Did you try using the LDO after the SMPS to improve it's performance? I'm not saying a linear supply isn't an improvement, but there's no way it's a 300$ upgrade. Simply put a 6W linear supply doesn't cost 300$ unless it's marketed to people willing to waste money for no reason.
I have seen a different path on some boutique commercial Hi-Fi audio devices that need 10W or less of power, such as phono preamps. The power supply is a toroidal transformer about 20 times bigger than required (to avoid any sagging due to load variation), followed by the rectifier and a string of RC filter cells. No voltage stabilization, because it is assumed that the AC supply to the device is coming from a power conditioner so its value is steady. The number of filter cells and the total resistive drop is high enough to lower the hum to a insignificant value. This "back to basics" approach may work well when the power supply voltage exact value is not overly important.
Can you be more specific? Which one below are you referring to?
The σ11 regulated power supply
Sigma11 Linear Regulated Power Supply PSU kit for DAC headphone amp L16-22 | eBay
Same question for the σ22. If you are using the e-bay clones, can you comment on how good they are compared to the genuine AMB power supplies?
Last edited:
The LDO IC comes in all colors. The LT1083CP is the cheapest and under $10. I used it with a LM317 style board. I used it with a 25VA toroidal transformer from Antek which costs $18 shipped in US. It is a $30, not a $300 upgrade.
I have not tried the ultra low noise TPS7A4701 which is usually about $20 on e-bay.
I can’t say I’ve never bothered with wasting my time toiling over improving a power supply for a lackluster device however, I just have since learned otherwise and now choose a better DAC to consume my efforts is all.
Different approach I guess.
I know that it is a common practice in high end DAC power supply and may be significant saving over transformer in making hundreds or thousands unit. But I do not know how to select a good SMPS for this purpose. I stay with what I am familiar with, which means transformer from reliable source.
Common sense actually ... If you have a low power device that reacts to it's power supply in a noticeable way, the problem is inside the device, not the power supply. Any good circuit design is going to be mostly impervious to minor issues like ripple, noise spikes and RFI.
Ha ha you and I know that is a too simple way of thinking but feel free to do so. In your way of reasoning one can have a mediocre PSU feeding a well designed device and still have good results. Keep fighting the obvious though!
I don't think there's a DAC out there that actually runs on 12VDC. They will be stepped down to 5V or 3.3V, normally with very good quality LDO regulators. The output buffer is where you might actually hear a difference in the supply though, but there again any competent design would have good PSSR there.
Hey,
I buy the boards and some of the parts directly from Ti Kan at AMB. As tempting as the Sigma stuff pricing is on eBay I strongly believe in supporting folks like AMB Labs. Parts from Mouser and Digital-Key.
I have bought some of the Chinese gear, recently a Jung Super Reg. It works well but I am still learning how to measure noise so am unsure how well it works. That’s another reason to by from AMB, proven to work!
I whole heartedly agree with the idea that a well designed lps is an easy inexpensive upgrade to almost any piece of audio kit that come with an smps. This is DIY so of course I’m going to build something even if it’s just the Psu!
In my case I know very little about switch mode power supply design. It seems relatively complex compared to linear PS design. I don't know how, for example, the ultrasonic switching frequency is reliably filtered from the output so that it does not cause interference in the audio of the associated components.
My main concern with SMPS is the possibility that the switching frequency might interact with the audio band to produce intermodulation distortion, that hard, brittle, crispy-crunchy effect that was associated with early digital audio and poor solid-state design. I am very sensitive to any emphasis in the upper midrange of the audio band and I suspect IM is a major contributor to that sound.
I would be interested to see comments from those with more experience with SMPS and its design. How are they filtered so that they don't cause audio interference from RFI, EMI, and IM? What is the expected service life of such power supplies? How is the performance measured, and how can one tell a good SMPS from a bad one?
I bought the V90 DAC because all the reviews I read were positive and it fit within my budget. I assume that all the reviews were performed with the stock factory SMPS supplying the V90, so it may be that my concerns are without cause and I am just being anal about the whole situation. But that's why we are all on this forum, is it not? We are all anal about audio?
