SparkoS Labs Dual Power Supply Review

Introduction

After previously building my own power supplies for projects I have built I recently learned that SparkoS Labs who make high quality audio components have brought out their own power supplies to aid the DIY community.

They come in two flavours, a Dual which provides a single positive and negative rail and a Quad which provides dual positive and negative rails. For the purpose of this review I will be taking a look at the Dual Power Supply. I would also like to thank Andrew at SparkoS Labs for providing the power supply.

Design

This is a high quality and well designed power supply. At the AC input you can see a small RC filter to help smooth the voltage source before it is rectified through the four high speed, low noise Schottky diodes into raw DC voltage before reaching the reservoir capacitors.

There are three 2200uf / 50v capacitors on each rail that provide 6600uf capacitance which aims to help limit the inrush current which in turn provides some additional protection to the voltage regulators. The regulators then transform the raw DC voltage into a stable, clean and regulated output to the desired voltage.

Features

The power supply features onboard terminal blocks to make connecting the inputs and outputs hassle free. Another handy feature is there are test points provided in case for any reason you need access to the raw unregulated DC voltage.

Large heatsinks have also been utilised which are capable of dissipating up to 3.5 Watts each with airflow holes underneath them on the PCB which help to keep the regulators running cool.

There are six 2200uF 50V Nichicon filter capacitors, charge current softening resistors to reduce transformer magnetic fields and noise, and 100uF Panasonic EEU-FR1H101B output capacitors on each regulated output rail.

Lastly these power supplies come with all required components excluding the regulators and also include full mounting hardware and standoffs.

Although extremely good specifications I personally would have liked to see an LED indicator for each rail and maybe a trimpot for each rail to finetune the output voltage. This however will only be a minor annoyance for some people. You could always hook up an LED with a series resistor to the raw DC test points for power indication if you really wanted to.

Testing

Using my limited skills and equipment I managed to take a peek at the ripple both on the raw unregulated DC output and the clean regulated DC output and the results were quite surprising.

It is clear that this was designed primarily to suit the designers own ultra low noise regulators due to the minimal amount of filtering on the raw unregulated DC output.

I personally think this power supply could have benefitted from a multiple RC filter to knock out that 50Hz ripple which would clean up the raw supply considerably. This would make the power supply more accommodating towards the use of the old monolithic 78xx/79xx regulators which would appeal to users with lower budgets.

With this in mind, if you plan to use this power supply for audio use then be sure to use the SparkoS Labs regulators to get the best possible performance unless you need to power something like a microcontroller, then use monolithic regulators.

• 

• 

• 

Here is a little extra info regarding measurements from Andrew at SparkoS Labs:

Our discrete regulators typically have single digit uV noise. For example, 12V regulators are spec’d at just under 6uV. That kind of stuff is impossible to measure with most scopes as they dont have that kind of resolution. So an external noise amplifier is required to get it up to measurable levels.

For the capacitors and such we are talking about the input ripple rejection of the regulators specifically at 120HZ. Our regulators have around 124dB or so ripple rejection. ICs have maybe 70 – 80 dB.

To put some numbers to that, 120dB is a factor of 1.5 million. So for every volt of ripple on the input you get 666nV of ripple coming through to the output which is well below the noise floor.

For an IC regulator with 80 dB rejection, that is a factor of 10,000. So for every volt of ripple on the input you get 100uV of ripple coming through to the output.

That is above the noise floor of our devices but not of the IC’s as their noise tends to be 75uV and upwards (as opposed to 6uV with our devices).

See the image below for some device comparisons.

Final Thoughts

This is a fantastic power supply and although I did not have the chance to test it properly, I certainly plan to use this in my next audio project.

Maybe in the future it would be nice to see a single rail version to accommodate users who wish to build external power supplies for DACs and headphone amps for example which would replace standard AC-DC wall adapters.

For comparison, my own Bi-Polar power supply that I used in Waynes BA2018 Linestage sounds excellent. Dead quiet and no audible noise. This is despite me measuring around 700uV ripple on the output. I measured around 70uV ripple on the SparkoS Labs power supply which I actually believe to be much lower as my scope has its limitations.

So if you need a Bi-Polar power supply for your upcoming audio project, then the SparkoS Labs power supply is certainly one you should consider when looking at your requirements.

For more information about this product and other products please visit https://sparkoslabs.com/