or during a charge.

For that, perfect calibration is not as required as much as other rides or other charges monitored with the same instrument. For example, one truck I owned had a gas gage that read zero with 15 gallons in the tank. Long as I knew this, I could deal with it. What I mean is, perfectly measured or not, no change from other rides is fine. Something changes, find out why.

Even my cheapest watt or volt meters haven't been that far off. But you don't want to use meters you know are more than .1v off. Cycleanalyst and a couple cellog 8's suit my needs fine. My DVM's are junk, but good for what I use them for, mostly just confirming + and -, or going Aha, there is or is not voltage here. About the limit of my skills anyway.

number1cruncher · Feb 10, 2015

dogman dan said:
Guys, you are going to let him trick you into saying shit that gets you banned. Be careful guys.

Is that his game? Has this happened to others? I trust your words more than most here, so I will take them at face value.

Maybe there needs to be the OPPOSITE of the GURU designation for those who mislead others, either intentionally or through their own disillusionment. That way someone trying to learn does not waste their time reading a bunch of nonsense that will just confuse them when they read something of value. Something like, (moderator edit)

It is okay for someone like you or I that has little to no formal training helping others with their questions so long as we do not mislead them into thinking we are experts, educated or whatever.

Everything he writes is Greek to me, so maybe I'm just missing all his great insights. :?

riba2233 · Feb 10, 2015

Nice idea, that would be funny :lol:

999zip999 · Feb 10, 2015

friendly1uk · Feb 10, 2015

Kiriakos GR said:
friendly1uk said:

You could measure the amount of charge in Ah using a $10 Ah meter. It need not be calibrated, because the exact result won't be acted upon.

Click to expand...

In such a case I will follow the average style of measurement and I will multiply 120mA X 3Hours so to get the sum of .36Ah, as answer in my question and I will keep the 10$ for another use.
When chargers do most work at constant current and just a bit of timing of this process can bring you close to an realistic estimation, then why paying even a dollar for a dedicated measuring tool?

The ones who care about quality of measurements is the ones who need true evidence so to challenge the battery manufacturer or the fraudulent battery seller who trying to dispose to consumers old stock as to were new.
The battery is the second most expensive part of one ebike, if something goes wrong with it, with out reliable proofs at hand, there is no chance to prove the truth and get your money back.

Was this thread not meant to talk about battery capacity? You don't seem to of got there yet. I know my $10 item sits constantly on my bike in a tidy fashion. It tells me more than your pair of much more costly meters. I see all the seasonal changes day by day, that will make your accurate meters pointlessly accurate. They are of limited use. Keep them locked up somewhere and fit a $10 meter. It will measure discharge, not melt like yours.

Where did the funny thread go? It had a lot more useful information in it. Delete this non-sense before anybody acts upon it, and brink back ykick's

emmetbrown · Feb 10, 2015

You are an electrician, how does this fact qualify you for electronics?
Why do you always speak that rude to people that are interested in this ELECTRONICS-hobby?
Why can all people here be happy about getting tips from you for free? It's an online-community here, giving tips
free of charge is normal.
Boah, you got meters? Me too, without beeing an electrician, and some of my equipment is even labeled R&S :shock:

.
How did others here build (not put together) whole controllers and other wierd stuff without beeing
a greece electrician?
My point: just be a bit more sympathic human and less greek electrician super-hero and most of us will
be interested in what you have to say and I guess you will be welcomed here everytime.
Never stop learning :!:

Greets,

Christian

liveforphysics · Feb 10, 2015

On a serious note, you have a very nice mutimeter. Agilent makes my favorite DMM's now, but oddly you're not taking advantage of the reason why.

Get the bluetooth module that snaps onto the back, download the keysight logger apk, and it will link to your phone and create a time-stamped excel file of data for you at whatever intervals you want, as well as giving you a graphical interface. Saves you from needing to waste your time manually.

For the folks who don't know, that meter is the cats pajamas. However, a tool like this is much better suited for data collection of batteries IMHO, and cheaper.

http://www.jun-si.com/EnProductShow.asp?ID=102

major · Feb 10, 2015

Kiriakos GR said:
All right !! finally my very first proofs at hand, that is a graph and artistic creation so to be more easily understood.
My very own question was how much time after the Green LED my very own battery needs to stay connected so to get fully charged?
The specific measurement gave me the missing clue which translates to three additional hours.

What my measuring equipment are unable to calculate is the portion of energy in Ampere-Hours which is received after the Green light and three hours later that that.
For such a measurement I would need a 3000$ USD equipment which is not for my wallet, especially when such evaluations does not belong to my major applications list.

I would like to point out an error on your graph and artistic creation. The value of 100 on the abscissa corresponds to 100 samples of 5 minutes (300 seconds) each taken from the ordinate at the start of the test. So that would be a duration of 500 minutes or 30000 seconds, which is equal to 8.3333 hours and that is equal to 8 hours and 20 minutes, not 8 H 33 Min 33 S as you have shown.

