SNVS700G December 2010 – October 2016 LM25066A
PRODUCTION DATA.
NOTE
Information in the following applications sections is not part of the TI component specification, and TI does not warrant its accuracy or completeness. TI’s customers are responsible for determining suitability of components for their purposes. Customers should validate and test their design implementation to confirm system functionality.
The LM25066A is a hot swap with a PMBus interface that provides current, voltage, power, and status information to the host. As a hot swap, it is used to manage inrush current and protect in case of faults.
When designing a hot swap, three key scenarios should be considered:
All of these scenarios place a lot of stress on the hot swap MOSFET and thus special care is required when designing the hot swap circuit to keep the MOSFET operating within its SOA (Safe Operating Area). Detailed design examples are provided in the following sections. Solving all of the equations by hand is cumbersome and can result in errors. Instead, TI recommends using the LM25066A Design Calculator provided in Development Support.
This section describes the design procedure for a 12-V, 45-A PMBus hot swap design.
Table 44 below summarizes the design parameters that must be known before designing a hot swap circuit. When charging the output capacitor through the hot swap MOSFET, the FET’s total energy dissipation equals the total energy stored in the output capacitor (1/2 CV^{2}). Thus both the input voltage and Output capacitance will determine the stress experienced by the MOSFET. The maximum load current will drive the current limit and sense resistor selection. In addition, the maximum load current, maximum ambient temperature, and the thermal properties of the PCB (R_{θCA}) will drive the selection of the MOSFET R_{DSON} and the number of MOSFETs used. R_{θCA} is a strong function of the layout and the amount of copper that is connected to the drain of the MOSFET. Note that the drain is not electrically connected to the ground plane and thus the ground plane cannot be used to help with heat dissipation. For this design example R_{θCA} = 30 °C/W is used, which is similar to the LM25066A EVM. It’s a good practice to measure the R_{θCA} of a given design after the physical PCBs are available.
Finally, it’s important to understand what test conditions the hot swap needs to pass. In general, a hot swap is designed to pass both a “Hot-Short” and a “Start into a Short”, which are described in the previous section. Also it is recommended to keep the load OFF until the hot swap is fully powered up. Starting the load early will cause unnecessary stress on the MOSFET and could lead to MOSFET failures or a failure to start-up.
PARAMETER | EXAMPLE VALUE |
---|---|
Input voltage range | 10 V to 14 V |
Maximum load current | 45 A |
Maximum output capacitance of the hotswap | 5600 µF |
Maximum ambient temperature | 55°C |
MOSFET R_{θCA} (function of layout) | 30°C/W |
Pass hot-short on output? | Yes |
Pass a start into short? | Yes |
Is the load off until PG asserted? | Yes |
Can a hot board be plugged back in? | Yes |
LM25066A can be used with a VCL of 25 mV or 46 mV. In general using the 25-mV threshold will result in a lower RSNS and lower I^{2}R losses. This option is selected for this design by connecting the CL pin directly to VDD. It is recommended to target a current limit that is at least 10% above the maximum load current to account for the tolerance of the LM25066A current limit. Targeting a current limit of 50 A the sense resistor can be calculated using Equation 5:
Typically sense resistors are only available in discrete values. If a precise current limit is desired, a sense resistor along with a resistor divider can be used as shown in Figure 39.
If using a resistor divider, then the next larger available sense resistor should be chosen (1 mΩ for example). The ratio of R1 and R2 can then be calculated using Equation 6:
Note that the SENSE pin will pull 25 μA of current, which will create an offset across R2. It is recommended to keep R2 below 10 Ω to reduce the offset that this introduces. In addition the 1% resistors will add to the current monitoring error. Finally, if the resistor divider approach is used, the user should compute the effective sense resistance (R_{SNS,EFF}) using Equation 7 and use that in all equations instead of R_{SNS}.
Note that for many applications, a precise current limit may not be required. In that case, it’s simpler to pick the next smaller available sense resistor. For this application, a 0.5-mΩ resistor can be used for a 50 A current limit.
It is critical to select the correct MOSFET for a hot swap design. The device must meet the following requirements:
For this design the CSD17556Q5B was selected for its low R_{DSON} and good SOA. After selecting the MOSFET, the maximum steady state case temperature can be calculated using Equation 8:
Note that the R_{DSON} is a strong function of junction temperature, which for most MOSFETs will be close to the case temperature. A few iterations of the above equations may be necessary to converge on the final R_{DSON} and T_{C,MAX} value. According to the CSD17556Q5B datasheet, its R_{DSON} is approximately 1.3x at 85°C. Equation 9 uses this R_{DSON} value to compute the T_{C,MAX}.
