Beo5/Beo6 battery

[matador]

Thanks Yohan, It’s a joy to have them back.

The flamingo’s name is Crockett and he keeps an eye on the house!

[matador]

Hi alls,

Two weeks of usual use without problem.

The sleeping room one notably ,that used to have problems with Apple Remote IR commands that drains the battery instantaneously is now working well.

Regards.

[lausvi]

Inspired from this thread, I went and bought a new battery from ElectroPapa, I received a white-label vhbw one (identical to the first picture in this thread). I was happy to see that one of my two Beo5s (both bought second-hand, both worked for a while but died soon after and were put away) did came up with the battery, but there is still a problem as it’s not charging. I am using a USB charger with a USB power meter, and I can see that when I first put the remote to the dock, the current goes up to 90mA (matches the USB charge current in the user manual) BUT drops down to zero after just 5-10 seconds. The Beo5 says ‘charging’ but it’s clearly not, however it keeps it barely alive (if taken out of dock, it gives ‘charge battery’ but while in charger keeps going).

I disassembled the remote (again) and experiment shows that if I disconnect the display ribbon cable, it keeps charging. So I assue the charging circuitry itself is working, but the CPU (inside the display housing) tolds it to turn off. The wiring diagram confirms that there are HIGH_CHARGE_STOP, VBAT_SENSE and CHARGE_ID* signals so clearly it’s doing something with the charging. My ribbon cable was already bent and the contacts don’t look too good so I am somewhat expecting a bad cable. I ordered some new ribbons (the service manual gave me the needed info: 30-pin 0,5mm pitch, contacts on same side) and I’ll need to dig up my other Beo5 and compare what it does with the new battery. The ribbon cable is detachable on both ends (for some reason the display unit has a nice locking-tab connector while the keypad has a push-in connector in my remote). I was able to open the display housing too.

*) CHARGE_ID appears to be also a signal from the dock, I presume this identifies the wall-charger with which the charging should happen at 400mA, ie. much faster. If I can find my wall-charger, it would be interesting to measure how the detection works to bring the remote into the higher current charging mode.

I am pretty sure this is NOT a fault of the battery, but my Beo5s/the ribbon cable, as this is very much the same I had before; the remote seemingly works but it’s not really charging and eventually the remote dies.

 

[lausvi]

Slight update; nope, I am at loss here. While the disconnected ribbon makes the keypad PCB show current draw, the battery is still not charging, I left it overnight and it’s still as low. My other Beo5 (with visually much better looking ribbon cable) shows the exact same issues with the new battery (shows 90mA current for a moment, then drops to zero) so either it is still the battery (incorrect temp sensor reading preventing the charging?) or both of me Beo5s have some identical fault with the charging.

[matador]

lausvi wrote:

(incorrect temp sensor reading preventing the charging?)

Very dumb and ignorant assumption from me but yes, I believe the main problem could be here. I’ve tried lots of batteries that didnt work inside the beo6 but that worked well outside as regular batteries.

 

[lausvi]

I got the new flat flex cables. A short 8cm one is now being used in my test-setup and the remote works, so it’s good to know the cables are replaceable. The cable didn’t make any difference on the battery issue, but I did found something else.

I measured the resistance of the temp sensor. In the new battery I have (that won’t keep charging) it’s about 8.6-8.9 kOhm (in room temperature), while an older Sanyo NiMh pack (which I bought some 5 years ago, but didn’t work) has much higher value of 36-38 kOhm. (Measured between the black negative wire and the white center wire on the battery pack, I had yet another one un-wrapped pack and I verified that the NTC is connected between there).

I connected the old battery pack and it stays on charging! I am not sure if the cells will actually hold charge (it has been sit flat on parts bin for years now), but at least the difference in the temp sensor seems to have some part in the charging.

Now, if the old battery pack won’t hold charge (I’ll leave it charging overnight in the slow 0.09A on USB) then the next experiment might be taking the “good NTC” from the old battery and replacing it into the new…

Does anyone have an original B&O battery (doesn’t need to be working one) for comparison on the NTC value?

[Axel73]

Hello,

I also think that the problem lies with the Thermistor (temp sensor):
I’ve had the exact same problem recently with a battery bought on the Internet: impossible to charge the battery with the Beo6 even if the status on the screen is “Charging”.
It seems that the Beo6 detects correctly the battery, tries to charge it but then stops immediately.
On the original B&O battery there is NTC Thermistor (between the black and white cables): its value is 50kOhm (at 25°C).
The battery bought on the internet also got a NTC Thermistor but its value (in my case) is 10kOhm at 25°C.
Since a NTC Thermistor resistance decreases when the temperature rises, it means that the Beo6 “thinks” that the new battery is very hot because its thermistor value is down to 10kOhm. That’s why it stops the charging process I guess (oddly enough  “Charging” is still displayed on the screen !?).
To solve the problem I’ve simply replaced the new battery thermistor (10kOhm) with the thermistor coming from the old B&O battery (50kOhm).
Now the new battery is charging correctly at last 🙂

I hope that this trick will help some of you solve their battery problem.

