EDITS: 6/11/19 - Fixed broken links
LTC1624 Calculations
I have spent a fair bit of time trying to get a good understanding of this part of the design (and to be honest, I’m still working on that!). I have used the data sheet to make the necessary calculations for the selection/confirmation of various components in the design as well as the power dissipation of MOSFET and Schottky Diode.
Rsense selection
This is the resistor used as an external current sense. Calculation is on page 7 of the data sheet.
Simply, it is 100mV / maximum current of output: 100mV / 3A = 0.333Ohms. However, using this value in the LTSpice simulation results in unstable output so I will stick with the original 0.01Ohms and test when in place.
Inductor Value - delta IL
It favours smaller inductor values but the choice has a direct affect on the ripple current. Calculation is on page 7 of the data sheet.
Input Values:
| Parameter | Value | Note |
|---|---|---|
| Vin | 24V | As measured in the LTSpice simulation |
| Vout | 15V | Max output load voltage |
| f | 200000Hz | Frequency of operation of LTC1628 |
| Vd | 0.5V | Schottky diode forward voltage |
The delta-IL is the expected ripple current from a selected inductance:
| Inductance (L) | Delta-IL | Note |
|---|---|---|
| 5uH | 5.69A | Delta-IL = ((Vin-Vout) / (f * (L*0.000001)) * ((Vout + Vd) / (Vin + Vd))) |
| 10uH | 2.85A | |
| 15uH | 1.90A | |
| 20uH | 1.42A | |
| 50uH | 0.57A |
Ferrite core inductor should be used. Simulation shows it working fine with 10uH through to 50uH but better characteristics at 10uH - in theory, the lower value of 50uH would make a better choice but the original design uses 10uH. I will stick with that and test.
Power MOSFET selection
This is a calculation on the expected power dissipation - the actual selection is based on logic-level threshold, a low Rds(on) as well as the power handling characteristics. Calculation is given on Page 8 of the data sheet and values are based on selected MOSFET (IRF3205).
| Parameter | Value | Note |
|---|---|---|
| Vin | 24V | As measured in the LTSpice simulation |
| Vout | 15V | Max output load voltage |
| Vd | 0.5V | Schottky diode forward voltage |
| Main switch duty cycle (MSDC) | 0.633V | Calculated as: (Vout + Vd) / (Vin + Vd) |
| iMax | 3A | Max output load current |
| Temp Dependency | 0.005 per degree C | An approximation for low voltage MOSFETs |
| Rds(on) | 0.008Ohms | |
| K | 2.5 | Constant given in the data sheet |
| Crss | 2.11E-10F | |
| f | 200000Hz | |
| Temp rise | 25C | 50C - ambient of 25C |
| Pmain (power dissipation) | 0.164W | Pmain = (MSDC * Imax^2 * ((1 + (Temp Dep * Temp rise)) * Rds(on)) + (K * (Vin^1.85)) * Imax * Crss * f) |
I will use the Pmain value in the thermal calculations.
Schottky Diode Selection
Selection is based on peak current and power dissipation. The most stressful condition is under short circuit where Vout = 0V
| Parameter | Value | Note |
|---|---|---|
| Vin | 24V | As measured in the LTSpice simulation |
| Vd | 0.5V | Forward voltage |
| Isc | 3A | Short circuit current |
| Pdiss | 1.469W | Pdiss = Isc * (Vd * (Vin / ( Vin + Vd))) |
The diode will need a low thermal resistance or heat sink - the Pdiss value is used in the thermal calculations.
INTVcc Regulator
An internal regulator produces a 5V supply. Need to determine what the maximum junction temperature might reach.
| Parameter | Value | Note |
|---|---|---|
| Ambient | 100C | I wouldn't want it to get this high but assume worse case |
| Vin | 24V | As measured in LTSpice simulation |
| Temp Co-efficient | 110C/W | From datasheet |
| Supply Current | 0.008A | From LTSpice simulation, current at Vin |
| Junction temperature | 121.12C | JT = Ambient + (supply current * Vin * Temp Coefficient) |
Tjmax is 125C so if ambient reaches 100C it's pretty much at the limit. However, other parts will fail at that temperature so I need to keep it lower. Monitor under test.