EEVblog #751 - How To Debunk A Product (The Batteriser)

[madires]

Yep, a very simple test that can be done by average Joe/Jane. Note the devices and times, and we'll get a nice list everybody should be able to understand.

<device>                   <runtime of batteries>     <additional runtime with Batteroo>         <gain>
-----------------------------------------------------------------------------------------------------------------------------

[Jay_Diddy_B]

Hi,

I thought I would try an experiment and measure the VI curves for the LED flashlights (torches for UK) that I have in the house. Then I thought I would do a Google search and see if it has already been done. Well it has been done and done really well.

Here is an example:



This is just one of the results, picked at random.

Here is a link to page that explains the equipment used and the test method:

http://lygte-info.dk/info/HowDoITestADriver%20UK.html

And then here is a link to the drivers that have been tested:

http://lygte-info.dk/info/indexLedDrivers%20UK.html

Some examples:




At the other end of the scale we have a 'sophisticated' buck-boost circuit:





What's the point of all this? to understand the test results with an LED flashlight as the load, you have to understand the flashlight circuitry too.
These LED drivers will give vastly different results.

With the Linear circuit, the brightness will be held constant when the cell voltage falls, but when the cell voltage is the battery discharge rate will be higher because the boost circuit in the Batteroo is causing more power dissipation in the linear regulator.

With the buck-boost circuit, The LED current and therefore brightness is independent of battery voltage. The Batteroo will add another stage of conversion efficiency and the battery life will be shortened by 7-10% Which may be difficult to detect.

It would be more useful to know:

1) Quiescent current, that is the battery current with no load.

2) The efficiency versus load current. This could be obtained from just be battery current with a known resistive load.

Regards,

Jay_Diddy_B

[Fungus]

The target market isn't engineers. The target market is people who just want devices to last longer.

Batteriser promised to save the world's landfills from being filled with batteries.

[EEVblog]

Yep, a very simple test that can be done by average Joe/Jane. Note the devices and times, and we'll get a nice list everybody should be able to understand.
<device>                   <runtime of batteries>     <additional runtime with Batteroo>         <gain>
-----------------------------------------------------------------------------------------------------------------------------

I have a spreadsheet for that:
https://docs.google.com/spreadsheets/d/18K9c2YAT0d0QABGYGpzItbvDcgfAQCRUtloEzzfXADU/edit?usp=sharing

[digsys]

...  UP TO 8x ..

ok, UP TO

And Bob does have a device that lasts 8x longer - the Garmin GPS. This performance was verified by Underwriters Laboratory, a very trustworthy organization.
(The fact that UL didn't sign the report, you can't look it up on their web site and they placed conditions on usage doesn't matter...the report is real!) 

So, are you suggesting (or implying) that IF it lasts 1.2x longer, that it meets the claim? SO .. all he has to do, is find ONE non-verified test result, and the deal is done?
Well, I prefer real intelligent users run tests on various common products and verify it.
Sorry, but an 1.2X figure sure wouldn't fly in OZ. I would expect an average of 4X on various products, some higher, some lower -  then I'm sure everyone would be happy.

[madires]

I have a spreadsheet for that:

KISS Of course, the EE likes to have more data points, but this is for non-EEs. In case more IGG backers get their Batteroos, we should keep the test as simple as possible to give them some motivation to participate. We shouldn't ask for too much. If they understand the test, they'll help. They also want to verify if the sleeves are working as promised. As more devices are tested as better the statistical value becomes. When forum members get Batteroos, we'll get the "proper" data.

[EEVblog]

I have a spreadsheet for that:

KISS Of course, the EE likes to have more data points, but this is for non-EEs. In case more IGG backers get their Batteroos, we should keep the test as simple as possible to give them some motivation to participate. We shouldn't ask for too much. If they understand the test, they'll help. They also want to verify if the sleeves are working as promised. As more devices are tested as better the statistical value becomes. When forum members get Batteroos, we'll get the "proper" data.

How can my tests get any simpler?

[Fungus]

So, are you suggesting (or implying) that IF it lasts 1.2x longer, that it meets the claim?

Not necessarily.

There's laws that prevent (eg) car manufacturers from saying, "Can do up to 1000mpg".

The average battery life would have to be a lot closer to 8x than 1.2x.

Would consumer protection go after Batteroo for making wild claims? Doubtful.

[madires]

How can my tests get any simpler?

By performing just the fresh-battery-until-dead and dead-battery-with-batteroo tests, which require just one set of batteries.

