Danny/SMC

iTrader: (2)

How much does a Carbon Pile tester cost ? Can it be set at any amp load and can you use 7.4V ? Will it stop the discharge at 6v and will it give you the capacity ? This is what is needed to do C rate testing. If it does and is cheaper than 2K I will have to look into this.

SPC

One thing I've not heard mentioned; what if a battery company does not like the results of the test/shootout? I know for a fact a company would have the right to seek legal retribution for any slanderous information that is published or released. And before you reply and say it is not slanderous if it is the truth, think again. In cases like this the person who published the info has to prove the truth (innocence) and will be assumed guilty until they do so. Even if they win, so what.... it will be very difficult or next to impossible to recoup costs from the battery company that lost. In other words, the battery company that feels they have been slandered has nothing to lose by taking legal action. I know this because I sought legal council after I was recently legal threatened by one of the larger LiPo battery RC companies when they accused me of publishing test results that did not make their LiPo packs look favorable. So if you decide to test battery packs be careful what you do with the results.

Tom Himes
Engineering Manager
SPC Racing

CamLight

Carbon-pile testers are very inexpensive and can be used as a load to discharge cell/packs but there are some things to consider...

- It's a fixed-resistance load and the discharge current level will drop as the pack voltage drops. This isn't a problem on its own but it's not a constant-current or constant-power discharge which are the standards for testing. For a lot of the tests, your results will not be able to be compared to the test results from anyone else.

- The unit-to-unit resistance of these testers varies. This will prevent comparison of results between people using the same model tester unless each unit is calibrated/tested.

- They're incredibly hard to adjust to a particular current value.

- The resistance varies as the load heats up. This isn't a problem on its own but must be taken into account.

- They're designed for short-term testing and will fry if used anywhere near their rated power/current levels for very long.

I'm not sure how much credibility the industry and customers will give to tests done with a carbon pile tester. But, for in-house testing/comparison, that's not a problem. All of the above doesn't prevent a carbon-pile tester from being used for battery pack discharging and testing with a logger. It's just that there are a few things to consider if you do so.

John M.
Owner
CamLight Systems

Francis M.

iTrader: (2)

I would even settle for Xtreme to just do GFX #'s comparison.....

the usual #'s runtime, avg voltage, and IR...

The only problem with GFX is the variance from machine to machine. With one source no matter what the actual #'s maybe it will give you an indication of who has the best packs out there....

Same thing happened to nitro.. None and I mean none of the engine manufacturer's claimed HP was reached but It still showed who had the most power...

SWTour

iTrader: (9)

I kind of agree with Francis.

I'm not so much interested in the OUTPUT numbers from a particular brand, but moreso interested in seeing the comparitive data from one type to another on the same test equipment under the same test conditions.

We all know SMC's and probably now Thunder Power will show the best performance, but seeing the numbers from a dozen or so of the different brands under simulated race conditions may show how much, or how little the difference is in usable terms.

For the testing you guys want to do, a Carbon Pile tester would NOT be the ticket. NOT because it's not heavy duty enough, it's TOO heavy duty to give fine details for this small of a capacity battery.

Testing Automotive type batteries you are looking for bad or weak cells at fairly high amp loads, generally it becomes quite obvious if a cell is BAD, but weak cells sometimes are a little trickier (they don't always show up on the first couple tests)

In general a Automotive Carbon Pile tester just has a DIAL and a digital display, YOU crank the knob until you get the desired amp load, read the on screen voltage and watch what it does during a 15 second load (as the MACHINE starts smoking and getting really warm)

This is also used to load a charging system on a car to see the output ability of the Alternator/Generator.

The issue of using a "CONSTANT" Amp Load (self correcting) I never understood when testing batteries, but there's a lot of things in this industry I don't understand why or how they became the "STANDARDS"

When you are racing, you're voltage drops, and as your voltage drops, so does your amp load, which is how I prefer to test stuff...

The Numbers on labels may be TRUE and CORRECT information, but IMHO too often the numbers aren't relevant to useful information.

CamLight

Constant-current testing is only one of the "standards" because it makes calculating a cell's capacity very easy (current x run-time = capacity). If the current was constantly changing (as when using a constant-resistance load), the value of the current would have to be measured constantly. The capacity would then have to be calculated for the period of time since the last sampling and added to the total. A new current reading is then taken and the above is repeated until the cell drops to the cutoff voltage. That's a lot harder to do than when using a constant-current load. And then there's the debate over how often to measure the current level...every second? Every millisecond? The total capacity rating becomes more accurate the more often the current readings are taken but how often the readings must be taken is, unfortunately like so much in this field, debatable.

A constant-current load makes it much easier to determine the capacity of a cell (under those conditions) and for manufacturers to rate their cells when being sold. It also make it easy for others to test cells against their ratings. Also, a lot of constant-resistance loads vary their resistance depending on their temperature. So the load will have one value for a 2S pack test and another for a 4S test using the same cells, which will be a much higher power test....not so good. And different types of resistors have hugely varying responses to varying temperature....also not good for comparing results between loads. A constant-current load keeps its value steady throughout the test no matter what the power and pack voltage levels are, making it easier to compare. Assuming that it's a decent quality constant-current loads, that is. But, that's true for any test equipment.

But you're right, racing uses a load closer to a constant-resistance load.

