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

[Kalvin]

It would be just great to be able to measure how much more energy they can get more from a battery with a Batteriser: Take a fresh set of batteries and discharge the batteries with constant power until the battery voltage drops to specific cut-off voltage, for example 1V. Now, attach the Batteriser to the battery and measure how much more energy you can squeeze from the battery until the output voltage of the Batteriser drops below that same cut-off voltage. You may also want to repeat the test with a constant current for the comparison purposes.

Yep, this will be one of the standard test I'll be doing when I get my hands on some.
And a parametric graph as well based on various constant current and constant wattages (close enough to how the majority of electronics products will behave)
Sadly though it would take far too long to do this for light loads. But even Batteroo have (now, not initially) freely admitted that it's not going to perform well on low current drain devices.

A simple led lamp would be a good candidate. It is very easy to measure how long it works without the Batteriser. Then attach the Batteriser and measure how much longer it will work. This is a simple, real-world test that can be done without any technical knowledge or special measurement instruments whatsoever. Only a clock is required. The guy in Greece might be able to do this kind of test so we could get the initial figures. I really do not care whether the led lamp has a dc/dc-converter inside or not at this point, just want to get some data.

[EEVblog]

Although I would encourage Wayne to actually test these things properly. With a conversion efficiency graph for different loads, and controlled product testing.

No need.
His "two identical flashlights" test is enough - so long as he's honest about the results.

Wayne has said anyone is free to come over and do more extensive testing on them. If he was in Sydney I'd take him up on the offer. Although decent product testing will take many days work just for starters.
He has admitted he's not good with the complex characteristic testing of such things and graphing etc. But he will release all his basic product test results.

[AndyC_772]

If these are just a very limited batch of samples, they probably don't need the approval mark

Well, technically they do, though pragmatically they can probably get away without them if indeed they're only planning to ship a handful of samples. A cynic might (and in this case, does!) infer that lack of approval marks is further evidence that the product is not "shipping", but is instead merely "sampling" to a select handful of people.

What's baffling is that CE and FCC aren't even difficult, or expensive, or time consuming to obtain.

[Kalvin]

Although I would encourage Wayne to actually test these things properly. With a conversion efficiency graph for different loads, and controlled product testing.

No need.

His "two identical flashlights" test is enough - so long as he's honest about the results.

Actually it may be best to use only one flashlight as you can be sure that the conditions do not vary. Take a set of fresh batteries, drain the batteries until the flashlight stops working or the light becomes very dim. Then attach the Batteriser to the batteries and measure how much longer the flashlight will work.

[jonovid]

why all the technical work just to joule thief an old battery. when you can have a all new product, like say a Li-ion battery inside a AA cell battery case
with its own buck converter & recharge port.
http://www.globalsources.com/gsol/I/Lithium-ion-battery/p/sm/1134386506.htm#1134386506

is there a prize for the longest forum thread ? or for the most revived AA battery from the dead?

[EEVblog]

Actually it may be best to use only one flashlight as you can be sure that the conditions do not vary. Take a set of fresh batteries, drain the batteries until the flashlight stops working or the light becomes very dim. Then attach the Batteriser to the batteries and measure how much longer the flashlight will work.

Ideally that's the case, but I think some of my initial experiments will be done with two identical products if possible just to speedup the testing time of initial results, with single product confirmation to come later. It also looks good on the time-lapse video with two products side-by-side.

[Fungus]

His "two identical flashlights" test is enough - so long as he's honest about the results.

Actually it may be best to use only one flashlight as you can be sure that the conditions do not vary.

No point. The Batteriser flashlight will die first.* The results/conclusions of the test will be obvious to even the most technically challenged.

(*) Assuming it's a LED flashlight with built-in boost converter. If it isn't then Batteriser will be brighter** but die sooner and more suddenly.

(**) No this isn't a legitimate case for using Batteriser. Buying a $10 LED flashlight is a much better option.

as you can be sure that the conditions do not vary.

You can swap the two flashlights around if you don't believe that Batteriser always dies first.

Take a set of fresh batteries, drain the batteries until the flashlight stops working or the light becomes very dim. Then attach the Batteriser to the batteries and measure how much longer the flashlight will work.

I don't think that's how most people would use Batteriser. Most people will just put the thing onto the new batteries.

(who wants to carry the batterisers around in their pocket then fiddle around in the dark trying to fit them to a dead flashlight?)

[Kalvin]

as you can be sure that the conditions do not vary.

You can swap the two flashlights around if you don't believe that Batteriser always dies first.

Take a set of fresh batteries, drain the batteries until the flashlight stops working or the light becomes very dim. Then attach the Batteriser to the batteries and measure how much longer the flashlight will work.

I don't think that's how most people would use Batteriser. Most people will just put the thing onto the new batteries.