[digsys]

It is a long time ago now but IIRC that 8x or 800% claim was regarding batteries people throw out with most capacity remaining. Again IIRC they claimed to have surveyed a range of batteries people discard and some still had most of their energy remaining. Yes they worded things a bit cheekily but I don't recall they claimed they would turn 1 battery into 8. Which is how that particular claim has morphed.

It may be so, that somewhere they made a cheeky claim, but did the common man-in-the-street IGG subscriber understand that? Clearly?
It sure has riled shltloads of engineers / techs :-) 2nd highest post topic? And the constant fubar comments since? .. last over and over .. 6 times ... ?? .. use only in space ..

[Fungus]

The actual claim was that 80% of our batteries goes unused, that Batteriser would tap into that unused 80%.

The "8x longer" was just bad math - it should have been "5x longer".

[Kalvin]

How can my tests get any simpler?

By performing just the fresh-battery-until-dead and dead-battery-with-batteroo tests, which require just one set of batteries.

Yes, I agree. Very simple to perform, the results do not need any interpretation, just hard facts.

1. Take a new set of fresh batteries.
2. Measure the time until the device stops working ie. the working time without the Batteriser.
3. Attach the Batterisers to the batteries.
4. Measure the time until the device stops working again ie. the extra time provided by the Batteriser.

[Blocco]

Their maths (or use of concise English) hasn't gotten any better since they confused 500% with 800%; their website now headlines with "tap into 80% more energy with Batteroo" which would only extend the battery life by 180%. While at the same time they indirectly suggest 500% with their original claim that only 20% of a batteries energy is normally used and Batteroo will tap into the remaining energy. Confused? Oh yes.

But of course, saying that Batteroo "lets you instantly tap into the 80% of energy that is usually thrown away." is not the same a saying it can extract all of it.

[Kalvin]

Their maths (or use of concise English) hasn't gotten any better since they confused 500% with 800%; their website now headlines with "tap into 80% more energy with Batteroo" which would only extend the battery life by 180%. While at the same time they indirectly suggest 500% with their original claim that only 20% of battery capacity is normally used and Batteroo will extract the remaining. Confused? Yep.

But of course, saying that Batteroo "lets you instantly tap into the 80% of energy that is usually thrown away." is not the same a saying it can extract all of it.

Claiming that only 20% of the battery capacity is normally used, and one can tap into the 80% of the energy that is usually thrown away cannot mean anything else than this:
- If a device runs for 1 hour with the fresh batteries and then stops working, there should be still energy left for 4 hours, which can be tapped into with the Batteriser.
- In other words, if a device would normally run 1 hour with fresh batteries, the device should run 5 hours with the Batteriser.
That is their claim.

[Blocco]

Their maths (or use of concise English) hasn't gotten any better since they confused 500% with 800%; their website now headlines with "tap into 80% more energy with Batteroo" which would only extend the battery life by 180%. While at the same time they indirectly suggest 500% with their original claim that only 20% of battery capacity is normally used and Batteroo will extract the remaining. Confused? Yep.

But of course, saying that Batteroo "lets you instantly tap into the 80% of energy that is usually thrown away." is not the same a saying it can extract all of it.

Claiming that only 20% of the battery capacity is normally used, and one can tap into the 80% of the energy that is usually thrown away cannot mean anything else than this:
- If a device runs for 1 hour with the fresh batteries and then stops working, there should be still energy left for 4 hours, which can be tapped into with the Batteriser.
- In other words, if a device would normally run 1 hour with fresh batteries, the device should run 5 hours with the Batteriser.
That is their claim.

With any battery I can tap into 100% of its stored energy every time I connect a load to it but that tells us nothing about how much of that energy I use or don't use or can't use. "Tap into" is not the same as "consume everything contained within".

[Kalvin]

Could someone explain me what Bob is trying to say in the video time 8:14 - 9:40:
https://youtu.be/622uCZ_pE0w?t=494
Yes, by applying a constant current of 100mA, we will get the red line displaying the battery voltage as a function of time. But then he draws the rectangular and says something that doesn't make any sense at all. That Bob's misunderstanding is the reason why he thinks that the Batteriser will work, and that is the reason why the Batteriser will fail. He even got it documented very nicely.

[IanB]

Could someone explain me what Bob is trying to say in the video time 8:14 - 9:40:
https://youtu.be/622uCZ_pE0w?t=494
Yes, by applying a constant current of 100mA, we will get the red line displaying the battery voltage as a function of time. But then he draws the rectangular and says something that doesn't make any sense at all. That Bob's misunderstanding is the reason why he thinks that the Batteriser will work, and that is the reason why the Batteriser will fail. He even got it documented very nicely.

