Thread Tools
Nov 09, 2012, 04:27 AM
aka JetMan Joe
MCSGUY's Avatar
Thread OP
Discussion

Battery Load Test Comparisons


(THREAD INDEX)

June 2019 Update

UNDER CONSTRUCTION

June 2018 Update

5 years of testing has revealed much about how Lipo's respond to sustained high loads. At the beginning standing over a pack pulling 180 amps was damn scary. (Apparently those things are full of smoke too ). So It's hard to believe how safe I felt yesterday as my meter, for the first time ever, pegged 302 amps!

We've explored resistive and dynamic loads in constant and fluctuating modes. A constant load of either type has proven an effective means of separating Lipo fact from fiction. I have attempted to acquire and present the data with the same level of detail expected of me by my various Aerospace contract employers. I am very grateful for the support and recognition many of you have given in response.

I want to sincerely thank WAYNE GILES for freely sharing his insights, knowledge and experience. He's been instrumental in helping me avoid a few pitfalls and has pointed me in directions I'd have missed otherwise. I don't want to get into too much detail here because it may negatively impact my carefully crafted image of infallibality and omniscient expertise

IR based calculations, conventional capacity checking and sub 100A power evaluations were around before my time and continue to have a place. However my goal has been to determine how the type of packs I use perform when powering my EDF's the way I fly:
As fast as I can go for as long as I can reliably stay in the air.

IR measurements are effective for evaluating pack health. Before Graphenes changed the equation they were the gold standard in determining maximum cycle life vs. load. (Report 5498, Page 367). However it is not an effective method for predicting performance under sustained high load conditions as pack chemistry interacts with raising temperatures (Ref. report 653, Page 44).


THINGS I'M CONVINCED OF BEYOND A SHADOW OF DOUBT

The best way to determine a given packs true C rating is to load the crap out of it to heat the guts up and determine the threshold at which the Voltage sags before recovering. As long as the voltage doesn't sag and the running temperature remains below ~140F (60C) the pack will live out whatever it's design potential is.

Excessive internal temperature kills packs. Cell Matching and QC are the holy grail for pushing the limits. (Ref report 941, Page 63). Running a pack beyond it's chemistry and structural support causes it to get hot (duh!). But what happens is excessive heat degrades the cell matching beyond the industry standard 0.100v/cell. A vicious cycle develops as excessive cell voltage spread increases internal temperature which increases cell voltage spread, etc. This can lead to a premature failure cascade.

The point is the metrics that influence this threshold are not readily measurable until the pack has been worked hard enough to reach EDF running temperature. So Yes, there is a reason I make all these boring graphs. They allow us to study a packs actual "C" rating break point which may not be readily apparent by excessive temperature. In other words a sucky 0.200v/cell spread at EDF temp may not decay to a noticeable degree during your first season but chances are by the second you'll realize you've been screwed by substandard factory QC practices or cells (RC Lipo users don't rate iPhone level SPC standards).

All RC Lipo users can profit from this data, not just the EDF enthusiasts it was developed for. Why over pay for your packs?

Finally I offer the answer to the age old question: "Should I break in my Lipos?" Unless you're pulling high amps out the chute, No. If you are then put on a few easy cycles first. But otherwise I don't worry about it unless there's a question regarding the shelf life or they've been sitting for a few months. Then you should put a couple cycles on them before expecting their best performance. I do that at the beginning of every season before using my EDF packs. I don't worry about my others.


HERE'S THE "CLIFF NOTES" VERSION OF THE TEST RESULTS OVERVIEW

All Lipo's have batch to batch variations and no absolute assurances regarding OTS products. However the source has a track record that influences the odds concerning what you can expect.

