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

[EEVblog]

A viewer sent in the real thing!

[pickle9000]

Awesome monkey!!!

[ccs46]

A viewer sent in the real thing!

So the question begs Dave... are you now "A monkey's uncle"? 

[FrankBuss]

What I understand is that "Volts x Time" is meaningless without understanding the current draw. So it really is "Amps x Time" that is important to rate capacity? MilliAmpHours? Is that what is going on here? And he is saying that he can just use the same 100 mA draw all along the entire voltage discharge curve (whether boosting or limiting) and expect the curve to follow the same curve?

Using the area under the current line would make sense only if the device uses constant current, and even their test with the Garmin device shows rising current, because probably the Garmin device has an internal boost converter and needs constant power, not constant current:

So for most devices it is wrong to use the constant current chart to estimate battery runtime, but you should use the constant power chart, which you can find in most battery datasheets, too.

They are trying to make the video look scientific, but they are just distracting from the main questions: they don't say anything about the batteriser performance. Just a simple test would be sufficient: How long runs the monkey with normal batteries and with the batteriser? I guess with the batteriser it would run shorter, if it runs at all because of the high current draw, and of course they can't show this.

[edy]

It does look scientific. They almost had me convinced!   

The most simple summary explanation I can come up with here is that the Batteriser smooths out current draw from the battery, limiting "spikes" or "glitches" as they call them, which in turn results in a SMOOTHER voltage discharge curve whose variations take longer to eventually get close to dipping below the cutoff voltage, thus saving the device from just "powering off suddenly".

But what happens when a device needs a certain current to function? They are basically saying the Batteriser acts like a "power supply" voltage source in that it keeps a constant minimal voltage while somehow being able to supply whatever variation in current is demanded by the product without any dip in voltage.  Assuming it could magically isolate the device and battery so that current draws don't affect battery voltage, where is the extra "energy" coming from?

The Batteriser is tiny, where is it storing a current source that it can release at will whenever the device demands it? So that it protects the battery from seeing that current draw and shields it from having a voltage drop? Or am I over-thinking this? Is it just a matter of boosting the voltage which in turn still results in the same variations in voltage as the current draw varies, but just higher (shifted up) so it doesn't hit a cutoff voltage? Why can I not understand this? 

[Bud]

[Jay_Diddy_B]

Hi,

I have a built a circuit that should be fairly representative of the performance of the Batteriser.

I have posted pictures of the board in this thread:

https://www.eevblog.com/forum/projects/new-project-a-batterruser/msg739648/#msg739648


At Vin =1.1, Vout =1.5V, Iout=250mA I measured 81% efficiency.

Regards,

Jay_Diddy_B

[EEVblog]

But what happens when a device needs a certain current to function? They are basically saying the Batteriser acts like a "power supply" voltage source in that it keeps a constant minimal voltage while somehow being able to supply whatever variation in current is demanded by the product without any dip in voltage.  Assuming it could magically isolate the device and battery so that current draws don't affect battery voltage, where is the extra "energy" coming from?
The Batteriser is tiny, where is it storing a current source that it can release at will whenever the device demands it?

Bingo.
You canna change the laws of physics captain.

[joseph.anand]

I think we need to summarize the issues with the Batteriser, and just let it die a natural death. I am attempting to summarize them here. Others can just add to this and we can be done with it.
1. With most decent devices already having boost regulators adding another one in the equation is just a waste of energy.
2. With devices using multiple cells, each cell having a boost regulator is a bad idea. This is why most devices have a single boost regulator for batteries connected in series (or parallel - yet to see devices using batteries in parallel).
3. Most devices use battery voltage to detect the battery level. Now using a batteriser you will have no idea of the actual battery level.

Now the issues with the technicality of their claims:
1. For all their "area under constant current curve nonsense", one needs to estimate the actual energy (in %) left. For a constant current load, you can instead estimate the area under the voltage curve. Simple basic maths W=VI. For those "PhD Professors who are consultants to Batteriser" this is Power= Volts x Current. Power = Energy/Time. 


Other readers can add to this, and bring this to a close. If you can put a point more succinctly than what I have, kindly replace yours with mine. This way  we can actually debunk this as and when the product releases and then we can all with our smug faces gift the Batteriser team with "We told you so" t-shirt.

[jancumps]

I'd like to add:

'There's bugger-all energy left'

[Halcyon]

Clapping cymbals monkey - A new design for an EEVblog shirt perhaps?

[daqq]

Their campaign was "successful" so the product will be out.

If they don't have any issues with production.

Or reality or the laws of physics... at the end of the day you'll get a bunch of crappy products that are useful for 20 ppm of products available, make good on their claims only with the right interpretation of weaselspeak.

I'm probably being naive, but it seems a wonderful object lesson in listening to engineers and peer review for the people who bought it, expecting their magic monkeys to last 8 times longer on just as crappy batteries. On the other hand, human stupidity is a VERY deep well...

[Fungus]

Other readers can add to this, and bring this to a close. If you can put a point more succinctly than what I have, kindly replace yours with mine. This way  we can actually debunk this as and when the product releases and then we can all with our smug faces gift the Batteriser team with "We told you so" t-shirt.

The biggest issue will be that it simply doesn't work as claimed. If it did, they would have shoved a couple up the monkey's butt and showed it coming back to life.

They didn't do that because they can't. This whole product is based on a lie. There's no need for maths and equations if it fails the basic sniff test. Everybody going into nerd-speak is exactly what they want from this video - the nerds aren't their target market.

