Nexus 7 - name that connector?

[siggi]

Hi y'all,

I'd like to play around with my Nexus 7's power supply a bit, put a current shunt in-line and such -  ideally without harming the dang thing. I figured it'd be easy to buy a few battery connectors, but for the life of me I can't figure out what manner connector is involved. The pitch is 1.5mm, six pins, blade-type, e.g. it pulls up from the motherboard. There's lettering on both pieces reading "TYV", in case that means anything to anyone.

iFixit have a decent picture of it: [http://www.ifixit.com/Teardown/Nexus+7+2nd+Generation+Teardown/16072]. I've searched digikey and molex until my eyes bleed, but I haven't found the exact thing.
It sorta kinda looks like Molex Pico-EZmate, but not quite - the latching and mating are quite different.
Can you name this connector, and/or a source for it?

Thanks,
Siggi

[sacherjj]

If they used the same 6 pin connector as the first gen, I didn't find a compatible connector when I tried digikey, etc.  I was able to do what I wanted by replacing the FFC button cable for my power needs.  You might do best to get a cheap replacement battery, just to get the connector.

[TVman]

Apples secret....


Have you ever wondered about why sometimes when you use an unofficial Apple charger, the device says "Charging Not Supported with this Accessory" when in fact... it sometimes still charges?


Or do you know why some outside chargers don't charge all together?

It has to do with the USB. USBs work because they have 4 pins - 1 for the 5V power, 1 for ground, and two for data (D+/D-).

When you use non-Apple devices and plug them into the wall or the computer, these devices are made for the standard 5V that comes standard with most USB cables. What happens is that if you are using the USB solely for power, the device will ignore the two data pins - the D+ and the D-. In that case, it works like any other wire... as it should.

With Apple devices, it's different. The engineers at Apple have made the process a bit more complicated.

For example, when you plug in an iDevice into a wall plug, the standard 5V and grounding still applies... but the device still uses the other two cables...


The Secret Sauce

While most devices ignore the D+/D- cables when only utilizing the USB for power, Apple devices use the D+ and D- as communication cables. The Apple cables have resistances on them (if you recall, resistance controls the Voltage and Amps). The resistances tell the Apple device what it's connected to and in return tells the device how much power to draw. This means that voltage still needs to run across the D+ and D- pins!

This is a cheap way for the Apple device to "guess" what it's connected to. ::)For example, if it thinks it's connected to a wall charger, it will draw more power and thus charge the phone quicker (usually at 1A). If the device thinks you're connected to the computer, it will draw less power (usually at 0.5A) and thus will take longer to charge. The resistances thus are the signals that tell the iDevices what to do.

This neat little trick by Apple has 2 main consequences:

    You can trick the charger into charging at different rates (faster or slower) by tweaking the resistances on the data pins
    A well-designed charger for Apple devices will take these resistances into consideration

Check out this video about Apple's Secret Charging Methods.


Hunor

[zapta]

I'd like to play around with my Nexus 7's power supply a bit, put a current shunt in-line and such ...

I wonder, what are you trying to do?

[siggi]

I wonder, what are you trying to do?

I'm mainly doing this out of curiosity.
I'm curious what is e.g. the range of power consumption, what are the time constants, how much goes on e.g. various radios vs. CPU, vs. backlight, etc. How much energy storage is there in practice in the various step-down regulators and filter/bulk caps, etc.
Does the device's consumption change when it can or can't talk to the battery, when it's powered from the wall.

[siggi]

Apples secret....

Thanks, yes I'm aware that Apple have a secret handshake in their chargers, this has been reasonably well documented for years.

Now, that connector, I've a really hard time believing it's proprietary, though I suppose it's possible. Do you know?

[zapta]

I'm mainly doing this out of curiosity.
I'm curious what is e.g. the range of power consumption, what are the time constants, how much goes on e.g. various radios vs. CPU, vs. backlight, etc. How much energy storage is there in practice in the various step-down regulators and filter/bulk caps, etc.
Does the device's consumption change when it can or can't talk to the battery, when it's powered from the wall.

I am also mobile battery drain curious ;-)

Here is an arduino based unit that measures the battery consumption.

https://www.eevblog.com/forum/chat/what-are-u-working-on-right-now-(electronics-related-and-with-picture-please)/msg438324/#msg438324

It provides 3.8V DC (adjustable) which is connected to instrumented devices with battery removed. The design is optimized for accurately tracking the charge (current integrated over time) but also tracks things like awake vs standby times and charges. The data is reported to a FTDI serial/usb port.  I am working now on a more stand alone one with a small OLED display.

If you want the real deal, this is a very capable product http://www.msoon.com/LabEquipment/PowerMonitor/

(I am not recommending modifying devices or messing with batteries, it can cause fire, injury and death, nor am I an expert for anything or representing any third party).

[Monkeh]

Apples secret....

Very nice C+P of a poor explanation of how Apple handle charger capability. What on earth does it have to do with this thread?

[TVman]

Apples secret....

Very nice C+P of a poor explanation of how Apple handle charger capability. What on earth does it have to do with this thread?

Connectors!

[Monkeh]

Apples secret....

Very nice C+P of a poor explanation of how Apple handle charger capability. What on earth does it have to do with this thread?

Connectors!

No.

[zapta]

Worst case you can get a spare battery and cannibalize the connector.

Sent from my Nexus 5 using Tapatalk

[amyk]

I actually managed to find the schematic but that wasn't too helpful: "WtoB 6P"

It looks similar to JST ADH and JST AUH.

[siggi]

I am also mobile battery drain curious ;-)

Here is an arduino based unit that measures the battery consumption.

https://www.eevblog.com/forum/chat/what-are-u-working-on-right-now-(electronics-related-and-with-picture-please)/msg438324/#msg438324

It provides 3.8V DC (adjustable) which is connected to instrumented devices with battery removed. The design is optimized for accurately tracking the charge (current integrated over time) but also tracks things like awake vs standby times and charges. The data is reported to a FTDI serial/usb port.  I am working now on a more stand alone one with a small OLED display.

Wow, that's a really cool project!
What sort of bandwidth do you get from the coloumb counter, etc?
Since you're reading the data through I2C, this is more a data logger than realtime measurement - e.g. you you'll get series of consumption averages over relatively large time periods I'd guess?

[zapta]

Wow, that's a really cool project!
What sort of bandwidth do you get from the coloumb counter, etc?
Since you're reading the data through I2C, this is more a data logger than realtime measurement - e.g. you you'll get series of consumption averages over relatively large time periods I'd guess?

The design is optimized to track accurately power consumption over time. If my calculations are correct, the  Coulomb counter gives one  tick for about ~0.166 uAH  (with 0.025 ohm shunt) and I read it 10 times a second (could go 50/100hz if needed). This provides the momentary consumption data (which I graph with KST) but the most important and accurate result is the charge over a longer period.

If you want higher frequency, a Coulomb counter is not a good choice because the resolution goes down with high sampling (you get less ticks per sampling interval). The LTC2943 has also a A/D for the current and voltage. Currently I am not using them but it should be a simple firmware change.

If you care only about momentary graph you don't even need to go digital.  A simple current probe connected to an oscilloscope will do. If you need an amplifier you can try hacking Dave's uCurrent to have the shunt resistor you need.