Removing 12V DC voltage fluctuation (Capacitor choice?)

[dcg9381]

This is an oddball one... Chasing a problem with a back-up camera on an RV for a family member. These connect to trailer running lights. Camera wouldn't power up on the trailer, took it down, powered it up with a battery just fine. Hooked it back up to the trailer, got nothing, check with DVM, new RV wiring was bad, fixed wiring. Then I could hear the camera click every second or so, but it still wouldn't work.

Got out the scope and I found that the trailer light 12V signal dips on a pretty substantial frequency (~2 times second or so), but appearing to last .25mS. It's dropping so much that it's shutting the camera down, which explains the "click" (camera was stopping and starting).




Chased it upstream, this is caused by 2024 BMW hybrid X7 "trailer wiring". The pulse is present at both the 7-way and 4-way trailer connectors. Running the BEV gas motor doesn't change it. It does seem to bother trailer incandescent or LED lighting, so there is that...

So most of my board level knowledge was forgotten and I think I just need to reduce or remove this little blip, which I "think" I can do with a capacitor in parallel?

I think it's:

C= I * t/V

C=capacitance (farads)
V=voltage drop (volts), I know it will function at 11V, so, lets say this is 3V.
t = time (seconds), which looks like 2 mS
I = amps, 2 amp (SWAG)

3V * 2/1000 (second) * 2A = .012F Capacitor, rated for 12-18V?

That seems to be a pretty big and somewhat unavailable capacitor. Suggestions? Anything that "buffers" power would work here...

 

[pancho400cid]

No comment on the capacitor value. Sounds like grounding issue could be possible? Check the trailer grounding wiring, and how well connected everything is. If the trailer is only grounded through the ball, add a hard-wired ground to connect it to the pull vehicle through the trailer power plugs.

...

 

[CoogarXR]

You could always go with a giant car audio capacitor, they are like 1F and rated for mobile applications.

 

[dcg9381]

No comment on the capacitor value. Sounds like grounding issue could be possible? Check the trailer grounding wiring, and how well connected everything is. If the trailer is only grounded through the ball, add a hard-wired ground to connect it to the pull vehicle through the trailer power plugs.

It's not grounding. It's present at the BWM trailer plug with the trailer disconnected, both 7-pin and 4-pin connectors. I'm probing off the trailer ground at the harness. Apparently BMW uses some sort of pulse width modulation to monitor bulbs.

You could always go with a giant car audio capacitor, they are like 1F and rated for mobile applications.

Yes, but they are quite large in size, not something I can stuff into the existing wiring harness unless I want to cut it up front.

The "off the shelf" solution seems to be what's called a "buck converter" which stabilizes unstable DC:
https://www.sparkfun.com/products/18376

 

[tool_scrounge]

For $16 I would go with the dc-dc converter. The question is will you get errors from the BMW bulb monitor?

 

[dcg9381]

For $16 I would go with the dc-dc converter. The question is will you get errors from the BMW bulb monitor?

Yep, maybe one over 3A though unless some EE can confirm my capacitor and suggest one. I think the answer to your question depends on how the DC-DC converter works internally.

 

[nadogail]

Capacitors are rated for Maximum Voltage, so a higher voltage rating should not be a problem.

 

[tool_scrounge]

The pulse in the scope trace looks a bit like a discharging capacitor. It makes me wonder if BMW has a switch upstream from a large capacitor and switches off the voltage while monitoring the capacitor discharge voltage to decide if the downstream Bulbs are working. If so, adding additional capacitance for the back-up camera may result in false positives for bulb failure.

Have you searched the BMW forums for this issue? i realize it is a new model year so others may or have seen it.

 

[dogdog]

I would agree with post #8 go find another source of constant power to power the camera. The camera didn’t shut down because of a bleep or dip in voltage... if PWM is send to the lights, you should be able to confirm that with your scope...should have PWM measurement in % some where in frequency measurement. and just use that PWM as a trigger signal. Besides that, tapping into tail light that gets PWM might overload something upstream ?

with something like these things or concoct something yourself with transistors/mosfets

https://www.amazon.com/Adjustable-Signal-Trigger-Switch-Module/dp/B08GZ1RHWT?tag=atomicindus08-20

 

[walta]

Taillight circuits are very different than they use to be.

