[TangerineSDR] 10 Mhz leakage

John Ackermann N8UR jra at febo.com
Tue Jun 2 11:18:24 EDT 2020

10 MHz leakage is a real problem.  cf my lab/ham shack. :-)

But I'm pleased to say that the current design for the Tangerine CKM
module will avoid, or at least won't add to, this problem.  We're going
to directly slave the 122.88 MHz ADC/DAC clock to GPS PPS, with no 10
MHz oscillator in the mix.  So with good practices applied to the
digital layout, we shouldn't create any HF spurs in the clock system.

We will have a 10 MHz output available (synthesized in the FPGA) but it
will be for optional uses and we can make it switchable.

BTW, I don't recall for sure but on the original Hermes board which had
a 10 MHz TXCO to which the 122.88 MHz was locked, I don't think there
was any bleed-through noticeable in the receiver.  So it's possible to
avoid the problem with good design.

The real problem is when the 10 MHz is running around on leaky coax
cables.  Then it gets into the air and is pretty difficult to suppress.
 (I found that cheap RG-58 jumper cables running in parallel in my lab
were porous enough to allow two OCXOs to injection lock -- I ended up
replacing all the jumpers with double-shielded cables and that problem
has gone away.)

On 6/2/20 11:04 AM, Engelke, Bill via TangerineSDR wrote:
> A little discovery that might have a bearing on something in the
> TangerineSDR…. Or maybe this is just my ignorance at work.
> We’re putting a lot of emphasis on being able to observe the doppler
> shift of WWV, often focusing on 10MHz. I have been looking at this using
> FlexRadio 6600 and Spectrum Lab. In observing the frequency moving
> around quite a bit (by this I mean 20 to 40 millihertz), I thought,
> maybe that is the Flex’s TXCO doing its thing. So I got a Leo Bodnar
> GPSDO and started using this as the Flex’s clock.
> What I noticed is that even with the antenna grounded and the rig
> connected to a dummy load, there is a pretty significant (and totally
> stable) 10 MHz signal shown in Spectrum Lab. I speculate that this is
> the clock signal finding its way into the front end of the receiver. So,
> when watching for WWV doppler shift, I see a 10 MHz signal (which is
> comparable in strength to WWV) superimposed on it.  The same signal can
> be seen on 15 MHz and 20 MHz – not surprising, since the Leo Bodnar
> GPSDO provides a square wave output (i.e., rich in harmonics). I
> recognize that the Flex is not a lab instrument, so I have no grounds on
> which to complain, but still…
> I wonder if this will play into the Tangerine SDR?  Are we able to keep
> RF from the clock from leaking into the front end and overlaying the WWV
> we’re trying to observe? Do we need to?
> Any thoughts?     -73- Bill AB4EJ

More information about the TangerineSDR mailing list