#211 BuckConverterModules
Test a range of low-cost buck converter modules
Notes
There are many super-cheap buck converter modules available, many based on the LM2596S-ADJ. This is the place I record any notes on the individual units I get to see.
Four so far:
- Module 1 - LM2596S 2.6-30V Buck Module
- Module 2 - LM2596S 1.23V-30V Buck Module
- Module 3 - XL4015 1.25V-36V Buck Module
- Module 4 - MP2307DN 1.0V-17V Buck Module
Module 1 - LM2596S 2.6-30V Buck Module
The example I’m using is similar to this seller listing. Typical price USD$0.93 with free shipping.
Buck circuit comprises:
- LM2596S-ADJ (as marked, but behaves more like an LM2576 - see note below)
- 330µH output inductor
- 220µF output capacitor
Performance:
- minimum voltage achieved in practice: 2.56V
- switching frequency appears to be in the ~50kHz range rather than the 150kHz per datasheet(??)
- switching spikes around 30-40mV
Conclusion
This is one of the most common modules, but quite noisy and can’t regulate to very low voltages.
It also behaves more like an LM2576:
- the 330µH inductor is also more suited to a standard LM2576 topology rather than the 33µF recommended for the LM2596.
- the apparent switching frequency is closer to the LM2576 52kHz rather than the LM2596’s 150kHz.
- so is this a fake or QC-fail LM2596S-ADJ? Possibly, if my scope traces are reliable.
Module 2 - LM2596S 1.23V-30V Buck Module
The example I’m using is similar to this seller listing. Typical price USD$0.79 with free shipping.
Buck circuit comprises:
- LM2596S-ADJ (as marked, but behaves more like an LM2576 - see note below)
- 470µH output inductor
- 220µF output capacitor
Performance:
- minimum voltage achieved in practice: 1.22V
- switching frequency appears to be ~50kHz, not 150kHz per LM2596 specs.
- switching spikes under 10mV
Conclusion
Performance is pretty good, but this does not appear to be a genuine LM2596S-ADJ.
- the 470µH inductor is also more suited to a standard LM2576 topology rather than the 33µF recommended for the LM2596.
- the apparent switching frequency is closer to the LM2576 52kHz rather than the LM2596’s 150kHz.
- so is this a fake or QC-fail LM2596S-ADJ? Possibly, if my scope traces are reliable.
Module 3 - XL4015 1.25V-36V Buck Module
The example I’m using is similar to this seller listing. Typical price USD$2.68 with free shipping.
Board comprises:
- XL4015E5 with toroidal inductor and capacitors as per datasheet recommendations
- digital voltmeter with pushbuttons to switch display on/off and select input/output voltage for display
Performance: up with the best. There’s no scope capture below because despite my best attempts, the output voltage remained flat as a board (no detectable ripples or spikes .. just the normal noise floor)
Module 4 - MP2307DN 1.0V-17V Buck Module
The example I’m using is similar to this seller listing. Typical price USD$0.51 (USD$5.06 for 10) with free shipping.
Board comprises:
- MP2307 monolithic buck converter with standard support circuits per datasheet recommendations
- pretty small package!
Performance: quite impressive for the size. There’s no scope capture below because despite my best attempts, the output voltage remained flat as a board (no detectable ripples or spikes .. just the normal noise floor)
Test Bed Construction
In the test configuration, I have the buck converters adjusted to ~3V output, and two blue LEDs in parallel with the load. The LEDs have a forward voltage Vf of ~3V.
