Batteroo testing

[daveake]

Btw. I just realized: It might be entirely possible that this was their Plan - Make ridiculous claims, get all the engineers riled up and then sell the product to them so they can disprove the marketing-claims

And with the bonus that eventually some actual workable use-cases turn up from all the testing ...

[SaabFAN]

One test that would be interesting would be to power the train with Daves Lab PSU (using a pair of long wires to allow it to continue to loop round the track a few times before getting twisted up!) and record the power consumption at various supply voltages!

You could use some solder wick as wipers that connect to the battery-terminals aluminium-foil glued to each side of the tracks that is connected to wires.
That way you can have the train run continuously with the PSU or with the batteries in a battery-holder.

[EEVblog]

And with the bonus that eventually some actual workable use-cases turn up from all the testing ...

Given that they just asked me to do formal contract testing for them, it get the impression they have done bugger-all testing. And that shows in their public examples. It can't be that hard to find a genuine product it would give good results in, especially given the years they have been working on this and the money invested.

[FrankD]

Hooking up a few coaches would have shortened the batterroochoochoo test considerably. A real train aficionado would have known this. 

[Brumby]

You could use some solder wick as wipers that connect to the battery-terminals aluminium-foil glued to each side of the tracks that is connected to wires.
That way you can have the train run continuously with the PSU or with the batteries in a battery-holder.

You would not need a lot of time to get some characteristic readings.  Sliding contacts would be a nightmare.  Several turns of twisted wire is easy to work with and the connections would be rock solid.  You could even do 4 wire measurements.

[FrankBuss]

You would not need a lot of time to get some characteristic readings.  Sliding contacts would be a nightmare.  Several turns of twisted wire is easy to work with and the connections would be rock solid.  You could even do 4 wire measurements.

Don't know if this model train has a junction track part, but you can avoid wire twisting with an 8-shaped track.

[Fungus]

From same Duracell datasheet capacity at 0.5W drops to about 550mAh

Batteriser is basically busted right there.

Even if Batteriser is 100% efficient/perfect: Boosting the voltage by X% will decrease available battery power by more than X%.

[TechnicalBen]

We need to know why the train/mp3 player do so badly with Batteriser.

If we take the amount of energy needed to get the train around the track as our unit of measure then the Batteriser is only about 60% efficient.

That's a plausible figure when you're working with low voltages and tiny components but is it really that bad after all the hype we've seen? No wonder Bob has been trying to keep them away from engineers.

There is likely to be a signficant reduction in the efficiency of the train as well as the voltage increases!  Small, cheap, brushed motors have terrible hysteretic iron losses, that increases rapidly with motor speed.  It's not unusual to see a 15% reduction in efficiency from just a 5% "Overspeed"

One test that would be interesting would be to power the train with Daves Lab PSU (using a pair of long wires to allow it to continue to loop round the track a few times before getting twisted up!) and record the power consumption at various supply voltages!


The biggest irony is that because batteries are electrochemical, ie, the movement of Ions within the compounds of the battery create the electron flow, this is a electromechanical process and hence it is "rate" dependent.  Pull lots of electrons (high current) from the battery, and there is simply not enough time for Brownian motion to stir the chemicals to provide enough ions, so the battery goes "flat" (ie, output voltage falls).    The Bateroo, being a boost convertor, actually loads the battery with a higher average current (on a passive load) and so requires a higher rate of ionic movement, and hence actually renders a proportion of the electrochemical energy un-useable!  From the train test, it looks like, for a typical AAA battery, that un-useable proportion is larger than the un-used (due to insufficient voltage) portion without the Bateroo.....

What we would need to see if the discharge curves for these batteries at a range of different discharge rates. In all cases, i'd expect the lowest possible discharge rate to be able to extract the largest total energy from the battery.

So your saying I need to plug in the Batterizer upside down to reduce the voltage draw and extend the battery performance... interesting.

(I hope you note the sarcasm! So they shot themselves in the foot with this one. )

[PeterL]

On the long thread someone mentioned that a Canon Powershot camera could be a candidate for a batteriser, since these work notoriously bad on alkalines.

Digital camera's were also a point of attention in the 'zinniker report' (http://www2.ife.ee.ethz.ch/~rolfz/batak/ICBR2003_Zinniker.pdf)

So I made some current measurements with my old 'Canon PowerShot S2 IS', just to see what currents this actually draws.

The camera has a 'ready' mode in which it draws approx 300mA, then when you do something like taking a picture, or start to zoom, current go op to 400 or 500 mA, with spykes up to 700mA. Highest currents are when taking a flash photo, at which point current tops at 800mA.

