Battery testing - choices for series circuits

[AriM]

@Stosh - Ni-Mn and Ni-Zn cells
At some point we could maybe start to advocate for mods built to use these 'safer' types of batteries. That seems a good move to me if these batteries really are a better prospect.


The Kick
I have been told the Kick, in practice, will deliver a maximum of 10 watts. If that is true, then perhaps it cannot overdrive a battery? Although I suppose that depends on what the definition of 'overdriving a battery' is. The little Tenergy Li-FePo4 cells seem to be able to take a huge overload without problems. Provided you don't get a counterfeit one.

My concern about the Kick, is that there is no conclusive data being posted, to show it's maximum draw....from ONE single battery. I am also concerned that Super-t would not answer the question I posed. Does he PERSONALLY ENDORSE a 10c rating as a vendor. I mean it's totally unfair for a vendor to come here and say "a battery can take 10c" and then not personally endorse that statement.

This is the real problem. People are throwing random statements out there, with no data to back them up.

 

[AriM]

I've noticed the flack you've gotten for recommending things as an 'official' of ECF and I understand your feeling of responsibility to the 'community'. I imagine that some of the lessor experts agree with you but find excuses to disagree because of your position, not your objectives, or your qualifications. Something has to be done, agreed, and the open discussion approach you are taking now has a good chance of getting more of the 'experts' to actually contribute rather than fight the concept of driving mod makers and resellers to build a safer product.

If 'we' can recommend a safer stacked battery combination and highlight that counterfeiting is rampant is the Li-ion market that would be a plus for the community. Then figure out a way to make this info highly visible to a 'Noob'.
and highly visible to our vendors, so they KNOW we are watching them

This is why I suggested early on, that this does not become a "discussion". Rather it just becomes a place to post real data. I can't see how anything else is relevant. Something is either safe, or it isn't.

It would be simple to determine this. Simply present a load to a cell (which simulates an average "vape") and then use a temperature probe to measure the temperature of the cell under load. It's not that difficult.

I have a new testing rig showing up this week, that can more accurately simulate the "duty cycle" of a vaper. Hopefully I can qualify some of this data, or disqualify it.

 

[forcedfuel50]

My concern about the Kick, is that there is no conclusive data being posted, to show it's maximum draw....from ONE single battery. I am also concerned that Super-t would not answer the question I posed. Does he PERSONALLY ENDORSE a 10c rating as a vendor. I mean it's totally unfair for a vendor to come here and say "a battery can take 10c" and then not personally endorse that statement.

This is the real problem. People are throwing random statements out there, with no data to back them up.

Super T and Evolv both endorse the use of the AW IMR's for use with Kick and absolutely i have the data to back it up and will post up shortly.

 

[forcedfuel50]

Here are some tests specific to the IMR 18490/Kick. I also have many more tests posted on the photo gallery page of my site. I also have many more not posted and can do any custom requests so just ask. I'll post up AW IMR 18350 tests later which were similar.

But pertaining to the Kick and the recommended AW IMR18490:

- Test one was a cycle test conducted at 4amps (The Kicks approx. max amperage draw) with 5 secs on, 20 seconds off to simulate vaping on the Kick. I came up with the 5 second on, 20 second off estimate based on vape times poll taken in my subforums about 6 months ago.

As you can see, the AW IMR 18490 far exceeds the Kicks demands:

Results Test 1:

AW IMR 18490
Test Amps: 4
Duty Cycle: 5 Secs. On, 20 secs. Off
Beginning Temp: 71F
Max Temp: 76F
Tested mAh: 948 (with a cutoff at 3.2 volts as the Kick shuts down automatically just before this voltage)

-Test two was to push the AW IMR to its maximum rating. Load was set at 10amps continuous which is far beyond any demands of the Kick or any Ecig for that matter and max recorded temp was 106F, which was just warm to the touch.

