[time-nuts] power supply design Re: Slightly OT: interest in a four-output, > ultra-low jitter, synthesizer block?

[Gerhard Hoffmann]

Am 26.01.2018 um 05:26 schrieb Mark Goldberg:
> On Thu, Jan 25, 2018 at 8:51 PM, jimlux <jimlux at earthlink.net> wrote:
>
>> What we've done is switcher from wide range bus (9-24V) to 8V, 60 dB
>> ultimate attenuation low pass, switcher 8V to 5V(e.g.), 60 db low pass,
>> linear with great HF rejection (i.e. the LT3042) to 3V
>>
> If it is any help to someone, I did some research on low noise regulators
> to drive a low phase noise oscillator. You will have to look at this in a
> fixed width font. LT3042 is indeed a spectacular part with way less noise
> than the baseline LT3009. Hopefully no mistakes below. Pricing is about 2
> years old.
>
Yes, the LT3042 is in a completely different league than the ordinary
LM317, LM350, LM2940, 7805. It has simply 40 dB less noise than the rest.

Don't overdo it with the output capacitor, 4.7u is ideal. 22u produces
a noise peak at 15KHz.

< 
https://www.flickr.com/photos/137684711@N07/24070698809/in/album-72157662535945536/ 
   >

The high current version with the external NPN is also fine.
< 
https://www.flickr.com/photos/137684711@N07/29452163806/in/album-72157662535945536/ 
 >

The test circuit is quite small, 1:1 from the data sheet.
< 
https://www.flickr.com/photos/137684711@N07/29197476530/in/album-72157662535945536/ 
 >

The steep rise below 100 Hz is an artefact of the too small coupling 
capacitor
of my preamp and the 1/f behavior of the 89441A that I used as FFT analyzer.