0
\$\begingroup\$

I am working on a project where I have used MC-34063 boost supply on 3 pcb's and all the 3 places I have the same issue ie

  • output voltage is steady on open circuit, but soon as ANY load is connected, MC starts drawing massive amount of current, and puts the Bench Supply from CV to CC mode. (it goes upto 1.5amp)

I am not sure what is the issue here as I have taken care of these precautions

  • 4 layer board
  • consistent ground plane
  • very small ground loop for any return paths
  • 100-150 mohm ESR caps and inductors

I am really clueless as to why the voltage is sagging, in all cases when I boost voltage from either 3.3v to 12 or 5v to 12v.

Current requirements in either case is less than 100mA and inductor is rated for 1.8amp.

if it helps, what I have noticed is disconnecting the 180ohm/drive resistor even though there is no boost, but there is no huge current consumption either!

Additionally, I tried

  • changing drive resistor to lower values of 10ohm and higher values of 1kohms
  • adding more capacitors at output
  • added additional LC filter at output

Please see below for circuit daigram, layout and pcb pictures.

lastly please note, all the schematics, layouts and PCB's are pretty similar for both 5V and 12v, except for

  • feedback resistors (1k, 9K) for 12V, and (1k, 3K) for 5v and
  • output capacitors ratings of 10V for 5v boost, and 16V for 12v Boost

PCB1 PCB2 PCB3 Board Layout - same for all 3 pcb's Schematic - same for all pcb's except for external resitors

\$\endgroup\$
4
  • \$\begingroup\$ Your startup currents may be huge too >1A \$\endgroup\$ Feb 24, 2021 at 6:59
  • \$\begingroup\$ when I supply the same load from bench supply, current draw in all cases is less than 0.1amps. I think it might have something to do with the inductors as when i upped the inductor to x10, atleast the draw goes down to <0.1amp, but output still sags, Looking at the inductor calculations now to see where I messed up! \$\endgroup\$
    – Amogh Jain
    Feb 24, 2021 at 7:44
  • \$\begingroup\$ idk why I used a 22uH inductor when my own calcs show a minimum of 47uH. I am very stupid! \$\endgroup\$
    – Amogh Jain
    Feb 24, 2021 at 7:45
  • \$\begingroup\$ not so stupid, it will give more boost current \$\endgroup\$ Feb 24, 2021 at 15:01

2 Answers 2

1
\$\begingroup\$

A few major issues with your specs.

The datasheet does not show it can boost 3V3 to 12V and output 1.5 for both conditions.

Looking at the Darlington saturation voltage of 1.5V and 3.3V input drops the input boost voltage of 1.8V so the input boost current is 12.5/1.8 almost an Amp (even if you cut all the other losses)

Your diode drop may also be excessive.

You ought to lower f for this chip which is over it's recommended operation due to the Darlington saturation recovery times even though for some uses it works.

Being a low cost may not be a good reason to choose this IC when there are far better ones using TI's Power Designer or similar.

DCDC design is hard and you should follow the TI design guide EXACTLY in every detail. (This old chip needs a lot of work)

\$\endgroup\$
2
  • \$\begingroup\$ can you please recommend a few ic's for the boost operation? diode drops about 0.3V -0.4V, and max output current draw is 0.1amps in either the small motor load, or the relay load. with BMS there is no draw as the supply is solar, with max amperage being 320mA. \$\endgroup\$
    – Amogh Jain
    Feb 24, 2021 at 7:49
  • \$\begingroup\$ No.. Go register and use the design tool. webench.ti.com/power-designer/switching-regulator/select.. Motor start current is much higher \$\endgroup\$ Feb 24, 2021 at 15:05
1
\$\begingroup\$

for anyone else working with MC34063 here is what I changed :- (basically followed the chip specs with as little deviation as possible)

  1. Reduced the value of R_Inb to 100 Ohms

  2. Increased R_Pk from A) 0.22 to 1.1ohms for 5V solar Booster, and B) 0.22 to 0.66 for for 12v Boost.

  3. Added the Timing Capacitors as per calculations (470pF, and 330pF).

  4. I ensured inductor size was the right and 20% oversized!

Now all works like a charm :-)

\$\endgroup\$

Your Answer

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

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