Automotive Current Sensing Auto Switch Project

Thread Starter

Hectors Dad

Joined Oct 3, 2018
6
Firstly, Hi Everyone.
I am totally new to this amazing site and look forward to trying to be part of the community.

So with the above in mind, I would like to ask for some help with a project please.

I have VERY little experience within this field but am a quick study so please bear with me if I don't use technical descriptors and jargon.

I have a dashcam in my vehicle that is powered from a purpose designed battery. This enables me to park for several days and still have my dashcam active without discharging my vehicle battery.
My problem is that by the time I go to use my car, the battery running my dashcam is invariably flat. This means that my dashcam wont power on until I have been driving sufficiently to charge the dashcam battery. This can take up to 10 minutes.

I would like to design a circuit that will sit in between the Dashcam and the dashcam battery that allows the following to occur:

On switching on the ignition, the dashcam is powered from a switched ignition source of between 11 - 15VDC. Upon the dashcam battery reaching a sufficient charge (12VDC) the circuit will auto switch to the dashcam battery input.
When the vehicle is switched off the dashcam will draw power from the dashcam battery until it reaches a critical level and then will disconect. It will then auto reset ready for when the ignition is switched back on again to repeat the process (if required as the dashcam battery may NOT be flat).

The switching between both sources would need to be seamless and appear to the dashcam to be uninterupted as it would switch itself off and back on therby interupting recording.

Long winded I know but I presume it would be feasible and would anybody be able to steer me in the right direction (whilst remembering my knowledge level).

Any help would be gratefully received.

Regards,

Simon
 

ebeowulf17

Joined Aug 12, 2014
3,307
Firstly, Hi Everyone.
I am totally new to this amazing site and look forward to trying to be part of the community.

So with the above in mind, I would like to ask for some help with a project please.

I have VERY little experience within this field but am a quick study so please bear with me if I don't use technical descriptors and jargon.

I have a dashcam in my vehicle that is powered from a purpose designed battery. This enables me to park for several days and still have my dashcam active without discharging my vehicle battery.
My problem is that by the time I go to use my car, the battery running my dashcam is invariably flat. This means that my dashcam wont power on until I have been driving sufficiently to charge the dashcam battery. This can take up to 10 minutes.

I would like to design a circuit that will sit in between the Dashcam and the dashcam battery that allows the following to occur:

On switching on the ignition, the dashcam is powered from a switched ignition source of between 11 - 15VDC. Upon the dashcam battery reaching a sufficient charge (12VDC) the circuit will auto switch to the dashcam battery input.
When the vehicle is switched off the dashcam will draw power from the dashcam battery until it reaches a critical level and then will disconect. It will then auto reset ready for when the ignition is switched back on again to repeat the process (if required as the dashcam battery may NOT be flat).

The switching between both sources would need to be seamless and appear to the dashcam to be uninterupted as it would switch itself off and back on therby interupting recording.

Long winded I know but I presume it would be feasible and would anybody be able to steer me in the right direction (whilst remembering my knowledge level).

Any help would be gratefully received.

Regards,

Simon
Do you have a datasheet or detailed manual for the dashcam? It would be helpful to know what range of voltages it can accept. Presumably it's "12V," but that generally encompasses a range, maybe 10.5-14.5 or something. Then again it might be pickier than that and expect well regulated input voltage.

Depending on the acceptable voltage ranges, this might be as simple as adding a few well placed diodes.
 

Thread Starter

Hectors Dad

Joined Oct 3, 2018
6
Hi,
And thanks for replying.
The datasheet I have just says 12VDC with a 4.2W mean consumption.

It is a Thinkware F800 Pro if that helps

Regards,

HD
 

ebeowulf17

Joined Aug 12, 2014
3,307
Hi,
And thanks for replying.
The datasheet I have just says 12VDC with a 4.2W mean consumption.

It is a Thinkware F800 Pro if that helps

Regards,

HD
Ok, I'll do a little brainstorming on two options - one that's simple, but with less voltage regulation, and another with tighter control, but more parts.

I'm busy right now, but I'll try to play around with simulations this evening.
 

ebeowulf17

Joined Aug 12, 2014
3,307
After a little googling, I discovered that they have an option to hardwire directly into the car's fuse box. This means the unit's power input can definitely handle whatever voltage the car battery puts out (my fear had been that there was some sort of regulation happening in the power adapter, and that bypassing it could be bad news, but clearly that's not the case, so that simplifies things.)

