LTO (Lithium Titanate Oxide) – The Ultimate Battery for Dash-Cam parking mode (DIY)

I got a clamp meter on the charge cord for my battery pack and the battery is indeed pushing about 12A out into the car’s utility port for up to 10 seconds once the ignition is turned on. This backflow behavior does not happen with my SMPS that I tested with, so it’s something specific to the way the car’s electrical system behaves. I’m wondering if the car activates the utility port with low current when the ignition is turned on, but holds back on delivering any real current for a few seconds until all of the car’s electrical systems are energized, run through their power on self tests, and things stabilize. This seems to take about 10s when the car has been off for 8-10hrs, but takes closer to 2-3s when the car has only been off for a few minutes. This means I can’t repeatedly test things in rapid succession. Rather, I need to wait for the car’s systems to fully discharge for half a day before I can repeat my measurements. Bummer…

Back to the BMS, the maximum discharge current setting seems to work, but I’m not sure if it is enough to make me happy. I set it so that if the BMS discharges more than 2A for more than 2s (those are minimum settings), it turns off the output for 10s. This part seems to work, as evidenced by the BMS log file:

Before [5M55S] - [Discharge overcurrent protection is released]
Before [6M5S] - [Discharge overcurrent protection]

It shows the protection was engaged and then released 10s later. During those 2s before the protection kicks in though, the BMS is still pushing out 11-13A at about 12-16v (determined by the battery state of charge). Thus, the function is not really “current” limit (it still pushes out 11-13A), but a “time” limit as it shuts off the output only after my 2s time threshold has been exceeded. It is still outputting whatever current it wants for those 2 seconds. So, I’ve limited the problem (to 2s), but haven’t eliminated the problem yet.
 
GPak said:
I just tested my 5S LTO battery, to see if I can replicate your backflow issue.
Click to expand...

Thanks for your efforts to replicate my observations! I'm beginning to suspect that my backflow issue is related to the specifics of my car's electrical system. At this point, I'm thinking my best solution is to add a variable time delay 12v relay circuit into my LTO box. If I set it so that the relay gives a 15-20s time delay before activating the LTO circuit, this should be enough time for things to stabilize. It seems there are quite a variety of them on Amazon for about $20.
 
Something is wrong, there should be no reverse current, period.
Backflow should stop at charger level, the only other possibility beyond that is if current flows thru dashcam connected to the LTO battery with red cable and car with yellow signal cable, I doubt it but who knows at this point
-Check the continuity of chargers input and output positives it should be open - infinite resistance,
if good
-Try different outlet or better different car
if all good
-Remove the dashcam, and test without it in your car.
Not sure what else to do.
 
Last edited:
This is what I'm confused about as well. I agree, I don't think there should be any backflow at all. The charger should stop this from happening, but apparently it isn't. I'll measure the charger board for continuity as you suggest. Sigh...

My last round of testing was without the dashcam connected to the battery pack.
 
If somehow charger has lost the backflow protection you can try this, I have used it and it works, it is very efficient.
https://www.amazon.com/gp/B07T558XTM
Different models are also available on amazon perhaps with faster delivery

It should be installed between LTO battery and the charger
 
Last edited:
I remember you using that diode in an earlier build. Does it generate any heat? I'm thinking I'll get one and just tape it to the top of one of the LTO cells if it stays reasonably cool.
 
On my lunch time:
I repeated the tests, the same setup and conditions, EXCEPT this time the LTO was fully charged to 13.5V (this is above the car battery voltage) Just for this test only, I increased my cell overcharge protection to 2.8V per cell or 14V total for LTO and adjusted charger to 14V (otherwise these protections will not allow charging above 13.5V)
And..... I was able to replicate reverse current flow!!!
It was not that big upto -3.5A, since my voltage difference for 5S is smaller then yours for 6S, but it doesn't matter.
The fact is that the charger does not have backflow protection! (need to read specs again)

I have this diode and will try to test it today after work.
The noted Diode generates the heat but it is much more efficient with much less voltage drop than regular backflow diodes.

@EricSan - Special Thanks for noticing this issue and sharing it!
 
