Stepping down 4 72V Battery Packs to a 48V vehicle

[nugget_handler]

I need to get a 48V vehicle to work with up to 4 72V battery packs. The packs are equipped with a BMS which already supports mating up to 4 of them together. I am mostly concerned with the DC-DC step down problem.

I found these ready-made converters which seem to have functionality I need, but I am not sure if I will need anything else:

1. Dc 60V 72V TO 48V 30A 1440W Dc Dc Converter IP68 Step Down Module, 12V
2. 72V to 48V 60A 2880W DC DC Step Down Converter Voltage Regulator w ACC Switch
3. تيار مستمر محول تيار مستمر 52-90 فولت 60 فولت 72 فولت إلى 48 فولت تيار مستمر محول تنحى باك 1A 2A 4A 10A 16A 20A 48 فولت تيار مستمر الجهد المنظم لجهاز السيارة - AliExpress
4. DC/DC converter 72V to 48V, 550W • DWE
5. DC/DC converter 72V to 48V – 500W • DWE
6. 72V to 48V 30A Step Down DC DC Converter Regulator 72 Volt to 48 Volt 1440W Buck Transformer Power Supply for Cars Toys
7. Multiple Alibaba Options
8. SD-1000-MEAN WELL Switching Power Supply Manufacturer

I think I will also have to design a PCB with the input protection, output protection, power management/control, monitoring/interface circuits, as well as the thermal management components.


Am I missing anything? Any guidance on how to start this project? Thank you in advance!

 

[Diggs]

I don't see anything about watts/amps. How many amps max is the work vehicle? The buck transformers all state their wattage. What amp rating are the batteries? I assume the work vehicle has a controller. What are its ratings?

 

[amberwolf]

Your DC-DC converter will have to be more capable than the controller itself, because like the batteries it must supply all the power the system demands during motor operation.

So it will probably be larger and more expensive than the motor controller.

It must be able to continuously handle the worst-case current draw the system will ever have, under any votlage conditions of the batteries.

It will waste probably a third to half of your range (battery capacity) as heat in the conversion process (and you'll need to add a cooling system to get rid of that heat).

Because the controller is designed to work with the impedance of the battery on it's inputs, not another electronics device like the DC-DC, then unless you can find a DC-DC that supports automatic two-way conversion and acts instaneously to detect regen and change modes (I don't know of one), you will not be able to use regenerative braking; if your controller has this you must disable it. If the controller can't disable it, you must replace the controller. Otherwise, the voltage your ocntroller produces during braking will have nowhere to go, and will instantly spike up far beyond what your controller can handle, and you'll blow up your controller (and probably the DC-DC).


Because it now replaces half the job the controller is designed to do, you'll need to pick one that:
-- has a voltage input range wider than the batteries' full-to-empty range.
-- has an LVC to shut it off that is well above the batteries' own LVC, as you never want to run the batteries down to that point.
-- has a current limiter in it to prevent it from drawing more current from what even a single battery can handle by itself, under that battery's worst-case low-current condition.
-- has an output voltage at the specific voltage you want the controller to operate at (depends on the max speed you need from the motor, and any other controller or system characteristics that are determined by voltage).
-- has an output current higher than anything the controller will ever draw under worst case driving conditions.

 

[nugget_handler]

then unless you can find a DC-DC that supports automatic two-way conversion and acts instaneously to detect regen and change modes (I don't know of one)

First, thank you so much for your thorough response. I truly appreciate it! Now, what if I wrote firmware for a microcontroller that has mechanisms to detect this switching from acceleration to regenerative braking (let's call this the translator module since it will "translate" messages to the ECU from the battery pack(s) and back)? Then I can use the microcontroller to handle this switching and won't have to find a specialty converter that does this two-way conversion automatically, right?

has a voltage input range wider than the batteries' full-to-empty range.

What does this mean exactly? May you please provide an example?

How many amps max is the work vehicle? The buck transformers all state their wattage. What amp rating are the batteries? I assume the work vehicle has a controller. What are its ratings?

I tried compiling all this data in the table below. Thank you in advance for any advice:

Item	Specification
Vehicle system voltage	48V nominal
Controller model	NEOS/Lvtong 48M350B (variants: 48M350B-FA-21, 48M350B-00-02)
Controller rated current	350A (some variants list up to 440–450A maximum)
Controller rated voltage	48V
Controller rated output power	5kW
Controller protection / environment	IP54–IP65 (varies by variant)
Max vehicle current (worst-case draw at 48V bus)	Up to controller limit: ≈350A class; some variants specify up to 440–450A maximum
Min vehicle/motor current (light load/idle)	Near-zero at no-torque command; exact idle/quiescent current not specified
Battery voltage class (vehicle)	48V system; typical 16S LiFePO4: full ≈58.4V, empty ≈40–44V (varies by pack/BMS)
Example battery spec (context)	60V 100Ah LFP, 5.5kW motor example in similar 4-seat category (illustrative only)

 

[amberwolf]

First, thank you so much for your thorough response. I truly appreciate it! Now, what if I wrote firmware for a microcontroller that has mechanisms to detect this switching from acceleration to regenerative braking (let's call this the translator module since it will "translate" messages to the ECU from the battery pack(s) and back)? Then I can use the microcontroller to handle this switching and won't have to find a specialty converter that does this two-way conversion automatically, right?

What converter are you going to use that can switch directions on command? (I don't know of any of those; you'd probalby have to design and build that, too, unless you know of one)

Does the converter present sufficient impedance to the controller to load down the regen voltage to prevent system damage? (the battery normally does this)

The MCU can't actually do the switching itself as it can't handle either voltages or currents anywhere near what you want.

If you need to use contactors or relays to do the switching, you will have to completely turn off the system at every changover from traction to regen and back.. If you don't, then when the regen engages and current needs to flow back into the battery via the converter, the voltage will spike from the disconnection of load from controller as it is doing the regen, and that will probably destroy the controller.

When you power off the system to change over, then the motor controller probably loses power completely and has to reboot, which with some controllers will require the motor to be stopped before the controller will finish a reboot / powerup sequence.

If the motor spinning is providing power back to the controller (not regen, just normal BEMF) to keep it on even with the converter disconnected, that spinning must be slow enough to keep voltage below the controller limits; if it isn't, the controlelr can be destroyed without the battery there to absorb this.

What does this mean exactly? May you please provide an example?

"has a voltage input range wider than the batteries' full-to-empty range. " means exactly that: The converter must be able to operate at a voltage higher than anything the battery will ever charge to, and below anything the battery will ever discharge to.

It must also be able to operate at a voltage higher than anything the motor/controller will ever generate during regen.

 

[amberwolf]

Basically, you're going to have to use a converter or set of them that will be larger and more capable and more expensive than just changing your controller out for one that will actually operate at the voltages of the batteries you want to use.

If you use the converter system, you have more points of failure, more problems that will happen because of the extra complexiity, greater system cost and size, much more loss (my guess is a third to half your battery power wasted in the conversion process, so losing up to half your range or requiring up to twice as big a battery system to compensate for that, greatly increasing the cost and size of the system).

Much cheaper and easier to change out the controller.