Battery isolator

Discussions about anything electronics, from altimeters to cameras.

Moderator: Moderators

strud
Engineer
Engineer
Posts: 1033
Joined: Thu Oct 02, 2008 7:37 am
Location: Central Tablelands, NSW

Battery isolator

Postby strud » Sun Jun 14, 2020 10:35 pm

So in conjunction with the Rockettracker thing and the 4" avbay, I have come up with a variation on an old battery isolation method use about 18 years ago on my first rocketry electronics bays/avbays.

There are two boards here, one very simple with only switching/isolation function (with output voltage taps) and the second using the same switching hardware but with the addition of voltage/current and power monitoring of the 3 channels via an I2C interface.

Image

Image

Both are using the 3.5mm stereo socket from RS as the main switching/isolator.

The monitoring IC is from Microchip and is rather nice.

https://www.microchip.com/wwwproducts/en/PAC1934

Has 4 channels (differential input) requiring a precision series resistor in the current path. Is able to measure instantaneous voltage & current, thereby calculating power consumption.
It can also accumulate so you can use it as a battery fuel gauge.

I'm keen to have not just voltage monitoring but instantaneous current consumption for all 3 channels since before you press the big red button, knowing that all systems are consuming the current they should be is another metric to give you confidence.

air.command
Astronaut
Astronaut
Posts: 1988
Joined: Mon May 14, 2007 11:18 am
Location: Sydney, Australia
Contact:

Re: Battery isolator

Postby air.command » Thu Jun 18, 2020 8:20 pm

Very cool Strud, I am not sure how the internal contacts are configured inside the 3.5mm plug, but is there any chance of a cross connect between the contacts as the plug is inserted or extracted?
Crop Circles: ... just a bunch of guys looking for their rockets ....

strud
Engineer
Engineer
Posts: 1033
Joined: Thu Oct 02, 2008 7:37 am
Location: Central Tablelands, NSW

Re: Battery isolator

Postby strud » Fri Jun 19, 2020 8:19 am

Hi AirCommand

I don't think so since the pitch of the conductors is the same and their width is narrower than the pitch.

However, if that is the case I had planned to use a non-conductive rod anyway since the jack will not be long enough when the socket is installed within the avbay behind the switch band.

User avatar
OverTheTop
It's only money...
It's only money...
Posts: 2867
Joined: Sat Jul 21, 2007 2:20 pm
Location: Melbourne, Australia

Re: Battery isolator

Postby OverTheTop » Fri Jun 19, 2020 5:49 pm

Don't forget if there is any risk of the phono plug shorting anything out you could machine up a piece of plastic, like nylon or something, and use that. If you want low-tech I think 1/4" dowel is "made to fit" perfectly. That's what 1/4" phone plugs are about :D
TRA #13430
L3
"Everybody's simulation model is guilty until proven innocent" (Thomas H. Lawrence 1994)

strud
Engineer
Engineer
Posts: 1033
Joined: Thu Oct 02, 2008 7:37 am
Location: Central Tablelands, NSW

Re: Battery isolator

Postby strud » Fri Jun 19, 2020 7:07 pm

Hi OTT

Yes that is my plan anyway, since as stated in my previous post, the phono jack will not reach with it being recessed.

I want to setup a test to see at what acceleration these contacts open, even though I expect it to be very very high.

They will be aligned to open in the horizontal plane (X or Y axis) Z being aligned with the rocket's long axis.

An iteration of this to handle higher currents and be completely vibration/shock immune is to use these contacts to drive a high side protected fet with some extra capacitance to give a time constant of say 1s. That way the isolator only needs to drive enough current to drive the fet (almost nothing) and the debounce is now electrical.

strud
Engineer
Engineer
Posts: 1033
Joined: Thu Oct 02, 2008 7:37 am
Location: Central Tablelands, NSW

Re: Battery isolator

Postby strud » Tue Jun 30, 2020 2:22 pm

I have the PCBs back for these now:

The basic version

Image

Image

Image

Image


The smart version (not assembled as yet)

Image

Image

strud
Engineer
Engineer
Posts: 1033
Joined: Thu Oct 02, 2008 7:37 am
Location: Central Tablelands, NSW

Re: Battery isolator

Postby strud » Sat Jul 04, 2020 11:07 pm

So I've been playing with the 'Smart Power Switch' / 'Smart Disconnector' today and finally worked out why I couldn't get I2C comms working.....

- Board error - no
- soldering error - no
- firmware error / config error etc - no (tested comms with another I2C sensor board and all good)
- Behavior not per data sheet - Bingo!

Image

Finally after trying many things (changing chips, changing board...), it turns out that if you use a 0R resistor/short to pull the address control input to ground, it doesn't work ie I2C comms does not work at all even with the supposed address of 0x10 you are meant to get by pulling to ground with 0R.

I put a 510R resistor there (table lists 499R to give 0x11 address) and everything works......

Anyway, happy to say the default Microchip lib /demo code for a single channel works giving the below debug out when supplying the RocTrak board from an 18650 Li Ion cell (input) and then to the board via a 5V boost converter.

--------------------------------------------------
Product ID: 5B
Manufacturer ID: 5D
Revision ID: 3


Read start:
Voltage (mV) = 4046.88
Vsense (mV) = 1.341248
Current (mA) = 134.124756
Raw Power (HEX) = 6F29A0
Power (mW) = 0.542786
Power Acc (mW) = 35572.031250
Energy (mWh) = 18.618435

--------------------------------------------------

I must say that the Power Acc number is rather curious.... since I would thing Power accumulated should be different units... maybe that should be mW seconds.

Anyway, I will modify the library next to support 4 channels and implement some form of semaphore so that I can use this data in the other tasks running on the ESP32.


Return to “Electronics”

Who is online

Users browsing this forum: No registered users and 10 guests