Ha - it was a fluke that I noticed at all. If you don't have the app open during the first few seconds when the ignition turns on, you won't see it. At least this means that I didn't screw something up with my wiring 🙂. Thanks for your time and efforts to duplicate my situation! Bummer that this situation exists, though. I remember reading somewhere that the charger included an anti-backflow diode, but looking at the product description now, I'm not seeing it... Wonder if it was "supposed" to be a feature that was somehow left out of the product version that you and I purchased.

A diode looks like a good solution, but I'm curious about the effectiveness of the diode that you linked. The amazon reviews (there are not many) indicate that it doesn't fully block reverse voltage.
 
With my 10 sec delay relay I would never have noticed this!

When testing the efficiency of the charger I connect wattmeteres to the input and output terminals of the charger.
With the power supply ON and the battery disconected (normal flow) you can see that both wattmeters show voltage.
With the battery connected and the power supply OFF (revers flow) you can see that the wattmeter on the input side of the charger is off, meaning there is no revers flow for this particular charger (check post #109, the first video)
This is how I usually test for the revers flow protection, apperantly it is not enough and I need to test the revers flow protection with a real car battery.
 
I tested my battery box with an outboard SMPS. Apparently, the construction of the SMPS output is more robust than the wiring configuration of the car's utility port. The backflow doesn't last long, but in my case, it's still pushing 12-13A into the car. That can't be good. I wonder if it is as simple as a relay, then a connection to the battery/alternator, or if it is more complex than that with some sort of voltage regulator or smoothing circuit to quell the EM noise from the alternator. I'm trying to locate a schematic for the utility port wiring to see how it is powered. Not sure if I'll be able to find what I want...
 
Tested the battery in my Jeep again, the same setup and conditions as previously, except with anti-backflow diode, and it works!
No more reverse current.

Even so 10 sec delay is enough time for the engine to start and alternator to engage, I don’t want to risk it.
I believe anti-backflow diode must be installed if charger doesn’t have one incorporated.

I tested with the red one linked above, with no issues.
I also have this one rated at 20A, and I will test it during the weekend.
https://www.aliexpress.us/item/3256..._main.123.21ef1802lLeQ7J&gatewayAdapt=glo2usa

Additionally I ordered this one (15A) to compere to the red one.
https://www.amazon.com/gp/B07V42TMFH EDIT: this one is not good, GPak
 
Last edited:
Score! So, your present setup uses a time-delay relay AND the anti-backflow diode? I'm pretty sure I have enough room in my box to add both...

Given what I'm measuring, the key is finding something that can handle a minimum of 15A, preferably closer to 20A for a bit more headroom.
 
EricSan said:
Score! So, your present setup uses a time-delay relay AND the anti-backflow diode? I'm pretty sure I have enough room in my box to add both...

Given what I'm measuring, the key is finding something that can handle a minimum of 15A, preferably closer to 20A for a bit more headroom.
Click to expand...
Yes, I think time-delay relay and anti-backflow diod, provide the best protection to the car battery and electrical system.

Headroom is always good, I like at list 50% headroom for the components.

All of these components affect the overall charging efficiency, which is not as critical when powered by a car battery/alternator, but is critical when using the cigarette lighter (CL) port.

In my experience, usually charging efficiency is about 75%-85% (depends on components efficiency)
So if CL is rated at 10A which is 125W at 12.5V for car battery, than the max charging rate with CL should be set at 125*0.8=100W max.
 
Yeah, with large current flows, I tend to like headroom that falls into the 50-100% range. This is always my goal while building power supplies for large class-A amplifiers. I agree with derating the utility port power draw to prevent problems.

While I was measuring my LTO box last night, I noted a charging loss of about 10w to inefficiencies: the BMS app showed 7.3A draw at 14.87v (108w), while the SMPS showed an output of 8.8A at 13.48v (118w), so 10w is being converted to heat in the BMS, charger board, LEDs, and misc wiring losses.
 
That is about 91.5% efficiency, I assume without relay and diode, with addition of relay and diode efficiency will drop to about 85% resulting in about 127W out of SMPS, I guess.
 