Is the pulse present on pin 4 of the 7 way plug +12 aux connection?

Walta

 

[rlitman]

The pulse in the scope trace looks a bit like a discharging capacitor. It makes me wonder if BMW has a switch upstream from a large capacitor and switches off the voltage while monitoring the capacitor discharge voltage to decide if the downstream Bulbs are working. If so, adding additional capacitance for the back-up camera may result in false positives for bulb failure.

Have you searched the BMW forums for this issue? i realize it is a new model year so others may or have seen it.

I don't see a sawtooth there, but if it were, it's probably the cap in the camera power supply.

Yep, maybe one over 3A though unless some EE can confirm my capacitor and suggest one. I think the answer to your question depends on how the DC-DC converter works internally.

First, the capacitor voltage would need to be the alternator voltage at a minimum, so no less than 15V would be safe.
Second, look for a supercapacitor. These are usually 2.7V, but come in series assemblies rates at 16V made specifically for automotive purposes. Like this:

https://www.amazon.com/SaiDian-Capacitor-Module-Capacitors-Protection/dp/B08CGXJY2J?tag=atomicindus08-20

However, if you are holding the line voltage up, you can rest assured that the bulb sensing will think your bulbs are out.

If it were me, I'd use a bridge rectifier between the vehicle and the supercap/camera. That way, the vehicle voltage can dip, but the camera won't see the dip. You don't actually need a bridge; two diodes would work also, but a bridge is cheaper.

 

[theoldwizard1]

Look up "pi filter".

 

[dcg9381]

Have you searched the BMW forums for this issue? i realize it is a new model year so others may or have seen it.

Not my overly-complicated BMW. I do know that that use PWM to control their LED lights and do some other fancy stuff. Here is another post with the same issue related to a back up camera powered by back-up lighting:

https://www.bimmerfest.com/threads/reverse-light-wiring.838994/

I think I'm going to split the trailer harness and do a DC-to-DC converter.

 

[dcg9381]

I would agree with post #8 go find another source of constant power to power the camera.

If it was on the car, a relay would be a way to solve it. The nearest "constant" 12V source in the RV is at the other end and is built into an inverter source. This RV has a fancy 4000 watt inverter that powers both 120V and 12V draws which has to be "on" for the RV to have any power. I think my solution will be to try a DC-to-DC converter that's designed to produce cleaner power.

The camera didn’t shut down because of a bleep or dip in voltage...

I disagree, look at the scale on that dip, it's pretty much "off" for a fraction of a second.

if PWM is send to the lights, you should be able to confirm that with your scope...should have PWM measurement in % some where in frequency measurement. and just use that PWM as a trigger signal. Besides that, tapping into tail light that gets PWM might overload something upstream ?

You might be right. I didn't see the size of the dip changing and I didn't spend time trying to lock on to it's specific frequency.
This is signal that BMW is using on their trailer connectors, I assume it's got circuit overload, but my guess is that it's also what they are actually sending to the vehicle running lights as some sort of "probe". I'm not worried about it overloading something - this is factory trailer 7 and 4 pin trailer connection.

with something like these things or concoct something yourself with transistors/mosfets

https://www.amazon.com/Adjustable-Signal-Trigger-Switch-Module/dp/B08GZ1RHWT?tag=atomicindus08-20

This is essentially a "smart relay"? It requires a 12V source, which again, is at the other end of the trailer. BMW guys indicate a regular automotive relay will work (as they are slow). Trying to avoid that for now. I guess I could add this at the other end of the trailer splicing the trailer harness to get close to the house battery.

 

[Bad Habit]

I had to put in a SSR (solid state rely) in order to trigger off-road lights from the high beams on our Jeep. A typical mechanical relay would just chatter because of the PWM.

Similar to this, pretty cheap.

https://www.amazon.com/SSR-25DA-3-32VDC-Output-24-480VAC-Plastic/dp/B08GPJ1V2J/?tag=atomicindus08-20

 

[theoldwizard1]

I think I'm going to split the trailer harness and do a DC-to-DC converter.