This maximum steady state case temperature indicates that a second MOSFET may be needed to reduce and distribute power dissipation during normal operation. When using parallel MOSFETs, the maximum steady state case temperature can be calculated using Equation 10:
Thus using two of the CSD17556Q5B in parallel will result in a steady state temperature of:
Note that the computed T_{C,MAX} is close to the junction temperature assumed for R_{DSON}. Thus no further iterations are necessary.
In general, a lower power limit setting is preferred to reduce the stress on the MOSFET. However, when the LM25066A is set to a very low power limit setting, it has to regulate the FET current and hence the voltage across the sense resistor (V_{SNS}) to a very low value. V_{SNS} can be calculated using Equation 12:
To avoid significant degradation of the power limiting a V_{SNS} of less than 4 mV is not recommended. Based on this requirement the minimum allowed power limit can be calculated using Equation 13:
In most applications the power limit can be set to P_{LIM,MIN} using Equation 14. Here R_{SNS} and R_{PWR} are in Ωs and P_{LIM} is in Watts.
So note that the minimum R_{PWR} would occur when V_{DS} = V_{IN,MAX}. We can then calculate the minimum R_{PWR} using Equation 15:
The next largest available resistor should be selected. In this case a 5.23-kΩ resistor was chosen, which sets a 112.72 W power limit.
The fault timer runs when the hot swap is in power limit or current limit, which is the case during start-up. Thus the timer has to be sized large enough to prevent a time-out during start-up. If the device starts directly into current limit (I_{LIM}) × V_{DS} < P_{LIM}) the maximum start time can be calculated using Equation 16:
For most designs (including this example) I_{LIM} × V_{DS} > P_{LIM }so the hot swap will start in power limit and transition into current limit. In that case the estimated start time can be calculated using Equation 17:
Note that the above start-time assumes constant, typical current limit and power limit values. The actual start-up time will be slightly longer, as the power limit is a function of V_{DS} and will decrease as the output voltage increases. To ensure that the timer never times out during start-up, it is recommended to set the fault time (t_{flt}) to be 2 × t_{star}t or 10.06 ms. This will account for the variation in power limit, timer current, and timer capacitance. Thus C_{TIMER} can be calculated using Equation 18:
The next largest available C_{TIMER} is chosen as 560 nF. Once the C_{TIMER} is chosen the actual programmed fault time can be calculated using Equation 19:
Once the power limit and fault timer are chosen, it’s critical to check that the FET will stay within its SOA during all test conditions. During a “Hot-Short” the circuit breaker will trip and the LM25066A will restart into power limit until the timer runs out. In the worst case the MOSFET’s V_{DS} will equal V_{IN,MAX}, I_{DS} will equal P_{LIM} / V_{IN,MAX} and the stress event will last for t_{flt}. For this design example the MOSFET will have 14 V, 8 A across it for 10.58 ms.
Based on the SOA of the CSD17556Q5B, it can handle 14 V, 10 A for 10 ms and it can handle 14 V, 3 A for
100 ms. The SOA for 10.58 ms can be extrapolated by approximating SOA vs time as a power function as shown in Equation 20 through Equation 23:
Note that the SOA of a MOSFET is specified at a case temperature of 25°C, while the case temperature can be much hotter during a hot-short. The SOA should be de-rated based on T_{C,MAX} using Equation 24:
Based on this calculation the MOSFET can handle 5.24 A, 14 V for 10.58 ms at elevated case temperature, but is required to handle 8 A during a hot-short. This means the MOSFET will be at risk of getting damaged during a hot-short. In general, it is recommended for the MOSFET to be able to handle a minimum of 1.3x more power than what is required during a hot-short in order to provide margin to cover the variance of the power limit and fault time.
For designs with large load currents and output capacitances, using a power limit based start-up can be impractical. Fundamentally, increasing load currents will reduce the sense resistor, which will increase the minimum power limit. Using a larger output capacitor will result in a longer start-up time and require a longer fault timer. Thus a longer fault timer and a larger power limit setting are required, which places more stress on the MOSFET during a hot-short or a start into short. Eventually, there will be no FETs that can support such a requirement.
To avoid this problem, a dV/dt limiting capacitor (C_{dV/dt}) can be used to limit the slew rate of the gate and the output voltage. The inrush current can be set arbitrarily small by reducing the slew rate of the V_{OUT}. In addition, the power limit is set to satisfy the minimum power limit requirement and to keep the timer from running during start-up (make P_{LIM} / V_{IN,MAX} > I_{INR}). Since the timer doesn’t run during start-up it can be made small to reduce the stress that the MOSFET experiences during a start into short or a hot-short.
