Linear Vs Switching Supplies In Audio Equipment

[Th3_uN1Qu3]

Re: Linear Vs Switching Supplies In Audio Equipment

Most laptop amps are junk and i can confirm so myself... but when the average Joe Consumer doesn't know the difference between 1024x768 and 1280x768 on his 16:9 LCD, it doesn't really matter, does it?

[Krankshaft]

Re: Linear Vs Switching Supplies In Audio Equipment

Audiophools would have a field day were talking home theater systems here not mainstream just bought at Wally World consumer electronics.

You know the same people that say Blackgates make a difference in audio quality.

[Th3_uN1Qu3]

Re: Linear Vs Switching Supplies In Audio Equipment

When you go past a certain level of quality (and obviously price) switchmode supplies are the way to go. But cheap switchers are noisy and unreliable hence the use of undersized toroids in most HTIB systems.

[Pyr0Beast]

Re: Linear Vs Switching Supplies In Audio Equipment

And toroids with unfiltered output are better ? I really doubt that.

[Th3_uN1Qu3]

Re: Linear Vs Switching Supplies In Audio Equipment

And toroids with unfiltered output are better ? I really doubt that.

Never seen unfiltered outputs on an audio device, ever. It would go all hummmmm and bzzzzzz if that were the case. Maybe you meant unregulated. And unregulated supplies have their place too.

[Pyr0Beast]

Re: Linear Vs Switching Supplies In Audio Equipment

Never ?

Toroidal supplys are unfiltered and unregulated.
They are merely rectified and smoothened to some extent.

[Trinite]

Re: Linear Vs Switching Supplies In Audio Equipment

Even linear supplies are at least somewhat filtered. You have to have at least a capacitor to smooth the waveform, otherwise you would have very choppy DC that would be essentially unusable in most applications (lighting would be one example of an exception). Otherwise, you will get 60hz/120hz (or 50hz/100hz) noise depending if a half or full wave rectifier is used.

[Pyr0Beast]

Re: Linear Vs Switching Supplies In Audio Equipment

Most, if any filtering in linear supplies is done by voltage drop on regulator and its noise rejection capabilities.

[Th3_uN1Qu3]

Re: Linear Vs Switching Supplies In Audio Equipment

Never seen any linear supply, regulated or not, without one or more filter caps. Try powering a home theater system without large caps in the DC bus and then tell me what happens.

[Pyr0Beast]

Re: Linear Vs Switching Supplies In Audio Equipment

There are capacitor-less linear supplies. Home theater system without large caps is possible if you take power from battery or have a switched power supply with extremely fast response.

[Th3_uN1Qu3]

Re: Linear Vs Switching Supplies In Audio Equipment

There are capacitor-less linear supplies.

Sure but we're talking high power here. I don't think that even the smallest set of computer speakers can do away with the primary filter cap, let alone a home theater.

or have a switched power supply with extremely fast response.

It actually only requires 50Hz response, almost any SMPS can in theory do away with the line filter caps after the bridge rectifier. However, this is impractical because without the caps the voltage at the input of the switcher will be drastically lower which makes the supply less efficient and for any application higher in power than a cellphone charger, will make it unable to operate at all.

[Trinite]

Re: Linear Vs Switching Supplies In Audio Equipment

Here is a good comparison of the two: http://en.wikipedia.org/wiki/Switche...ply_comparison

[Pyr0Beast]

Re: Linear Vs Switching Supplies In Audio Equipment

Same stuff, higher power is just upscaled version.

PSU is as efficient as it can be, with or without caps. Just its power output is lower without.

[Logistics]

Re: Linear Vs Switching Supplies In Audio Equipment

I thought one of the main reasons for a linear supply versus a regulated or switching supply was the transients. Linear supplies have the ability to deliver spikes of power on low-frequency beats when necessary, given the supply bank of caps is sufficient and fast enough. Regulated and Switch-mode supplies, which are also essentially regulated, only deliver up to X point and that's it. But this is all a moot point if the topology in your amp was never meant to require higher than X point, anyway.

[Th3_uN1Qu3]

Re: Linear Vs Switching Supplies In Audio Equipment

I thought one of the main reasons for a linear supply versus a regulated or switching supply was the transients. Linear supplies have the ability to deliver spikes of power on low-frequency beats when necessary, given the supply bank of caps is sufficient and fast enough.

There are several designs using unregulated SMPS supplies...