I believe the preponderance of information below supports a thesis we can hedge our bets on: Thunder Power is unable to correct fundamental flaws in their assembly process, Revo has inconsistent quality, GensAce (Pulse, Tattu, Et al.) places little value on quality integrity, Roaring Top is a better made version of TP, Dinogy values consistent quality over performance, Hyperion is a mid level supplier masquerading as a premium brand and MaxAmps appears to be a border line criminal enterprise with sales chops. Hobby King has surprisingly emerged as the real innovator doing more to elevate RC lipo standards then any other entity. CNHL and Admiral have hitched their wagons to their success so can hopefully provide a balance to HK's potential dominance.


TEST RESULTS OVERVIEW (IF YOU DON'T WANT TO WADE THROUGH THE NUMBERS)

Lipo Suppliers have a "culture" that effects what we can expect from them. There are variations, better or worse batches, etc. but in some way they bear the corporate "stamp" that effects our odds of getting good product. Enough years have transpired to "sample" a stream of product over time from various suppliers so I've developed some opinions I believe are supported by the test results overall.

THUNDER POWER trades on their past glory and in large part is responsible for the spread of bogus C ratings. At one time they labeled their 6S/5000 for Continuous 325A output even though their split pack connector always (everytime) de-soldered at half that. I assume they rely on "Ignorance is Bliss 20C users" who believe TP is the "best" and are willing to pay the big bucks for the illusion.

5 years of testing has firmly established Thunder Power as a lower tier supplier, usually ranking in the bottom third in all test metrics. Their consistent failure to match the voltage between the 3rd and 4th cells on their 6S split assembly bears some responsibility (Ref. Reports 376, 417, 585, 653, 3212, 3520, 5404 and 5405). The "80C" Rampage tested well at 25C based on reasonable life expectancy. It will marginally handle 35C applications.
Competitively this pack should sell for $140 at most.

GENSACE PULSE Glacier Flourene Tattu, etc. Because it's their "premier" product the Pulse Ultra 65C was tested to represent GensAce. Historically their "C" rating labels are fiction. My quess is they fill the C label bags with whatever happens to be coming off the line. For example their
early run 5500mah/25C's were the finest 35C performers money could buy. A subsequent release of ~5000/60C units bearly tested at 20C's. In keeping with this legacy the respected and costly Pulse "65C" tested as a weak 25C unit and suffered from terrible cell Matching QC. At best the unit I received is adaquate for 20C applications and should command no more then $85 competitively. It failed the 35C run due to excess temperature (190F/88C), one of only two 60C+ labeled packs to do so.

Speaking of which: MAXAMPS, The most expensive 6S 5450mah on the planet was by far the worst pack I've ever tested. It failed in every metric so badly I actually feel sorry for the company more then the customers they should be issuing recall notices to. This "True 120C" unit is not even a decent 10C product. Originally I purchased the 5000mah "True 100C" version at a cost higher then TP's Rampage. It tested so bad it almost popped before completing a standard 25C run. Figuring it was simply defective I returned it for the recommended 5450mah "True 120C" version.

Here's where it gets weird because I've always riled against over priced packs: I LOVE THESE GUYS! I received some of the best support I've ever encountered, on par with Dinogy's MarkF. If the $250 5450mah would have compared favorably with his $140 70C unit I would have considered buying more even at their horrendously inflated price. No amount of service or salesmanship could have built a successful company using this caliber of cells, right? They must have received a very bad shipment and simply need to overhaul their inhouse QC procedures. Otherwise theirs is the worst kind of scam. Competitively both units tested are not worth $50 each. It's the only one I've ever filed a Paypal refund claim against because they didn't send what I ordered.

REVOLECTRIX used to be a favorite of mine. The first to offer Graphene "GO" packs (which I initially loved). As time passed the frequency of badly QC'd packs appeared to increase. Other problems arose. The covering is too brittle by far and I'm sure many remember the solid wire non-flexible leads? The HV "435 Blend" can perform well in general however a 4400mah pack rebadged as 5000mah (due to 4.35v charging) does not do well in this weight class. It literally ran out of gas towards the end of the 35C runs. Poor Cell QC contributed to a 30C capability (at best). However the Cell Spread increased noticeably from the beginning to the end of the testing probably because of it being under sized. On the other hand the 5000mah "420 Blend" tested as being capable of 35C use. But even it's cell matching spread increased under the higher loads which does not happen with the best Graphenes.

