Bosch powerpack 400 opening

[Voltron]

For anybody wondering.....you don't need to saw it apart. There are 4 tabs per side... I just brute forced this one apart, cracking two tabs in the process, but now that it's open it looks like the properly bent piece of thin metal could be used to pop the tabs one by one, starting from the bottom.

View attachment 1


In the last pic, you can see the corroded trace the killed the BMS and discharged the cells to 1.5 ish volts.

If I owned one of these systems, I would definitely look into caulking the seam or covering the whole battery....the waterproofing seal on the seam is not very robust.

 

[kdog]

Cheers voltron,
I had a very gentle attempt at opening one of these recently, didn't want to damage it so I gave up...thanks for the peek inside.
Are you going to repair it? How do you think the (I assume)proprietary bms etc will go being swapped out if that's what you're planning. I wonder if this Bosch system is so integrated that it won't allow for tinkering with- like some of the power tool stuff.
Keep us posted!
K

 

[Voltron]

Slow reply, sorry. Yes, with some prying and poking at the right spots it opens up fairly cleanly, and even if one broke all the tabs it locks together enough for taping closed. And yes, planning on pulling the BMS and replacing it, as I won't be using it on a Bosch system there won't be issues with the system trying to talk to the BMS. I have a second one, so they'll be paired together for a 72v pack. I like 72v... it's a nice voltage for getting a slower motor to turn some rpms, it's easy to find 36v packs to start with, and you can make them easily separable for charging with two cheap 36v bulk chargers....

 

[Voltron]

The pack out of the plastic shell... slight construction differences between the two...

 

[tlg]

I also recently had the opportunity to look at the Bosch 400 battery pack close up after the pack failed on a friend's bike.

As Voltron mentioned, there seems to be no gentle way to open these packs due to the four clips down each side of the case. So if you plan to open one up, be prepared to do some damage to the edges of the case. This probably also indicates that Bosch have no interest in servicing the batteries beyond complete replacement.

The failure of this battery is interesting. The bike has done a total of 1300km and the battery hours showed as around 200hrs on the Intuvia display. The original owner of the bike used it for about a year before becoming ill and the bike was stored for some period of time (6 months to possibly a year) before my friend bought it. The bike was "asleep" after being stored and was taken to a Bosch dealer before sale and they charged it and updated the firmware.

The battery in the Intuivia display had gone flat so that the bike had to be switched on from the button on the battery. Replacing the battery solved this issue.

Other than that, the bike performed as expected, although we did notice that the Range display (ie remaining distance estimate) showed a noticeably lower value that other bikes with the same system.

A few weeks and a couple of hundred kilometers later, the bike was used for a total distance of around 75km without recharging, at which point the display indicated that the battery was flat (last charge bar showing) and the range was a few kilometers. No power was being delivered to the motor, but the instrumentation was working eg speed and distance.

At this point the battery was put on charge, but failed to charge. There was initially a slow flashing LED on the battery display which disappeared after a while. The display then behaved as if there was no battery fitted to the bike.

The bike was taken to a Bosch dealer who said there was nothing they could do and the battery would need to be replaced (very expensive!)

With nothing to lose at this point, we opened the battery to see what had happened to it.
The battery is arranged as 10 groups of 4 cells connected series-parallel to get 36V. There is a BMS board at the end of the battery pack and each group of four cells has a monitor lead going to the BMS board, as well as the main + / - leads.

Of the 10 groups of cells, six groups have failed completely ie they will not take a direct charge at all and are effectively open circuit.

It is interesting to speculate on how and why the failure of so many cells has occurred, apparently quite suddenly, as the battery had been delivering power right up to the end.

But the bottom line is that buying a second hand e-bike (Bosch system or otherwise), particularly one that has been stored unused for a while, is probably fraught with risk, and a battery replacement should be factored into the price. (No real surprise I guess)

For the Bosch system, the (well known) failure of the display battery, as well as the reduced value of the Range display, are probably good indicators that there are likely to be major issues with the main battery.

I am also suspicious that the Bosch BMS (and others) may not protect the battery very well from excessive discharge. It might be worth avoiding getting down to the last bar on the battery charge display.

Comments and suggestions welcome.

