0
\$\begingroup\$

I am trying to design a switch-mode boost converter using a MAX1771. As far as I know, when designing this kind of converters there are a few parameters that must be set. In my case these are:

  • Vin = 12V
  • Iin (max) = 2A
  • Vout = 210V
  • Iout = 5-6mA (with pretty much no variation)

Once this is set it is important to decide whether we want the Boost converter to work in DCM o CCM (trying to avoid to work on the frontier between both modes). Accordingly to this, an inductor value is chosen. Let's say that in my case a 100uH inductor with a switching frequency of 300KHz would work.

This is how I would design it with an IC like TPS4021x or a 555 timer (just to stay a couple of examples). Please correct me if it is a bad method. The problem is that I can not work out a design with MAX1771 using this process since I can not choose the switching frequency nor duty cicle.

my questions are the following.

  1. Which process do I have to follow to design a switch mode boost converter using a MAX1771 integrated? How can I control the circuit parameters and know how is it working?
  2. How do I choose a proper Rsense value for such a high output voltage? (No information is provided in the datasheet for these Vout values).
\$\endgroup\$
3
  • \$\begingroup\$ The first step is choosing a chip that can support 210 V. According to its web page, the MAX1771's maximum is 100 V. Use something likle the MAX17597 or LM3488. \$\endgroup\$
    – CL.
    Feb 14, 2022 at 19:55
  • \$\begingroup\$ Strangely, the output voltage of any circuit using this chip never reaches the chip so I'm unsure why it should say that. \$\endgroup\$
    – Andy aka
    Feb 14, 2022 at 20:06
  • \$\begingroup\$ @CL. I saw it but I have seen many people doing it with max1771 and certainly I do not really know why those designs work properly. I am not sure which is the limitation they think of when they set a maximum Vout of 100V. Maybe because is the maximum voltage you can get working on CCM? Is just a guess, I do not really know it. Despite this, say I chose a MAX17597. I would have the same problem. \$\endgroup\$ Feb 14, 2022 at 20:08

1 Answer 1

1
\$\begingroup\$

Once this is set it is important to decide whether we want the Boost converter to work in DCM or CCM

The chip will handle both as per this: -

enter image description here

When the load gets heavier it runs in CCM and, should be fairly seamless so no need to worry about that: -

enter image description here

Regards operating read this: -

enter image description here

And that tells me you can use "normal" boost converter formula for this or, my crappy website calculator to get an idea of what mode you might be in for a given set of input parameters: -

enter image description here

As you can see, the peak inductor current is 0.826 amps so, set Rsense accordingly to about the value that gives this current. CS pin voltage level is 100 mV. I chose an operating frequency of 50 kHz and an inductor value of 68 μH for this simulation.

I chose 68 μH because that inductance shouldn't be so high that self-resonance prevents the high output voltage being reached. Concentrate on finding a really good inductor that has a self-resonance at least ten times higher than the operating frequency (if not more).


Conclusion

In any circuit design you have to do checks once they are built and, this is particularly important here. Personally, I'd be thinking of modelling the circuit if I wanted it to run in DCM else, I'd force it to run in CCM by choosing an inductor that was hundreds of micro henries.

Either way, I'd be wanting to check this circuit out very carefully when it was built. I say "very carefully" because this particular device is not that clear on what it actually does i.e. no timing waveforms or timing examples (unlike other devices that appear more well-defined).

\$\endgroup\$
10
  • \$\begingroup\$ Thank you. I have already ben playing around with your boost calculator. I would like to know why did you choose a switching frequency of 50KHz. I don't really yet understand whether is gonna switch at 300Khz or at 50KHz. \$\endgroup\$ Feb 14, 2022 at 20:15
  • \$\begingroup\$ That's a really cool crappy calculator. \$\endgroup\$
    – TypeIA
    Feb 14, 2022 at 20:45
  • \$\begingroup\$ I chose 50 kHz because I know that the inductor is going to be a potential problem i.e. getting a decent inductor with a high self resonant frequency (preferably no lower than 1 MHz) hence I've tried to consider running the switcher at 50 kHz. It won't switch at 300 kHz - that's the maximum it could ever switch at. \$\endgroup\$
    – Andy aka
    Feb 14, 2022 at 20:50
  • \$\begingroup\$ @Andy aka I am still having a problem determining the switching frequency. I understand that the 50KHz you chose is just an example you think could work. But how do I have to design my SMPS without really knowing the exact switching frequency? Is there any problem if I design it for it to work in DCM at 50KHz and then it turns out that it is switching at, i.e. 100KHz? \$\endgroup\$ Feb 14, 2022 at 21:01
  • \$\begingroup\$ Yes, the switching frequency isn't all that easily decided from the data sheet. I would work to try and reduce inductance as much as possible then breadboard the design on a proper pcb. Maybe drop to 47 uH. Alternatively try and simulate what the chip supposedly does from the block diagram. This can appear daunting but might be the one way to uncover what might happen. I'll have another think tomorrow and please do remind me in a comment. \$\endgroup\$
    – Andy aka
    Feb 14, 2022 at 23:56

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.