buck converter

[duncan webster 1Participant @duncanwebster1]

I've bought one of these, initially to use as a battery charger

**LINK**

Is there a good reson I can't replace the on board potentiometers with panel mount ones and use it as a bench power supply? I can fit a digital amp/volt meter easy enough.

[duncan webster 1Participant @duncanwebster1]

[Martin ConnellyParticipant @martinconnelly55370]

Application 1 in the description seems to be using it as a variable power supply so I think you would be ok using it as you suggest.

Martin C

[SillyOldDufferModerator @sillyoldduffer]

The main objection to using them directly as a bench supply is lack of filtering on the output. Mucky DC isn't a problem for their intended purpose (battery charging and LED lighting), but dirty power upsets electronics in odd ways. It would be interesting to check how noisy the DC of a particular unit is with an oscilloscope. If it's horrible, not difficult to improve the filtering with external components, but I still prefer old fashioned linear power supplies for bench testing amplifiers and similar.

Dave

[Joseph Noci 1Participant @josephnoci1]

As Dave said , not to difficult to filter decently – 2 capacitors and an inductor will help a lot. A more concerning issue for use as a bench test power supply is the output voltage overshoot on switch-on. Not normally a problem when charging batteries, etc, but not great when powering 'sensitive' bench projects. The XL4016 ( the PWM DC-DC used in that module) is quite good – I have used a lot of them – but the application note suggests at least 1000uF of filter cap on the output side – the module you have linked to has only 220uF, so the output may spike a little more, and may have more noise on it.

The bigger issue when using switch-mode supplies as variable bench supplies is that the design parameters are rapidly exceeded when using it to give output voltages 'close to zero volt' up to 'near the input voltage' .

The buck inductor is optimum over a reasonably narrow input/output voltage versus current range, so one size does not work very well over an infinitely variable input voltage and output voltage range – the result is that output ripple and output transient response suffers. At high end input voltages, with low output voltage, output voltage regulation will suffer and ripple is hugely increased – easily 300/400 fold ( due to the then very small PWM mark/space ratios) – exactly what the 'sweet spot' is for that module depends on the chosen inductor value and output capacitors..

Bottom line – if the application is not to demanding – go for it – it is inexpensive enough to ignore!

Else, just be aware of pending limitations at the top and bottom of the range.

Joe

[SillyOldDufferModerator @sillyoldduffer]

Found an 12V LED power supply and looked at it with my oscilloscope.

As can be seen about 160mV of noise on 12VDC which is poor. However zooming in on the waveform shows it to be decently sinusoidal (ie clean). The not very stable base frequency is about 700kHz and it's modulated at about 70kHz, both easy to filter.

I used one of these on an Arduino project without bothering to clean up the PSU. The Arduino and a motor controller both work fine – their input filtering is good enough.

Dave

[Russell EberhardtParticipant @russelleberhardt48058]

I have a similar one wired up to a laptop charger and digital I/V meters. Extra filtering easily reduced the ripple to below 10 mV p-p. The most difficult part of the project was removing the control pots from the board!

Russell

[duncan webster 1Participant @duncanwebster1]

Thanks for the replies, I've got both a scope and some big electrolitic capacitors, so I'll have a play

[Author]

Posts