Results Test 2:

AW IMR 18490
Test Amps: 10
Beginning Temp: 72F
Max Temp: 106F
Tested mAh: 1067


Test 1:

Test 2:

 

[AriM]

GOOD STUFF!!!

I can't agree that 106deg F is "warm to the touch". IMO that is a bit too hot for longevity of the cell. Certainly the results at 4 amps are within safety/longevity limits.

Can you please PM me the CSV data for the graphs (or post larger pics?). I am going to duplicate the tests you ran to back them up.

Also I will run the duty cycle test at 10amps. I am curious to see how consecutive switching at 10 amps affects the discharge results.

Not to badger you, but let me just pose the question one more time. For the public record..

Does Super-t manufacturing personally endorse a rating of 10c discharge for the AW IMR 18490?

I ask again because I am trying to get people to make legally binding statements about safety limitations. Specifications seem to magically change when you put people on the spot. Not doubting your professionalism, or integrity. I would like to see other vendors make binding claims about specs. This is the only way to "weed out" false information, and keep people safe.

Thanks again for posting the data. I look forward to verifying it later this week.

P.S. I hope none of this comes off as overly aggressive. I am just trying to keep things cold, clear and factual. I will hold myself and my statements to the same level of scrutiny. It's never personal.

 

[AriM]

Let's start with 3v cells for 6v vaping, then.

These have been nominated as candidates for discussion or testing:

Tenergy Li-FePo4 rcr123a 3.2v
AW Li-FePo4 r123a 3.2v

Any more?

Where could you buy these with a reasonable expectation they are not counterfeit?

Roly,

I have both of those cells set aside for testing. 4 of each to be exact. They are sourced straight from the vendors. I have no reason to believe that they are counterfeit.

I would like to agree on testing methodology, so that I can start running a full analysis on these cells.

Suggestions?

 

[rolygate]

Arim,
With regard to SuperT, perhaps he only needs to state that a 4C or 5C rating is supported, as that is what it will be used at? It would be nice to have a safety factor of 2 but it does not necessarily mean it will be used at that rating - surely?


3 volt batteries
It looks as if they would be used with an average load of 2.5 amps, perhaps under extreme conditions 3.5A. So perhaps they could be tested at 3A, 5 seconds on, 20 seconds off - for enough cycles to kill the cell.

Then, see how it performs after recharging. I expect the AW will die but the Li-FePo4 might survive. It would be useful to know if it becomes unreliable after being drained.

 

[AriM]

Arim,
With regard to SuperT, perhaps he only needs to state that a 4C or 5C rating is supported, as that is what it will be used at? It would be nice to have a safety factor of 2 but it does not necessarily mean it will be used at that rating - surely?


3 volt batteries
It looks as if they would be used with an average load of 2.5 amps, perhaps under extreme conditions 3.5A. So perhaps they could be tested at 3A, 5 seconds on, 20 seconds off - for enough cycles to kill the cell.

Then, see how it performs after recharging. I expect the AW will die but the Li-FePo4 might survive. It would be useful to know if it becomes unreliable after being drained.

I asked for the 10c rating endorsement, because he mentioned that as a specification. I think if any of us are going to make claims AT ALL. We need to endorse them, and take responsibility for those claims. That is the sure way to get real safety guidelines, we have to request that people to endorse their claims (any claim). I will personally never endorse a claim of greater than 2c. I agree that an endorsement of 10c is not necessary, but if the claim is going to be made...it must be backed up with some form of accountability.

I am on board with the test you described above. I would like some input from a few other members, just to make sure we are all on the same page. I am willing to perform any test requested. It looks like Super-t is willing to do the same. So that is a good way to vet each others data.

 

[Rocketman]

Ari,
True believable C rating are needed. It won't be long before the 25mm length cells are used with a Kick or two of them stacked for a smaller 6 volt mod. Granted that would be a smaller explosion. Real specs need to be put out front. A recommended lower C rating could be also given.
But lets be consistent.
State 4 C for a cell and someone will go buy the cheaper cells with a better label.

