Author Topic: Batteroo testing  (Read 172615 times)

0 Members and 1 Guest are viewing this topic.

Offline FrankBuss

  • Supporter
  • ****
  • Posts: 2092
  • Country: de
    • Frank Buss
Re: Batteroo testing
« Reply #200 on: December 26, 2016, 07:42:18 pm »
Any trusted EEVBlog members are welcome to become collaborators and/or I can share login credentials
Thanks for adding me. I created a new repository, see here:

https://github.com/batterootesting/OpenCV-tests

It contains the lap counter Python script, with a step-by-step manual how to use it. Could be enhanced with a nice GUI (OpenCV has some useful integrated functions, like anything required to use the mouse to draw rectangles), but should be already useful, not only for Batteroo sleeves testing.
So Long, and Thanks for All the Fish
 

Offline FrankBuss

  • Supporter
  • ****
  • Posts: 2092
  • Country: de
    • Frank Buss
Re: Batteroo testing
« Reply #201 on: December 26, 2016, 09:18:20 pm »
Windows version I hit a problem, I think it's a threading issue when running the GUI loop and the Hook Keyboard loop and trying to update the keystroke info in real time. Too hard basket for now. For anyone interested in the exact issue, it's here: https://github.com/batterootesting/wirelesskeyboard/blob/master/Windows/tester.py on line 57
Works now. I tested it and it was the same thread, but looks like Tk doesn't like to be called from within a pyHook event.
So Long, and Thanks for All the Fish
 
The following users thanked this post: quad

Offline quad

  • Regular Contributor
  • *
  • Posts: 112
  • Country: au
Re: Batteroo testing
« Reply #202 on: December 27, 2016, 06:29:40 am »
We have a Windows 0.1 version! Thank you to Frank for fixing the not updating keystroke problem and adding the last keystroke time.



I have latched the start button, and added a pause/resume toggle button. Also if you try close the window it will prompt you if you really want to quit.

It should work now as long as there isn't a power outage, Windows doesn't crash or restart to do updates or anything silly like that. Maybe I can add a log file and a resume from log file feature, but I need to have a think about how to implement that   

If anyone wants to play around with it, here is the easy to use binary release: https://sourceforge.net/projects/wireless-keyboard-batt-testing/

Just extract the .zip and double click tester.exe -- it *should* work, but I have only tested on one other computer

Source code is at the GitHub page: https://github.com/batterootesting/wirelesskeyboard/tree/master/Windows
« Last Edit: December 27, 2016, 06:33:05 am by quad »
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #203 on: December 27, 2016, 09:20:25 am »
Haven't been following every post here. Who did that other (not Frank's) toy train video test?
 

Offline snik

  • Contributor
  • Posts: 30
  • Country: de
Re: Batteroo testing
« Reply #204 on: December 27, 2016, 09:49:36 am »
Haven't been following every post here. Who did that other (not Frank's) toy train video test?
Do you mean the test from Revive Batter(oo)ies Facebook Site ? It has done a "reliable and honest person with experience in electronics" :

Video : https://www.facebook.com/revivebatteries/videos/1303234933076207/
Post : https://www.facebook.com/revivebatteries/posts/1303252479741119
 

Offline samgab

  • Frequent Contributor
  • **
  • Posts: 422
  • Country: nz
Re: Batteroo testing
« Reply #205 on: December 27, 2016, 09:51:44 am »
Haven't been following every post here. Who did that other (not Frank's) toy train video test?

That related to something your old friend at Revive Batteries posted on facebook. No one knows who actually did the test or any real technical details about it at all.
https://www.facebook.com/revivebatteries/posts/1303252479741119
https://www.facebook.com/revivebatteries/videos/vb.576137725785935/1303234933076207/?type=3&theater
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 9134
  • Country: 00
Re: Batteroo testing
« Reply #206 on: December 27, 2016, 11:45:11 am »
Haven't been following every post here. Who did that other (not Frank's) toy train video test?