The D2 prevents the charge of C_{dV/dt} from interfering with the power limit loop during a hot-short event and Q3 discharges C_{dV/dt} when the hot swap gate comes down.
The inrush current should be kept low enough to keep the MOSFET within its SOA during start-up. Note that the total energy dissipated in the MOSFET during start-up is constant regardless of the inrush time. Thus, stretching it out over a longer time will always reduce the stress on the MOSFET as long as the load is off during start-up.
When choosing a target slew rate, one should pick a reasonable number, check the SOA and reduce the slew rate if necessary. Using 0.25 V/ms as a starting point the inrush current can be computed as follows:
Assuming a maximum input voltage of 14 V, it will take 56 ms to start-up. Note that the power dissipation of the FET will start at V_{IN,MAX} × I_{INR} and reduce to zero as the V_{DS} of the MOSFET is reduced. Note that the SOA curves assume the same power dissipation for a given time. A conservative approach is to assume an equivalent power profile where P_{FET} = V_{IN,MAX} × I_{INR} for t = t_{start-up} /2. In this instance, the SOA can be checked by looking at a 14 V, 1.4 A, 28 ms pulse. Using the same technique as Check MOSFET SOA, the MOSFET SOA can be estimated with Equation 26.
This value has to also be derated for temperature. For this calculation, it is assumed that T_{C} can equal T_{A,MAX} when the board is plugged in. This would occur if a board is plugged in at an elevated ambient temperature environment.
Based on this calculation the MOSFET can handle 4.44 A, 14 V for 28 ms at elevated ambient temperature, and is required to handle 1.4 A. This indicates the MOSFET will stay well-within its SOA during a start-up if the slew rate is 0.25 V/ms or less. Note that if the load is off during start-up, the total energy dissipated in the FET is constant regardless of the slew rate. Thus a lower slew rate will always place less stress on the FET. To ensure that the slew rate is at most 0.25 V/ms the CdV/dt should be chosen with Equation 28.
The next largest available C_{dV/dt} is chosen to be 100 nF. Then the typical slew rate and start time can be computed to be 0.22 V/ms as shown below, making the typical start time 55 ms, assuming 12 V input.
In certain applications, T_{C,MAX} may be used for temperature derating instead. This would only occur if a hot board is unplugged and then plugged back in before it cools off to ambient temperature. This is worst case and for many applications, the T_{A,MAX} can be used for this derating.
Based on this calculation using T_{C,MAX} for derating, the MOSFET can handle 3.15 A, 14 V for 28 ms at elevated case temperature, and is required to handle 1.4 A. This indicates the MOSFET will stay well-within its SOA during a start-up if the slew rate is 0.25 V/ms or less.
When picking the power limit, it needs to meet 2 requirements:
Thus the minimum allowed power limit can be computed with Equation 31.
Next, the power limit is set to PLIM,MIN using the equation below. Here RSNS and RPWR are in Ωs and PLIM is in Watts.
The closest available resistor should be selected. In this case a 5.23-kΩ resistor was chosen.
Next a fault timer value should be selected. In general, the timer value should be decreased until there is enough margin between available SOA and the power pulse the FET experiences during a hot-short. For this design a 22-nF CTIMER was chosen corresponding to a 420-µs fault time. The available SOA is extrapolated using the method described earlier.
Next the available SOA is derated for temperature:
Note that only 4 A was required, while the FET can support 17.17A. This confirms that the design will be robust and have plenty of margin.
By programming the UVLO and OVLO thresholds the LM25066A enables the series pass device (Q_{1}) when the input supply voltage (V_{SYS}) is within the desired operational range. If V_{SYS} is below the UVLO threshold, or above the OVLO threshold, Q_{1} is switched off, denying power to the load. Hysteresis is provided for each threshold.
The configuration shown in Figure 40 requires three resistors (R1-R3) to set the thresholds.
The procedure to calculate the resistor values is as follows:
The lower OVLO threshold is calculated from:
As an example, assume the application requires the following thresholds: V_{UVH} = 8 V, V_{UVL} = 7 V, V_{OVH} = 15 V.
The lower OVLO threshold calculates to 12.03 V and the OVLO hysteresis is 2.97 V. Note that the OVLO hysteresis is always slightly greater than the UVLO hysteresis in this configuration. When the R1-R3 resistor values are known, the threshold voltages and hysteresis are calculated from the following:
If all four thresholds must be accurately defined, the configuration in Figure 41 can be used.