Measurements were taken with a linear power supply, set to 5.5V or thereabout.  (the camera needs 4 AA's)
Some screenshots are attached below for those interested.

It would be interesting to see how the batteriser handles these currents and spykes.

I can't image the batteriser will fit in this compartment btw, see last picture.

[eeFearless]

Hi!

I'm new to the forum.  Interesting thread, but TL;DR.

I wonder if they had a different system design, if the thing could have added sufficient value.?

Pass-through voltage until the battery droops to 1.1V, then boost to only 1.2V as the battery drops to 0.6V?

An improvement, but still probably wouldn't have been worth it ... though ...


But, if you're really motivated by the environment, then use rechargeable batteries ...

[6581]

They will not tell you the battery could still be useful, because that destroys the purpose of their marketing.
Likewise they will not tell you that the Batteriser will give you less life in many products

They also don't mention:
a) That it doesn't actually fit in many products and might break unless you use special tools to insert it.
b) That it could be dangerous in any critical products that require a working battery gauge (eg. smoke alarms, CO sensors).

Just in case anyone is interested - just replaced battery in a smoke detector and measured the open circuit voltage of the battery that was continuously (from new) in that detector since may/2012 (4,5 years) - it measured 8,02v. It didn't sound the beep to change (!!!), but obviously it was way over due. Battery was Varta Longlife alkaline, manufactured in 2011. My ZTS Mini MBT battery tester indicated "dead" (flashing 20% light):
https://www.ztsinc.com/minimbt.html

Smoke detector manufacturer Marelco (model unknown.) I didn't test it with the self test before taking the battery out, but after measuring open circuit voltage and using battery tester, went to place the battery back to the unit and the self test was ok (!!!) it sounded the alarm very, very loud. (Obviously I did put the new battery back in afterwards.)

New battery happens to be the same model - open voltage 9,66v, in device it WAS 9,66v and dropped to about 9,2 and then started climbing back up - during self test the battery voltage dipped perhaps 0,02v. I believe there's a cap inside and proper testing would require discharging it. Thought this might be interesting. :-)

[mikeselectricstuff]

And with the bonus that eventually some actual workable use-cases turn up from all the testing ...

Given that they just asked me to do formal contract testing for them, it get the impression they have done bugger-all testing. And that shows in their public examples. It can't be that hard to find a genuine product it would give good results in, especially given the years they have been working on this and the money invested.

seriously..? By "they" you mean Batteroo? They're talking to you now?

[rollatorwieltje]

Just in case anyone is interested - just replaced battery in a smoke detector and measured the open circuit voltage of the battery that was continuously (from new) in that detector since may/2012 (4,5 years) - it measured 8,02v. It didn't sound the beep to change (!!!), but obviously it was way over due.

8.02V is nowhere near dead. A 9V battery is just 6 cells in series, it's dead when it's near 6V (when you consider 1V per cell as cutoff). Granted, you don't want to run them completely down in a smoke alarm. My smoke alarms start beeping when it drops below 7.4V, that's still more than 1.2V per cell.

[drussell]

Pass-through voltage until the battery droops to 1.1V, then boost to only 1.2V as the battery drops to 0.6V?

This was all discussed at length in the other thread....

A booster for rechargeables to mimic carbon-zinc/alkaline would be more useful...
A boost/buck converter with an output of 1.2 volts could extend the life of carbon-zinc/alkaline cells in some products where a rechargeable works satisfactorily, etc...

The product they built, however, is virtually useless.

Given that they just asked me to do formal contract testing for them, it get the impression they have done bugger-all testing. And that shows in their public examples. It can't be that hard to find a genuine product it would give good results in, especially given the years they have been working on this and the money invested.

Indeed...  Despite the fact that they have kept saying that they have tested it in many, many devices and they supposedly fit in almost all devices and provide great gains in longevity in all these devices, as expected, the reality is nothing of the sort! 

[drussell]

seriously..? By "they" you mean Batteroo? They're talking to you now?

Yup...   

https://www.eevblog.com/forum/blog/eevblog-751-how-to-debunk-a-product-(the-batteriser)/msg1101276/#msg1101276

[Poe]

The amount of energy you get out of a battery decreases if you draw more current.

That's true... I forgot about that.

It's only going to draw about 30% more though, is that enough to explain the difference? I guess it's a combination of everything.

My Olympus SP350 only lasts 50 flash pictures on a fresh set of batteries (<30min). 