DerekB

I'm not worried about legal action, and I'm sure whoever threaten with a law suit was just a threat. There are countless sites and consumer report style magazines and such that do independent testing.

My honest opinion now is that there isn't a company out there that has something to judge and print out usable data. Most will use what the battery provider gives them and I can sort of understand that. It was similar to HP on engines where people would just market what they are told. There are still companies that question how a dyno works, but when I ask them if they understand how HP is calculated they dont' have an answer.

I'm willing to work with anybody, and we have a good group here to come up with a repeatable test that gives data. I don't like using something that doesn't print data to a graph. I know we had some software that was used on a PC that graph data....but only went to 10amps. Maybe I can look up the test we did on that and see if they offer more.

I'm not really sure we need to test at 150 amps (hell NIMH was only really at 35 and that was the high side) It's not really my job to rank cells. Like in NIMH they changed so quickly and had different characteristics every week it was hard to tell what the better cell was.

Like I mentioned, I'm not sure that the better cell is the one that delivers the most power but doesn't last 3 charges.

I'm in the middle of deadline with this issue, but I'll email Danny and look into what we can do to set up a standard.

DerekB

Much of testing is just being able to repeat the process. So I think just doing a constant amp draw is the easiest way to have it happen as close to the same every time. Standards come from somebody that either took the time or did something first. Novak tried to label their batteries different without including the first second of discharge. What that did was lower the #s on the cells. While it might have been more accurate the normal guy looking at #s probably thought they were lower performance. So even when somebody tries to do something more realistic...the market might not like that way.

We always want to feel we bought the fastest/best. Like when people complained about engine HP, when we came out with #s that were half of what was being told....people started to say WE were wrong and the #'s can't be that low. But in the end they are just #s.

CamLight

The West Mountain Radio CBA II has graphing built into their software and the T35-GFX can upload data for graphing in Excel so I think that can be taken care of pretty easily. Exceeding 10A is easy too with the CBA's ability to use their own "Amplifier" or a third-party electronic load to increase the power and current handling abilities. The T35-GFX's power/current handling can also be extended with third-party loads.

Setting up the standard will be the real "fun" part.
Constant-current or bursts? Testing only new packs or only after 10, 50, or 100 cycles (or more?), or both? One current level for all or multiples (1C, 5C, 10C, etc.)? Also include "real world" load profiles that emulate actual conditions (similar to the GFX's profiles) a pack might see?

The list goes on practically forever.

DerekB

The software and data is the hard part. The standard is easy, I'd base it off what we've all used in the past.

35amp constant load. We can daydream about fancy race profiles but the truth is it don't matter. What one person races on doesn't mean jack for another one. If we overly complicate it becomes too hard to do and too long of a process. The old Keep It Simple Stupid theory applies here. If we really just want to test labeling that's fine. Any other testing really comes down to testing quality and race characteristics...which in some cases don't apply universally.

Also my concern is that LIPO's don't like to be dropped below their set voltage, doesn't that mean you can't discharge them fully to test capacity? I know I'm sounding dumb but I was never sure how that capacity was rated. I sort of assumed it was capacity to the lowest voltage...even though the battery will provide more?

Jube

iTrader: (43)

So stop talkin and start testin already!

BatteryWarLord

I was looking into the WestMountian setup for testing the batteries. I am curious what SMC is going to be using.

The WestMountian is rated for 1000watts when using the expanded setup with amplifiers. That should be able to test 7.4V batteries over 100A.

Pulse discharging would require new software and maybe new equipment. The GFX has a laid out approach to pulse discharging and that would be a good standard to expand for any pulse discharge interval. This pulse discharge method of either using the GFX or custom discharger would help see which batteries have better recovery under dynamic loads. I added some suggestions to how change the settings to better reflect extreme testing.

GFX OVAL Setting
45 amps for 1.25 seconds (Change Test to: 50, 75, 100A)
0 amps for 1.25 seconds
3.6 amps for 1.25 seconds (Change Test to 10, 15, 20A)
Repeat Cycle

I feel C rating is really bogus marketing, the testing should be done under steady amp discharge not the C rate. 30A and 35A is a little low for todays batteries and mod motors. The steady rate standard should be increased to 50A, with 75 and 100A rates possible. The testing standards should only be lowered if the battery is not rated for the 50A dischrage.

DerekB

I'll look into that system. 50A 100A constant? what the hell are we driving these days? I don't think we get near that on a 4-5 minute continuous. I don't think destroying a battery really is a good idea.

CamLight

Danny's system is capable of 2000W at not much more than the cost of doing 1000W with the CBA Amplifiers. And Danny's system will be able to do burst profiling at two independent current levels too (or use the GFX's profiles)...something the CBA system can only do full on/off.

His loads are compatible with the CBA and the GFX (expanding the discharge power/current ratings of either) and are expandable to the multi-kilowatt level (important for testing higher-voltage packs being used flying, etc.).

I'll let him give you any additional info as he's the one using the system.

BatteryWarLord

You are only think in the relm of R/C racing not extreme robots or full scale land speed cars or electric scooters. What good is a big beefy motor with wimpy batteries?

A 50A constant discharge would be about the limits of what sedans need. What about 1/8 off-road using 10,000+ Mah packs? The equipment needs to handle larger amploads or the only information that can be gained is for sedan/oval use.

I'll be emailing Danny of SMC on his setup. I have not committed myself to any equipment so I want to learn a little more about what can be done.