Our friend Wayne from Darwin.
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #207 on: December 28, 2016, 03:20:10 pm »
Ok, I have 4xAAA Batterisers and just tried one on my two discharged batteries (cutout voltage is 0.91V) from the MP3 MuVo TX FM player.
I tried one flat battery without the Batteriser, and one with the Batteriser.
Test without the Batteriser the unit switched straight back off, so zero extra run time.
First Batteriser would not fit and I broke the contact clean off the PCB!
I hacked the case to make it kinda-sorta fit and the 2nd went in with great force (and stabbing myself on the pos terminal!).
It ran for a fairly impressive 45 minutes extra (seems to be no audio noise issues). That's about a 14% extra run time, or x0.14. So much for x8 or x5. Pretty much just as predicted.
I'll upload the raw clip to my 2nd channel.

UPDATE: I put the initial "dead" battery back into the unit and it has started playing. Perhaps it was a PEBKAC, me pressing the button too long and turning it back off. Perhaps it was marginal startup voltage threshold. But without the Batteriser the "dead" battery lasted an extra 30 minutes!

So it seems the Batteriser does it's basic job of boosting a flat battery when that same flat battery no longer works in a product.
So the fans will no doubt claim victory. However, the real test is how much longer it runs for. And my first answer is x0.14. Meh.
But it's important to note:
a) The battery did not fit in what I thought was a fairly lose compartment.
b) The positive tab clip broke off with ease.
c) Even after modifying the case it would not fit properly (length issue) and had to be jammed in at the right angle to prevent positive tab clip damage.
d) Yes, the battery gauge showed full bars until it just suddenly died. Not a desirable thing.
e) An identical "Dead" battery if left to recover and without the Batteriser gave an extra 9.5% or x0.095 battery life.

So the difference the Batteriser made was about 4.7% extra life, or x0.047  :--
« Last Edit: December 28, 2016, 04:41:44 pm by EEVblog »
 
The following users thanked this post: Kean, djos

Offline samgab

  • Frequent Contributor
  • **
  • Posts: 422
  • Country: nz
Re: Batteroo testing
« Reply #208 on: December 28, 2016, 03:45:16 pm »
Cheers for the initial result update.  :-DMM
EDIT: Yeah, I thought that seemed a bit odd that it wouldn't even power up after resting the battery for several days. In my experience with Alkaline battery powered devices, after the batteries die the first time, they always work again for a fair bit after being given a few days to recover, or even an hour or two.
« Last Edit: December 28, 2016, 04:30:46 pm by samgab »
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #209 on: December 28, 2016, 04:20:56 pm »
Remember, the "Dead" batetry without the Batteriser has lasted for 30 minutes extra compared to 45min with the Batteriser.
That's probably within a margin of error for battery capacity, I'd expect those time to vary a fair bit if the test is repeated.
But in any case you don't need the Batteriser in order to get extra battery life from your product, just sit the battery on the shelf for a while and let it recover.
The difference the Batteriser gave was an extra 4.7% or x0.047 battery life.  :--

« Last Edit: December 28, 2016, 04:38:51 pm by EEVblog »
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #210 on: December 28, 2016, 04:48:30 pm »
EDIT: Yeah, I thought that seemed a bit odd that it wouldn't even power up after resting the battery for several days. In my experience with Alkaline battery powered devices, after the batteries die the first time, they always work again for a fair bit after being given a few days to recover, or even an hour or two.

Yeah, not sure what that was. Needs another few rounds of testing to see if it happens again.
 

Offline ez24

  • Super Contributor
  • ***
  • Posts: 3092
  • Country: us
  • L.D.A.
Re: Batteroo testing
« Reply #211 on: December 28, 2016, 04:48:41 pm »
In order not to break the Bateroo, insert the positive end like you tried but at the same time take a small blade and compress the negative spring and insert the Bat.  Then when the Bat is in, pull the blade out.  Your knife would do.  Save the Bats.

You will have to reverse the procedure to take them out.

Batteroo will have to insert a small blade with their Bats.
YouTube and Website Electronic Resources ------>  http://www.eevblog.com/forum/other-blog-specific/a/msg1341166/#msg1341166
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #212 on: December 28, 2016, 04:54:50 pm »
In order not to break the Bateroo, insert the positive end like you tried but at the same time take a small blade and compress the negative spring and insert the Bat.  Then when the Bat is in, pull the blade out.  Your knife would do.  Save the Bats.
You will have to reverse the procedure to take them out.
Batteroo will have to insert a small blade with their Bats.