Most two cell, non-LED based flash cameras are like this. 

Would be interesting to see how it handles this massive load considering the miniature size of their magnetics.

[FrankBuss]

I updated the spreadsheet. I did the test with the MP3 player again, this time fresh battery without Battery until it was dead, then waiting for 2 days, like for the Batteroo test, but using this battery again without the Batteroo sleeve. Extra time was 71 minutes, compared to 109 minutes with Batteroo. So Batteroo has a little bit of use for this device, maybe because of the high cutout voltage of 1.1 V. Dave's MP3 player extra time was 30 minutes and 45 minutes with Batteroo, and cutout voltage 0.9 V. Of course, doesn't matter much if you compare this with the 10 times higher initial time with a fresh battery, still far away from the 8x claim. And the drastically reduced time if you use Batteroo from the beginning with a fresh battery doesn't help either.

And I added the tests from Ysjoelfir: more than 8 hours without Batteroo, multiple tests with Batteroo, where the light bulbs lasted between 60 and 90 minutes until they were destroyed.

[6581]

Just in case anyone is interested - just replaced battery in a smoke detector and measured the open circuit voltage of the battery that was continuously (from new) in that detector since may/2012 (4,5 years) - it measured 8,02v. It didn't sound the beep to change (!!!), but obviously it was way over due.

8.02V is nowhere near dead. A 9V battery is just 6 cells in series, it's dead when it's near 6V (when you consider 1V per cell as cutoff). Granted, you don't want to run them completely down in a smoke alarm. My smoke alarms start beeping when it drops below 7.4V, that's still more than 1.2V per cell.

Well aware of that (6 cells in series) - although we all probably agree that open circuit voltage is not a good measure of the energy left in a battery... just saying that after 4,5 years one should definitely change the battery. Didn't recycle it either - it may still find its way to some less safety critical application.

Edit: sorry, probably going too far off topic.

[ez24]

Hooking up a few coaches would have shortened the batterroochoochoo test considerably. A real train aficionado would have known this. 

I asked that the cargo wagon be connected and have a load so a load test could be done.  I wonder why the testers don't use the cars that come with the train because that is what a child would do  ie a child would do a load test.

[StillTrying]

You'd need a miracle to get 4 batteroo slaves in and out of there.

[dcac]

For reference, it would be interesting to run the train test with a good quality NiMH rechargeable AAA battery, without the batter wasterisers! as they will only have the same negative effect as on the Alkalines.

NiMH has comparable mAh to Alkaline, though the voltage is lower, in the beginning, but it would even out at the end of the run. And the NiMH could possibly last longer than Alkaline. 

[dcac]

You'd need a miracle to get 4 batteroo slaves in and out of there.

I have an old Canon Powershot A640 which also uses 4 AA and they fit as snugly as cartridges in a gun. I use black Envelopes in it, seems to last forever.

[Fungus]

NiMH has comparable mAh to Alkaline, though the voltage is lower, in the beginning, but it would even out at the end of the run. And the NiMH could possibly last longer than Alkaline.

Realistically... it's looking like NO battery/device is going to last longer with Batteriser than without it.

Passive devices: So far have only lasted 50% to 60% as long as a bare battery.

Active devices: No possible gain, probably a loss because a) Batteriser isn't 100% efficient and b) Betteriser isn't optimized for each individual device's power needs like the built-in circuitry is.

Also: Batteriser doesn't appear to be using up 100% of the Battery's available power. Dave ran a torch for 16 hours on a Battery that Batteriser had finished with.

Fail, fail, fail.

[Stuart Coyle]

I just did a back of the envelope energy calculation based on Dave's train test. I ignore the static friction and only take into account kinetic friction, as the train only starts once. The work the train has to do to keep a constant velocity is the friction force F times the distance travelled S, (Yes, yes I know its an integral but this is an envelope with limited space...). So in the Batterizer case we get only about 351/508 * 100% = 70% of the energy.

Ah, but there is also kinetic energy! With the Batterizer the train runs 10% faster. So the energy (mv^2)/2 is 1.1^2 = 1.21 times as much as the non batterizer case.

This overall gain or loss of energy (-30% vs +20%), depending on the friction force and the mass of the train. In the case in the video it seems that friction dominates, and it would be a loss. Kinetic energy would be a very small proportion of total energy over such a long test. A measurement of kinetic friction and mass of the train would clarify this.

[djos]

You'd need a miracle to get 4 batteroo slaves in and out of there.

Those little gaps joining each of the cells could allow them to be installed if you align the sleeves up with them.