LTC4020 Inductor Calculation

Hello,

I am currently involved in the design of an universal and microcontroller based battery charger system.

The key requirements are:
- 10A charge current
- VIN ranging from 9V to 48V
- Battery Voltage Range from 3V to 53V
- Adjustable Charge current
- Adjustable Input current
- Adjustable Battery Voltage
- Adjustable Battery Chemistry

All adjustable parameters should be easily changeable by an embedded microcontroller.

First i tried to calculate the min. inductor value according to datasheet page 23.
The values i got seems to be a bit to small to me, so i searched the analog forums and found this two entries:

https://ez.analog.com/power/f/q-a/91126/ltc4020-no-load-stability/303105#303105

and https://ez.analog.com/power/f/q-a/100595/ltc4020-unstable-operation/303106#303106

Both are saying, that the formula specified on page 23 use an extra 'Imax' term in the denominator, which shouldn't be there.
Does this mean this is an errata in the datasheet?

I defined the worst cases and calculate it exactly according to the datasheet:

- Buck:
VinMax -> VinMax/2 (48V -> 24V)

Lmin = (Vout x (1- (Vout/VinMax)) / (fo x IRipple x IMAX) = (24V x (1- (24V/48V)) / (120kHz x 5A x 10A) = 2µH


- Boost:
VinMax -> VoutMax (48V -> 53V)

Lmin = (Vin x (1- (VinMax/VoutMax)) / (fo x IRipple + IMAX) = (48V x (1- (48V/53V)) / (120kHz x 3,23A x 11,042A) = ~ 1,06µH


If i remove the "Imax" denominator i get this values:
LminBuck  = ~ 20µH
LminBoost = ~11,7µH

This is quite a massive difference - so wich value is the correct one?
For me the second values seems to be more common in such power devices?

Another questions regarding the adjustable option:
1. ILIM and RNG/SS is adjusted by an DS4432 current sink DAC -> could this be suitable for this application?
2. Battery Voltage are adjusted using an 12bit voltage buffered DAC which makes an offset to the feedback node -> possible and accurate enough?

Kind Regards
Danie

Parents
  • 0
    •  Analog Employees 
    on Jul 12, 2019 4:23 PM

    You will not be able to use a current-sink DAC to control ILIM and RNG/SS because these pins are driven internally by a 50µA current source.  Typically, a resistor from ILIM or RNG/SS to SGND is used to establish a voltage at the pin with 1.0V (corresponding to 20kΩ) as full scale.  Any external circuit connected to either of these pins must be capable of sinking 50µA and establishing a 0-1V voltage at the pin.  A digipot (RDAC) may be suitable for this purpose.

    It should be possible to adjust the battery-charging voltage by adding a DAC in series with the VFB pin. Note that typically, the power-stage output is connected to VFBMAX through a resistor divider.  The voltage output of the power stage will always be limited to the voltage corresponding to VFBMAX=2.75V, so be sure the maximum output voltage of the power stage is adequate for your full-charge battery voltage.

  • Thank you for your reply.

    After your reply we select an inductor with 6.8uH nominal value.
    Its a PQ2614BHA-6R8K from Bourns - datasheet: 
    PQ2614BHA-6R8K Datasheet
    I think this should suite our requirements?

    Current adjustment is now implemented with two 20k digital potentiometer AD5274.
    One for inductor current limit and one for charge current limit.


    Regarding the DAC and battery voltage adjustment:
    What do you mean with "adding a DAC in series with the VFB pin"? Why in series?
    Could you please provide a short example?

    Our general idea was to read out an EEPROM placed in the battery pack and then adjust the voltage and charge current regarding to the battery pack parameters stored in the EEPROM.
    Currently we have no idea how to implement this features. Do you have any suggestions for us?

    We thought about a few ways, but we are not really sure if were going the right way...
    So it would be great if you can point us in the right direction!

    Thank you very much!


    Kind regards
    Danie 
     

Reply Children
No Data