ROARING TOP has become a definite player with generally decent quality. Their 80C pack is solid at 25C but struggled at 35C. The Cell QC could be better but appears superior to TP's Rampage which is probably the reason it developed higher amps under similar loads.

TURNIGY HD was included to keep the heavy carbon packs honest. Turnigy "Blues" have developed a reputation for delivering one DOA for every ~5 units shipped. Also the last time I flew a set of the HD's there was a definite reduction in the motor RPM over my other brands. However this Turnigy Blue HD was a star. So good it was the only non-Heavy Graphene to make the 45C test pool before it maxed out. It's weight is the same as the Roaring Top and only 25g over the Hyperion G7. Not only does it cost considerably less then both it cleaned their clocks at every performance turn! In fact it was so uncharacteristically good I strongly suspect HobbyKing's need to ride a tight herd on their lipo supplier to maintain Graphene QC and performance is trickling throughout their lipo product line.

Hyperion G7 has never impressed me, mainly because of it's cost vs. performance relationship. Plus the "silicone" (whatever that is) runs hotter then the norm. However if the price was on par with the HK Bolt and Revo 435 HV's it would be my choice for 25C applications. However it
does not handle 35C well and also shares the "HV" labels propensity to run out of gas during the higher amp runs.

For me the HK Bolt HV is the bridge between the Heavy Carbons and the light weight Dinogy V2. It's an inexpensive solid performer that testing over time has increased my respect for. It's a great 35C capable pack even if is suffers a bit from the "HV" label tendancy to short change MAH. It's only competitive draw back is it's relatively high weight. But it is now my choice over everything not heavy carbon or V2.

DINOGY V2 is my "go to" pack when weight is an issue. It's not long term 35C capable but it's a solid 30C unit that tolerates overcharging well. Plus, aside from some mislabeled 8000mah FPV packs I've have never recieved a bad Dinogy product, ever. I can't say the same for Revo, Thunder Power, Turnigy or Pulse/GensAce even though I've literally purchased 10 Dinogys for every one of the others. So in my hanger if it's not heavy carbon it's V2.

XPS 50 offers a very well built 25C capable pack at the lowest price available in it's class ($85). It would be my choice for 20C applications. It is not suitable for 30C-35C requirements but then neither is the Pulse Ultra at twice the price.

HK GRAPHENE Never in my wildest fantasies 5 years ago would I have thought today I would be singing Hobby Kings praises. And I am. Their Graphene line is the best commercially available RC lipo in the world, Period. They pulled a coup and raised the performance bar to a standard their competitors can only hope to match. Because Hobby King laid the supplier "pipeline" they now have competitors in this space. However I don't expect any will copy the "Rhino Hide" HK utilizes for pack covering. I can now prove that not only do the HK Graphene's deliver on a life sustainable 50C the skin can handle it, something not true for the others.

The track line of cell voltage readings (the spread) look like a rope. Any better they would appear as a single line instead of 6 separate ones. Simply Perfection. To avoid sounding completely like a freaking Fan Boy I have to point out all this performance comes at a weight penalty (So there )

CNHL G+ performed as good in every way and in every test as the HK Graphene except for the heavy duty skin. The conventional shrink wrap covering accumulated some heat damage around the power lead entry points at elevated pack temperatures. This lack of "Rhino Skin" is probably why the CNHL is 10g lighter then the HK Graphene. I'll wager half my net worth they both came out of the same factory.

ADMIRAL CARBON is as good in every way and in every test as the CNHL G+ and HK Graphene. I skinned these myself so I can't predict what covering MotionRC intends to use but I'm confident it will be something great. My wager still stands regarding the factory source for these packs.