T

 

[neptronix]

I have one at half charge and the BMS just died and there's no physical indication of any damaged components on the BMS, and every cell is between 3.8 and 3.9v.

 

[tlg]

You might be lucky enough to get a BMS board from someone (in the country) who has one with failed batteries.

 

[Hillhater]

You might be lucky enough to get a BMS board from someone (in the country) who has one with failed batteries.

another case of Battery Murdering Systems !
these packs are the ones that member Tumich was/is selling as bare packs without the BMS.
https://endless-sphere.com/forums/viewtopic.php?f=31&t=61608
$125 per 10s, 4p pack.
I have had 2 of those for a few years now, one of which has never been used,..just stored.
It still reads 3.65 v/cell, as delivered, without ever being recharged or balanced.
Ditto for the 2nd pack which is used infrequently ..maybe 25 cycles..and only ever bulk charged.
but the cells have stayed in balance with no need to adjust any.
My cells are all Sanyo GA 3500.
Every commercial pack i have had problems with, has been caused by the BMS.

 

[amberwolf]

FWIW, there are "laptop case splitters", which are generally a plastic (ABS, nylon, etc; something slightly flexible in thinner sections but fairly stiff in thicker ones) rectangle an inch or two in width, and a couple inches long, around 1/4" thick, tapering to a wedge at one end. The wedge is usually at around 45 degree angle, and these are used to pop case latches that look pretty much like the ones on this case.

They're also used for LCD monitor cases, and other similar things.

Toshiba, Compaq, HP, and IBM supplied these to techshops when I did that sort of thing earlier this century and last. Apple used to use them for the old iMacs, first the bondi-blue CRT ones, then later the white LCD types. Dunno what they use now.

For a while with some of them, they had a tool that looked like a very wide paint scraper; stainless steel, pretty thin, long handle for leverage. Width was to pop all the latches on one side at the same time, to lever the case open without damaging it, and not let any of the latches slip back closed. Some cases required two or even three of them used at the same time, like tire levers. Insert and set one with latches open, then go to the next side around and do that, then the third side and then it'd hinge open on the fourth side's latches and you could put all three tools away.


This image search has an assortment of similar types to those described above, as well as some I haven't seen before (and the usual irrelevant garbage).
https://www.google.com/search?q=laptop+case+opener&btnG=Search&newwindow=1&tbm=isch

 

[tlg]

These cases are seriously tough and it is difficult to even get a gap to insert a tool.
I have the standard tools for laptops etc, but they weren't any use.
There is only a narrow gap between the case and the batteries inside, and putting a metal tool in there is courting disaster...

 

[amberwolf]

In that case, then as long as you know the length and location of the tabs, you could use a tiny drillbit to make a hole big enough for a pin or paperclip or similar, that lets you loosen the clip without damaging the case in any other way.

If the holes compromise water resistance, you can fill them with something afterward.


If they really intended the case to be unserviceable without damage, they'd've used ultrasonic welding to seal it up, as that would also greatly increase the waterproofing of it. Since they didnt', it's likely that there is a method or tool designed to open the case.

If the tabs are in the right place, it's possible they might even have a fixture that squeezes the tab-side of the case inward on the sides, while lifting the other side away from it. Possibly squeezing that other side by the ends (if there are no tabs there), to force it's sides even further away from the tabs.

Or just use such a fixture to squeeze it enough for the "prybar" tools to go in there and release the tabs.

 

[tlg]

For the record, the centreline of the tabs are located at the following distances from the rear edge of the housing (ie at the case join above the plug):

57 mm
106 mm
157mm
207 mm

The bottom edge of each tab is located 25 mm below the case join and there is 5mm of tab above this to push on (below the hole in the tab).
There is only 2-3mm gap between the tab and the cells, so drilling will require care.

 

[amberwolf]

Yes, drilling would require care anyway, as you wouldn't want to drill any further than the thickness of the outer case. Drilling into the tab itself would defeat the purpose of the drill hole.

Putting a straw or other tube (couple layers of heatshrink) over the drill bit, covering all of it past the chuck except the 1-2mm needed to get thru the casing would be one way to prevent a drill-thru.


