[Power Supply] Switching/hybrid vs Full Linear

[dougduck]

Among the best DAC models reviewed on the forum is the TOPPING D90 III Sabre, a switching-mode power supply-based/hybrid device that performs as well as the SMSL SU-10, which is a full linear PS.
Is the performance of DC-DC converters like the TPS5430 in the D90 really so good that a single LDO in the last stage can suppress all switching noise?
What do you guys think?

 

[sam_adams]

This may come as a shock to you, but with audio devices, we don't listen to the power supply.

Oh, and by the way, let me extend a warm, Hank Scorpio welcome to ASR.

 

[xanalog]

Among the best DAC models reviewed on the forum is the TOPPING D90 III Sabre, a switching-mode power supply-based/hybrid device that performs as well as the SMSL SU-10, which is a full linear PS.

It's a baited question that you ask. But the shortest answer is evident above:
Depends on the topology of the specific design/implementation/verification.

 

[dougduck]

This may come as a shock to you, but with audio devices, we don't listen to the power supply.

Oh, and by the way, let me extend a warm, Hank Scorpio welcome to ASR.

Thank you.

It's really a shock.
I learned in college that any conversion/amplification was just managing the voltage/current of the power supply.

 

[dougduck]

It's a baited question that you ask. But the shortest answer is evident above:
Depends on the topology of the specific design/implementation/verification.

I worked for a while developing hospital equipment, and they avoided the use of DC-DC converters because they modulated the signal from sensitive converters.
It's been a few years, and I haven't really kept up with the "evolution" in audio.
But apparently people around here are a bit "bitchy," and they get offended by this kind of question?
My doubt is real, it is not any kind of provocation.

 

[JSmith]

suppress all switching noise

Well yeah... or we'd see it in the measurements;

TOPPING D90 III Sabre DAC Review

This is a review and detailed measurements of the TOPPING D90 III Sabre balanced stereo DAC. It was sent to me by the company and costs US $899. As you can see, the design is what we have know about most of the Topping products. I do wish at this price that we had a high resolution display that...

www.audiosciencereview.com

JSmith

 

[Doodski]

I worked for a while developing hospital equipment, and they avoided the use of DC-DC converters because they modulated the signal from sensitive converters.
It's been a few years, and I haven't really kept up with the "evolution" in audio.
But apparently people around here are a bit "bitchy," and they get offended by this kind of question?
My doubt is real, it is not any kind of provocation.

I serviced thousands of expensive car amps up to the huge monster sized humongous things and most all of them had major switching power supply noise. The power supply transformers where constantly whining, peeping, buzzing, very loudly hissing and all sorts of audible issues that where right in one's face and clear as a bell loud. I replaced many many switching power supply toroidal transformers. The transistors and FETs where always hissing too, peeping and generally the entire system from the power supply to the amp output section was a noise maker. It also became more and more common for devices like camcorders and other snazzy rare personal electronics to have fancier and fancier switching power supplies & battery chargers. They had all sorts of new for the day monitoring systems and over-this sensing and under-that sensing and temperature sensing and all the new stuff of the day. They where costly, they had service manuals, full parts availability, we actually unscrewed the covers and repaired them when they failed for awhile until the costs of them came down. All this switching power supply stuff has improved majorly compared to not long ago. Even the desktop PC power supplies are way way better. I really don't get why people are determined with today's gear to find fault with switching power supplies used in audio gear. Why are these people magnified focused on this and where are these ideas coming from?

 

[xanalog]

But apparently people around here are a bit "bitchy," and they get offended by this kind of question?
My doubt is real, it is not any kind of provocation.

I may have given you the TL&DR answer and on your behalf I went to three different aliexpress store fronts and searched:
1 >> Topping LPS

2 >> FiiO-K7

3 >> SMSL RAW

These were just arbitrary first hits and screen captures from pages the 3 searches landed me on.

Depends on the topology of the specific design/implementation/verification.

...each with exampelary and verified test results.
I hope you sympathize, now.

Welcome aboard.

 

[dougduck]

I really don't get why people are determined with today's gear to find fault with switching power supplies used in audio gear.

I don't know if people are determined to find fault with switching power supplies.
Switching power supplies are excellent in many ways, but one of their characteristics is the noise generated by their switching. And this noise is very damaging to sensitive signals like converters.
This is a characteristic that's there, like it or not. They've been around for decades; it's not a recent development.

At least in my experience, it was with life-saving equipment, so every detail mattered.
Perhaps it's simply not relevant with audio, and that's what I'm trying to understand.

 

[dougduck]

2 >> FiiO-K7
View attachment 483406

These were just arbitrary first hits and screen captures from pages the 3 searches landed me on.

...each with exampelary and verified test results.
I hope you sympathize, now.

