Here is my first PSU mod
I have ordered the reg board here:
LT3042 PSRR RF Linear Regulator Power Board Module Ultra low High Noise Board Reverse Battery Current Protection 20V 200mA-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
As a pre-reg, i have used a 7808 attached on flying leads to the back of the reg board
once I receive this it will look much leaner.
5PCS LM78XX LM7805 LM7812 L78XX PCB Fixed Voltage Regulator Prototype PCB Board FAST DISPATCH-in Instrument Parts & Accessories from Tools on Aliexpress.com | Alibaba Group
I was able to get a signal lock, I take my coffee now and put some music on, to see if there will be any improvement
I have ordered the reg board here:
LT3042 PSRR RF Linear Regulator Power Board Module Ultra low High Noise Board Reverse Battery Current Protection 20V 200mA-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
As a pre-reg, i have used a 7808 attached on flying leads to the back of the reg board
once I receive this it will look much leaner.
5PCS LM78XX LM7805 LM7812 L78XX PCB Fixed Voltage Regulator Prototype PCB Board FAST DISPATCH-in Instrument Parts & Accessories from Tools on Aliexpress.com | Alibaba Group
I was able to get a signal lock, I take my coffee now and put some music on, to see if there will be any improvement
@KCHANG:
Oscillation isn't the only possible consequence. Ringing is another. Also, the combination of caps is not random, there are reasons for the particular recommended values which is that they complement each other in a useful way to keep impedance low over a wide range of frequencies. At least low enough to get sound quality to conventionally accepted measurement standards and widely accepted human hearing abilities. Since IMHO a lot of old hearing research is flawed and at least some people can hear much lower level aberrations than research has thus far indicated, we also need to take some further steps if achieving best sound quality is a concern.
Oscillation isn't the only possible consequence. Ringing is another. Also, the combination of caps is not random, there are reasons for the particular recommended values which is that they complement each other in a useful way to keep impedance low over a wide range of frequencies. At least low enough to get sound quality to conventionally accepted measurement standards and widely accepted human hearing abilities. Since IMHO a lot of old hearing research is flawed and at least some people can hear much lower level aberrations than research has thus far indicated, we also need to take some further steps if achieving best sound quality is a concern.
Last edited:
@markw4
Completely agree with you, now even knowing what oscillation is. Producing audio gear and not only audio but much more, is just releasing a product that would fit into certain acceptable parameters so there is almost always an area for improvement.
I'm impressed by the change that the regulator introduced for not even 10 usd. There is so much more happening in the mid to high range. That distortion is still there, even more, highlighted now but happening in a much narrower frequency range. My gut says that bypassing the ops will sort that out.
Completely agree with you, now even knowing what oscillation is. Producing audio gear and not only audio but much more, is just releasing a product that would fit into certain acceptable parameters so there is almost always an area for improvement.
I'm impressed by the change that the regulator introduced for not even 10 usd. There is so much more happening in the mid to high range. That distortion is still there, even more, highlighted now but happening in a much narrower frequency range. My gut says that bypassing the ops will sort that out.
By "ringing," are you talking about the kind of ringing caused by, e.g., a step function? That's very interesting. Such ringing is different from (albeit related to perhaps) the oscillation I was thinking about. Would you be so kind to point me to a few references that discuss how power rail bypassing affects such ringing?
Usually, the issue is referred to as 'stability.' In the math of linear systems theory, something very applicable to electronics, oscillation is closely related to ringing. For a lot of useful and very practical discussion the book, Electromagnetic Compatibility Engineering, by Henry Ott is very good.
Today I tried to listen to some hi res audio, and was not able to. the DAC would play nicely up to 48 kHz 96 kHz pauses the payback here and there and DSD does not play.
I realized that the LT3042 provides a max current of 200 mA. And the ES9038PRO AVCC can Draw a maximum of 120 mA per channel, therefore 1 regulator per channel is OK but 1 for 2 joint AVCC is not a good idea.
At the other side, AVCC is the analogue part of the chip and so sampling rate should not affect current draw here...
anyone any hints?
I realized that the LT3042 provides a max current of 200 mA. And the ES9038PRO AVCC can Draw a maximum of 120 mA per channel, therefore 1 regulator per channel is OK but 1 for 2 joint AVCC is not a good idea.
At the other side, AVCC is the analogue part of the chip and so sampling rate should not affect current draw here...
anyone any hints?
I will use this little board it has the same reg, but can provide 1A current.
Low Noise LT3042 Linear Regulator Power Supply Board DC Converter Overvoltage-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
Low Noise LT3042 Linear Regulator Power Supply Board DC Converter Overvoltage-in Integrated Circuits from Electronic Components & Supplies on Aliexpress.com | Alibaba Group
@idir this 9038 DAC plays all sorts of dsf files in direct mode all day every day as long as you use the Amanero card. YOu can use theXMOS card for some, but not all, at higher bit rates it emits a terrible white noise. But it's the USB card's problem not the DAC.
You might have broken something with recent mods. Revert and see if you can play them.
You might have broken something with recent mods. Revert and see if you can play them.
I removed completely the PSU from the board
The rectifier and reg on the left side in only to power the MCU, therefore I plan to remove it completely from the board and place a small one near the display. For now, I made a quick temporary PSU that is near the transformer (outside the case in the pic below). So the board will have left only what will replace the 2 x 1963 regs. and the I\V stage PSU
On the right, there is a bipolar PSU dual 15 V for the I\V stage.
In the left 1A LT3042 PSU
The problem I had with DAC stability is solved by providing more power than the small 200mA LT3042 was able to produce.
The rectifier and reg on the left side in only to power the MCU, therefore I plan to remove it completely from the board and place a small one near the display. For now, I made a quick temporary PSU that is near the transformer (outside the case in the pic below). So the board will have left only what will replace the 2 x 1963 regs. and the I\V stage PSU
On the right, there is a bipolar PSU dual 15 V for the I\V stage.
In the left 1A LT3042 PSU
The problem I had with DAC stability is solved by providing more power than the small 200mA LT3042 was able to produce.
Hi idir,
Great to see that you are completely redoing the power supplies for this board. I'm anxious to know what kind of improvements in the sound you get from the power supply modifications.
Are you going to try to add big fat PP film capacitors as recommended by Markw4?
Also, from my own experience (mainly the "Band-Aid" approach of changing one cap here and adding another cap there), soldering bypass caps (small film caps or X7R ceramic caps) directly to the power pins of the opamps makes noticeable differences in the sound. Are you going to try that too?
Great to see that you are completely redoing the power supplies for this board. I'm anxious to know what kind of improvements in the sound you get from the power supply modifications.
Are you going to try to add big fat PP film capacitors as recommended by Markw4?
Also, from my own experience (mainly the "Band-Aid" approach of changing one cap here and adding another cap there), soldering bypass caps (small film caps or X7R ceramic caps) directly to the power pins of the opamps makes noticeable differences in the sound. Are you going to try that too?
Last edited:
My experience too - though I use low ESR electrolytics mostly (Nichicon HZ). Opamps turn out to be surprisingly sensitive to small power supply perturbations.
I have been told to avoid electrolytics as much as possible in audio application
Did you ask whoever told you this why? Perhaps he/she only meant to avoid 'lytics when coupling signals, rather than when decoupling power supplies?
Are you sure Tantalum caps are ok for this application ?
4 types of capacitor for filtering applications in switch mode power supply systems – European Passive Components Institute
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.