My main concern with SMPS is the possibility that the switching frequency might interact with the audio band to produce intermodulation distortion, that hard, brittle, crispy-crunchy effect that was associated with early digital audio and poor solid-state design. I am very sensitive to any emphasis in the upper midrange of the audio band and I suspect IM is a major contributor to that sound.
I would be interested to see comments from those with more experience with SMPS and its design. How are they filtered so that they don't cause audio interference from RFI, EMI, and IM? What is the expected service life of such power supplies? How is the performance measured, and how can one tell a good SMPS from a bad one?
I bought the V90 DAC because all the reviews I read were positive and it fit within my budget. I assume that all the reviews were performed with the stock factory SMPS supplying the V90, so it may be that my concerns are without cause and I am just being anal about the whole situation. But that's why we are all on this forum, is it not? We are all anal about audio?
My apologies regarding my comments about the V90, am aware of the diminishing returns with audio. I just don’t think that a much better power supply will necessarily be a good use of your time/resources. I described what I had done with a similar setup to eliminate the possibility of a potentially noisy switcher.
Many here would laugh at my projects I’m sure, mostly kit boards with random parts dangling...
Maybe a line conditioner that can be used upstream to help with potential mains noise?
Or an isolation transformer on it’s own even if you’re not into messing with a project like that.
Many here would laugh at my projects I’m sure, mostly kit boards with random parts dangling...
Maybe a line conditioner that can be used upstream to help with potential mains noise?
Or an isolation transformer on it’s own even if you’re not into messing with a project like that.
First of all the switching frequencies in modern SMPS are well above the audio band of frequencies. 45Khz is typical, some go over 100k.
Filtering in an SMPS is done the same way it's done in a standard linear supply... take the output of the switcher, run it through a transformer, rectify it and smooth it using capacitors. The only real difference is that with the frequency being so much higher you can use a smaller transformer and capacitors to get the same result.
As for ripple on the DC, which would you prefer at max output... 3 or 4 volts of unregulated 120hz ripple from a linear supply or a few milivolts of tightly regulated 45khz ripple from an SMPS.
For reasons wholly unfathomable to me, there are a number of audiophiles bent upon trashing these new technologies. Please don't get caught up in their muppingsnarf.
I suspect its because of some high visibility failures of some audio power amps with SMPS.
BS. It is a PSU that has to deliver 12V at 0.5A. No way a linear PSU with adequately sized filter caps will have 3 to 4V ripple. Even with a single 2200 uF cap ripple will be way less.
Besides that a wise man would use a regulator there (albeit a linear one) just like he would regulate an SMPS. Many A branded SMPS I have deliver tens of mV of high frequency ripple after the regulation. That high frequency ripple will go through many old fashioned regulators. Amongst the SMPS there are some that have decent specs, these mostly have a decent modern LDO in them 🙂
Are you really a tech guy? This has nothing to do with audiophiles bashing new technology. In fact many will recommend the most recent developments in LDO regs. If you would open your eyes you would notice high end streamers very often have linear PSU's. Why would a manufacturer use more expensive technology?
Last edited:
Like jean-Paul said I don’t think new tech being bashed is at all what’s going on. If it sounds good or better to me I will use it.
Subjective, heck yes, but, a good Sigma 11 12v 5amp power supply made such a marked difference that my wife noticed it on my Moon LP3. I am not an engineer or have the knowledge and equipment to measure it all, I know what I hear and if I like it I pursue it. . . My fast, deep, clean linear power supplies sound better to me and I enjoy building them.
Subjective, heck yes, but, a good Sigma 11 12v 5amp power supply made such a marked difference that my wife noticed it on my Moon LP3. I am not an engineer or have the knowledge and equipment to measure it all, I know what I hear and if I like it I pursue it. . . My fast, deep, clean linear power supplies sound better to me and I enjoy building them.
Are you?
A moderator tossing shade on a new member like that certainly doesn't look good.