(who wants to carry the batterisers around in their pocket then fiddle around in the dark trying to fit them to a dead flashlight?)

Yes, you are correct. My suggestion was just to make an initial test, to get rough figures. For example, if the flashlight will run for 8 hours with fresh batterires until the flashlight stops working, attaching the Batteriser to the dead batteries should make the flashlight running for more than 2 days. That is their claim.

[amspire]

Some of the things I am interested in include:

• What is the standby current (as the Batteriser is always on).
  
• If there are multiple Batterisers, do you end up with an audio frequency noise on the power to the device due to mixing of the switching noise.
• If you are powering an AM radio and the Batteriser is switching at perhaps 1.5MHz, how does the radio behave?
• What happens when one Batteriser shuts down - does it go open, or is there a forward biased diode junction from the flat battery?

If it goes open, what voltage can it withstand? Some old CB walkie talkies take 8 to 12 AA batteries, so the OFF Batteriser would have to be able to withstand up to 15V reverse voltage across it without failing.

If there is a forward biased diode from the battery to output or a normally reverse biased negative to output diode, and the current is 1A, will the diode overheat after that Batteriser shuts down? It could be using the latent substrate reverse diode junction of the switching IC to carry current when the battery goes flat. The signifigance of this is that if the contacts to the battery is dodgy, then when the battery disconnects, the switching IC gets reverse current from the other batteries. When the battery reconnects to the Batteriser, the switching IC may go to latch-up mode.

[timb]

Notice in the Aussie dude's pic, the packaging says Batteroo, whereas the Greek dude's says Batteriser. That makes very little sense to me.

Both sets of casings say Batteriser, however, which does make sense if you buy my theory of these being assembled from pre-production prototypes, which I think we can safely assume now based on the phone call from the Greek guy *and* Bob's insistence in the shipping email for the recipients to post photos.

I think it's pretty clear now that they're only shipping these to potential resellers and not general backers now.

[Brumby]

There isn't a single engineer on the planet that even remotely doubts the Batteriser will "work" as a boost converter and give you 1.5V output from a "dead" or "used" battery.

Absolutely.  The question is: For how long?

We anxiously await test results for some answers.

[jonovid]

Some of the things I am interested in include:

What is the standby current (as the Batteriser is always on).
If there are multiple Batterisers, do you end up with an audio frequency noise on the power to the device due to mixing of the switching noise.
If you are powering an AM radio and the Batteriser is switching at perhaps 1.5MHz, how does the radio behave?
What happens when one Batteriser shuts down - does it go open, or is there a forward biased diode junction from the flat battery?

personally I would never buy this product for the reasons listed.
but to give Credit where credit is due, the technical work in fitting electronics over a battery like a snake skin,
can be adapted for other uses or products.

[EEVblog]

Notice in the Aussie dude's pic, the packaging says Batteroo, whereas the Greek dude's says Batteriser. That makes very little sense to me.

Bob explained this in an update. They have permission from their Lawyers/Energizer to ship some existign units with the Batteriser name on it. Some people will get Batteriser, some will get Batteroo.

[Jay_Diddy_B]

Hi,

There have been a number of tests suggested for the Batteroo to see how it lives up to the claims made by Bob. These test involve time, waiting for the cells to discharge. The outcome of these tests can be predicted with a simulation like this:



This is a macro model of a generic boost converter. Three parameters are required to fit the model to the Batteroo.

The parameters are:

Iq - This is the quiescent current for the boost converter, measured with the output in regulation, but no load. In the model the Iq load is placed on the output side of the boost converter. This is because I believe the 'stuff' inside the chip will be powered from the output side.

The Iq parameter shapes the efficiency plot at light loads.

Rl - This parameter represents losses that are proportional to input current squared. It represents losses in the inductor and switch inside the boost converter.
The Rl parameter shapes the efficiency curve at the higher output currents.

Eff - This is the mid current efficiency. This determines the efficiency at medium currents.

Here is the efficiency plot, using typical parameters:



When these parameters have been determined empirically, the model can be fitted to the measurements. Once the model has been tuned in the performance of the Batteroo with various loads can be predicted without having to wait for the batteries to discharge.

I have attached the LTspice model.

Regards,

Jay_Diddy_B

[EEVblog]

When these parameters have been determined empirically, the model can be fitted to the measurements. Once the model has been tuned in the performance of the Batteroo with various loads can be predicted without having to wait for the batteries to discharge.

Yep, but it's like the audiophool snake oil claim "you have the listen to it". With the Batteriser true believers it's "you have to try it in the product".

But as I explained above, there is some merit to this as Batteroo's big claim is that of battery ESR and pulse currents causing products to stop working prematurely. You can't model this on the battery and a static load alone, you have to actually test individual products.