(drilling thru the battery would negate the need to continue any further, as it might just turn into slag anyway. )

 

[stlu]

Hi
While biking, a battery of this type, fell down from its holder, obviously not locked since there are no problems to mount it now.

The interesting part is that the battery does not work anymore, not even the test button.
The battery has not been used many times and had a good status, about 55 km after charging, before it fell down. I have no warranty since we bought the bike for more than three years .
So we decided to open it, thanks for the photos and instructions above. All cells seems to be ok an the voltage is about 38V.
While studying the PCBA I found a broken IC, probably generating 5V or 3.3V for the MCU and other parts of the electronics. The problem is that I can't read out the part number since a part of the package is missing.
We would appreciate a lot if some one can help us to identify the IC, I hope we can repair it by changing this IC.
View attachment 1
I think the IC is from Texas Instruments, on top row it is marked with 41 then the TI logo
Second row 5**60, ** represents the missing characters.
I think it can be TPS57160 or TPS54160 or...


Many thanks in advance to the one who can help us.

 

[amberwolf]

FWiW, when chips blow their guts out like that, it usually means a severe over current (high load) or over voltage (high input) occured, which usually means a short somewhere else in the circuit before or after that part.

Impacts can often cause momentary shorts between things like cell casings and wires during the impact, that may not remain shorted after impact, but still leave damaged insulation that makes it much easier for the same short to occur later from vibration or other environmental conditions, but which can then remain shorted and cause more damage (sometimes fires), instead of being just a momentary short like during the impact.

So I would recommend carefully examining every potential shorting point everywhere within the pack, wherever wires or other conductors cross each other or come near anything metal, etc.

 

[tlg]

situ

I had a look at the IC on the board in my battery unit.
This is what appears to be written on the IC:

41 (TI)
57160

I found a reference to the Texas Instrument TPS57160-Q1 chip here:

http://www.ti.com/lit/ds/symlink/tps57160-q1.pdf

It is described as:
The TPS57160-Q1 device is a 60-V 1.5-A step-down
regulator with an integrated high-side MOSFET.

Good luck with the repair. Please post back and let us know how you go.

As an alternative approach, you may be able to get hold of another unit where the batteries have failed and take the board from that.

T

 

[stlu]

Thanks to both of you!
I will carefully study the battery pack and also measure components nearby the regulator. I hope I will not find any other problems. I will also test the board using a Power Supply to be able to find other broken components without damaging the replaced IC.

Regarding the IC, I removed the conformal coating and I found out that the last letter/digit is not a 0) (Zero) but a Q. So I think the code written on the IC is 5716Q which is TPS57160QDGQRQ1.

I will order tomorrow and exchange the IC next week.

 

[Speady]

Thanks to both of you!
I will carefully study the battery pack and also measure components nearby the regulator. I hope I will not find any other problems. I will also test the board using a Power Supply to be able to find other broken components without damaging the replaced IC.

Regarding the IC, I removed the conformal coating and I found out that the last letter/digit is not a 0) (Zero) but a Q. So I think the code written on the IC is 5716Q which is TPS57160QDGQRQ1.

I will order tomorrow and exchange the IC next week.

You are right

 

[flippy]

for the people interested i got a brand new one open, also rated 400Wh.
NCR186650PF cells, 10S4P.
conformal coating on the BMS is -as usual for bosch- a utter shitshow.
also no drain holes, they put all their money on the 1mm rubber edge they put on the edge of the case. it will fill up with water/condenstation wich will cause poorly coverd parts of the BMS to oxidize and kill it.
BMS is also on the bottom of the case so that is some shit engineering, it will flood the bms if any water comes in....
BMS is basically identical from the older version just smaller and mounted upside down, so opening the pack and you get mooned by the PCB.


https://photos.app.goo.gl/TvqHzPbc6cwB7ikB7

 

[tris11]

Hi, I have a battery pack which does nothing. No lights, no charging etc.

I have now opened the battery up, I cannot see anything untowards.

Whats the best process to test the batteries & the BMS board please

Thank you

 

[flippy]

measure voltages of each group.
there is probably a group below 2.5V.
charge the groups manually beyond 3V and make sure they keep floating above that when disconnecting the power supply.
if the board does not come back with a original charger hooked up (and pressing the butting for 30+ seconds) then bin it, it's junk at that point.