One option will simply require two diodes, but will mean battery run time is reduced due to the voltage drop in one of the diodes. The other option I'll need to simulate to make sure I've got it right. It will use a MOSFET to handle switching with less wasted power. I'll update after simulating.
 

ebeowulf17

Joined Aug 12, 2014
3,307
OK, the sims seem to confirm that I've figured out the MOSFET arrangement I had in mind. This concept is a little new to me - I borrowed it from something @crutschow shared as a reverse voltage protection circuit with minimal voltage drop. Hopefully he can take a look at what I've done and make sure I haven't done something silly here.

Anyway, I've attached two images. The first one is simply using two diodes so that the car power (~12V with key in "ACC" position, ~14.4V with engine running) powers the camera directly when it's on (through D1) and the rechargeable camera battery powers the camera any other time through D2. The diodes simply prevent voltage from back-feeding through unintended paths (if you just tied the car power and battery power together, the battery would try to power all car accessories as soon as the engine turned off.) The downside of the simple diode circuit is that you lose some voltage through the diode, which means that the camera will detect a low-voltage condition and shut down that much sooner. You can use Schottky diodes with low Vf specs to minimize this loss, but it will still be there.

The second image shows a circuit which still uses the same diode between the car power and the camera (because losses there aren't hurting us,) but uses a P-MOSFET instead of the 2nd diode. The MOSFET is configured such that when car power is on, the MOSFET is inactive, preventing backfeeding of the car's 14.4V power directly into the camera battery. When the car is off, the MOSFET turns on, providing power for the camera with essentially no wasted power, so the camera will run longer before shutting down.

The schematics look busy, but most of what's there is simulating existing components. The only new parts are either 2 diodes for the first plan, or 1 diode, 1 MOSFET, and 1 resistor for the second plan. Ignore the part numbers in the schematics - I only have standard library models, so I just use whatever I can find for proof-of-concept. To choose parts in real life, we'd want to select parts with appropriate heat dissipation, low Vf and low RDS-on.
MOSFET-diode-OR.png
MOSFET-diode-OR_2.png
 

Attachments

Thread Starter

Hectors Dad

Joined Oct 3, 2018
6
Hi again,
Apologies for the delay in my response, I am in the UK and so working at GMT +1.

The second circuit using the MOSFET looks to be the best in terms of preserving battery power.

The specs for the battery I use is:

Type: LiFeP04 which is 12.8V / 9Ah
Input: 12V - 14.6V DC @ 10A or 15A
Output: 12.8V DC @ 2A
Capacity is: 115.2Wh

This gives me about 30 hours recording time before shutoff.

Power feeds are currently taken from the rear fuse box on the car giving me 1 x Ignition switched +Ve and 1 x constant +Ve.

So it looks like it could be as simple as a MOSFET switched by the ignition switched +Ve feed to switch over to the battery when ignition is off.

Huh - I thought it would be considerably more complicated than that. I suppose it is just a case of working out the MOSFET type and rating and building a sample.

Thanks for you hard work on all this for a stranger (me), it is much appreciated.

Regards,

HD
 
Last edited:

ebeowulf17

Joined Aug 12, 2014
3,307
Starting to look at parts selection: it looks like diode selection isn't terribly critical. Standard picks will work.

The 1N4001-1N4007 series should be ok. If your camera draws ~4.2W max, that would be 350mA at 12V. Forward voltage should be less than 1V, so dissipation at the diode will be 350mW or less. The thermal resistance of 100K/W means that the diode will run at up to 35 degrees C above ambient temp under that load. I'm pretty sure that's all fine.

I had been thinking of suggesting a Schottky like the 1N5819 which has lower Vf and better heat dissipation specs, but the reverse voltage blocking specs are lower on those, and I don't think there's a heat problem regardless.
 

ebeowulf17

Joined Aug 12, 2014
3,307
One thing I forgot to consider until just now is that automotive environments are notorious for noisy power, voltage spikes, etc. and MOSFET gates can be rather sensitive.

We may need some more parts to provide protection to the MOSFET gate. Hopefully someone with more expertise than me can help with that... I'm still learning and I have low confidence in my ability to make sure I've got that part totally right.
 

ebeowulf17

Joined Aug 12, 2014
3,307
I think I followed MOST of that (newbie remember).