Welcome aboard.

Oh yep!
That's impressive, several regulators, including the negative rail, are generated from the positive rail itself.
There's nothing sophisticated about that, you know?
Is a good PCB layout alone responsible for the good performance measured?
That's what I want to study.

best regards.

 

[solderdude]

The thing is that SMPS can be lower noise than linear power supplies.
SMPS handle input voltage sagging and peaking better.

Both SMPS and linear can have hum on the DC out but some SMPS also can have HF switching noise.

The thing with amplifiers is that CMRR is usually better at low frequencies than high frequencies.

Differential noise is easily lowered with an inductor and capacitor.

Amplifiers using an SMPS can have smaller sized smoothing caps.

The biggest possible issues with SMPS, however, is common mode noise and possibly leakage currents.
This can be lower in linear power supplies but depends on the physical construction of the transformer.
When for instance the primary coil is wound on top of the secondary coil and the L is close to the secondary coil you can have substantial HF noise capacitive coupling.

For SMPS leakage currents can be low to (depends on the usage of Y-caps).

Sooo ... there both are advantages and downsides to both power supply types and which one is 'better' in which aspect depends on the actual construction and usage as well as conditions they are used in.

For longevity the linear ones are usually 'better' than cheaper SMPS but a well designed SMPS can also have a long service life.

 

[dougduck]

Differential noise is easily lowered with an inductor and capacitor.

No, if that were true, any cheap SMPS would be noise-free because they all have LC filters.
The most easily controlled factor is ripple, but in this case, ripple isn't important because the current consumption is low.
Switching noise is also differential and not easy to filter, especially when there are more than one dc-converter in the same circuit sharing the same GND reference. Any implementation detail affects the feedback of these step-down; they don't like that.

I'm currently considering the possibility that these new converters(DAC/ADC) have an aggressive filter in the audible range, but not in the modulation range.
That way, it doesn't have to be perfect, just appear perfect.
It looks beautiful in the pure sine wave test, and everyone is happy.

**Should I spend $1,000 to measure rail noise?
Maybe not; I'll leave that to the expert reviewers.

regards.

 

[solderdude]

No, if that were true, any cheap SMPS would be noise-free because they all have LC filters.

Differential noise can be lowered with a sufficiently large L and C.
The issue, however, is that manufacturers only apply enough LC filtering internally to pass EMC testing which is higher up than the audible range.

**Should I spend $1,000 to measure rail noise?
Maybe not; I'll leave that to the expert reviewers.

regards.

Why would you want to spend money on measuring rail noise when all that matters is the analog out ?

 

[Snickers-is]

The rectifiers are an important source of current transients in a power supply. The higher the frequency of the current pulses, the lower the energy of the current spikes in the rectifiers will be. Since the rectifiers draw current from the transformer, the transformer will give off high bandwidth electromagnetic transients corresponding to these impulses. A larger transformer will typically act as a larger antenna. The number of turns and gauge of wire also plays a role here.

A switching power supply do come with its own possible challenges as well. Some generate common mode switching noise. They may have a non fixed switching frequency as well. When used with gear that rely on signals in the same frequency range, this can cause unpredictable interference. This is especially true when a switching power supply is really made at a budget.

However, if what you are looking for is the best possible performance, you would not make an SMPS at a budget. You do not have to go very far to achieve a level of performance that might be challenging to achieve with a 50/60 Hz transformer.

We want the PSU to be invisible in the product, so the power is just delivered to the circuits like it came from an outside source. This requires some effort regardless. Both ways are workable. But there are several benefits you can have with an SMPS, like PFC and active output voltage regulation.

 

[marin.weigel]

It's been a few years, and I haven't really kept up ...

You actually answered your own question right there, although I'd say it's rather been a few decades.

 

[MaxwellsEq]

I worked for a while developing hospital equipment, and they avoided the use of DC-DC converters because they modulated the signal from sensitive converters

I know this is off-topic for an audio forum but can you explain this? Is there something specific about the signals or even the noise being processed in medical gear? Why might this translate to audio?

 

[dougduck]

The rectifiers are an important source of current transients in a power supply. The higher the frequency of the current pulses, the lower the energy of the current spikes in the rectifiers will be. Since the rectifiers draw current from the transformer, the transformer will give off high bandwidth electromagnetic transients corresponding to these impulses. A larger transformer will typically act as a larger antenna. The number of turns and gauge of wire also plays a role here.

A switching power supply do come with its own possible challenges as well. Some generate common mode switching noise. They may have a non fixed switching frequency as well. When used with gear that rely on signals in the same frequency range, this can cause unpredictable interference. This is especially true when a switching power supply is really made at a budget.

