On Feb 6, 5:50 pm, K7ITM wrote:
On Feb 5, 4:46 pm, Andrew VK3BFA wrote:



On Feb 5, 3:52 am, K7ITM wrote:

On Feb 4, 3:55 am, Andrew VK3BFA wrote:

On Feb 4, 10:13 pm, Leon wrote:

On 28 Jan, 11:27, "Richard" wrote:

I have an ex AM PMR radio that I wish to use on 144Mhz. I need a PSU feeding
24V @ about 300mA.

Are switch-mode powers supplies alright or are they too noisy to be used for
receivers? TIA.

I use a 12V switcher with my FT-817D, it's very quiet.

Leon

You were lucky - you got a quiet one. And its the luck of the draw,
unfortunately.

How come we have all been conned into these switchmode things? - by
their very nature, without a lot of effort, they are noise generators.
Their cheap to make, thats their only advantage - everything else is
part of the spin cycle. And if it works on 2m and above, fine - you
will just be adding to the already horrendous RF pollution from
thousands of the bloody things already on HF - the suburbs are getting
unbearable, S9 plus noise, every night, on 80m.....160 aint to crash
hot either...

I refuse to have to work on a power supply to get it quiet enough to
be used with a radio. I am interested in the radio, not fixing
shortcomings in design of things that , by their very nature, are
unsuitable to be used near radios. That sucks. Linear technology is
far easier and quieter...

Andrew VK3BFA.

There's another advantage of a switching supply that is very important
to some of us: efficiency. The receiver I've been working on runs
from 5VDC, but I have to supply power at 1.0V, 1.2V 1.8V, 2.5V and
3.3V, and each one of those uses a switcher. A linear regulator to 1V
from 5V will result in a supply that's 20% efficient. The switcher
I'm using for that supply is about 87% efficient under its operating
conditions. If I tried to run all my supplies with linear regulators,
I'd have to supply over twice the power; it would come close to
violating the power rating of the 5V supply, and power dissipation in
the receiver would result in excessive heat rise.

But even with all those switchers, the only spurs greater than about
-145dBm (antenna-input-referred) and greater than 1.5MHz are from
digital circuits, not the power supplies. -145dBm is, I believe, 20dB
_below_ S1 by usual definition of the S-meter units. Even the power
supply fundamentals are only at about S1. And the supplies are not
themselves shielded. (The receiver RF circuitry is, of course, as is
the whole receiver module.)

I would be much more circumspect about using a switching supply in a
homebrew receiver that was going to be used where power dissipation
wasn't an issue, but then I'd do a lot of things differently than I
have for such a receiver. It wouldn't keep me from using switchers;
I'd just look at the whole design in a different way. On the other
hand, finding ones that would be adequately quiet for a 144MHz
receiver should not be difficult at all.

Cheers,
Tom

Hi Tom,
all points noted, and readily conceded - if the things didn't have
SOME advantages besides low cost, well....
You've got yours down to S1 - mm, well,......so thats the noise floor
you are limited to. And thats probably on a small range of
frequencies, gawd knows where else it is radiating...(and adding to
phase noise in your receiver...)

And, lets be honest, if it isn't causing ME QRM, on the frequency I
want to operate, then its OK......fine for VHF/UHF unless a spur is on
your IF frequency...

The only application I can see for them is, as you say, where low
power consumption/efficiency is paramount - but if you can plug your
switch mode power supply into a source of AC power, then whats the
problem with lugging a quiet linear supply as well as the generator
required to run it?

To me, they are a flawed technology, and in a better place would not
be allowed to be used because of their inherent design flaws...

Andrew VK3BFA.

(PS - I do a good "back in the good old days" rant if you like.....)

Oh, well, the receiver is 100kHz (nom.; actually is -3dB at about
18kHz) to about 38MHz. It also alternately does 70MHz IF at 36MHz
bandwidth, at about the same sensitivity. Noise figure on HF is
around 9dB at max sensitivity; since switchers are not stable sources,
their output becomes broadband noise by the time you get to bandwidths
low enough to see -145dBm (e.g., 100Hz), and there's essentially no
indication of increased noise floor at the harmonics (past maybe the
third) of any of the supplies, even with lots of averaging. The
master clock is a low phase noise VCTCXO, and it IS run from a linear
supply, with scrubbing added. That oscillator really does set the
system phase noise. There are no other oscillators to worry about in
the signal chain.

You're welcome to rant about the "good old days," but I probably won't
hear it. I'm too busy enjoying the "even better new days" to bother
looking back.

Cheers,
Tom

Hey - I agree - I like the modern DDS chips as well, fantastic things.
(Note - AD specify separate ANALOG supply for them...Mmm?)


I try again - my point is, why bother to stuff around getting a "new"
technology to work when an "old" technology is cheap, easy,simple, and
it works with no extra fiddling.

And if you need a spectrum analyser to check your power supply for
suitability, - that is a bit ludicrous, isn't it? Mad Scientist in
attic room / Tony Hancock sketch sort of thing. Why make things so
difficult? KISS.

Never said there were no quiet switchers - just that the majority were
not....Google on SMPS noise to prove.....

Andrew VK3BFA.