0
\$\begingroup\$

I have a DC-DC buck boost converter (TPS55288) that I was planning on using for a project. I'm making a revision to my current heated gloves, and my circuit needs to be as small as possible since the electronics will be stored inside the gloves themselves.

I need the converter to heat up a 5 ohm heating element with a output voltage ranging from 5V to 8V, an input voltage ranging from 8.4V to 6V, with a switching frequency of 750KHz.

I was going to use a 4.7uH inductor, but I noticed that my board got uncomfortably hot (around 60 degrees C probably, I didn't measure it) when running it at a 5V output. When I changed to a 1uH inductor, the board was a lot cooler (probably around 40 degrees C.) With this in mind, I was thinking of using the 1uH inductor for the sake of thermals.

So I wanted to ask: Why does a lower inductance value generate less heat on my PCB? What are the positive and negative effects of having a lower inductor size with a large current ripple on a static load?

\$\endgroup\$
15
  • 1
    \$\begingroup\$ Do you really need something so fancy for a heating element? You could probably get away with 30Hz and zero inductance. The heater won't care and thermal characteristics smooth out themselves without the need for electrical intervention. Lower inductance = shorter wire = less resistance all other things equal. But lower inductance = higher ripple = higher peaks which should mean more heat due to I^R so it does seem a little strange. at least for a buck. Yours is buck boost though. \$\endgroup\$
    – DKNguyen
    Oct 2, 2021 at 20:18
  • 2
    \$\begingroup\$ 1) It is possible that lower inductance value by itself didn't "solve" your thermal issue. Suppose the 4.7 uH inductor had a saturation current that is too low, that will make the converter ineffient and heat up. If your 1 uH inductor didn't saturate then that could result in a better efficiency, less power loss and less heating up. \$\endgroup\$ Oct 2, 2021 at 20:23
  • 1
    \$\begingroup\$ You are putting almost 13W of heat into a glove? Ouch. \$\endgroup\$ Oct 2, 2021 at 20:26
  • 1
    \$\begingroup\$ I agree with @DKNguyen that for a heating element, such a fancy DCDC might be overkill. Why not use a heating element that can generate enough heat even at the lowest voltage. Then use PWM to switch it on/off to regulate the power. Again: the thermal capacitance will make the response of the heating element slow so even a low frequency PWM control (say 100 Hz) will be more than fast enough. \$\endgroup\$ Oct 2, 2021 at 20:27
  • 1
    \$\begingroup\$ @Jay The solution to that is use less resistance so more heat is produced with less voltage and throttle back on the duty cycle when the battery is full, not step it up because it's just a heating element. And use zero inductance because the heating element isn't picky. You can outright pulse power to it it but thermal mass keeps the temperature smooth. \$\endgroup\$
    – DKNguyen
    Oct 2, 2021 at 20:39

1 Answer 1

3
\$\begingroup\$

For heating, all you need is a resistive element that will generate enough heat at minimum input voltage, and PWM to control the average power. The PWM doesn't need to be fast, only quicker than the thermal time constant of the load, even the Hz range would be quick enough for a glove heater.

In any design, the smallest, cheapest and most reliable components are the ones you don't fit, because you don't need them. This is the most important learning experience to take from this design. You don't need the inductor, or a fancy high frequency DC-DC chip.

\$\endgroup\$
1
  • \$\begingroup\$ To add to that: what Neil_UK means is that you don't need an inductor at all, nor the output capacitor, nor the flyback diode. All you need is a MOSFET. Easy! \$\endgroup\$ Oct 2, 2021 at 22:31

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.