HK PANTHER GRAPHENE My initial tests indicated these wouldn't offer much more then the first generation of Heavy Carbon pioneered by Hobby King. By and large that is true. There are minor improvements which are evolutionary whereas the original HK Graphenes were revolutionary. However that doesn't negate the fact that while the others "grazed" the 300A mark the Panther banged solidly off the scale. Also none of the others displayed the voltage authority and certainly didn't run as cool as the Panther did in the 55C test. So I can say with confidence the HK Panther is a true 55C pack. The others are 50C capable. So the only question is whether using arguably the most powerful pack in the world is worth the additional one ounce (28g) of weight?


(PROOF CHARTS FOR 2018 TEST SUMMARY'S)
(PROOF CHARTS FOR 2019 TEST SUMMARY'S)
Last edited by MCSGUY; Today at 04:07 PM.
Sign up now
to remove ads between posts
Nov 09, 2012, 03:38 PM
Space Coast USA
hoppy's Avatar
Interesting the comparison between C ratings - The 25C rated packs are right up there with the 60C rated packs.

What is "reserve voltage? How is it calculated?
Nov 09, 2012, 04:02 PM
Dude, I do fly all day long!
rcalldaylong's Avatar
nice setup! on the gens ace 25c vs 60c...If I remember correctly I think everydayflyer's test echoed the same thing.

it's also interesting to see how well the nanos held up though.

I recently bought a couple of Haiyins and they do hold up fairly well....
it looks like the Haiyins are right up there and even a tad ahead of the comparable 25C gens ace.
Nov 09, 2012, 04:46 PM
aka JetMan Joe
MCSGUY's Avatar
Thread OP

Thread index


Page 6:
Data acquisition hardware
Velox w/ Schubeler DS94-HST as 12S Test Rig

Page 25:
12S Test Rig Setup
Thrust Sensor Hysterisis & Dead Weight Validation Graphs

Page 26:
12S Test Series Graphs
Product Review and Pricing Comparison

Page 28:
Thunder Power's new G8 Test Results

Page 34:
Why Measured IR is Useful, courtesy of JJ604

Page 39:
The Damage Caused by Trusting Thunder Power

Page 42:
6S EDF Based Test Rig

Page 44:
6S Test Results

Page 53:
10AWG Heat Calculation, courtesy of WAYNE GILES

Page 56:
7S Test Results

Page 61:
40C 12S Comparisons

Page 63:
Why Cell Matching is Critical, courtesy of WAYNE GILES

Page 65:
30C Comparisons

Page 66
Compare Revo SilverHV 4.27v vs. 4.20v

Page 72
https://www.rcgroups.com/forums/show... tested at 40C

Page 73
China Hobby 50C packs tested at 30C

Page 73
Dynamic EDF Based Test Rig

Page 74
Brand Evaluation Synopsis

Page 75
20C Test Comparisons

Page 78
40C Test Comparisons

Page 79
Test Comparisons

Page 84
Dynamic (EDF) Thrust & Efflux Comparisons

Pages 78, 85 & 119
Chinese Mfg Facts 1, courtesy of RevoJohn
Chinese Mfg Facts 2, courtesy of RevoJohn
Chinese Mfg Facts 3, courtesy of RevoJohn

Page 109
Life Cycle Test System

Page 111
Glacier, GensAce ,Pulse - Rebadge?

Page 119
Brand Evaluation Synopsis

Page 179
A Lipo Behavior Thesis, courtesy of JJ604

Page 182
2015 Comparison Roaring Top vs. Revolectrix

Page 211
IR Readings vs. Break In Cycle Comparison

Page 215
8000mah Pack Tests

Page 233
Retest of Failed NeuEnergy 8300

Page 235
Update on 8000mah Test Series

Page 243
Revo HV, Pulse Ultra and Hyperion G6 Comparison

Page 243
5000-6000mah Pack ESR Measurements

Page 245
Standardized Procedure for Measuring ESR

Page 257
5000-6000mah Tests

Page 277
12S EDF Based Pack Tests

Page 335
Performance Test, Courtesy of SockRat

Page 352
Performance Tests, Courtesy of FeloPaul

Page 361
6S/5000 Tests Summary
4.20v Charts
4.35v Charts

Page 367
How Graphenes Changed the IR Landscape

Page 376
Over Charging Your Packs? Read This

Page 418
12S vs. 13S Comparison in "12S" EDF

Page 420
Evaluation: HK 10K 15C Pro's as Saddle Packs?

