Viofo HK3-C: Disable Low Battery Cutout for External Battery Bank Operation

[SPL15]

Pretty simple: Remove Q3 from PCB to disable the automatic low-battery cutoff if you’re wanting to be able to discharge dashcam batteries to 0%.

DO NOT PERFORM THIS MOD IF YOU ARE RUNNING OFF LEAD ACID BATTERIES, OR YOUR CAR’S STARTING BATTERY. Do not do this mod unless you know exactly what this mod is going to do to your batteries without having to ask someone else what it will do to your batteries. If you perform this mod w/ your car’s battery as the power source, you will more than likely irreversibly damage the battery due to way over-discharging it, and could possibly damage expensive electronic control modules in your vehicle.

Disclaimer: Perform at your own risk. I’m just a random stranger on the internet, I could be full of BS and stupid like most random strangers on the internet are.

Notes / Observations:

With Q3 removed, the buck-regulator inside the HK3-C power module will regulate down to around 7 VDC from the battery pack(s). Operation below 10 volts or so is unknown, & could potentially cause issues w/ component heating / unreliability due to the higher peak current pulses required for a given power output. It’s likely OK to run low voltage, but I would try to stay around 10.5 - 11 VDC minimum to be on the safe side with a 3-ch setup. If there are issues running low voltage, increasing inductor size & value would most likely get things working reliably & more cool.

Transistor Q3 is what switches the NDP1335KC PMIC into standby mode on pin 7 (FS). Without Q3, the step down converter PMIC never receives the “low-battery voltage” standby signal from the ChipSea CSU32P10 MCU.

The voltage selection switch effectively has no effect w/ Q3 removed; the MCU still monitors +12bat & fires the standby signal when low, but there’s no Q3 to pull down Pin 7 on the PMIC to put it into standby mode. I removed the voltage level switch & jumpered it, but this isn’t necessary (I only removed it because I don’t want the lever sticking out of the housing).

MCU logic for the ACC input & output signal remains the same; there is no difference in operation w/ Q3 removed.

Because there’s no shutdown signal with Q3 removed, the dashcam will receive power & be put into parking mode whenever 7 volts or more are applied to the HK3-C’s +12VBat input. ACC output to dashcam works fine down to 7 volts input as well.

The PMIC is rated 7 - 32 VDC input, up to 3A output. Output voltage is set via resistor network, there’s also the ability to add output voltage resistive loss compensation (I might play w/ this later on). There’s reverse input voltage protection via series diodes on the +12vbat & ACC inputs.

Current limit of 2.0 amps stated by Viofo is likely due to size of the buck-regulator inductor (both value wise, wire gauge, & core saturation), as well as input & output filter cap values. Could easily increase max current output to the 3 amp limit of the PMIC by increasing inductor value, gauge, & core size, as well as input & output cap values (need to be low ESR caps w/ high ripple current rating).

The low voltage cut-off is set via a resistor network w/ a 5 VDC reference that’s also the MCU’s 5 volt power source (MCU is powered directly off the +12bat input via a basic 3-pin linear regulator). The voltage level switch connects ADC 0 input to different voltage potentials on the resistor divider network, the resulting voltage to the MCU is used to compare against the +12bat voltage that’s also divided thru a resistor network. It’s not easily feasible to change the low-bat cutoff voltage to be below 11.8Vdc due to the logic being internal to the MCU & the lowest setting of 11.8vdc being referenced to 5Vdc (which is the highest voltage available). Could potentially play w/ the +12Bat resistor network to fool the MCU.

 

[marcomrctavares]

Thank you so much, this is exactly what I was looking for! I didn’t know why but sometimes my dashcam shuts off when my battery pack was 40+% and this solved the issue. Now I can drain the battery pack to 0.
Thank you!

 

[SPL15]

Glad this was useful.

Having 40% capacity left is pretty high for the dash cam to shutoff due to low battery voltage w/ liFePO4 batteries. I’m guessing you have excessive voltage drop on your cabling from the battery to the HK3-C power module, both due to length of cabling & due to the tiny 24 gauge wire used in the cabling supplied by all of the Egen manufactured dashcam battery banks. I’m guessing you’re running 3-channels as well? I’m also assuming you had the HK3-C’s voltage cutoff setting to the minimum at 11.8 volts?

This cabling induced voltage drop caused by copper resistance will reduce the voltage seen at the HK3-C’s power module, causing it to prematurely shutoff power to the dashcam before the battery voltage actually reaches the cut-off voltage setting of the HK3-C hardwire cable. In addition, the HK3-C power control module will draw increased current as the input voltage from the battery reduces, further reducing power efficiency & parking mode runtimes (this increased current draw w/ low input voltages is due to how switch mode dc-dc converters work, not a flaw in the design). This same phenomena is the reason why all dashcam batteries that use an internal switch-mode DC-DC converter to charge the battery bank will draw higher current from the car, sometimes beyond their listed specification, if the input voltage from the car is low due to thin wiring / poor installation.

What I did to drastically reduce this parasitic voltage drop loss & significantly increase efficiency/ runtimes was to make my own 5 foot long 16 AWG wire harness from the dashcam battery bank & soldered it directly to the HK3-C power module’s PCB, completely removing the 20AWG portion of the HK3-C’s wiring harness, as well as use the car’s chassis ground for HK3-C’s ground connection, instead of running it to the battery bank’s negative output terminal. I also made my own 2 ft long 10AWG wire harness for the dashcam battery’s power input & used a 70/80 amp relay to toggle ignition power directly from my car’s rear batteries to minimize voltage drop during charging, as well as when the dashcam is in parking mode (dashcam will sink negative thru battery bank’s 12V power input ground connection to chassis if HK3-C’s ground connection is also terminated to the chassis).

