[hpsdr] Demeter PSU design

Tue Jul 1 16:48:34 PDT 2008

Jeroen

Enjoyed talking with you and the others on Saturday at 19:30. I'll be around
again at 19:30 and hopefully we will get a few more folks to the east of the
USA!

Thanks for running with the Tiny Demeter! It has gotten my reading juices
flowing again. Open design discussion has been the lifeblood of H in HPSDR!
I think there are now over 700 on this list. There is a LOT of expertise out
there even amongst the lurkers!

I did a search of earlier e-mails and Bill had posted the current
requirements of Janus/Ozy early last year. I guess they are still valid:

Janus/Ozy current draw:

***************************
Bill Tracey - KD5TFD schreef:

Current draw I'm seeing with Ozy/Janus playing radio with an SDR 1000 is:

      +12v: 200 ma
      +5v: 180 ma
      -12v: 70 ma

***************************

I'm not sure about Penelope running a full .5 Watt, but perhaps Kevin or
others might measure Penelope. Also measure Ozy alone without Janus to get a
baseline, where we can determine other boards current requirements by
difference, since OZY is always required to even test. Can someone help with
Penelope minus Ozy?

I have always looked at every project from 10,000 ft. as they 'fit' with
other projects. Bottom line is that all the HPSDR boards and external
projects should 'play nicely' together and avoid duplication in keeping with
not duplicating functions on other logical 'building blocks'. For instance
having to duplicate clocks or oscillators serving the same function on other
boards. 

HPSDR design should accommodate other planned projects, but not
'overcompensate'.

My suggestion for Tiny Demeter, whether the board be mounted on the end
mounting screws of Atlas or the ATX connector itself (If we can find one),
that the board extend the same distance rearward from atlas, in the same
plane as other buss mounted boards.

The object is that we mount protruding LED power indicators, and a switch on
the 'rear' edge of Tiny Demeter such that it would extend through the
rear/front panel of Pandora or other enclosure.

Further: That this rearward extension have 2 x 10 pin .1 inch headers mated
with each other with a 'cut' line between them so a daughter board with
indicators could be produced. This could be cut off the regulator board, and
mounted on front panel of an enclosure, via a commonly available 10 pin
ribbon cable.

Just my early 2 Cents!

Thanks
Eric

-----Original Message-----
From: hpsdr-bounces at lists.hpsdr.org [mailto:hpsdr-bounces at lists.hpsdr.org]
On Behalf Of Jeroen Bastemeijer
Sent: Tuesday, July 01, 2008 11:37 AM
To: Phil Harman; HPSDR
Subject: Re: [hpsdr] Demeter PSU design

***** High Performance Software Defined Radio Discussion List *****

Dear Phil, Graham and others,

Thank you for all the E-mails with regard to the Tiny Demeter design! I 
will read them and respond to them in the coming days.

Phil, at this moment I'm very curious about the performance of Mercury! 
I would like to give you the following suggestion for doing more 
"scientific" noise measurements on Mercury. It is possible to do these 
experiments with a simple setup.

What you need:
* Power supply (which you already have)
* Signal generator
* Toroid or transformer with two coils (primary and secondary)

The transformer is the most tricky part: with succes I used a, so 
called, common mode choke. These kind of chokes can usually be found in 
the input-filter of switching power supplies. So, if you have an 
old/defective computer power supply there is probably one in there. The 
common mode choke is basically in series with the mains connection. It 
passes the mains-power but it represents a short for "RF"-common mode 
signals on the power line and vice-versa.
These chokes have a pretty high power handling capabillity and they have 
two sets of windings (ratio 1:1). The higher the current handling 
capabillity, the better.

For the measuremnts you can put one side, let's call it the primary, in 
series with the power supply line (e.g. the 5 Volt line). The secondary 
winding can be connected to the signal generator. In this way it is 
possible to inject an AC-voltage on the power supply line. The generator 
will be  loaded with the circuit as well, so, when you turn up the 
output level of the generator, a much smaller waveform will be present 
on the power supply (due to the 50 Ohm internal resistance of the 
generator). By adjusting the frequency, amplitude and the waveform you 
could see if there is any influence on the performance of Mercury.

Hope you are willing to do these tests, this will help us enormously in 
determening the finals specs for (Tiny) Demeter.

One more thing, on one of the boards a LT3080 is used. If I'm correct it 
is used with a gyrator like circuit in series. Is the schematic where it 
is used somewhere online?

Thank you, 73 Jeroen PE1RGE

Hope you are willing

:
>
>>
>> Do you have any means to test the sensitivity of the HPSDR and/or
>> Mercury for noise on the supply lines? Especially for "high frequency"
>> noise (at the switching frequency (and harmonics) of a swicther) on the
>> +5 Volt and the -12Volt rail?
>>
>
> Nothing beyond using it to power Mercury and looking at the bandscope 
> - not very scientific I'm afraid!
>
> I think the picoPSU will be a good test and we should be able to try 
> that in the next few days.
>
> 73's Phil...VK6APH
>
>
>

-- 
Ing. Jeroen Bastemeijer

Delft University of Technology
Department of Electrical Engineering
Electronic Instrumentation Laboratory
Mekelweg 4, Room 13.090
2628 CD Delft
The Netherlands

Phone: +31.15.27.86542
Fax: +31.15.27.85755
E-mail: J.Bastemeijer at TUDelft.nl
GPS: Lat N52.00002 Lon E4.37157 Alt 46.2m

_______________________________________________
HPSDR Discussion List
To post msg: hpsdr at hpsdr.org
Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
HPSDR web page: http://hpsdr.org
Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/

 1214956114.0