Page 455
HYPERION G8's vs. DINOGY Tests

Page 469
IR Essentials Courtesy of JJ604
Last edited by MCSGUY; Today at 02:03 PM.
Nov 09, 2012, 10:27 PM
Space Coast USA
hoppy's Avatar
Quote:
Originally Posted by MCSGUY
The lipo voltage checker shows the value as a percentage of full charge. I believe an average of 4.20 volts per cell would be 100% and 3.20 volts per cell would be 0%
So that was the voltage of the pack under load? 5% reserve would mean the underload voltage was about 3.6V. Did you record the voltage under load or just the % from the cell checker?

BTW, thanks for posting this data, it's very informative.
Nov 09, 2012, 10:31 PM
aka JetMan Joe
MCSGUY's Avatar
Thread OP

Proof charts for 2019 test summary's


Summary's Located on Page 1
Last edited by MCSGUY; Today at 04:02 PM.
Nov 10, 2012, 12:10 PM
Shut up and fly
papabatman's Avatar
very interesting
Nov 11, 2012, 05:21 AM
Registered User
AaahhhhhhhHHH... now THIS is what we want to see manufacturers SHOW us... Not some hyped thumb sucked figure !!!!
Nov 11, 2012, 08:00 AM
Registered User
For under $50 you can build a constant current discharger: https://www.rcgroups.com/forums/show....php?t=1107072
Then you can directly compare your results with other peoples' results.
Nov 11, 2012, 02:22 PM
Registered User
It is a very correct aproach
The big load comparsion is giving the real imfo about the battery. not the stickers on it!
I have posted before the discharged curves at big loads.



It is 4000 80C made 2011 weight 127 gr/cell full discharged 200 Amps



It is 5000 65C made 2012 weight 133 gr/cell full discharged 275 Amps
You can see the difference betwin the original and the copy



It is 6600 65C made Oct 2012 weight 133 gr/cell.
Nov 11, 2012, 04:43 PM
aka JetMan Joe
MCSGUY's Avatar
Thread OP
Thank you Ivaylo, you have provided very good information, I appreciate it.
Dec 26, 2012, 09:19 PM
Сделайте Америку Великой Снова
7oneWo1f's Avatar
Awesome thread, looking forward to the brushless or pulsed inductive load testing, just to learn what sort of differences there might be.

Did you use the same resistance for all of the tests, or did you pick different taps for each battery?
Dec 26, 2012, 10:12 PM
They Call him Dead!
YellowJacketsRC's Avatar
Yeah, I'm in!

Thanks!
Dec 27, 2012, 01:28 AM
Registered User
Quote:
Originally Posted by LoneWolfRC
Awesome thread, looking forward to the brushless or pulsed inductive load testing, just to learn what sort of differences there might be.

Did you use the same resistance for all of the tests, or did you pick different taps for each battery?
No the load is not the same. It can not be the same if the IR of the Cell are not equal. But if it is programed, could be the same.
The tests are with purpously designated devices Load tester with heat extraction and calculas


The devices allow to test the battery performance with algoritm and/or governor controlling the battery figures and change the resistance as per load function programmed.

One example for such a test. (It is a racing battery for Boat 6800 30C, developed over 6600 65C)




On some of battery the loads aplied are bigger then the C-rate to test the strenght of the tabs, or analyse the electrodes after the test.
I have posted the curves for 5000 25C discharged with 150 amps (30C)
https://www.rcgroups.com/forums/show....php?t=1786416
One more good project 8000 25C (but in weight limits)


Quick Reply
Message:

Thread Tools