Pretty small requirements. I’d like to DIY a “clean side power” for an Allo Digione Signature player. There are several options for kits on eBay and Alispress, I’m not opposed to that route - if someone can recommend one. Or a design. Thanks!
DigiOne Signature Player SPDIF Out
They also have two power supplies
Nirvana $60
Nirvana SMPS
And Shanti $160
Shanti Dual Linear Ultra Low Noise PSU
DigiOne Signature Player SPDIF Out
They also have two power supplies
Nirvana $60
Nirvana SMPS
And Shanti $160
Shanti Dual Linear Ultra Low Noise PSU
Attachments
Old smartphone charger will work.
Scrap value.
Use filter caps, 6.3 or 16 V from a scrap motherboard, that works.
Alternately, use a 7805, the input should be about 1.5 above the output, so use a PSU with 7-15 volts, and filters of course.
Use pF too for high frequency noise.
Scrap value.
Use filter caps, 6.3 or 16 V from a scrap motherboard, that works.
Alternately, use a 7805, the input should be about 1.5 above the output, so use a PSU with 7-15 volts, and filters of course.
Use pF too for high frequency noise.
Last edited:
A standard linear PSU 5V 1A would suffice but I guess you want lowest noise. Therefor I would steer away from scrap switcher stuff with a 265 Euro Digione player.
The ready made LM723 based industrial PSU's are often very good quality with regards to noise. They are cheap and only need a recap if they are very old. I recently found new in box ones. Fully protected, built in transformer and integrated heatsink/frame as well.
Power One ha5-1.5/ovp-a+ DC Netzteil 5v 1.5a | eBay
This one is also good and very low noise, relatively small and affordable. It needs a 6V transformer which could be an R-core type. I use these on some devices with good results.
TPS7A4700 Ultralow-noise (4mVrms) Power Supply 1.4-20.5V,1A for Audio DAC | eBay
The dutch Delta PSU's are excellent but difficult to find in the US I think. These are discretely built, small and have a built in transformer. I sell these new in box but I won't promote my own stuff here.
The ready made LM723 based industrial PSU's are often very good quality with regards to noise. They are cheap and only need a recap if they are very old. I recently found new in box ones. Fully protected, built in transformer and integrated heatsink/frame as well.
Power One ha5-1.5/ovp-a+ DC Netzteil 5v 1.5a | eBay
This one is also good and very low noise, relatively small and affordable. It needs a 6V transformer which could be an R-core type. I use these on some devices with good results.
TPS7A4700 Ultralow-noise (4mVrms) Power Supply 1.4-20.5V,1A for Audio DAC | eBay
The dutch Delta PSU's are excellent but difficult to find in the US I think. These are discretely built, small and have a built in transformer. I sell these new in box but I won't promote my own stuff here.
Last edited:
If really good ones are needed, try Meanwell, they are good, and should be easily available.
Or just make one like described above, for higher ratings a LM317 can be selected.
The 7805 has good ripple and noise rejection.
Or just make one like described above, for higher ratings a LM317 can be selected.
The 7805 has good ripple and noise rejection.
OP wants a silent PSU. Dinosaurs LM317 and 7805 are not ultra low noise. Dinosaur LM723 is already factors better but TPS7A4700 and LT304x based PSU's are even more silent.
Any switcher will need a linear regulator to be called silent. This seems counterproductive so I would go fully linear then as results are often best possible. This involves strange primitive devices containing steel and copper called "transformers" 🙂 Heresy in todays switcher world but way better for audio despite the negative propaganda.
Any switcher will need a linear regulator to be called silent. This seems counterproductive so I would go fully linear then as results are often best possible. This involves strange primitive devices containing steel and copper called "transformers" 🙂 Heresy in todays switcher world but way better for audio despite the negative propaganda.
Last edited:
So it is possible for him to make one using a higher voltage input from a linear supply, and converting to regulated supply.
Linear will be more electrically quiet than switcher, and the regulator will take care of fluctuation that will occur on the transformer.
Simple, should work nicely I think...
Linear will be more electrically quiet than switcher, and the regulator will take care of fluctuation that will occur on the transformer.
Simple, should work nicely I think...
Some that require a silent PSU go for sub 10 µV noise numbers over a broad bandwidth and low output impedance.
Last edited:
First off, nowhere in the OP does the word silent appear. "clean side power" does though.
Secondly, I think you'll find 100 microvolts is 0.0001 volts. Most batteries will fluctuate more than that with loading, no?
Just saying. What is the point of diminishing returns for this PSU? OP listed a linear PSU that costs 160$. That's a lot of money to spend for a 2.5 Watt PSU. That's about what I pay to build a 500VA supply for a power amp.
The LT3045 would work if you parallel two of them. 0.6 microvolts noise. https://www.analog.com/media/en/technical-documentation/data-sheets/lt3045.pdf
Personally, I'd probably tap the 5V line on my computer. Or the 12V line and use an LDO like MIC29300. MIC29300-5.0WT
Secondly, I think you'll find 100 microvolts is 0.0001 volts. Most batteries will fluctuate more than that with loading, no?
Just saying. What is the point of diminishing returns for this PSU? OP listed a linear PSU that costs 160$. That's a lot of money to spend for a 2.5 Watt PSU. That's about what I pay to build a 500VA supply for a power amp.
