0
\$\begingroup\$

Im running a few simulations with an energy harvesting project using Peltier devices as a Thermo-Electric Generator with PSPICE for TI. The chip powering the boost device doesnt have a pspice model and I don't have all the data handy (or able to be produced yet for that matter) to generate a complete behavioral model. Are there any general rules for a fixed output voltage boost converter? I was thinking of using a voltage dependent current source since that is really what varies dependent on the input and load but I'm not sure how to use that while still having a steady 5V as output.

In summation, is it possible simulate (for given input range) a dc-dc boost converter(fixed output) with a combination of (voltage or current) regulated (voltage or current) sources with "reasonable" accuracy? Quotes because thats purposely ambiguous

Ckt specifics if it matters: TEGs will be arranged to produce a minimum of 500mV and about 300mA Boost module is Pololu U1V11F5 which uses the TPS61202

\$\endgroup\$
6
  • 2
    \$\begingroup\$ Focus on a question and not on trying to get opinion based feedback. That's how this site works. Questions seeking opinions will be closed. Ask a question that can deliver a fairly decent chance of a specific answer. Your previous question also lacked a goal-orientated question and nobody answered so... \$\endgroup\$
    – Andy aka
    Commented Feb 9, 2021 at 19:17
  • \$\begingroup\$ I didn't think I was asking for opinions or starting an open-ended discussion but hopefully I cleared it up with the edit. \$\endgroup\$
    – Darius
    Commented Feb 9, 2021 at 20:05
  • \$\begingroup\$ You can model it in a simulator as an ideal converter with your real input source so, yes. You can make the inductor have resistance losses and use a real diode model. You can even use a simulated mosfet. The answer is still yes. \$\endgroup\$
    – Andy aka
    Commented Feb 9, 2021 at 20:35
  • \$\begingroup\$ @Darius If what you need is just the DC voltage, varying according to whatever source you need, besides simple approximations with sources, you could actually model a boost converter as an averaged model, which eliminates switching and gives (virtually) the same voltage, variation, etc. \$\endgroup\$
    – a concerned citizen
    Commented Feb 9, 2021 at 22:32
  • \$\begingroup\$ Because you're using Peltiers as a source and they don't generate a ton of power even after you maximize the thermal difference across them, you should consider Maximum Power Point Tracking(MPPT). Note that the output voltage can only be continuous if the load is small enough and also that your converter has low efficiency at low input voltage(although the graph only shows down to 1V, the trend is clear). \$\endgroup\$
    – K H
    Commented Feb 10, 2021 at 4:54

1 Answer 1

Reset to default
0
\$\begingroup\$

The best way to do this in spice is with a PWL source (where you can define what the source needs to be in a transient simulation)

But with fixed DC/DC converters a current source to simulate loading is usually sufficient. A good load condition would be to start at zero at the beginning of the simulation and ramp up to its full load current.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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