1
\$\begingroup\$

I am working on a boost converter design using the LTC3788. I would like to take advantage of the dual output capabilities of this chip to produce 24 VDC at 4 amps on one output and 48 VDC at 2 amps on the other output.

Linear Technologies provides an example of this circuit in Figure 10 on page 29 of the LTC3788 datasheet.

High Efficiency Dual 24V/48V Boost Converter with Inductor DCR Current Sensing Figure 10: High Efficiency Dual 24V/48V Boost Converter with Inductor DCR Current Sensing

I was able to simulate this circuit in LTspice but I am having trouble interpreting the simulation output. Some of the components referenced in the figure were substituted with equivalent models that are either defined in the schematic or were found in the LTspice database.

I have read through a lot of the datasheet describing how to choose the external components for a particular design and have chosen components accordingly. I tried specifying models for all diodes, MOSFETs, and capacitors which improved the response of the circuit but the output still fluctuates more than intended.

Other simulations of this boost converter shows a response that reaches steady state in less than 2 mS which make me think there is much to improve on with this design.

What could be causing this circuit to produce the corresponding response? Are there any particular components that could be causing the output to behave in this way?

Circuit design: simulated circuit

Simulation results: enter image description here

Update (Solved):

It was pointed out to me that the SENSE+ pins on the controller were not properly connected to the inudctors. In addition, the inductors that I originally chose for this design did not specify a DC resistance which is particularly important for this application since DCR sensing is used to measure current to the load. My understanding is that it is important to chose inductors with low DCR values to maintain high efficiency in the controller. A sense resistor could be used to measure current more accurately but, will consume more power in a high current application.

After running a new simulation with an updated schematic the outputs behaved as expected. Using low ESR capacitors also helped with reducing the output ripple voltage.

Updated schematic: enter image description here

Updated simulation output: enter image description here

\$\endgroup\$
  • \$\begingroup\$ What are you using for loads? \$\endgroup\$ – Voltage Spike May 18 '17 at 19:51
  • \$\begingroup\$ The 24 volt output is used to supply a PCB with LEDs and various low power I/Os. The 48 volt output will be used for PoE. Both outputs only require half of the designed current (i.e. 2 amps for 24 VDC and 1 amp for 48 VDC). \$\endgroup\$ – kprince May 18 '17 at 20:03
  • \$\begingroup\$ So do you have a 12Ohm resistor connected to out in your simulation? \$\endgroup\$ – Voltage Spike May 18 '17 at 20:25
  • \$\begingroup\$ The above circuit does not have loads connected to the outputs. I have simulated the circuit with 12 ohm loads on the outputs and observed a similar response to what is presented in my simulation results. \$\endgroup\$ – kprince May 18 '17 at 20:41
  • \$\begingroup\$ But your RUN pins are floating? Will it even start regulating then? \$\endgroup\$ – Dejvid_no1 May 22 '17 at 21:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.