In the case of this product that was not possible as there is a partial plastic cover over the negative end of the battery.
 

Offline ez24

  • Super Contributor
  • ***
  • Posts: 3092
  • Country: us
  • L.D.A.
Re: Batteroo testing
« Reply #213 on: December 28, 2016, 05:09:14 pm »
In order not to break the Bateroo, insert the positive end like you tried but at the same time take a small blade and compress the negative spring and insert the Bat.  Then when the Bat is in, pull the blade out.  Your knife would do.  Save the Bats.
You will have to reverse the procedure to take them out.
Batteroo will have to insert a small blade with their Bats.

In the case of this product that was not possible as there is a partial plastic cover over the negative end of the battery.

Oh.  |O

An idea for a new invention - a device that helps install a Bat  :-DD

If they decide to continue making these, I imagine they will change the positive end and do away with the hooks.

I enjoyed the live feed, it was the first one I viewed.  It is mind boggling that I can see someone on the other side of the world in real time.  Amazing  :-+
YouTube and Website Electronic Resources ------>  http://www.eevblog.com/forum/other-blog-specific/a/msg1341166/#msg1341166
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #214 on: December 28, 2016, 05:14:51 pm »
So this is hilarious:
I take the battery that was discharged in the Batteriser, and it still measures 1.26V. So I put it in a desk temperature monitor and it works just fine as you'd expect with a low drain device.
So I just put it in my Thrunite Ti keychain torch and it works just fine!
Currently recording how long the torch will last on the battery that has all the juice squeezed out of it by the Batteriser!  :-DD

Of course there is an engineering explanation for this, but lets not throw Batteroo any more bones shall we  ;D
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 8270
  • Country: au
Re: Batteroo testing
« Reply #215 on: December 28, 2016, 05:31:51 pm »
I wouldn't be cheering just yet.

Certainly the result is pretty much what was expected, but there will need to be many more tests conducted by a variety of testers before a consistent profile of performance is established.

The other thing that will be necessary is to define what constitutes as success and what counts as failure.

This is one absolutely critical parameter in the whole testing process - and it is one that has NOT been agreed upon by all camps.  As a result, there are going to be claims and counter-claims that will never be clarified and the battle will descend into a street brawl.

As an example:
1. Claim by Batteroo - originally "up to" 800%
Engineering assessment predicts "anywhere remotely near 800%" can never be achieved.  Initial tests support this.  If Batteroo have specific cases where such performance is found, then they should be forthcoming with details.
2. Claim by Batteroo - "up to" 80%
Engineering assessment again predicts this as being, at best, unlikely.  If Batteroo have specific cases where such performance is found, then they should be forthcoming with details.
3. Observation by first known Australian recipient
Batteriser maintains a more consistent output over the earlier part of the life of a battery as compared to a battery without.  The observed difference being little Johnny will have his train run around his track at a goodly speed until it stops dead, rather than have it start off ok and then get slower and slower until it's moving as fast as frozen treacle.

As I see it, the first two would count as "fail", because - very simply - they can be shown to not meet the manufacturer's claims.  It is simply comparing numbers, which is how an objective conclusion can be made.

The third case, however, does show a tangible difference. The first point, however, is that Batteroo made no reference to such behaviour, nor promote it as a feature or benefit of their product - so claiming it as a "successful" test is still a bit inappropriate as it is outside of the claims made by Batteroo.

Having said that, however, it cannot be denied that such behaviour is significant, but the question that then arises is whether this is good or not.  This now enters a subjective phase where two people might have opposite opinions - and each is valid in their own way.

Take the case of the classic flashlight with a filament bulb.  With a Batteriser, it will shine more consistently but die suddenly.  Without, it will gradually decrease in brightness over time, but will run for longer than the unit with the Batteriser.  Which is better? - The brighter light or the longer run time?  How do you choose?  What criteria do you go by?  If there is no specific way to define how to evaluate which is better, this argument will never be resolved.