HD.
Ha! No worries. I tend to over explain, partially in the hopes of educating, but also because talking through the entire thought process helps me check my own work, and makes it easier for anyone else reading this thread to catch my mistakes. If you get something useful from it, great! If not, ignore it.

I'll do some MOSFET shopping as time permits, and hopefully I'll get my head around the noise issues just a little more while contemplating MOSFET specs.
 

ebeowulf17

Joined Aug 12, 2014
3,307
Any chance I could get a double check from some experts? @crutschow , @ebp , @OBW0549 ?

Referring to the second schematic in post 7, I think I've got the MOSFET and pull down resistor configured properly, relying on the body diode to pass voltage from drain to source and turn the MOSFET on, but this little "trick" is still new to me.

Also, I'd appreciate any advice on handling automotive electrical noise and protecting the MOSFET. My gut feeling is that either a Zener or a TVS diode in parallel with the existing diode will protect against negative voltage spikes. I could also see using a small cap there.

Any thoughts? Thanks in advance!
 

Thread Starter

Hectors Dad

Joined Oct 3, 2018
6
Any chance I could get a double check from some experts? @crutschow , @ebp , @OBW0549 ?

Referring to the second schematic in post 7, I think I've got the MOSFET and pull down resistor configured properly, relying on the body diode to pass voltage from drain to source and turn the MOSFET on, but this little "trick" is still new to me.

Also, I'd appreciate any advice on handling automotive electrical noise and protecting the MOSFET. My gut feeling is that either a Zener or a TVS diode in parallel with the existing diode will protect against negative voltage spikes. I could also see using a small cap there.

Any thoughts? Thanks in advance!
Hi again and thanks for your continued work on this.
I just wanted to chip in and say this is now WAAAAY beyond my comprehension so can't really have an input here - didn't want you to think I was just resting on my laurels

HD
 

OBW0549

Joined Mar 2, 2015
3,566
Any chance I could get a double check from some experts?

Referring to the second schematic in post 7, I think I've got the MOSFET and pull down resistor configured properly, relying on the body diode to pass voltage from drain to source and turn the MOSFET on, but this little "trick" is still new to me.
I haven't been following this thread so I'm unclear what you are trying to achieve here, and I don't want to go back and slog through the whole thing to find out; but I have to wonder why you've got R1 hardwired across R3.
 

ebeowulf17

Joined Aug 12, 2014
3,307
I haven't been following this thread so I'm unclear what you are trying to achieve here, and I don't want to go back and slog through the whole thing to find out; but I have to wonder why you've got R1 hardwired across R3.
Sorry, most of what's in the dashed-line boxes is just for simulation. R3 doesn't exist in reality - it was just there for simulation purposes so that when the switched 14.4V line turns off, it reads 0V instead of holding a residual charge and showing 14.4V forever.

Of course, after I added R1, I suppose I could've removed R3...

As for the purpose of the circuit, it's just powering a camera from two separate power supplies: the vehicle's 12V switched (accessory) power, and a rechargeable battery that gets charged when the car is running. Of course it could be done with two diodes, but the idea with the MOSFET was to eliminate the voltage drop so that the camera lasts longer on the rechargeable battery after the car is turned off.
 

OBW0549

Joined Mar 2, 2015
3,566
As for the purpose of the circuit, it's just powering a camera from two separate power supplies: the vehicle's 12V switched (accessory) power, and a rechargeable battery that gets charged when the car is running. Of course it could be done with two diodes, but the idea with the MOSFET was to eliminate the voltage drop so that the camera lasts longer on the rechargeable battery after the car is turned off.
It seems to me that both the vehicle power and the rechargeable battery are going to be in the neighborhood of ≈14V, so my choice would be to supply the camera through two Schottky diodes, one from each power source. Using Schottky diodes with a high current rating will minimize the voltage drop.
 

ebeowulf17

Joined Aug 12, 2014
3,307
Well, dang. I thought saving half a volt might translate into a lot of extra runtime with a voltage that was ramping down steadily (like lead acid batteries.) But now I just looked at discharge curves for LiFePO4 batteries, and it looks like voltage stays fairly high until the end, then drops suddenly.
6E2A5C44-96B0-4B86-A2F2-8E255A15F786.jpeg
https://www.batteryspace.com/prod-specs/9055.pdf

Based on that, the small voltage difference might not matter, and the two diode solution might make more sense.

I was kind of excited to figure out another applicator for the MOSFET trick, but maybe it's not that helpful here after all.
 
Top