Hey everyone!
Running into some issues with carloop draining my battery… I perform some safety checks in the code and have logged the number of times the device goes into deep sleep based on battery voltage (after calling carloop.update of course) throughout the course of an evening. Everything is as expected compared to my sleep duration.
I do have a gps connected as well as a micro sd card reader. Is there a chance that either one of these devices is draining my battery over the course of 2 days?
EDIT - I just looked up the current draw of the gp-735 module. 37ma. Seems likely that if this is continuously powered then it could pull my battery down over the weekend. Any way to put the GPS module to sleep or switch the power to the jst pin?
From the GP-735 data sheet -
“GP-735 supports the easy power saving control mechanism via a GPIO pin. To control the power of GP-735, connect the PWR_CTRL pin to a GPIO of micro-processor. To cut off the power of GP-735 (VCC is always connected to power source), just pull the PWR_CTRL pin low (in this case, GP-735 only keeps power of RTC and RAM). During normal run, pull it high or leave it floating (in this case, GP-735 is fully powered).”
What pin is pin 6 of the gps connected to on the carloop?
Sorry for not reading the carloop schematic. Looks like I can shut off the GPS by pulling A0 low.
@egarl004,
My apologies for not responding sooner. I do not have the ability to respond from my workplace; I respond when I have spare time at home.
One other thing to look into for current draw is the micro SD card slot. Some micro SD cards consume a lot of power when the card is being accessed. If you are storing data to your memory card frequently, then you may have a higher current draw than you expect.
Are you successfully shutting off the GPS? That looks pretty straight-forward now that you figured out how it should work.
I also want to comment on the general design of Carloop. It was designed to be always on when the car is supplying 12V to the OBDII port. The assumption, which seems to hold true for the majority of cars, is that the 12V to the OBDII port is only ON when the key is in the ignition; sometimes to the accessory position or sometimes only when in the run position. Given that design assumption, the power regulator on the Carloop works nicely with noisy power, but is not necessarily the most efficient.
One final note is that car batteries are not designed for deep cycling; they are not meant to be drained down. The batteries are designed to provide a whole lot of current quickly, which is what is needed to start the engine. If your battery is still functional but on the weak side, then drawing a deeper cycle might mean it can no longer provide starting current from a partially drained state. If nothing else works, maybe get your battery load tested.
Let us know if you get it sorted out or you need further help!
I’ll definitely get it tested. Thanks for the input.
As far as 12v cutoff based on key switch, if you look at a diagram for the obd II standard pin out, it’s unswitched power. I’ve tested this on probably 20 vehicles from Ford, Mazda, Kia, BMW and Chevrolet with the same results (power always on) all of course 2006 and up, however.
Over the weekend I let my car sit and low and behold— dead battery again. I am successful in getting the gps to sleep, however I’m deep sleeping here which means the pins are floating. I just tied it to ground for now. Im going to attempt to find the internal resistance and pull the gps low by getting close to matching that internal resistance so that I still have control of turning the gps on and off programmatically throughout deep sleep.
Im going to explore the sd card component next and I’ll report my findings. My current code only initializes the microsdcard if battery voltage is > 13v.
Again, thanks for the feedback!
I’m going to attempt to wire in a 2N7002 mosfet per particle recommendation for any peripherals. If that resolves the issue, maybe we can include this in whatever new hardware carhop produces?
Alright… so quick update. I did install the 2n7000 mosfet and it definitely works as far as shutting down power to the microsd card. But… battery is still draining. Everything now points to the buck converter. @cyclin_al1
Looking further into the data sheets from carloop and the buck converter. It looks like there is an enable pin. I’m wondering if there is a way I could put a battery at vbat and then have the particle sleep and set the enable pin to low, then on awake, set the enable pin to high. Any thoughts?
@egarl004,
Sorry for taking a while to get back to you.
I was about to say that switching off the enable of the buck converter would remove power from the Photon and so it would not be sleeping … it would be unpowered. However, I realized you planned on providing a battery to keep the Photon alive.
However, if the Carloop is drawing that much current to drain the car battery, then how large of a battery do you need to keep the Photon alive? It would have to be a bigger battery than the car battery.
At this point, I think there is something else going on. Have you been able to measure the current draw to the Carloop? I wonder if there is hardware damage causing excessive current draw?
One final thing that we have not explored is the current draw of the CAN transceiver chip (TJA1049)?
No worries @cyclin_al1
I’ve looked at the schematic some more and had the same thoughts with the enable pin. I can’t see that its terminated to any pin so I’m assuming I’d have to tap into on the board. (could at least be a good safety countermeasure) Today I actually just got some more efficient dc dc converters from digi-key I’m going to try.
3.3 V 1.65W
My next step beyond this will be a power shield from adafruit. So run from the LiPo and charge the LiPo when the car is on.
Thanks for the continued interest.
@egarl004,
Just a heads up about voltage! The Carloop circuit board runs on 5V, not on 3.3V.
The Photon runs on 3.3V, but has its own built-in regulator to convert from approximately 5V to the 3.3V.
The CANBus transceiver chip runs on 5V (VCC), but also has a 3.3V reference pin (VIO) to set the levels for communicating with the photon.
As a result, you might want to look into this one from the same series instead:
https://www.digikey.com/products/en/power-supplies-board-mount/dc-dc-converters/922?k=&pkeyword=&pv1525=62&FV=ffece3f6&mnonly=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25