The four resistor values are calculated as follows:
As an example, assume the application requires the following thresholds: V_{UVH} = 8 V, V_{UVL} = 7 V, V_{OVH} = 15.5 V, and V_{OVL} = 14 V. Therefore V_{UV(HYS)} = 1 V and V_{OV(HYS)} = 1.5 V. The resistor values are:
When the R1-R4 resistor values are known, the threshold voltages and hysteresis are calculated from the following:
The minimum UVLO level is obtained by connecting the UVLO/EN pin to VIN as shown in Figure 42. Q_{1} is switched on when the VIN voltage reaches the POR threshold (≊ 2.6 V). The OVLO thresholds are set using R3, R4. Their values are calculated using the procedure in Option B.
The OVLO function can be disabled by grounding the OVLO pin. The UVLO thresholds are set as described in Option B or Option C.
For this design example option B is used and the following values are targeted: VUVH = 10 V, VUVL = 9 V, VOVH = 15 V, VOVL = 14 V. R1, R2, R3, and R4 are computed using the equations below:
Nearest available 1% resistors should be chosen. Set R1 = 43.2 kΩ, R2 = 6.49 kΩ, R3 = 43.2 kΩ, and R4 = 3.65 kΩ.
When the voltage at the FB pin increases above its threshold, the internal pulldown acting on the PGD pin is disabled allowing PGD to rise to V_{PGD} through the pullup resistor, R_{PG}, as shown in Figure 44. The pullup voltage (V_{PGD}) can be as high as 17 V, and can be higher or lower than the voltages at VIN and OUT. VDD is a convenient choice for V_{PGD} as it allows interface to low voltage logic and avoids glitching on PGD during power-up. If a delay is required at PGD, suggested circuits are shown in Figure 45. In Figure 45A capacitor C_{PG} adds delay to the rising edge, but not to the falling edge. In Figure 45B, the rising edge is delayed by R_{PG1} + R_{PG2} and C_{PG}, while the falling edge is delayed a lesser amount by R_{PG2} and C_{PG}. Adding a diode across R_{PG2} (Figure 45C) allows for equal delays at the two edges, or a short delay at the rising edge and a long delay at the falling edge.
Setting the output threshold for the PGD pin requires two resistors (R4, R5) as shown in Figure 43. While monitoring the output voltage is shown in Figure 43, R4 can be connected to any other voltage which requires monitoring.
The resistor values are calculated as follows:
Choose the upper and lower threshold (V_{PGDH}) and (V_{PGDL}) at V_{OUT}.
For this example PGDH of 9.25 V and PGDL of 8.75 V is targeted. R5 and R6 are computed using the following equations:
Nearest available 1% resistors should be chosen. Set R5 = 10 kΩ and R6 = 1.5 kΩ.
Proper operation of the LM25066A hot swap circuit requires a voltage clamping element present on the supply side of the connector into which the hot swap circuit is plugged in. A TVS is ideal, as depicted in Figure 46. The TVS is necessary to absorb the voltage transient generated whenever the hot swap circuit shuts off the load current. This effect is the most severe during a hot-short when a large current is suddenly interrupted when the FET shuts off. The TVS should be chosen to have minimal leakage current at V_{IN,MAX} and to clamp the voltage to under 24 V during hot-short events. For many high power applications, 5.0SMDJ13A is a good choice.
If the load powered by the LM25066A hot swap circuit has inductive characteristics, a Schottky diode is required across the LM25066A's output, along with some load capacitance. The capacitance and the diode are necessary to limit the negative excursion at the OUT pin when the load current is shut off.
Figure 38 shows the schematic used to implement the requirements described in the previous section. In addition, Table 45 provides the final component values that were used to meet the design requirements for a
12-V, 45-A hot swap design. Figure 47 to Figure 54 are based on these component values.
COMPONENT | VALUE | |
---|---|---|
R_{SNS} | 0.5 mΩ | |
R1, R3 | 43.2 kΩ | |
R2 | 6.49 kΩ | |
R4 | 3.65 kΩ | |
R_{FB1} | 10 kΩ | |
R_{FB2} | 1.5 kΩ | |
R_{PWR} | 5.11 kΩ | |
Q1 | CSD17556Q5B | |
Q2 | MMBT3904 | |
Q3 | MMBT3906 | |
D1, D3 | 1N4148W-7-F | |
D2 | SK153-TP | |
Z1 | 5.0MDJA15A | |
C_{dV/dt} | 100 nF | |
C_{TIMER} | 22 nF |
V_{IN} = 12 V |