What's even more interesting, IMHO, is that - if the engineers had been presented with the scenario of two flashlights or two trains - one with and one without a Batteroo sleeve - they would have come up with exactly the same predictions on the behaviour of these devices - as the devices have exhibited in real world testing.

These behavious are not a surprise - but they were not considered by the engineering minds, because they were not in the claims from Batteroo.


The other - rather telling - thing is that, so far, there has not been much indication that the reduction of battery use is going to be significant.  This has to be the biggest fail on the part of Batteroo - as it has been their fundamental claim.

If anything, the higher output for a shorter time would seem to be the one feature of the Batteriser that may actually make it a marketable product (Bravo to our Darwin friend for that discovery), however I can see battery usage going up - not down.
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 8270
  • Country: au
Re: Batteroo testing
« Reply #216 on: December 28, 2016, 05:34:35 pm »
So this is hilarious:
I take the battery that was discharged in the Batteriser, and it still measures 1.26V. So I put it in a desk temperature monitor and it works just fine as you'd expect with a low drain device.
So I just put it in my Thrunite Ti keychain torch and it works just fine!
Currently recording how long the torch will last on the battery that has all the juice squeezed out of it by the Batteriser!  :-DD

That is gold.  ;D   ;D   ;D


Quote
Of course there is an engineering explanation for this, but lets not throw Batteroo any more bones shall we  ;D
No.  Besides, Engineers don't know everything (well, that's what I've been hearing).
 

Offline usagi

  • Frequent Contributor
  • **
  • Posts: 343
  • Country: us
Re: Batteroo testing
« Reply #217 on: December 28, 2016, 06:51:22 pm »
As I see it, the first two would count as "fail", because - very simply - they can be shown to not meet the manufacturer's claims.  It is simply comparing numbers, which is how an objective conclusion can be made.

they could still claim success, as "up to" includes zero.

Offline Towger

  • Super Contributor
  • ***
  • Posts: 1418
  • Country: ie
Re: Batteroo testing
« Reply #218 on: December 28, 2016, 06:54:59 pm »
I have seen the 'power management' on some high tech gadgets requiring a higher voltage to 'reset' before they restart, but once going keep on going for a while.

Have a look at the solder pads under the microscope, when Frank broke his off, another member pointed out it was an example of a classic dry joint, and Foxconn should have known better.

I hope you have the same song at the same volume going on the player during all tests.  Don't want the naysayers claiming there is a difference in power consumption between heavy rock and classical music!
 

Offline FrankBuss

  • Supporter
  • ****
  • Posts: 2092
  • Country: de
    • Frank Buss
Re: Batteroo testing
« Reply #219 on: December 28, 2016, 07:00:00 pm »
But it's important to note:
a) The battery did not fit in what I thought was a fairly lose compartment.
b) The positive tab clip broke off with ease.
c) Even after modifying the case it would not fit properly (length issue) and had to be jammed in at the right angle to prevent positive tab clip damage.
d) Yes, the battery gauge showed full bars until it just suddenly died. Not a desirable thing.
e) An identical "Dead" battery if left to recover and without the Batteriser gave an extra 9.5% or x0.095 battery life.

So the difference the Batteriser made was about 4.7% extra life, or x0.047  :--

Exactly my experience. I was careful, because I did have only 2 sleeves for testing, and still broke off the positive tab clip from one of it. We can expect many end users will break it, too (and puncture their fingers :rant: ). And your MP3 player extra time percentage is very similar to my MP3 player test, which is no wonder, because probably similar cut-off voltage and current consumption.

Good idea with the extra column in the spreadsheet for the dead battery extra time without the sleeve. This will demonstrate that it has no advantage at all for many products.