What he said there was completely wrong and crazy. I won't waste my time watching even a second more of that video. He drew a line representing 100 milliamps on the voltage axis of his graph, for heaven's sake.

To explain why the area under the curve is the correct measurement, we just need to consider the fundamental equation:

  energy = voltage x current x time

If we integrate the area under the voltage vs time graph we get (voltage x time). Since the current is constant, we can just multiply by the current to get the total energy.

[IanB]

I have a clock that stops working when the battery voltage gets below about 1.25 V. I think this would be the perfect application for the Batteriser. A voltage sensitive device with a very low current draw and a high cut-off voltage.

Did they invest all that money to solve the problem of my clock?

(I solved the problem by using lithium batteries instead of alkaline. Then the clock runs for a decent time between battery changes.)

[6581]

I'm still trying to solve the snail riddle, but my snails keep starving to death before they reach the top! It'd help if it wasn't a vertical wall, but a much more gentle voltage curve, but it turns out my snails keep falling to the voltage-drop crevasses. I'm now back to using monkeys - they climb much faster and can jump. Downside is that their cymbals are noisy! This whole batteroo thing is making me lose my mind.

[drussell]

Again IIRC they claimed to have surveyed a range of batteries people discard and some still had most of their energy remaining.

No, they quoted someone else's reasearch into "discarded" cells without any regard for the reasons why those cells had been discarded or properly analyzing the data. 

[SL4P]

Testing in a motorised toy is only slightly better than the 'regulated' LED test... for a very limited test time.

Boob's kids may turn the toy on, and wait a couple of hours for it to run down, but more likely, they'll turn it off and on, stall it or cause the mechanism to stop for many reasons.
The point where Betterpoo is unable to restart the mechanism will be much sooner with the start current demand...

Let's just say we got something out of this whole debacle - a new word.
BATTERROOTED

BTW if Boob feels hard done by these pesky injuneer fellas, he should reflect back in time and reconsider what point may have been a good time to come clean on his mistaken claims.  Mea culpa.

Up to that point he could have been forgiven for moving forward based on incorrect calculations and advice, and retained a shred of his reputation, but now he's Batterooted it completely.

[PA0PBZ]

I have a clock that stops working when the battery voltage gets below about 1.25 V.

Did you measure that while the battery was still inside the clock? 

[drussell]

So, are you suggesting (or implying) that IF it lasts 1.2x longer, that it meets the claim?

LOL...  Well, that kind of deceptive claim is used all the time these days in all sorts of products across many diverse industries and it is rarely called out.  You can even rig up a test that will show some ridiculous number of times longer battery life (like the GPS test for exmple) and then claim that you were correct all along.

The simple fact of the matter is that with most products, unless the Batteroo Sleeve is delivering significantly less power to the load than a plain cell would deliver, (and this may be the case if the sleeve cannot deliver the current attempting to be drawn by the device!) over the operating time the sleeve will actually DECREASE the total battery life, not increase it at all, despite the fact that you can show a gain in certain situations.

If a flashlight draws more current than the Batteroo Sleeve can muster (so it is now operating as a current limiter) and thus the actual operating power of the lamps is less, it is possible for it to extend the life, though the luminosity would be less, which is opposite to their claims of providing maximum performance for a longer time.  You cannot "have your cake and eat it too" in this situation.  Either it ends up running the flashlight for a longer time at lower brightness (where you could have bought a "dimmer torch" in the first place; you're not the getting the full output that you expect from your device) than without the sleeve or it is going to run it for a shorter period of time. 

Especially if the flashlight already has the same kind of driver circuit in it...  Which is what many people don't understand... The devices where the designer and manufacturer decided it was worth it to put in a converter, they did.  Are there some devices whose performance characteristics will be altered by the addition of that circuitry?  Certainly.  Will it deliver anywhere near the claimed increase in runtime at full performance that they state, for the vast majority of products?  Not a chance.

Arrrrrr...  The laws of physics be a harsh mistress, Matey! 

[Fungus]

I have a clock that stops working when the battery voltage gets below about 1.25 V. I think this would be the perfect application for the Batteriser. A voltage sensitive device with a very low current draw and a high cut-off voltage.

Sure.... unless the quiescent current of a Batteriser would drain the battery first.

[IanB]

Sure.... unless the quiescent current of a Batteriser would drain the battery first.

That would suck