However, if what you are looking for is the best possible performance, you would not make an SMPS at a budget. You do not have to go very far to achieve a level of performance that might be challenging to achieve with a 50/60 Hz transformer.

We want the PSU to be invisible in the product, so the power is just delivered to the circuits like it came from an outside source. This requires some effort regardless. Both ways are workable. But there are several benefits you can have with an SMPS, like PFC and active output voltage regulation.

yep! agree.
But I don't think I expressed myself correctly.
Are you a designer for any brand?
At no point did my question mean to impose a topology as good or bad; that's not the point.

As you mentioned: "However, if what you are looking for is the best possible performance, you would not make an SMPS at a budget."
I agree, and that's precisely what impresses me when I see these published results.
Seemingly simple implementation of traditional converters.
So what's truly responsible for these excellent results? Just a good PCB layout? That's what I want to understand.

I know this is off-topic for an audio forum but can you explain this? Is there something specific about the signals or even the noise being processed in medical gear? Why might this translate to audio?

They also use AD/DA converters, but for higher frequencies and lower-amplitude signals, which makes implementation more difficult, as small uV impairs the conversion and alters the result.
With this type of equipment, signal fidelity is more important than the best-looking signal.

Why would you want to spend money on measuring rail noise when all that matters is the analog out ?

Because engineers are curious and want to understand how things actually work, it's different from the end consumer.

I think I asked the question in the wrong place, but I thank everyone who contributed their answers; I'll take them all into consideration.
Especially to the members who post teardown photos of these devices—that's amazing!

The models mentioned in #1 are just because they were models published here; it's not a direct comparison between brands/models, by any means.

 

[MAB]

yep! agree.
But I don't think I expressed myself correctly.
Are you a designer for any brand?
At no point did my question mean to impose a topology as good or bad; that's not the point.

As you mentioned: "However, if what you are looking for is the best possible performance, you would not make an SMPS at a budget."
I agree, and that's precisely what impresses me when I see these published results.
Seemingly simple implementation of traditional converters.
So what's truly responsible for these excellent results? Just a good PCB layout? That's what I want to understand.


They also use AD/DA converters, but for higher frequencies and lower-amplitude signals, which makes implementation more difficult, as small uV impairs the conversion and alters the result.
With this type of equipment, signal fidelity is more important than the best-looking signal.


Because engineers are curious and want to understand how things actually work, it's different from the end consumer.

I think I asked the question in the wrong place, but I thank everyone who contributed their answers; I'll take them all into consideration.
Especially to the members who post teardown photos of these devices—that's amazing!

The models mentioned in #1 are just because they were models published here; it's not a direct comparison between brands/models, by any means.

You can test this out.
Get a high performing device with an external power supply and swap supplies. You can try linear, low cost SMPS, batteries, etc. RME has several supplies available for their gear, people have even tested them even if I can't find the test someone posted at ASR, there are differences but both definitely not in a transformative way, especially in the audio range. The device retains it's limitations irrespective of the power supply. I imagine some devices will be much more sensitive to power supply than an RME.

I do have a range of power supplies. I test some things, like a DIY preamp, in order to debunk all of this OpAmp rolling business. I have done tests with an SMPS, a lab bench supply, an old linear supply I made, and a battery, to name a few. They all perform about the same, unless I introduce a ground loop in my measurement setup. I haven't posted a test of the supplies specifically, was too busy testing the nothing that is OpAmp, but in all of my power supply wanderings, I don't find them transformative.

I recall someone replaced an amplifier's internal supply with an SMPS. They changed test hardware between initial and final test though, so hard to make an apples to apples comparison. The SMPS amp looks cleaner for mains, but hard to tell the rest of the difference because of the test residual differences between the two measurements.

I don't think much about the impact of supplies, more on the grounding. These are audio devices, if they are somehow sensitive to supply it may be curious, but would be an odd sensitivity for a consumer audio device. I think Naim specialized in devices that were intrinsically sensitive to the power supplies, so that they could sell lots of expensive upgrades. I don't own any, but would be interested in seeing why those were legendarily sensitive to all sorts of things like power supplies.

edit: typo

 

[xanalog]

I am guessing that a lot of the members have a fetish for beefy reserves (aka big*ss Capacitors), as an integral part of our power-sources.

 

[Doodski]

I am guessing that a lot of the members have a fetish for beefy reserves (aka big*ss Capacitors), as an integral part of our power-sources.

I had a major fetish for coffee can sized capacitors for decades. I had a drawer full of hand sized and larger caps for a bunch of years and I was even able to use a few of them to modify stuff with better smoothing caps. But... On consideration of the actual reality of high frequency switching power supplies and the smaller caps required and the major improvements that SMPS gear has received in recent time I am very gung ho on switching power supplies.