If you are going to use your 53131A counter, do you plan to connect the GPIB interface? Would be nice to have a spreadsheet with the individual round times. The Voltnuts Raspberry Pi distribution supports some GPIB USB interfaces and has already Python test scripts for other Agilent gear, which should be possible to adapt to your counter easily. But maybe even easier to hookup an Arduino and print the Arduino system time (using millis(), don't need to run for more than 50 day :) ) on the serial port for each detected round.
So Long, and Thanks for All the Fish
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #220 on: December 28, 2016, 07:33:07 pm »
As an example:
1. Claim by Batteroo - originally "up to" 800%
Engineering assessment predicts "anywhere remotely near 800%" can never be achieved.  Initial tests support this.  If Batteroo have specific cases where such performance is found, then they should be forthcoming with details.
2. Claim by Batteroo - "up to" 80%
Engineering assessment again predicts this as being, at best, unlikely.  If Batteroo have specific cases where such performance is found, then they should be forthcoming with details.
3. Observation by first known Australian recipient
Batteriser maintains a more consistent output over the earlier part of the life of a battery as compared to a battery without.  The observed difference being little Johnny will have his train run around his track at a goodly speed until it stops dead, rather than have it start off ok and then get slower and slower until it's moving as fast as frozen treacle.

I'd also add a 4th test, and it just occured to me durign my initial testing. That is, does the Batteriser actually extract all the energy from a battery as it claims?
The answer is no. I drained a battery using the Batteriser and it still had enough energy to run a torch (yet to verify the duration).
I find this utterly hilarious, we need another Batteriser design in order to exact all the unused energy left over from using the Batteriser!  :-DD
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 27691
  • Country: au
    • EEVblog
Re: Batteroo testing
« Reply #221 on: December 28, 2016, 07:35:49 pm »
Good idea with the extra column in the spreadsheet for the dead battery extra time without the sleeve. This will demonstrate that it has no advantage at all for many products.

That didn't occur to me before, but it is an obvious test. The Batteriser is of no use if the battery naturally recovered and gives a similar extra life anyway.
In my case 30 minutes extra without the Batteriser, 45 minutes with the Batteriser, making it of marginal value, and possibly within the margin of test variability.

Quote
If you are going to use your 53131A counter, do you plan to connect the GPIB interface?

No, I don't have a proper GPIB setup, nor the time to get that all sorted and set up.
 

Offline TechnicalBen

  • Contributor
  • Posts: 37
  • Country: gb
Re: Batteroo testing
« Reply #222 on: December 28, 2016, 08:23:51 pm »
So basically, if this was snake oil, it would be very good snake oil for cooking, moisturizing and attracting snakes. As it's oil, it does everything oil does. No problems...

... except it is marketed as a "cures all and makes you live 800% longer"? Yeah, not cool.
 

Offline PeterL

  • Regular Contributor
  • *
  • Posts: 178
  • Country: nl
Re: Batteroo testing
« Reply #223 on: December 28, 2016, 08:41:33 pm »
There is this recurring idea on the batteriser threads that batteroo once changed their 800% to 80%, but they did not.

They have said: "We can extend battery life up to 800%"
And they have said: "devices usually only use 20% of a battery, so 80% is left"

So the 80% does not refer to the same thing as the 800%

Now if the latter statement was actually true, and batteriser gave you all of the remaining 80% this would give an improvement of 80/20 = 400%. Still not 800%, but not an order of magnitude at least.
Or for the first statement to be possible a device would have to leave 89% of energy.

It's all a bit off a confusing maths trick, but I think we should have this clear.
 

Offline razvanme

  • Contributor
  • Posts: 31
  • Country: ro
Re: Batteroo testing
« Reply #224 on: December 28, 2016, 09:08:46 pm »
Can we have the first Post with the claims / results? Or a new topic with one post that only Dave can edit.

Quote
Claims
- 800% extra life
     Not going to happen
- 80% more - maybe kindoff in some extreme situation not likely
- Extend battery life - Not really
- Less batteries used - Actually more batteries used, since they "extract" all the energy faster, it will take less time to discharge - So much for saving the dolphins / planet.

Passive devices
- Toy train - less time - possible it runs faster?  << insert youtube link >>
- Toy monkey -
Active devices
- MP3 player - no extra time - 
- Camera ----
- Wireless keyboard ---- 

PROS
.... Boosts to 1.5V
CONS
 - Cannot be used in flashlights
 - Tight fit
 - Breaks easily
 - Hot batteries ??
 - Gauge not working correctly


It just occurred to me that the "hot" complaints involved multiple batteries (candle light, flashlight). So it might get hot when used in series / parallel.


 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf