The first time I tried using switching power with an LM3886 amp, one workshop was amazed at the kind of punch is was delivering. I am sure there are lots of variables in how the power supply interacts with the amp when playing music, it would be very interesting to actually find what effects sound. I have played around with some ideas, and it would be interesting to see how these effect this amp. The initial idea would be to have two different power modules in two channels.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
A composite amp like Tom's doesn't react to any power supply characteristic as long as you're not clipping it.
Different stray currents may have an effect, though. Not for the amp itself, but for the source and the interconnect if unbalanced.
Different stray currents may have an effect, though. Not for the amp itself, but for the source and the interconnect if unbalanced.
My Terms of Sale (including currency information) are here: https://www.neurochrome.com/terms/
All prices are in USD as 99.9 % of sales go outside of Canada and over 50 % to the US. If you're in Canada and would like to pay in CAD, just toss me an email.
Tom
My apologies if there was a language foobar, which was probably my fault. However, know that I believe that all measures in audio need to be done by several different means, and do passable every time. It is unfortunate if all of the answers come up different. But, it is proof if all of the answers come up similar (and so handy because that job is finished).
If you said that my measures alone will not do, I would agree with you and defend your position on that. I had not meant to say that only my measures should be used. What I would have liked to say is that they should be considered in with several different means of measuring.
To me, a device works when it passes ALL means of measuring, not just a tiny few.
Oddly indirectly, that approach works fantastically for layout. It is otherwise inconvenient.
In Tom's case, I think that he may have already checked for suitable practical output before offering any product for sale. By my measuring, he might be an alien or a superhero, probably both.
He and his wares do pass multiple divergent measuring means, excellently. I can't imagine how much labor was involved.
I think the way PSRR is measured is not the way current is drawn from the power supply...
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
On Tom's amp, it looks like the stabilized power for the small signal op-amp, will sound the same with an SMPS as it does with a linear supply. This applies to midrange and treble. It sure does make the power supply shopping simpler and easier to do (most options will work fine if one simply gets the current provisions and voltage in range).
KSTR probably said that better.
However, soongsc, at the bass, post #3301: It is not terribly expensive to purchase an SMPS that has a current rating at least 3/4ths more than the need; and, had one done that, the bass dynamic will compete or exceed the linear supply. That is cohesive with the list of what goodly amplifiers should do. Instead of aiming for lowest; had one aimed high instead, then the rather bigger and heavier SMPS becomes admirable. If there was far more current capacity at the supply than the need used, then the bass dynamic from the speaker will be more effective and noticeable. That is expected for overbuilt supplies; however overbuilt SMPS costs less than overbuilt linear supply.
KSTR probably said that better.
However, soongsc, at the bass, post #3301: It is not terribly expensive to purchase an SMPS that has a current rating at least 3/4ths more than the need; and, had one done that, the bass dynamic will compete or exceed the linear supply. That is cohesive with the list of what goodly amplifiers should do. Instead of aiming for lowest; had one aimed high instead, then the rather bigger and heavier SMPS becomes admirable. If there was far more current capacity at the supply than the need used, then the bass dynamic from the speaker will be more effective and noticeable. That is expected for overbuilt supplies; however overbuilt SMPS costs less than overbuilt linear supply.
I like SMPS, I just think people are incorrectly relating how PSRR associates with certain performance of the amplifier.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
SMPS are smaller, lighter, cheaper and regulated. What's there not to like?
Your amplifier doesn't care where you get it's power from.
Actually big Iron, big capacitors, big heatsinks and chassis work are sexy and have been used successfully to sell many big $$$ amplifiers over the years. There is little correlation with performance of a well executed modern design though.
Your amplifier doesn't care where you get it's power from.
Actually big Iron, big capacitors, big heatsinks and chassis work are sexy and have been used successfully to sell many big $$$ amplifiers over the years. There is little correlation with performance of a well executed modern design though.
Really? How's the output voltage controlled then?
Every SMPS I've built and all I've seen schematics of, including the Connex Electronic supplies, have been regulated. They have a feedback loop from the output back to the switching regulator IC. That's the definition of a regulated supply right there.
The block diagram for the IRM-series from Mean Well (attached) shows a feedback loop as well. That is also regulated!
Tom
Attachments
regulated is very common. That's because most of them drive digital circuits that are sensitive to significant changes in supply voltage.
But smps is simply a switching power supply. It does not by definition include a regulation stage.
But smps is simply a switching power supply. It does not by definition include a regulation stage.
If its sold as a smps what's the chance it doesn't have some way of setting a target voltage? Anyone ever seen one that doesn't include regulation?
Actually, few linear power supplies in power amps have regulation. I have one series that do. Switching power I have used seem to use the feedback for short circuit protection and recovery, complicated designs seems to have regulation. LT has some interesting demo boards on switching power application.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
Here an example of an unregulated SMPS (for audio):
https://www.hypex.nl/product/smps400/12#tab_downloads
By the way, Connexelectronics also sells unregulated SMPS.
Last edited by a moderator:
Bruno set out to make transformer replacements and designed a supply where the output voltage is proportional to the input voltage. First off, that doesn't mean the supply isn't regulated. Secondly, it also doesn't mean that all SMPSes are made this way. In fact, the Hypex supplies are the only ones I've come across that are made this way.
Really? Which one? I'm not saying I know different, I just haven't noticed them in their portfolio.
All the Connex Electronic supplies I've recommended for use with the MOD86/286 are regulated.
Tom
Regulation doesn't need be obvious i.e an opto to allow directly sensed feedback to be applied to the primary switcher, it can also rely on primary derived voltages to apply basic load/voltage regulation.
Can't recall any running open loop... how could they when most cope with 90 to 265 vac input.
Silpad vs mica/goop
In my Para 86 builds I used 1500ST as recommended in the manual. I recently came across this :
http://www.diyaudio.com/forums/newreply.php?do=newreply&p=430976
In my Para 86 builds I used 1500ST as recommended in the manual. I recently came across this :
http://www.diyaudio.com/forums/newreply.php?do=newreply&p=430976
Last edited: