Hello,

I have recently had a crank-mounted cycling power meter go pop when I experimented with using a LIR2032 instead of a CR2032 battery. I figured there would be some sort of voltage regulation going on in there... and it seemed to be working normally for around 5 minutes.
After getting to the circuit board, there was an obvious blown SMD component, which I desoldered and, hey presto, it works again!
I really want to identify the blown component so that I can decide whether I need to replace it or possibly upgrade them so that I can use the LIR2032 batteries. I'm also interested in whether there is some form of voltage regulation on the board. I think its a triple layer board, making following the tracks a bit tricky...

currently I'm thinking the popped component might be a tantalum capacitor of some sort, but I haven't found a manufacturer with components that look exactly like that and the codes seem to vary a bit between manufacturers. Also, could that be an LDO regulator directly beneath it?

many thanks
Dan

 

Hi, after googling LIR2032 for stages, I was wondering the same thing.
Did you ever resolve this issue?
Do you have some more high resolution photos of the circuit board of the power meter? I'm curious what's in there.
If they're running the NRF51422 BLE SoC, which has integrated regular and DC/DC, I'm not sure that they'd have an external regulator since it'd negatively impact battery life. This part has a maximum Vdd of 3.9V, so it's not very Li Ion friendly.

 

Hi Kulivontot,
I didn't see your reply straight away, so hopefully your still somewhat interested!
Unfortunately, while the power meter was in pieces I was unable to test some of the functionality, as it was too delicate to attach the battery, temporarily hold everything in place, and test it on the bike. After I had replaced the blown capacitor and reglued the case back together, it only partially works- the soc, wireless connectivity and accelerometer work, but some part of the circuit that works the strain gauge has broken

I think I have some pics of the other side of the board, it turned out to be only a double layer board and you can figure out most of the tracks, so that might be some help with identifying the components, if you like i could upload them when I get home.

I don't think they used an external power regulator, but considering the battery life I was getting i thought it would be far more economical to simply charge the lir2032 every day, and put up with a bit of extra current leakage.

 

Hey,
thanks for responding. I would be super interested to see more pics of the circuit board and components. Devices like this are rarely given a public teardown, so it's interesting to see how the industry puts things together in a manufactured design.
What I was getting at with the regulator is that most of the chips on the board are not rated to operate above 3.6V. If you're applying a LIR cell, that could be upwards of 4.2V in fully charged condition, which means that the actual SoC's on there may either A) break down in an unexpected manner, or B) start dumping copious amounts of current as ESD protection is activated. In short, without regulation, I expect that very bad things will happen.
I think the bigger problem with the Stages battery design is that the contacts on both the battery door and the PCB are not very robust, so if you were replacing the battery at 10x the rate, you would end up breaking one of the contacts, and get intermittent dropouts (which perhaps you're already getting).

 

Ok, so the pictures I took weren't as good as i'd remembered them, and there aren't any clearer pictures of the front of the board, but you can make out most of the writing on the components. In this picture I've overlayered the tracks from the other side of the board.

Yeah, in the past I had resorted to adding some extra spring to the contacts by bending them a little. Well, this was a mark 1, maybe stages have sorted some of these problems in the newer products.

 

Thanks for posting this. It looks like they're using a TI MSP430 processor and a pair of NRF SoC's for the BLE/ANT+ transceiver. All three chips are internally regulated, so I don't think there is an external LDO on the board as you had initially thought. I think the 3 pin device is probably just an NPN transistor to prevent the board from frying if the battery reverse polarity is applied. The two chips to the bottom right are either flash or EEPROM, although, perhaps one is a temp sensor or accelerometer? I believe the chip on the far left is the ADC, as the built-in ADC on the MSP430 has all its inputs on the top side to which are not routed anywhere on the bottom side of the board (which is where I assume the strain gauges are attached). What's not clear to me is where the antennas are, and how the strain gauges are attached. If the antennas are sandwiched between the bottom layer and the crank arm, this makes sense to me as to why the RF performance of stages is so poor.
In actuality, it's not so complicated a system, I figured there'd be more precision components in it than this but I guess the timescale of cadence is super long as compared to that of a microcontroller. It's a bummer there's not really anything you can modify or salvage on the board. It'd be neat to sniff the ADC data, but I'm guessing at that point it'd probably be less of a hassle to just build your own board at that point.

 

Sorry about reviving this old thread, but I couldn’t find any other pages on this topic.
What was the rated voltage of the LIR2032 battery? Mine are rated 3.7 V but I obviously want to avoid damaging my power meter, so I haven’t tested them yet.

 

According to the data sheet the LIR2032 can be up to 4.2 volts when fully charged.

Les.