 

[AriM]

Ari,
True believable C rating are needed. It won't be long before the 25mm length cells are used with a Kick or two of them stacked for a smaller 6 volt mod. Granted that would be a smaller explosion. Real specs need to be put out front. A recommended lower C rating could be also given.
But lets be consistent.
State 4 C for a cell and someone will go buy the cheaper cells with a better label.

It's a good point. I think we should give a lower recommended C rating. To limit liability. If people choose to go above, they are on their own. We can reach 2.1 amps draw with each spec I listed on page 2. They are all capable of giving great vapor. I, of course will provide data to show more difficult loads. I think we can show safe levels on the basis of temperature.

Tenergy's official rating for their lifePO4's is 700ma. We are already pushing that by endorsing 2 amps. I suppose it's all conjecture until we have the data.

My new load testing rig should be showing up on Thursday. So hopefully I will be able to start collecting data this weekend. Including the 18650 (low end) shootout.

P.S. I have done testing on cr-2 sized cells, both LiFe and Li-ion. Neither are up to the task of being suitable for PV's. Aside from the fact that the current availability isn't there, they will only give you about 45 minutes of usable vapor production.

 

[AriM]

Ok, the new lab gear is here and set-up. I am going to build a test jig that houses the battery inside a tube mod. That way we can simulate the real world. I also have been thinking about the duty cycle test. It's not really accurate. A more accurate test would be a constant resistance test. That would simulate an atty load. The duty cycle test is inaccurate, because it is constant amps draw. This is not accurate because as the battery voltage goes down, the atty stays constant, so therefore the amp draw goes down with voltage. So instead of the duty cycle test, I will do these as constant resistance, with random duty cycles....

any ideas or thoughts on that?

 

[rolygate]

Well maybe if we can do it, a range of tests would be a good idea - then you don't have to choose one over another.

Take a battery and mark it for identification.
Put a constant load of 3A on it and drag it down to a sensible minimum voltage.
Put an intermittent load of 3A for 5 sec then 20 sec rest and drag it down to a sensible minimum.
Repeat tests x 10.
Repeat tests, this time take battery down to dead.
Examine results and design any further test.
Put a dead short on it for 10 seconds, do not recharge, allow to cool, then run tests again.
Put a dead short on it for 10 seconds, recharge, then run tests again (expect possible fail event).
Does the battery reliably supply the current required to drive an atomizer?
Can it do that after heavy loads or abuse?
Does the battery fail if abused?
Does the fail happen without significant events, or is there a physical event of some kind associated with failure?

These might be some of the considerations but there is always more that can be added after testing starts because other things become clearer at that time.

 

[AriM]

Well maybe if we can do it, a range of tests would be a good idea - then you don't have to choose one over another.

Take a battery and mark it for identification.
Put a constant load of 3A on it and drag it down to a sensible minimum voltage.
Put an intermittent load of 3A for 5 sec then 20 sec rest and drag it down to a sensible minimum.
Repeat tests x 10.
Repeat tests, this time take battery down to dead.
Examine results and design any further test.
Put a dead short on it for 10 seconds, do not recharge, allow to cool, then run tests again.
Put a dead short on it for 10 seconds, recharge, then run tests again (expect possible fail event).
Does the battery reliably supply the current required to drive an atomizer?
Can it do that after heavy loads or abuse?
Does the battery fail if abused?
Does the fail happen without significant events, or is there a physical event of some kind associated with failure?

These might be some of the considerations but there is always more that can be added after testing starts because other things become clearer at that time.

That all sounds fine to me. So I will do duty cycle tests. Constant current. Constant load (resistance) and dead shorts. I will graph temperature as well. I will allow for specific amounts of time for "recovery" in the dead short and temperature tests.

I will do all of these test in a 2 cell "stacked" configuration, inside of a vented tube....with a tape or mylar/cardboard sleve insulator.

