I have this little phono preamp which has an integrated psu, converting 12-0-12 ac coming from a center tapped transformer, to regulated +/-12Vdc with a 7912/7812.
I need a 5Vdc in this chassis as well, so I used a LM2596 DC to DC converter I had lying around which I plugged on output pin of the 7812 and ground, in parallel, the 12Vdc is stepped down to 5Vdc.
I have tested everything and it works fine, the phono preamp seems to do its job and the little ADC on the 5V supply is working fine as well.
Is there any downside to this setup? Should I try to build a psu which will be also fed from the 12-0-12 AC in parallel to convert it to 5V dc without connecting to the phono preamp psu?
Thanks!
I need a 5Vdc in this chassis as well, so I used a LM2596 DC to DC converter I had lying around which I plugged on output pin of the 7812 and ground, in parallel, the 12Vdc is stepped down to 5Vdc.
I have tested everything and it works fine, the phono preamp seems to do its job and the little ADC on the 5V supply is working fine as well.
Is there any downside to this setup? Should I try to build a psu which will be also fed from the 12-0-12 AC in parallel to convert it to 5V dc without connecting to the phono preamp psu?
Thanks!
Electrically what you have done is fine. Possible issues are that DC/DC converters are inherently noisy, both in output ripple and radiated noise and also possibly contaminating the once clean 12 volt rail.
Why not use a 7805 from the 7812. Nice and quiet albeit with increased power dissipation. If the currents are low that is not an issue.
Why not use a 7805 from the 7812. Nice and quiet albeit with increased power dissipation. If the currents are low that is not an issue.
DC/DC convertors are great for many applications but DC/DC convertors and Phono Preamps are something that doesn't really go together. It all needs to be electrically quiet with clean rails and no radiated hash and noise.
The 7805 will work well connected to the output of the 7812, because of the preregulation.
👍 and as rayma mentioned you now have a clean 12v feeding the input to the 5 volt reg so the output of that one is as clean as can be now.
You could connect the 7805 to the input side of the 7812. That way the 7812 doesn't have to provide the current for the 5 V circuit. The two ICs have the same maximum input voltage so assuming you have sufficient heat sink on the 7805 there should be no issues with that approach.
There's much hand-wringing and fear about DC/DC converters. Most of it is unfounded in my experience. Any modern DC/DC converter IC will switch either around 400 kHz or 2 MHz. This is to ensure that it doesn't interfere with AM radio reception. Both of those frequencies are well above audible, so any noise will be inaudible unless your circuit manages to pick it up and down-convert it to the audio band. That's pretty unlikely. Also, any competently designed DC/DC converter will be designed to minimize EMI radiation. I.e., it will be designed such that it couples as little into the surrounding circuitry as possible. Modern DC/DC converter ICs use spread-spectrum clocking and other tricks to minimize any impact on surrounding circuitry.
Now, I have used some qualifiers here. "Modern". "Competently designed". There's no doubt that a garbage design using obsolete parts will deliver garbage performance.
People also forget that a traditional "linear" supply is not linear at all. Any power supply will be a switching power supply. A traditional "linear" supply switches at twice the mains frequency, so 100 Hz in Europe and 120 Hz in North America. Unlike the switching frequency of a DC/DC converter, these switching frequencies (and many, many of their harmonics) are audible. So I would argue that we should fear the traditional "linear" supply to a much greater extent than any DC/DC converter.
Tom
There's much hand-wringing and fear about DC/DC converters. Most of it is unfounded in my experience. Any modern DC/DC converter IC will switch either around 400 kHz or 2 MHz. This is to ensure that it doesn't interfere with AM radio reception. Both of those frequencies are well above audible, so any noise will be inaudible unless your circuit manages to pick it up and down-convert it to the audio band. That's pretty unlikely. Also, any competently designed DC/DC converter will be designed to minimize EMI radiation. I.e., it will be designed such that it couples as little into the surrounding circuitry as possible. Modern DC/DC converter ICs use spread-spectrum clocking and other tricks to minimize any impact on surrounding circuitry.
Now, I have used some qualifiers here. "Modern". "Competently designed". There's no doubt that a garbage design using obsolete parts will deliver garbage performance.
People also forget that a traditional "linear" supply is not linear at all. Any power supply will be a switching power supply. A traditional "linear" supply switches at twice the mains frequency, so 100 Hz in Europe and 120 Hz in North America. Unlike the switching frequency of a DC/DC converter, these switching frequencies (and many, many of their harmonics) are audible. So I would argue that we should fear the traditional "linear" supply to a much greater extent than any DC/DC converter.
Tom