Cheers,
Axel

[lausvi]

Thanks Axel for confirming the theory.

I just went from two non-working Beo5s to two working Beo5s 🙂

I tested with parts from two remotes, new flat flex cables (seen here) and three battery packs: two older Sanyo packs (blue cells with yellow wrapping, bought years ago, suitable NTC, one dead pack, one working) and the new one (green wrapper).

I replaced the flat flex on one remote as the old one had a sharp bent and the connectors were somewhat fiddly too after several insertions and removals.

New flat flex next to the old one. I bought a 8cm one (link for the current item on eBay at the time of writing) which does fit inside the case (needs a bit of fitting) and is easier to handle during disassembly. Shorter cable would be ideal but I couldn’t find one. Cable specs: 30 pins, 0.5mm pin spacing, Type A (contacts on the same side).

This is the new battery pack into which I replaced the “more correct” NTC from the dead older pack. While I presume this is not a good practice (the NTC should be matched to the specific cells), I expect the range to be about there for the remote to detect a hot cell, and at least with the USB charging the current is so small that unless the cells go short, I guess it’ll be fine.

For reference, this is how the display module (actually the CPU board) looks inside. The clip-up connector for the flat flex is so much easier to work with than the push-to-insert one in the keypad board.

The display case is held with clips on the edges, it can be opened and closed with not too much marks on the casing.

Both remotes appear to work now, and least they both charge (first with USB as I could plug in my USB power meter to keep checking that the charge kept going and then with the original charger (400mA), keys work and with camera I can see the IR leds beaming.

Neither remote is in too good condition, both have worn key legends and scratches on the aluminum parts etc and one has a bit loose display mounting. Both remotes have my somewhat hacky configuration from years ago (that january 2017 as seen in the display is probably correct), when I only started to grasp the configuration tool before the first remote died (I presume it was a bad battery, then the later remote I bought probably had a bad ribbon cable and/or bad battery too, and I couldn’t figure this all out then).

Now I need to get the configuration utility installed and start hacking with it again to see if I can further polish the configs!

[matador]

A little feedback on the CS batteries,

They are still running fine but they discharge a lot faster than in the first weeks of use.
To the point that I need to have two Beo6: one on the table the other on the charging dock. Average swapping is two days.

I’ve also noticed that it never charge at full: on the Beo6 red battery indicator there is always one or two millimeters black even though the remote sits on its craddle for days.

For the record all Beo6 are full of third party IR codes (Samsung & Apple) and all screens use full screen images.

Nevertheless, those batteries are still the better option to date.

[matador]

Breaking News (like if my last post brought me bad luck)…

The remote does’nt even last few hours nor even minutes. I used it today in Apple TV mode and after few clics both went dead.

I suspect this being related to the aTV IR command but I dont know how to tell for sure.

[Tignum]

Hey Matador, that is a shame… My (only) Beo6 has been sitting on its cradle since it received its CS battery. Reason is that I messed the configuration up, and am now waiting to get an old Windows computer to enable me to access it again. It shows “No valid configuration!” only, and I also cannot get it to display battery status. When I will have it reanimated, I will update about battery behavior. How can these batteries be so full of magic? Weird… Regards, Johan

[Axel73]

Hello matador,
In my opinion, the first thing to check is the value of the CS battery NTC thermistor. Simply measure the resistance between the white wire and the black wire of the battery.

I think the Beo6 microcontroller was programmed to work with a “50kOhm NTC Thermistor”. By measuring thermistor resistance, the microcontroller can deduce the temperature using the 50kOhm NTC Thermistor correlation table:
at 20°C the resistance is roughly 62kOhm, at 25°C the resistance of the NTC is 50kOhm, at 63°C it will be more or less 10kOhm and so on (as the temperature increases, the resistance of the NTC decreases).

If you measure the resistance  of the CS battery NTC and  it’s something like 10kOhm at room temperature (20°C), then the microcontroller will think that the current temperature is 63°C rather than 20°C since in its correlation table 10kOhm = 63°C.
As a result the battery charging process will often be stopped because the microcontroller will think that the battery temperature is too high.

My advice would be that if the resistance value of the CS battery NTC is not close to 62kOhm at room temperature (20°C) as expected by B&O, it is better to replace it with the NTC of the old B&O battery or to buy a new “50kOhm NTC thermistor”.
If an incorrect NTC is fitted in the CS battery, it can explain why the battery is not charging correctly or enough.
Best regards,
Axel

[Author]

Posts