[hpsdr] Penelope measurements

[Phil Harman]

Hi Mark,

Sorry about the delay in replying - I've just arrived home from Dayton.

Following Dayton I spent a fascinating afternoon with John, N8UR, doing 
measurements on Penelope - more on this later.

We did the same measurements as you with similar results. The on board 
TCXO is really intended to remove drift from the 122.88MHz clock rather 
than a highly accurate frequency reference. It's a relatively low cost 
device and the sort of frequency error you found is consistent with 
what John measured on his Penelope. We also found the TCXO drifted some 
which made phase noise measurements difficult.

Could you please measure the 10MHz TCXO frequency on the Atlas bus at 
C16. If you select Penny as the 10MHz source in PowerSDR then Penny 
will send this signal to the bus. You can then measure the frequency 
error. If you get a similar frequency error then this means it's the 
TCXO; else there may be a problem with the PLL.

We also tried injecting a very high performance 10MHz source into C16 
and phase locking the 122.88MHz clock to it. We found the same issue - 
the PLL does not lock correctly and we had a frequency error. The 
reason for this was that the FPGA needs a nice sharp edge 3.3v square 
wave on C16 and by the time we have a suitable level when feed through 
10'  of coax the waveform was sufficiently distorted to cause a problem.

I did test feeding the 10MHz TCXO on the (Alpha 1) Mercury board into 
C16 to lock the 122.88MHz clock on Penny and it worked fine.

John is working on a small board that will plug into the Atlas bus that 
will provide the correct drive levels etc. The board can also be fitted 
with a high performance 10MHz OCXO/VCXO and also take external clocks. 
These will be able to be divided/multiplied to 10MHz so external 
standards (2/5/20MHZ) can be used. There will also be a GPS locking 
facility.

I'll publish all the results of the tests that John did on Penelope 
ASAP, just to say at this stage that the results are very gratifying.

73's Phil...VK6APH




Quoting Mark Amos <mark.amos at toast.net>:

> ***** High Performance Software Defined Radio Discussion List *****
>
> Folks,
>
> I did some more playing around with Penelope this weekend testing the 
> frequency of the transmitted output.
> Penelope had been powered up for about a half an hour. These were 
> tests run at 10.101000 MHz.
>
> I found that when I selected Penelope for the 10 MHz clock source the 
> transmit signal was low by 133 Hz
> (10.100867).  When I selected Atlas as the 10 MHz clock source (using 
> an an external GPSDO fed to pin C16 on
> the Atlas bus) the measured transmit frequency was about 24Hz high at 
> 10.100024.
>
> I expected it to be spot on. (A similar experiment with my SDR1000 
> showed the output to 10.101000 using the
> same GPSDO as with Penelope and same counter.)
>
> I believe the measurements to be accurate - I'm using an HP 5328A 
> counter with a GPSDO external clock that
> measures the GPSDO's to within 1 Hz at 10 MHz.  Also, when I zero 
> beat the GPSDO driving Penelope against 10
> MHz WWV with the SDR1000 and Ten Tec Orion it seems to be dead on 
> when it shows 10.000000 on the counter.
>
> So, I think there's something wrong with my expectations.  Shouldn't 
> the transmitted frequency be closer than
> 2.4 PPM?
>
> Just to be certain that my question is not misunderstood - I'm not 
> complaining about this; I'm really just
> trying to learn something!
>
> Thanks in advance!
>
> Mark
>
>
> _______________________________________________
> 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/
>



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.


 1211434289.0