Something to try that is not too expensive.

Get 5 2.7V 1F capacitors. Hook them in series. That should be you about a 13.5V 1F capacitor.

Biggest problem may be, that will look like a dead short the first time you power it up.

 

[AffableCurmudgeon]

This is an oddball one... Chasing a problem with a back-up camera on an RV for a family member. These connect to trailer running lights. Camera wouldn't power up on the trailer, took it down, powered it up with a battery just fine. Hooked it back up to the trailer, got nothing, check with DVM, new RV wiring was bad, fixed wiring. Then I could hear the camera click every second or so, but it still wouldn't work.

Got out the scope and I found that the trailer light 12V signal dips on a pretty substantial frequency (~2 times second or so), but appearing to last .25mS. It's dropping so much that it's shutting the camera down, which explains the "click" (camera was stopping and starting).




Chased it upstream, this is caused by 2024 BMW hybrid X7 "trailer wiring". The pulse is present at both the 7-way and 4-way trailer connectors. Running the BEV gas motor doesn't change it. It does seem to bother trailer incandescent or LED lighting, so there is that...

So most of my board level knowledge was forgotten and I think I just need to reduce or remove this little blip, which I "think" I can do with a capacitor in parallel?

I think it's:

C= I * t/V

C=capacitance (farads)
V=voltage drop (volts), I know it will function at 11V, so, lets say this is 3V.
t = time (seconds), which looks like 2 mS
I = amps, 2 amp (SWAG)

3V * 2/1000 (second) * 2A = .012F Capacitor, rated for 12-18V?

That seems to be a pretty big and somewhat unavailable capacitor. Suggestions? Anything that "buffers" power would work here...

Rather than trying to stabilize the pulse, use the existing harness connector to drive a mechanical relay and use that relay to power your camera. With the frequency you stated, it may give you stable and clean power. Test it out using a relay and a battery.

 

[dave*99]

If it were me, I'd use a bridge rectifier between the vehicle and the supercap/camera. That way, the vehicle voltage can dip, but the camera won't see the dip. You don't actually need a bridge; two diodes would work also, but a bridge is cheaper.

I'd try this. The diode isolation between the camera and the "off pulse" may be all you need.

In looking at the oscilloscope photo...... I wonder if the BMW is not only interrupting the power but clamping the line to ground??? Thereby pulling the camera power lead down as opposed to letting any internal filter capacitance in the camera power supply support the camera for a few mS. The diode or bridge rectifier would change that. And adding capacitance at the camera terminal will help if the rectifier does not fix the problem.
YMMV

 

[dcg9381]

First, the capacitor voltage would need to be the alternator voltage at a minimum, so no less than 15V would be safe.
Second, look for a supercapacitor. These are usually 2.7V, but come in series assemblies rates at 16V made specifically for automotive purposes.

There we go... that gets me desired capacitance.

If it were me, I'd use a bridge rectifier between the vehicle and the supercap/camera. That way, the vehicle voltage can dip, but the camera won't see the dip. You don't actually need a bridge; two diodes would work also, but a bridge is cheaper.

This is the device I think I was forgetting and looking for, just a regulator/rectifier should do it.. Seems like I can get a pretty small one!


Thanks for the help,guys.

 

[Innovate1]

I would put a diode and then a capacitor for the camera. Use a 3A diode for some margin - they are common. Make it at least 100V for any transients - probably overkill but they are cheap. The diode will allow the existing system to function for whatever the pulses are for. Is the camera really 2A. Seems like a lot but not that familiar with this stuff. You could measure the current of the camera when running directly from the battery. May not need as big a capacitor as you think. It looks like it lasts about 1 to 2 mS. Your calculation is wrong - you divide by voltage drop not multiply so the capacitor is about 10 times smaller than you calculate in the original post. About 1200 uF. That's not going to be hard to find.

 

[Innovate1]

If it were me, I'd use a bridge rectifier between the vehicle and the supercap/camera. That way, the vehicle voltage can dip, but the camera won't see the dip. You don't actually need a bridge; two diodes would work also, but a bridge is cheaper.