As for transient response, a unregulated mains-powered supply still has the 50/60Hz limit as to how fast the caps can be charged, so exceeding a certain point will simply cause mains frequency modulation of the output signal. A SMPS has the potential to recover much faster, though, for cheapness, low current transistors are chosen which need peak current limiting so as not to blow up... which again, limits transient response.

[cadiman]

Re: Linear Vs Switching Supplies In Audio Equipment

As said above, Thanks for everyone`s input. This has been a great topic and I learn new things everyday from everyone here!

[hardwareguy]

Re: Linear Vs Switching Supplies In Audio Equipment

As long as the current delivery is adequate and the ripple and noise are low it should sound excellent.

The QSC Powerlight series uses an SMPS and its a very nice professional amp.

Shield the SMPS in a grounded cage if you have it near a tuner. AM radios might not like it much. A good LC filter on the input and output are a must for low noise on any source, not just radios.

[japlytic]

Re: Linear Vs Switching Supplies In Audio Equipment

Switching controllers with a spread spectrum (or frequency jitter) function would help reduce RF interference, and would be good for TV antenna amplifiers.

[999999999]

Re: Linear Vs Switching Supplies In Audio Equipment

Not really. More likely an ideology issue.

Laptop amplifiers have all switched power supplies. So do LCD TVs and other consumer junk and I see no one complaining.

Linear regulators are simple but can also be garbage as SMPS can be. If done properly there shouldn't be any or very small difference.

That is not quite accurate. Many laptops and some other consumer gear that have AC mains to lower voltage DC will do this with a switching PSU, BUT they also have a linear regulator or tweaked filter circuit for the audio amp stage. We can say "if done properly" but usually the situation is either "done cheaply" or with the typical DIY'er, just plugging the SMPS into the amp.

See what I wrote previously about PSRR factors for integrated chipamps, the truth is that most SMPS have output noise that chipamp datasheets clearly suggest will contaminate the output to a larger extent. Until your amp is outputting too near the power rail voltages, you can literally have a linear or nonregulated PSU with 5V changes at 120Hz and it is inaudible, but a SMPS with 200mA changes more audible... because it's causing amplification errors potentially 50 thousand times each second.

AFAIK, there is no linear regulator sold today that is "garbage", on the contrary even old tech like LM317 has vastly cleaner output at $1 a piece than the typical high end switching PSU large and costly enough you wouldn't ever find them in a TV, laptop, etc.

[999999999]

Re: Linear Vs Switching Supplies In Audio Equipment

There are several designs using unregulated SMPS supplies...

As for transient response, a unregulated mains-powered supply still has the 50/60Hz limit as to how fast the caps can be charged, so exceeding a certain point will simply cause mains frequency modulation of the output signal. A SMPS has the potential to recover much faster, though, for cheapness, low current transistors are chosen which need peak current limiting so as not to blow up... which again, limits transient response.

The difference depends on amp design and input voltage margins. Take for example what I would consider a better design - Your amp input is +-45V, but to reach the required output voltage to the speakers (per volume level required by listener) it only has to output +-30V. Even if you factor for the voltage drop across BJT output transistor pairs in the amp output stage you still have 90V rail to rail input from the PSU versus (60V - 2 * 1V) = 58V output from the amp.

That 90V - 58V, 32V difference simply means that so long as your filter capacitors never drop below 58V, it does not matter that you have 120Hz (60Hz mains frequency doubled by the bridge rectifier) changes in voltage, only that the voltage stays above the 58V threshold. So long as it stays above the 58V threshold what you want to avoid most is high frequency changes in voltage because every time it changes the amp has an error, it output a value based on the input voltage which has now changed when it compares in the opamp-like comparitive loop in the amp (some amps are open loop but it is very uncommon in commodity consumer gear like we have been discussing).

I hope I have conveyed this concept in a coherent manner. Powering an analog amp it is not as important how close the PSU stays to the design input voltage, rather it is important that the PSU voltage never drop below the minimum threshold (58V in my example above) and that it change values with the least frequency possible... but that is also assuming the unregulated PSU wasn't designed with grossly undersized capacitors which could cause significantly increasing voltage change rate.

On the other hand, small differences which can be reduced to negligible or even inaudible differences in sound, can be outweighed by the other benefits of switching PSU such as size, weight, heat, power consumption, etc. That an amp is "high end" and has a switching PSU, that does not necessarily validate the SMPS as being as accurate, it might or might not sound as good or better which is a subjective issue in audio, rather it can be that the other factors were decided by the design team to be more important... and many people do claim a SMPS powered amp sounds "livelier" and they prefer that (while other people don't) even if it is a less accurate representation of what the live audio was.