After the above modifications, with a PowerCell 8 & Cellink Neo EXT7, I get around 24 - 26 hours of parking mode runtime w/ 3 channels & IR enabled on the A139 Pro, & around 45 - 48 hours of runtime if I run 2 channels (front & rear cameras). Both scenarios are with “Station Mode” WiFi enabled & connected to my home’s network. The above modifications I described reduced voltage loss to the HK3-C power module in my installation to a calculated 0.015 - 0.025 volts (15 - 25 mVdc). I calculated the stock component wiring harnesses voltage drop to be around 0.250 - 0.500 volts (250 - 500 mVdc) if everything is connected optimally (for the factory supplied PowerCell 8 & Viofo HK3-C wiring harnesses).

In your case, if your dashcam was dropping out at 40% battery charge, I’d estimate more around 1.0 volt of drop on the wiring / connections between the battery bank & the HK3-C power module, which is excessive & might suggest a poor connection somewhere (or my calculations / assumptions for the stock supplied wiring harnesses are way off). To put things into perspective for how much power loss / waste this is: 1.0 volt of drop on the 12 volt side of the cabling between the battery bank & HK3-C power module equals approximately 6+ watts of power lost just in cabling alone (assuming 3 channel operation @ 0.5 amps current draw), where the dashcam itself in 3-channel parking mode takes around 6 - 7 watts (depending on WiFi & parking mode settings). You can significantly increase parking mode runtimes by reducing cabling losses.

If I were you, I would go over all of the connections & wiring to make sure they’re secure & well connected to one another. Fixing the excessive voltage drop will increase power efficiency of the HK3-C power module, & thus also increase parking mode runtimes significantly, as well as ensure reliable & safe operation long term. Simply terminating the HK3-C’s ground termination to chassis ground alone will reduce voltage drop to the power module by approximately 140 mVdc. Further easily gotten gains can be realized by also upgrading the battery bank wire gauge to 12, or 10 AWG in my case, as well as removing unneeded cabling length.

In the future, I may re-locate the HK3-C’s power module closer to the dashcam by removing some length of the black USB-C portion of the cabling & extending the 16 AWG 12 volt section of cabling; this will increase the voltage the dashcam unit sees & should theoretically reduce dashcam temps slightly due to slightly better efficiency of the Dashcam’s internal DC-DC regulator w/ the higher effective voltage input.

Wiring voltage drop w/ DC voltages is purely a function of current, not voltage; the regulated 5 volt USB-C cabling experiences higher current than the unregulated 12 volt input cable due to ohms law for a given wattage / power with a switch mode DC-DC regulator (this rule does not apply w/ linear regulation, due to how linear regulators work). Example: 0.5 amps @ 12 volts from the battery bank will be converted to roughly 1.33 amps @ 5vdc assuming a typical 90% conversion efficiency of the HK3-C’s DC-DC converter. The 10 feet of 24AWG conductor used on the HK3-C’s USB-C cable will drop approximately 1.05 volts to the dashcam from the power module, causing the dashcam to receive only 3.95 volts, instead of 5. This will cause the dashcam’s internal DC-DC converters to be less efficient due to higher peak current pulses needed for a given wattage / power output at the reduced input voltage.

 

[dimeloco]

EZ calp

 

[ly-70mai]

Thank you so much, this is exactly what I was looking for! I didn’t know why but sometimes my dashcam shuts off when my battery pack was 40+% and this solved the issue. Now I can drain the battery pack to 0.
Thank you!

How do you know that your battery pack still has 40% power, and the power of LFP usually has a large error

 

[Nigel]

How do you know that your battery pack still has 40% power, and the power of LFP usually has a large error

It quite likely was 40%, in cold weather the voltage of LiFePO4 cells drops, and the low voltage cut off will turn off with more power left in the cell, at -20 degrees C it may turn off with 90% left! The people having issues do live in places that sometimes get cold. You can't estimate power remaining by looking at the cell voltages!

 

[ly-70mai]

It quite likely was 40%, in cold weather the voltage of LiFePO4 cells drops, and the low voltage cut off will turn off with more power left in the cell, at -20 degrees C it may turn off with 90% left! The people having issues do live in places that sometimes get cold. You can't estimate power remaining by looking at the cell voltages!

I am thinking that the battery pack should provide a stable 13V voltage output (may require a voltage regulator circuit) until it is cut off after the power is exhausted, or let the voltage drop naturally from 14.8V.

 

[Nigel]

I am thinking that the battery pack should provide a stable 13V voltage output (may require a voltage regulator circuit) until it is cut off after the power is exhausted, or let the voltage drop naturally from 14.8V.

Problem with a regulator is that it "wastes" some power, maybe 10%, so the battery does not run for as long. Also it sometimes needs to increase the voltage, but if the engine is running then it needs to reduce it, so a high cost regulator. All the battery packs just provide the cell voltage, except when they output 5V USB, which does not waste power since you then don't need the hardwire kit (so the same amount of power is wasted as if you used the hardwire kit instead).