Diode is a good idea but no reason for more than one. Just one in the positive line. Need a capacitor after the diode to reduce the dip.

 

[dogdog]

This is essentially a "smart relay"? It requires a 12V source, which again, is at the other end of the trailer. BMW guys indicate a regular automotive relay will work (as they are slow). Trying to avoid that for now. I guess I could add this at the other end of the trailer splicing the trailer harness to get close to the house battery.

I don’t own a bmw so can’t say for sure. But a PWM or buck converter usually resets when it is being overloaded as part of the circuit protection I guess, can’t say about the bmw but 100% sure on the house hold chargers.. Now as far as the capacitor solution.. might not work if the source is not able to support the load. PWM as far as I understand is the period of on/off time pulses so… it’s not exactly a frequency.

As far as that relay from Amazon. Yes it’s a fancy relay that is requires a constant power and triggered by a signal and have a delay on/off of 0-25 secs In the description.

 

[Worsedog]

If it was on the car, a relay would be a way to solve it. The nearest "constant" 12V source in the RV is at the other end and is built into an inverter source. This RV has a fancy 4000 watt inverter that powers both 120V and 12V draws which has to be "on" for the RV to have any power. I think my solution will be to try a DC-to-DC converter that's designed to produce cleaner power.

I disagree, look at the scale on that dip, it's pretty much "off" for a fraction of a second.





You might be right. I didn't see the size of the dip changing and I didn't spend time trying to lock on to it's specific frequency.
This is signal that BMW is using on their trailer connectors, I assume it's got circuit overload, but my guess is that it's also what they are actually sending to the vehicle running lights as some sort of "probe". I'm not worried about it overloading something - this is factory trailer 7 and 4 pin trailer connection.


This is essentially a "smart relay"? It requires a 12V source, which again, is at the other end of the trailer. BMW guys indicate a regular automotive relay will work (as they are slow). Trying to avoid that for now. I guess I could add this at the other end of the trailer splicing the trailer harness to get close to the house battery.

The 7 pin connector should have +12v to charge the breakaway or house battery. With the reverse signal and power side by side it use the +12 to power the camera and reverse to trigger the relay. The pulse looks short enough that run of the mill automotive relay won't react fast enough to power down the camera. I think you may find the solid state relays are capable of reacting the short pulse.

 

[American Locomotive]

I think you're way overestimating the camera's power consumption. 2A would be crazy, that'd be almost 30W dissipation. The camera would need a big heat-sink, probably with a fan to keep that cool.

The camera is likely using a few watts tops, probably ~150mA or so or less. I bet a simple diode inline with its power supply (to prevent the cap from back feeding) with a 25v 470µF cap is all you would need.

https://www.digikey.com/en/products/detail/rubycon/25ZLH470MEFC10X12-5/3563614

 

[dcg9381]

The 7 pin connector should have +12v to charge the breakaway or house battery. With the reverse signal and power side by side it use the +12 to power the camera and reverse to trigger the relay. The pulse looks short enough that run of the mill automotive relay won't react fast enough to power down the camera. I think you may find the solid state relays are capable of reacting the short pulse.

That's right. I'd have to probe it to make sure BMW didn't put some other digital signal in there somewhere, but if it's clean, that could be used for a power source. I agree that a standard automotive relay wouldn't react fast enough. I got under the trailer today and that 7-pin harness terminates in a box, which I sure is a hot mess, but could house a relay.

I think you're way overestimating the camera's power consumption. 2A would be crazy, that'd be almost 30W dissipation. The camera would need a big heat-sink, probably with a fan to keep that cool.

I can't find any watt or amp specs on most of these wireless China cameras. You might be right, but actual power draw is something I could test. Google says 1-2A, but that could be bogus... They are wifi devices, aluminum body...

 

[Wrench97]

Sounds like a body/lighting module sense circuit, try running it off a 5 post relay using the reverse light wire to trigger the relay and run a power wire and fuse to the supply camera voltage through the relay.
Alliteratively you could check to see if there are options in the module for lighting types incandescent draw more amps then LED for example if set to LED set it to incandescent.

 

[American Locomotive]

I can't find any watt or amp specs on most of these wireless China cameras. You might be right, but actual power draw is something I could test. Google says 1-2A, but that could be bogus... They are wifi devices, aluminum body...