[Fungus]

Notice in the Aussie dude's pic, the packaging says Batteroo, whereas the Greek dude's says Batteriser. That makes very little sense to me.

Well spotted!

Both sets of casings say Batteriser, however, which does make sense if you buy my theory of these being assembled from pre-production prototypes

That's now my theory too (see post on previous page).

These aren't production, they're cobbled together from the prototypes Bob was showing last year. There can only be a few dozen units total.


(And I'm still not convinced the Greek guy has any PCBs in his. Greek guy could be a complete lie, Aussie Guy is simply being duped)

[timb]

Notice in the Aussie dude's pic, the packaging says Batteroo, whereas the Greek dude's says Batteriser. That makes very little sense to me.

Bob explained this in an update. They have permission from their Lawyers/Energizer to ship some existign units with the Batteriser name on it. Some people will get Batteriser, some will get Batteroo.

Yes, I can understand that for the shells of the units themselves, since they're stamped with a die they would have already paid for. It doesn't make any sense for the packaging, since they wouldn't have had much of that on hand with the Batteriser name on it.

There's zero evidence they had any Batterisers in production when the trademark issue was settled, so I highly doubt there are any "exsisting" units to send.

Now, one possibility is they may have had a bunch (hundreds? thousands?) of shells stamped out as early as last year, expecting everything to go smoothly with the electronics. Then they ran into an issue, current delivery most likely, that they couldn't easily solve.

So, seeing "Batteriser" on the units themselves isn't what I'm questioning here. It's the fact it's printed on the cardboard packages themselves.

Nobody has the cardboard packaging made six months to a year before the actual product goes into production. You might have a couple of them produced as tests, but that's it. In reality, you'd have them start rolling off the presses anywhere from a month before production to the actual time of packaging (print on demand).

[Fungus]

The whole thing still smells fishy to me.

Bob confidently predicts there'll be a few photos in the next few days but was unable to provide any himself. Now we have Bobs cronies saying they got theirs and two (count them) people posting pics, one of whom admits that he's visited Batteroo personally.

Google can't find anybody else with Batterisers, there's nothing outside of Indiegogo comments. No bloggers??

The package we've seen appears to be mailed by hand using a very expensive shipper (USPS) despite Bob blaming a "Logistics Company" for the latest delays.

[jonovid]

Yep, but it's like the audiophool snake oil claim "you have the listen to it". With the Batteriser true believers it's "you have to try it in the product".

B + I Call on all batteries who are Dead, and Low I will give you resurrection in the name of Batteriser 

[Fungus]

I'm calling bullshit on this.



Either:
a) The batteries weren't nearly dead (in which case there wouldn't be a 'noticeable increase in brightness')
or
b) They didn't go for another hour.


...and
(c) What happened after an hour? You stopped the test because it was finished??   
(d) Who on earth writes "two(2)", apart from a lawyer?

[daveshah]

Either:
a) The batteries weren't nearly dead (in which case there wouldn't be a 'noticeable increase in brightness')
or
b) They didn't go for another hour.


...and
(c) What happened after an hour? You stopped the test because it was finished??   

It could just be a really crap flashlight (say 2 AAs driving a white led with a Vf around 3V)

[Fungus]

It could just be a really crap flashlight (say 2 AAs driving a white led with a Vf around 3V)

Does that exist? I've never seen one (2xAA).

[Quarlo Klobrigney]

The words are correct. This place is full of wankers trying to sully the good name of Mr. Roofvarnish.
It's apparently clear to many learned individuals that the principle is a sound one.
Using an Over-Unity proprietary IC, it is possible to tap in Zero-Point energy to boost the battery's already existing remaining power. Shame on you all, including myself for disbelieving....

PS I just realized that the IoT future of these devices is enormous!

[FrankBuss]

Wayne has said anyone is free to come over and do more extensive testing on them. If he was in Sydney I'd take him up on the offer. Although decent product testing will take many days work just for starters.
He has admitted he's not good with the complex characteristic testing of such things and graphing etc. But he will release all his basic product test results.

Shouldn't be difficult to do some basic tests. Like use fresh batteries in a flashlight without Batteriser, stop the time until it gets dark, then the same test with Batteriser with fresh batteries. Bonus: use the dead batteries from the first test with Batteriser and stop the time again until it gets dark again. The results should be pretty unambiguous.

[LabSpokane]

I'm fairly certain that what has been sent to these initial recipients are not production units, but the pre-production samples that we've been seeing all along. But, finally, there are units in the wild to test claims against. I really wish Wayne would have a change of heart and loan one to Dave.

The bounty program is now closed. I might revisit this in order to convince someone to send a unit for testing. If you're out there and have a Batteriser, please contact me.

My eevblog username at gmail.com.