3
\$\begingroup\$

In this LT3652 solar charger application: enter image description here

the BAT pin first connects to a resistor of 30k, then to the Schottky diode's anode, then to a filter capacitor and finally to a current detector resistor and a battery. How crucial is it that we maintain this order while designing the PCB?

\$\endgroup\$
1
\$\begingroup\$

Order and placement of components does matter, a lot. Read the layout considerations section from the datasheet carefully, and follow it.

\$\endgroup\$
  • \$\begingroup\$ But apart from the layout consideration section of the datasheet, there must be some standard thumb rules or practices that are followed in these cases. \$\endgroup\$ – NRai Mar 29 '19 at 16:16
  • 1
    \$\begingroup\$ Yes, the good practice is to put the components to the schematic in the same order as they shall be in the layout. The problem is that you never know if the schematic is done following the best practices, unless you know the designer. \$\endgroup\$ – TemeV Mar 29 '19 at 16:26
6
\$\begingroup\$

It really depends on the circuit. In good schematics the components are arranged in such a way that it exposes their intent. You can gather this information from the datasheet. For example, C3 is said to be a decoupling capacitor. The closer you put this capacitor to the battery, the better decoupling you will have, since the loop for RF to ground will be shorter, and therefore parasitic inductance of the PCB trace will also be lower, which in effect will increase your decoupling efficiency.

So it definitely does matter. The hard task (for me personally) is to recognize the reason why the component is on a board, and design the layout in a way that maximizes the usefulness of this component and its purpose.

\$\endgroup\$
2
\$\begingroup\$

You want to minimize this area

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
  • \$\begingroup\$ I understand that the inductor is responsible for the magnetic flux but how does smaller loop area assists in its cancellation? \$\endgroup\$ – NRai Mar 30 '19 at 5:51
  • \$\begingroup\$ ALL thei wiring are inductors, as approximately 1mm / 1nanoHenry. Smaller loop area will linearly reduce the stored energy. And smaller loop area linearly reduced any primary-secondary coupling. The "Vfb" pin of the IC is how the IC monitors the output voltage; magnetically-induced trash onto Vfb causes uncertain control-loop behavior (servo-loop), and may produce chaotic (irregular) output voltages. \$\endgroup\$ – analogsystemsrf Mar 30 '19 at 17:25
  • \$\begingroup\$ Thanks for clarifying. What should I do if I could not decrease the loop area. Would adding more decoupling capacitors over the track would help? \$\endgroup\$ – NRai Mar 31 '19 at 6:55
  • \$\begingroup\$ Adding more decoupling caps "over the track"? Interesting thinking. And since you need to handle the very fastest of slewrates (200 volts in 200 nanoseconds, or 32 volts in 32 nanoSeconds), you can use 0.01 or 0.1uF caps. Let us know how this behaves, once you get a PCB; evaluate the switching edge jitter, over the Vin and Iload ranges, with and without the additional small capacitor that is intended "to reduce the loop area{. \$\endgroup\$ – analogsystemsrf Mar 31 '19 at 21:23
  • 1
    \$\begingroup\$ Ah! I see now that adding a decoupling capacitor is actually a way to keep the effective loop area minimum. Also, sorry for my vocabulary. I am not a formally trained EE. \$\endgroup\$ – NRai Apr 1 '19 at 10:11

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.