I wouldn't even bother trying to rest, really. Go find some random electronic device like an old computer or something. They almost always have large-value capacitors in the power supply area, in the 200-500µF range. Hack it in with a diode to act as a check valve, see if it works.

 

[BillK]

There is a similar issue with newer GM trucks that is caused by the vehicles trailer brake computer looking for trailer lights. There is an updated controller for the GM stuff but that wont help you. Putting a filter cap in might actually cause other problems because the brake controller or ecm will see something unusual.

I believe your best bet is going to be a dc-dc converter but you still might not be able to wire it into the trailer lights.

 

[CoogarXR]

Just stick a solar panel on the roof and only drive it on sunny days

 

[cycle61]

I would find a source of switched 12v power other than the trailer harness.

 

[dcg9381]

Just stick a solar panel on the roof and only drive it on sunny days

There are "solar" cameras and it was suggested that I install one. I said "no".

I would find a source of switched 12v power other than the trailer harness.

Which would mean patching into the inverter's 4000Ah 12V battery and giving them a reason to deny warranty. Nope.

There is a similar issue with newer GM trucks that is caused by the vehicles trailer brake computer looking for trailer lights. There is an updated controller for the GM stuff but that wont help you. Putting a filter cap in might actually cause other problems because the brake controller or ecm will see something unusual.

I believe your best bet is going to be a dc-dc converter but you still might not be able to wire it into the trailer lights.

Thanks. DC-to-DC is where I'm leaning. I'm familiar with it for adjusting vehicle charge for newer lithium batteries. I have to cut the trailer/vehicle harness, but I think that's OK.

 

[Innovate1]

There are "solar" cameras and it was suggested that I install one. I said "no".

Which would mean patching into the inverter's 4000Ah 12V battery and giving them a reason to deny warranty. Nope.


Thanks. DC-to-DC is where I'm leaning. I'm familiar with it for adjusting vehicle charge for newer lithium batteries. I have to cut the trailer/vehicle harness, but I think that's OK.

A diode and cap is likely to fix the issue and can be put at the camera so no cutting into wiring, just a few parts where you already tapped for power. Your calculation of size in the first post is wrong and about 10 times bigger cap than the proper calculation.
You posted: 3V * 2/1000 (second) * 2A = .012F
Should be 2/1000 * 2A /3V = 0.0013 F or 1300 uF
And that is with the other bulbs in the circuit loading it down and the current estimate is likely high. Both make the actual cap needed even smaller. All a DC/DC converter is going to do is add some hold up time which the cap will do. And you need a DC/DC converter that is 12V in and 12V out - not a common item although they do exist.

 

[SlappyWhite]

Like others noted you should put a diode in series (on positive) then the cap in parallel. If you do not the cap will back feed into the vehicle both after the car is shutdown and also during the blip. If the blip is there for some diagnostic reason it may throw things way off if it back feeds. On shutdown it won't back feed for very long but again you also don't want it to.

It will depend a bit on the diode selected but it will drop the voltage to the camera by around 0.7v.

 

[dcg9381]

A diode and cap is likely to fix the issue and can be put at the camera so no cutting into wiring, just a few parts where you already tapped for power. Your calculation of size in the first post is wrong and about 10 times bigger cap than the proper calculation.
You posted: 3V * 2/1000 (second) * 2A = .012F
Should be 2/1000 * 2A /3V = 0.0013 F or 1300 uF
And that is with the other bulbs in the circuit loading it down and the current estimate is likely high. Both make the actual cap needed even smaller. All a DC/DC converter is going to do is add some hold up time which the cap will do. And you need a DC/DC converter that is 12V in and 12V out - not a common item although they do exist.

Thanks for the math assist, missed that. I'm cutting harness either way, I'd just rather do up at the box for the 7-pin rather than stuff components back in the rear wall of the trailer... At least that's where I'm at right now. It's already cut in both places, really.. RV manufactures just crimp stuff.

The current estimate was just a SWAG on cameras of this type per google. I can find a few that are slightly under 1A.

Appreciate the ideas.