Stevil_Knevil · Feb 10, 2015

major said:
Kiriakos GR said:

All right !! finally my very first proofs at hand, that is a graph and artistic creation so to be more easily understood.
My very own question was how much time after the Green LED my very own battery needs to stay connected so to get fully charged?
The specific measurement gave me the missing clue which translates to three additional hours.

What my measuring equipment are unable to calculate is the portion of energy in Ampere-Hours which is received after the Green light and three hours later that that.
For such a measurement I would need a 3000$ USD equipment which is not for my wallet, especially when such evaluations does not belong to my major applications list.

Click to expand...

Kriakos,

Ever heard of a Kill-A-Watt?
20 bux USD ->

http://www.newegg.com/Product/Product.aspx?Item=N82E16882715001&nm_mc=KNC-GoogleAdwords-PC&cm_mmc=KNC-GoogleAdwords-PC-_-pla-_-Home+Gadgets-_-N82E16882715001&gclid=CMGbmpK-2MMCFc9ffgodOIIAhA&gclsrc=aw.ds

It even measures how long you have been "METERING" :lol:

dogman dan · Feb 11, 2015

Okey dokey this has gone far enough.

Before I lock this thread, I owe you a sincere apology Kriakos.

I've had the flu for a few days, but it's still no excuse for saying you could be trying to trick people into getting them banned. You aren't doing anything, except being yourself, and having your own personal point of view. I see NO trolling on your part Kriakos, and I apologize for making a statement that implied that.

But you others, you guys are doing the name calling, and are going to be the ones getting the ban stick if this doesn't stop. It won't be Kriakos fault, it's yours. You wrote that shit. Ridicule the graphs, the math, but not the person. When did ES become a place where being a bit different was not allowed? Picking fights though, that's never been allowed.

dogman dan · Feb 11, 2015

Thread unlocked, now that I got a chance to tidy up. Again, my apologies for my inappropriate choice of phrasing.

windtrader · Feb 21, 2015

I am joining late and seemingly missed the major dustup earlier; thank goodness. I want to thank Kiriakos for all his time and knowledge he shares with us. Virtually we all benefit from those who bring more schooled and science based knowledge to us here about electrical stuff. I want to thank Kiriakos especially as English is not his native tongue yet takes the extra extra effort and time to communicate in English above and beyond his time to present his views, whether they are any more or less valid than anyone else is not the issue.

I'm sure the true EE and ME here who put food on the table with such knowledge must throw their hands up everyday about all the BS that gets posted and I'm sure the actual truth is held in a very small number of posts by those that really do know. Assuming Kiriakos has made his living for a few decades doing electrical stuff, he goes to the top band of expertise around here.

Let's not run him off as he has been dignified and respectful, more than seems to be said for others here. Thank you again, Kiriakos.

teklektik · Feb 21, 2015

Kiriakos GR said:
While recording memory in this meter is huge, Microsoft EXCEL has a limitation and can not handle unlimited samples.
60 samples per hour for about 8 hours is a safe bet. :wink:

Nonsense.

Code:

+-----------------+-----------+--------------+---------------------+
|                 | Max. Rows | Max. Columns | Max. Cols by letter |
+-----------------+-----------+--------------+---------------------+
| Excel 365*      | 1,048,576 | 16,384       | XFD                 |
| Excel 2013      | 1,048,576 | 16,384       | XFD                 |
| Excel 2010      | 1,048,576 | 16,384       | XFD                 |
| Excel 2007      | 1,048,576 | 16,384       | XFD                 |
| Excel 2003      | 65,536    | 256          | IV                  |
| Excel 2002 (XP) | 65,536    | 256          | IV                  |
| Excel 2000      | 65,536    | 256          | IV                  |
| Excel 97        | 65,536    | 256          | IV                  |
+-----------------+-----------+--------------+---------------------+

At one sample per second or 3600 samples per hour:

Excel 2007 (8 year old version):

Worksheet size = 1,048,576 rows
- (1048576 rows) / (3600 samples/hr) = 291 hours = 12 days

Excel 97 (18 year old version):

Worksheet size = 65536 rows.
- (65536 rows) / (3600 samples/hr) = 18 hours

Even 'free download' Open Office handles 65536 rows...

teklektik · Feb 21, 2015

Kiriakos GR said:
Bill gates become rich because many like you they believe everything they read. :lol:

"People like me?" I can only guess that you mean engineers who use spreadsheets regularly as part of their job.

Anybody with even the most basic Excel skills (or Google) can demonstrate the limits called out above....

But - perhaps your remarks about this common data analysis tool explain your lack of familiarity and your 'misstatement' above.

Stevil_Knevil · Feb 21, 2015

Have you figured out your charger's efficiency and battery capacity yet, K's?

How an electrician evaluates a DC charger & battery capacity

1 MW

100 MW

1 MW

1 kW

100 kW

1 MW

100 MW

1 PW

1 PW

1 PW

1 kW

100 kW

100 TW

1 MW

10 W

100 TW

10 kW

10 kW

1 PW

1 PW

1 kW

100 MW

100 MW

10 kW

Similar threads