I can already tell you what will happen in the dead short tests (have done them many times). The shrink wrap will melt off and the cell will get roasting hot. With LiFe cells I have dead shorted them down to 2 volts...let them cool, and charged them back up...I have a set that I have done that to about a dozen times (cheap tenergy's) and they still work fine.

On the other hand, I did the dead short on an "exotic" 18650 LiFe cell (high discharge cell). The shrink wrap caught on fire....the battery vented and destroyed itself...it actually wnet into thermal runaway, and continued to self discharge with no load/short. Go figure. That was a $14 cell. The $1.67 Tenergy took it though!!!

We also have to consider what constitutes "failure". IMO a battery that has any visual signs of damage (even melted shrink wrap) should be considered "destroyed/trash". I am sure they will continue to function, but I would never stamp my name on one and call it "safe"

penny for your thoughts?

I will also print labels for each tested cell....stating the test dates and peak voltages and temperatures. That way we can re-visit them if necessary (let's say a year from now...if necessary)

P.S. I am going to have to refuse doing the dead short tests on Li-Ion (CO) cells. As this is not safe. I have done those tests, and the battery does explode. I have a fire box here to do those tests, but I don't want to endorse that testing methodology....not because it's unwarranted. I just don't want some other member out there to go out and try to duplicate that test (w/o proper safety gear), blow their finger off, and then try to blame me and ECF for endorsing that testing methodology. If some other brave soul wants to do that test....they are on their own...and I strongly advise them not to do so (and you will notice there is NO *wink* *wink*). Seriously don't do it....

 

[AriM]

Also, Roly. Would you be so kind as to generate a proper checklist for the tests. So that strict guidelines can be followed by myself and anyone else willing to do these tests. I would like to have that come from someone, other than myself, just to be sure that it's impartial and 3rd party to the test. Any good testing methodology should include vetting and scrutiny by ones peers.

A simple outline form checklist would be fine with me. This can be used in the next series of tests as well....that way everything stays consistent and repeatable.

It would be nice if that checklist was on "official" ECF letterhead. I will print the list and include it with each tested 2 cell "stack". I will label the cells and put them in a temperature/humidity controlled enclosure, along with the printed data and forms. Once the tests are considered "finished", I would be happy to mail the boxed items to "ECF" for long term storage.

 

[rolygate]

OK the comments on the testing I agree with but we should probably wait 1 day to see what other comments are made.

Agree that Li-ion testing to destruction is not a good idea.

I will wait a day then start to create a form/checklist for the battery test. This would need changes/edits to get right but should still be available fairly soon.

I would like to be sure we have got all necessary input before we go to that stage.

This is looking very promising. It will allow us to be able to say, "This battery was tested and appears fit for purpose". Also, "However you must guard against purchase of counterfeit items and are advised to buy from a vendor where the supply chain can be verified". If anything this is a big a problem as the batteries themselves, a fake AW or Pila battery is probably worse than a good basic Li-ion battery as it could be reject junk.

 

[AriM]

OK the comments on the testing I agree with but we should probably wait 1 day to see what other comments are made.

Agree that Li-ion testing to destruction is not a good idea.

I will wait a day then start to create a form/checklist for the battery test. This would need changes/edits to get right but should still be available fairly soon.

I would like to be sure we have got all necessary input before we go to that stage.

This is looking very promising. It will allow us to be able to say, "This battery was tested and appears fit for purpose". Also, "However you must guard against purchase of counterfeit items and are advised to buy from a vendor where the supply chain can be verified". If anything this is a big a problem as the batteries themselves, a fake AW or Pila battery is probably worse than a good basic Li-ion battery as it could be reject junk.

I am on board with all of the above. I am in no rush to get the tests started. I have to calibrate all of the gear that will be used, and build the test jig. I also need to source more tenergy cells, as I only set aside 4 for evaluation. I would be happy to accept any donated cells for testing. My requirements are as follows...