The LT3045 would work if you parallel two of them. 0.6 microvolts noise. https://www.analog.com/media/en/technical-documentation/data-sheets/lt3045.pdf
Personally, I'd probably tap the 5V line on my computer. Or the 12V line and use an LDO like MIC29300. MIC29300-5.0WT
Last edited:
Not a native speaker but I think "clean side power" means a silent PSU is required. This was supported by the given suggestions by the OP for relatively expensive ultra low noise PSUs. Add to this the Digione picture and PCB that both speak of "clean power". "Clean power" is generally seen as a thin line of DC without noise or any pollution. This would trigger to first search for lowest noise/best performance and only then look at costs. Whatever.
The MIC29300 is on of the noisiest regulators on this planet. It varies from 260 to 400 µV of pure noise!!! One almost can't choose worse and no sane person would use this one for audio. A linear regulator like that makes SMPS look excellent 😀 I would choose lowest noise/best performance at most affordable price so the suggestion of the LDOVR TPS7A4700 comes very close to "clean" with a factor 100 less noise for 36 $ all in. Just saying. When using the noisy 12V line of a computer PSU one also does not need an LDO for 5V. Even 7805 is then a way better choice.
The MIC29300 is on of the noisiest regulators on this planet. It varies from 260 to 400 µV of pure noise!!! One almost can't choose worse and no sane person would use this one for audio. A linear regulator like that makes SMPS look excellent 😀 I would choose lowest noise/best performance at most affordable price so the suggestion of the LDOVR TPS7A4700 comes very close to "clean" with a factor 100 less noise for 36 $ all in. Just saying. When using the noisy 12V line of a computer PSU one also does not need an LDO for 5V. Even 7805 is then a way better choice.
Last edited:
15mV isn't noisy on 12V IMHO (The noise my PSU makes outputting 50 amperes at 12V).
Also, 500uV is half of one thousandth of one volt. I hardly think it matters?
Then of course, I'm used to making 300V with 500mV of ripple and that makes a silent PP output stage. Maybe the device the OP is running is super sensitive (AKA a bad PSRR)?
In that case use 4 nickel metal hydride batteries to power it. Or is a set of batteries too noisy?
Also, 500uV is half of one thousandth of one volt. I hardly think it matters?
Then of course, I'm used to making 300V with 500mV of ripple and that makes a silent PP output stage. Maybe the device the OP is running is super sensitive (AKA a bad PSRR)?
In that case use 4 nickel metal hydride batteries to power it. Or is a set of batteries too noisy?
Last edited:
You are comparing apple with pears. With low signal levels and low working voltages noise figures matter more than in Stone Age 300V stuff. Feed a DAC with 5V requirement a 500 mV ripple voltage and understand.
It is simple. OP has a device that requires "clean power" so one searches/designs for "clean power". In prehistoric times we had 7805 with an average noise number of 40 µV in the 10 to 100 kHz range. So we search/design for lower noise numbers over a broader bandwidth at lower output impedance preferably at an affordable price. That is defined as "better" and "silent".
It is simple. OP has a device that requires "clean power" so one searches/designs for "clean power". In prehistoric times we had 7805 with an average noise number of 40 µV in the 10 to 100 kHz range. So we search/design for lower noise numbers over a broader bandwidth at lower output impedance preferably at an affordable price. That is defined as "better" and "silent".
Last edited:
I'm comparing percentages. 500mV on 300V is more noise than 500uV on 5V as a percentage.
The point is millionths of a volt don't matter very much in the grand scheme of things.
5V is also a stone age voltage - Aren't they using 1.8V these days? LOL I think my CPU is down to under a volt now.
The point is millionths of a volt don't matter very much in the grand scheme of things.
5V is also a stone age voltage - Aren't they using 1.8V these days? LOL I think my CPU is down to under a volt now.
Because I know my stuff and design to the requirements of both the device and the customer.
Last edited:
If that was the case, I'd see a schematic of your design, no?
How much noise is there on the average battery? Not much apparently... https://tf.nist.gov/general/pdf/1133.pdf
How much noise is there on the average battery? Not much apparently... https://tf.nist.gov/general/pdf/1133.pdf
Funny guy, you just spoke of a 260 to 400 µV regulator that measures worse than the average switcher and of which I would even doubt to use it for supplying power to a power LED. Batteries are of no importance as the OP clearly speaks of 5V 500 mA (see the thread title) and suggested 2 different ultra low noise PSU's both mains connected.
There are a few designs by me around somewhere here that measured quite well. AFAIK some measured best in their class. That does not have anything to do with the issue in this thread though.
There are a few designs by me around somewhere here that measured quite well. AFAIK some measured best in their class. That does not have anything to do with the issue in this thread though.
Last edited:
If you think the average switcher has less than 1mV noise, I have some oceanfront property for you in Saskatchewan.
4.8V from NiHM batteries is close enough, it's quieter, and it's cheaper and simpler. Get 8 batteries, charge four while using the other four. Just offering the option as most people don't think about batteries.
4.8V from NiHM batteries is close enough, it's quieter, and it's cheaper and simpler. Get 8 batteries, charge four while using the other four. Just offering the option as most people don't think about batteries.
As a reminder post #1. Read carefully and find out what OP wants. Hint: both Nirvana and Shanti are below 2 µV of noise over a broad frequency range. The MIC29300 would be a factor 130 to 200 more noisy. In the semiconductor world we call that "worse".
I suggested linear PSU's exclusively. SMPS are generally good for the bin, exceptions like the Nirvana confirm the rule. Nirvana has lower noise than 1 mV so the oceanfront property in Saskatchewan is accepted 🙂
Last edited: