[time-nuts] Thunderbolt Power Supply Question

Ed Palmer ed_palmer at sasktel.net
Mon Apr 26 17:53:36 EDT 2010

Ouch!  Something else to trip me up.  Thanks for protecting me from 
myself, Bob. :-)


Bob Camp wrote:
> Hi
> One gotcha on your proposed circuit:
> The -12 supply is used for the RS-232 and the DAC that controls the OCXO. I
> suspect you can indeed mess things up with negative supply when the supply
> gets within some magic range of the DAC output. What that range is - no
> idea. 
> Bob
> -----Original Message-----
> From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
> Behalf Of Ed Palmer
> Sent: Monday, April 26, 2010 5:25 PM
> To: Discussion of precise time and frequency measurement
> Subject: Re: [time-nuts] Thunderbolt Power Supply Question
> Hi Robert,
> Robert Atkinson wrote:
>> Hi Ed,The better quality "industral" power supplies normally have a
> regulator for each rail. Some specifications will put a minimum load
> requirement on the primary supply though. 
> You're right that better quality power supplies have regulation on all 
> outputs.  I don't think I'm comfortable using any other kind on a 
> Thunderbolt.  I was just surprised to see some 'close enough' power 
> supplies in use.
>> I've been looking at a simple supply for my TBolt. I wanted to avoid
> switchers or multiple transformers and think I have a solution. I propose a
> 15V 30VA toroidal transformer (if dual secondary windings, connect in
> parallel) . This feeds a full wave bridge and capacitive input filter
> followed by a 12V linear regulator (your choice, I was looking at a LM317T).
> It also feeds a half wave rectifier and capacitive input filter followed by
> a 5V linear regulator. For the -12V there is a capacitively coupled voltage
> doubler feeding another regulator. I've attached a sketch of the circuit I
> used to simulate this arrangement. R1, R3 and R5 represent the 12, 5 and -5V
> loads for the simulation. I think this gives a good repeatable solution
> using standard parts. I've plenty of odd
>>  multi-winding transformers that I could have used, but that would not
> have helped others.  I hope to get it built this weekend.
> I was thinking of something similar, but I've never been a fan of 
> half-wave rectification.  I was thinking of running the +5 supply off 
> the output of the +12 supply.  Total current for +5 and +12 is less than 
> 1 amp at startup and less than 0.5 amp normally.  Power dissipation in 
> the 5V regulator will only be about 2 watts.  For the -12 supply, you 
> could just use a max232 and pull off the -V from the capacitor.  After 
> all, it's just going to be used for RS-232.
> Ed
>> Robert G8RPI.
>> --- On Mon, 26/4/10, Ed Palmer <ed_palmer at sasktel.net> wrote:
>> From: Ed Palmer <ed_palmer at sasktel.net>
>> Subject: [time-nuts] Thunderbolt Power Supply Question
>> To: "Time Nuts Mailing List" <time-nuts at febo.com>
>> Date: Monday, 26 April, 2010, 9:02
>> Many low cost triple output power supplies are designed to power digital
> circuits (e.g. +5V) with maybe some RS-232 or analog circuitry (+- 12V).
> Since the only 'important'  voltage is +5, that's the only voltage that's
> regulated.  The others are designed to be within maybe 5% up to the rated
> load.  I see various power supplies on fleabay that are sold for use with
> the Thunderbolt that seem to fit this model.
>> In the Thunderbolt the +12 runs the oscillator.  Won't an unregulated, but
> relatively steady, +12 supply degrade the performance of the oscillator or
> does the Tbolt have a built-in regulator to deal with this?
>> Ed

More information about the time-nuts mailing list