Just checked the LTO box in my son's Mustang for backflow. It's wired directly to his battery with a 6-8s time delay relay that is ACC switched. With the app open while I started the car, there was no backflow reported by the app. Current draw stayed at 0.0A for the first few seconds, then showed a quick ramp up to about 8.4A once the relay engaged. Overall charging rate at this voltage level is reported as 130w by the BMS. I don't have any plans for changes to his setup at this point in time.

For my LTO box, I'm going to install a time delay relay (I should probably order another one or two to keep as spares) and I'll grab an anti-backflow diode to add into the mix while I'm at it.
 
Just made another interesting discovery about battery charge rates in my Volvo:

As noted above, the efficiency rate of charging my LTO battery pack is about 91.5% (measured a 118w overall power draw, with 108w of it going into the battery). I've been able to repeat this measured efficiency level, not really surprised there. But then I discovered something that was surprising. When I turned off the ignition, charge voltage out of the utility port drops from about 13.4v to about 12.66v (not a surprise). At this lower voltage level, though, the current draw from the car jumps higher than expected: up to 10.57A, representing a is total power draw of 134w. With this increased power draw from the car, the BMS still reports only 108w of power going to the battery pack. This is an efficiency rate of only 80.6%...

The Volvo's utility port stays powered up for about 2-3 mins after the ignition is turned off. I noticed in the Mustang that the power supply to the utility outlet is more "pulsed" with a high current draw while the ignition is turned off, so the overall power supply fluctuates a bit up and down, kind of like a sine wave (though I didn't put a scope on it to visualize the voltage/current pattern). I'm wondering if there is some active circuitry that is performing pulse width modulation or otherwise manipulating the duty cycle of power that comes from the port when the ignition is off in order to manage the current flow. I suspect most low current draw devices would not "notice" the PWM approach to power deliverty...

I'm thinking I need to wire the Volvo's LTO battery right to the main battery with an ACC time delay relay as well. These utility outlets have too many unknowns in their circuitry, are physically problematic in terms of maintaining a good electrical connection, and seem to use wires that are universally too thin (for much other than cell phone charging)...
 
Last edited:
I was curious to learn about what kind of circuits I'm back-feeding current into, so I spent some time digging through the 178 page PDF schematic that I have. I had to stitch this together because the PDF images were rather small. Anyhow, I was able to identify the circuit that feeds my power outlets. Looks like that 12-13A is finding its way back to the battery. I'm wondering what little details of the circuit are obscured with the block-style diagram.

S60 Power Outlets p15 p132.webp
 
Yehh, "smart cars"

With my 5S LTO without time delay relay and without anti-revers diode:
I can’t duplicate your situation exactly, as my CL port turns off as soon as ignition is turned off.
The closest I could do was to turn the ignition back on ASAP without starting the engine.
Interestingly, as in your case, the charger continues to charge LTO (no back-flow) despite the LTO being at higher voltage (I guess the charger capacitors did not have enough time to discharge and that somehow makes the difference), however, I did not see any power increase out of CL port, as in your case.
I got slightly less voltage with slightly more current and basically the same power.

If I wait a little longer, to turn the ignition back on without starting the engine, then I see back-flow as we noticed before (I guess the charger capacitors had enough time to discharge).
Going directly to the car battery/alternator is the best solution.
I think even with time delay relay it is better to use anti-revers diode.
 
Ha- smart indeed... 🙄

Looks like my images has been down-res'd to the point where it's not readable anymore. Oh well...
Yeah, it looks like there are two problems with using the Volvo utility port: 1) backflow for about 10s upon ignition on, and 2) excess current draw for about 2 mins after ignition off. I have to admit that I don't understand the change in efficiency from ~90% to ~80% when the engine turns off.

The time delay relay combined with the anti-reverse diode does seem like the best way to go. In my case, the main battery and an auxiliary fusebox are already in the trunk, so I can tap the battery without going through the firewall. I can tap the fusebox to turn the relay on and off with the ignition (hopefully, the fusebox is not on a similar 2min time delay circuit).

I've broken the schematic up into two smaller images, left side and right side. Maybe this works without being down-sized...
S60 Power Outlets Left Side.webp
S60 Power Outlets Right Side.webp
 
You must log in or register to reply here.
Back
Top