[time-nuts] Thunderbolt Supply
lists at rtty.us
Mon Feb 14 19:45:07 UTC 2011
Ok, since this keeps coming up, here's the full blow by blow of everything
you need to do. Nothing left to the imagination :)
You have a switcher running into a filter running. The filter feeds into
three terminal regulators. The regulators are running into the TBolt. Power
to the switcher is from the wall or in some cases from a battery.
First issue is inductors for the filter and why you don't want to just use
Finding an inductor that will handle a couple of amps, and have enough
inductance to be useful down to 30 Hz is a tough proposition. They are going
to be big, heavy, and expensive.
A three terminal regulator, properly chosen, will do a great job down at
audio. Your switcher takes in 60 Hz, it's got 60 Hz ripple on it's output.
You need to get rid of it. No they don't work to hundreds of MHz, but with
care you can find ones that do well over the audio range.
Next step is the first part of the filter. It needs to take over where your
particular three terminal regulator starts to have issues. That could be
lower or higher depending on what you use.
Finding inductors that will work from a few tens of KHz to a MHz while
moving over an amp - much easier than at 30 Hz. The switcher market has us
flooded with cheap (like 10 for a dollar) little parts that do that. Three
caps, two coils = nice filter. If size is not a big issue, getting low DC
drop as well as some inductance is quite doable.
Next stop above that - something designed for RF. Might or might not have a
chunk of ferrite / powdered iron for a core. If you wind your own, very low
cost (maybe a nickel). Likely a double L section, two caps, two coils.
With all the coils, you need to look carefully at saturation. A coil that's
rated for an amp, but that drops to 10% of it's inductance while doing it is
not what you want.
That gets you half of the filter. You need to do the same rock and roll with
the capacitors. The stuff that works at 10 KHz likely isn't going to do much
good at a 200 MHz. Even if they all are ceramics, your dielectrics likely
will change between the sections.
The net result is a gizmo that's quiet, small, and low dropout. None of the
coils will drop much voltage at all. The LDO will be in the sub 1 volt
range. If it drops by a volt with the OCXO full on, it should drop back to
0.1 V after warm up. Works fine.
Hopefully that clears things up.
From: time-nuts-bounces at febo.com [mailto:time-nuts-bounces at febo.com] On
Behalf Of Larry McDavid
Sent: Monday, February 14, 2011 1:32 PM
To: Discussion of precise time and frequency measurement
Cc: EWKehren at aol.com
Subject: Re: [time-nuts] Thunderbolt Supply
I think you should look at the specifications for the noise-rejection
capability of those three-terminal regulators. They are simple and
inexpensive but not particularly good at noise rejection. Some of the
noise from the switcher supply will pass through. Is it enough to cause
a real problem? Unknown to me as yet.
The power requirements of the Thunderbolt are not the usual ones you see
standard supplies designed to meet. I have yet to find a small linear
supply that is adequate that does not absolutely dwarf the Thunderbolt
Back in The Day, one could find a good assortment of multiple-secondary
power transformers that would allow us to build a simple linear supply
with the three-terminal regulators; there would no switcher to generate
noise. Alas, these transformers are no longer easily available today. I
think it can be done with two separate transformers but with the
filters, regulators and actual wiring, this adds up to a lot of work!
TAPR does offer a 13 vdc input, three-output power supply kit for their
software-defined radio kits, but its -12 vdc output comes from a
switching dc-dc inverter. That -12 vdc does supply the Thunderbolt OCXO
so I have some concern about noise on that power feed.
One could get a custom-designed PC board mount transformer and develop a
board to supply adequate power. I doubt there is enough interest or
purchase volume to justify this, however.
So, I'm still looking for a compact linear power supply. I welcome your
On 2/14/2011 10:06 AM, EWKehren at aol.com wrote:
> Yes I am doing it with some very low cost switchers off ebay, followed by
> linear regulators as I mentioned previously. Works great.
> Bert Kehren
Larry McDavid W6FUB
Anaheim, CA (20 miles southeast of Los Angeles, near Disneyland)
time-nuts mailing list -- time-nuts at febo.com
To unsubscribe, go to
and follow the instructions there.
More information about the time-nuts