1. Must be factory new cells
2. Must be from a verified source, in original packaging
3. Must be understood that the cells will not be returned
4. I take no warranties or claims of liability for the cells...in other words I will not put my name on the cells as the verified source
5. I will accept no "compensation" for the items sent to me
6. I will not pay for any associated costs for shipping, purchasing, handling or liability
7. I will not remove any liability from the original source (if you send me a "bad" cell and it blows my hand off out of the box....you will be hearing from my lawyer)

Sorry, I suppose I am not making a great case for donations. I also do not absolve or endorse any claims made by ECF or the general public.

Now that we have the formalities out of the way....please feel free to PM me with any concerns or suggestions. I am happy to listen, oblige etc etc etc...

P.S. I am really a pretty nice guy when the "science" hat is off....so sorry if I come off as rude, smug or "cold". I am just trying to cover my own ... here...

 

[rolygate]

No problem, this is just common sense I reckon.

Would it be easier if we could send you some funds for purchasing batteries?

 

[AriM]

No problem, this is just common sense I reckon.

Would it be easier if we could send you some funds for purchasing batteries?

That would also be acceptable. I am an AW re-seller, so no issues on those. I have plenty and the source is verified as AW. The tenergy's are another matter. I can buy them from my local Fry's in official tenergy packaging. The only problem is I have to buy them with a "charger". So they are $19 for 2 cells. I will do my best to find an alternate source that is verified genuine (I will contact tenergy monday). I think there is a local rep.

I think a truly fair evaluation would span 10 cells of each. Let me do some more homework on the tenergy issue, and then I will try to ballpark some prices. I am not terribly worried about the cost (should be under $50 for all of the above). I am happy to contribute, as I will also put the data in my own database. In fact, this is a personal mission of mine and my business. I want this information to be out there and endorsed. It only comes back to us all ten-fold, by avoiding liabilities from potential patrons/users.

My main request was for cells that I don't have time to source, or don't know about. I would love to have more than just the 2 brands in the test. I know of quite a few other brands, but I just haven't had time to source/evaluate them.

If any other members have cells that they want to include in the testing, then I am happy to accept those....just as long as they follow the guidelines I set above.

Would you agree that 10 cells from each make/model should enter the testing? My thinking is that I do a 4 cycle charge/discharge evaluation of each individual cell (of course I will record/log the data). Then I can weed out any bad cells, and give the cells a chance to "break in". I will pick 4 each out of the 10, based on average peak voltages. That way our sampling represents what a consumer could expect to purchase/use.

 

[Rocketman]

Comment on your cell selection criteria.
Peak voltage may indicate something, but mid charge voltage shows more about the entire cell, internal interconnects, cell resistance, etc. A partial dump from full charge and loaded voltage might be better to sort cells. AC impedance testing is used by the industry to measure internal resistance but voltage with a medium load vs time (2 amps for 5 minutes for example) would be a quick way to weed out poor cells. If a new cell gives a voltage outside the group distribution at 70 to 80% charge, it is an outlier. This way you wouldn't be 'cherry picking' the 4 best cells. The sample from within the group should represent the group.

Be aware that short circuit test results may be skewed by PTC recovery. PTC first action currents may change with subsequent actions and become less predictable. A dead short that opens the PTC won't "heat the cell" past the trip temperature. That's why you see test results with shorts for 30 minutes, and the cell doesn't blow up. If fact, instead of completely draining the cell ( a short for 30 minutes should empty the cell ) but the cell still has a charge when the short is removed. An 'almost short' will likely heat the cell the most.



EDIT: maybe 1 amp for 5 minutes. I matched some 18650 cells using 2 amps but that is probably too high for little cells, just to match them.

 

[AriM]

oh also, I do not want to include any "exotic" cells. I consider "exotic" anything that costs more than $4 per cell (wholesale).