[time-nuts] Power supply for Thunderbolt
quenbob5 at pacbell.net
Tue Aug 5 19:58:20 EDT 2008
I had a similar problem with my home-brew design. I found a "blip" with 1
second period on the output of the error amplifier. I couldn't eliminate
it with extra power supply filtering. The high gain of the error amplifier
picked up a very small voltage that developed on "ground". I solved it by
supplying the phase detector and error amplifier with a separate 5 V supply.
I tied the ground pin of the new 5 V regulator very close to the phase
detector ground pin.
----- Original Message -----
From: "Bruce Griffiths" <bruce.griffiths at xtra.co.nz>
To: "Discussion of precise time and frequency measurement"
<time-nuts at febo.com>
Sent: Tuesday, August 05, 2008 3:43 PM
Subject: Re: [time-nuts] Power supply for Thunderbolt
> David C. Partridge wrote:
>> I'm using the Meanwell T30-B, but Tom van Baak's test on this shows a bit
>> phase noise at 1Hz offset which is a tad nasty:
>> I'm wondering if this can be reduced with a bit of extra de-coupling on
>> 5V supply, but haven't got round to scoping the 5V supply to see what
>> happens every second (presumably as the output is sent to the serial
> If phase noise at 1Hz offset is a problem you will probably need a
> power supply filter with cutoff frequency well below 1Hz to make any
> discernable difference.
> Since the an RC filter will produce a voltage drop in the 5V supply the
> resistor value will have to be kept small and the capacitor value
> correspondingly large (several farads).
> It may be possible to stack 3 low ESR super capacitors in series
> (maximum voltage rating is around 2v -2.5V those with higher ratings are
> actually series connected stacks) to do this but this is a relatively
> expensive proposition.
> Alternatively a modified version of the Wenzel active power supply
> filter could be used if a series stack of 2 low ESR supercaps were
> substituted for the capacitor connected from the supply to the
> transistor base. The modified circuit stabilises the transistor's
> effective emitter resistance and keeps the active filter dc current low.
> It may also be possible to use a variation of Wenzel's opamp augmented
> active filter scheme, however the filter low frequency cutoff will have
> to be well below 1Hz for it to be effective.
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> and follow the instructions there.
More information about the time-nuts