Keep the argument against them simple. They still haven't actually shown a batteriser working in one of their videos (apart from the Apple keyboard, but they didn't show extended life, only the battery indicator level going up for a few seconds).

[AmmoJammo]

https://youtu.be/95AAOZGjuIc

[Galenbo]

Do they mention or show the pulse response of the Batterieser?
Do they mention or show the efficiency response curve of the Battersier?
I fully expected this "be-all end-all" technical videos for "real engineers" to include at least that stuff.

They show some story about a snail in a pit and some numbers.
He clearly does not target engineers and other (useful) higher educated people.

He keeps on targeting his customer base, like other salesmen do.

[Galenbo]

My main problem with Bob and his brother (or even the Chris one) is that they seems to really believe what they are saying...

There are courses for that. Called Law/Sales/Banker/Commercial engineer/Marketing/Poker/Management.

[AmmoJammo]

https://youtu.be/95AAOZGjuIc

Umm... I made a few more videos with this setup...

Firstly, testing without a boost circuit... the boost circuit I ended up using is found here:
http://www.ebay.com.au/itm/201397293506

Modified to output 3.2volts, it happily did 500ma, with down to 1.8volts input (two cells in series) ( I didn't test it beyond this current), I'd measured it to do between 60 and 90% efficiency or so, depending on the input voltage, and output current.

Without the boost circuit, the camera, with forced flash, took 90 photos...

But now, guess how many it too with the boost circuit, using brand new cells?

(I'll upload the first video now, its not very interesting)

[lpickup]

The biggest issue will be that it simply doesn't work as claimed.

Well it does, but you have to decode the marketing B.S. from their claims.

What they are saying is that the Batteriser can tap into the 80% of energy (or voltage using their parlance) still left in alkaline batteries that are thrown away.  Sounds great until you read this paper that they threw at us:

http://www2.ife.ee.ethz.ch/~rolfz/batak/ICBR2003_Zinniker.pdf

that does actually claim that SOME batteries contain over 80% of usuable energy left (and actually using accepted scientific principles)

of course they fail to mention the conclusions reached in this paper about WHY practically new batteries are thrown out which have nothing to do with monkeys failing to clap or GPS units screens' dimming.

So technically speaking, using the Batteriser can help you tap into this 80% of remaining energy.  Hell, not sure why they said 80%, the Zinniker paper found that 10% of the batteries in the study had 103% of energy left in them, which by Batteriser standards comprises a "significant" number (you'll have to read the paper for an explanation of why it's >100%)

Of course not doing anything other than taking that recycled battery and sticking it into a TV remote or even a clapping monkey can also help you tap into this 80% of remaining energy.  But that's not the point.  Batteroo is not obligated to tell you that.

Forget the talk about miniaturization and boost converter efficiency and the size of their inductor and whether 0.1mm clearance is enough...they could build a simple battery sleeve and nothing more and it would still meet their claim on the Indiegogo page, about 20% of the time if you get your batteries out of recycle bins in Switzerland from the year 2002.

Hold the presses!

I just went to Rolf Zinniker's site and while I don't speak German (so I have no idea what this is referring to), he's got this image on his site:



WTF is it with these battery guys and monkeys?  Seriously!  Dave, are you SURE that the proper way to test a battery is NOT with a clapping monkey?

[firewalker]

The above power converter ha a huge inductor/capacitors compared to batteriser.

Alexander.

[AmmoJammo]

The above power converter ha a huge inductor/capacitors compared to batteriser.

Alexander.

Yes it does, impossibly big to fit on top of a aa cell, let alone a AAA...

So do you expect more photos? Less, or the same?

[lpickup]

Without the boost circuit, the camera, with forced flash, took 90 photos...

I hope you are using a camera from no later than 2001 to replicate the data the Batteroo people use to justify their claims.  Wouldn't want any of this modern battery management system rubbish, or lower power memory cards or anything like that.

[Kalvin]

So do you expect more photos? Less, or the same?

The boost-converter's efficiency is about 80% - 90%, which can be compensated by the fact that the boost converter can drain battery a bit more than the camera can. However, from the battery discharge curves we can see that as the current is increased the total energy available from the battery will be less. So, in this sense you will get less photos with the boost-converter unless the cameras's cut-off voltage is insanely high. I would guess that you got 75 photos instead of 90.

[AmmoJammo]

Well, with the boost converter, I got zero photos... well, technically one photo, but it never managed to save it, as the current spike from the capacitor charging up for the flash caused a dip in voltage, and the battery cut off to be prematurely triggered... 

Brilliant!

Of course, the butteriser might be able to do 2 amps without dropping below 1.1volts

[lpickup]

I hope you are using a camera from no later than 2001 to replicate the data the Batteroo people use to justify their claims. 

Jesting aside, I don't necessarily have a problem with using older hardware as the basis for this kind of test.  After all, people actually do have older cameras around that they still use, so a theoretical device that could extend the life of those products would in fact be useful.  But even by my standards (I'm cheap and keep stuff forever), 2001 is ancient, and more importantly this was pretty much at the beginning of the digital camera revolution.  Purpose built digital cameras only first came out in the very late 90's, and they were probably far more concerned with getting the image sensors and picture storage working, not power management.

It looks like the camera you are using is a probably very adequate for the purpose of this test.

[EEVblog]

Of course, the butteriser might be able to do 2 amps without dropping below 1.1volts

They briefly mentioned 1.5A in a video, but they have provided no evidence of that, no formal specs, no test data, no efficiency curves, no pulse response data, nothing.
We haven't even seen them use it in a high pulse drain device.
So we have absolutely nothing to go on.