Sujet : Re: Distorted Sine Wave
De : jeroen (at) *nospam* nospam.please (Jeroen Belleman)
Groupes : sci.electronics.designDate : 03. Jun 2024, 19:38:21
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <v3kuuh$36c$1@dont-email.me>
References : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
User-Agent : Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.13.0
On 6/3/24 17:43, Cursitor Doom wrote:
On Mon, 03 Jun 2024 07:48:32 -0400, Joe Gwinn wrote:
On Sun, 2 Jun 2024 20:58:45 -0000 (UTC), Cursitor Doom
<cd999666@notformail.com> wrote:
>
On Sun, 02 Jun 2024 14:08:48 -0400, Joe Gwinn wrote:
>
On Sun, 2 Jun 2024 16:55:28 -0000 (UTC), Cursitor Doom
<cd999666@notformail.com> wrote:
>
On Sun, 02 Jun 2024 12:19:05 -0400, Joe Gwinn wrote:
>
On Sun, 2 Jun 2024 11:31:33 -0000 (UTC), Cursitor Doom
<cd999666@notformail.com> wrote:
>
On Sun, 2 Jun 2024 11:17:58 -0000 (UTC), piglet wrote:
>
Cursitor Doom <cd999666@notformail.com> wrote:
On Sat, 1 Jun 2024 22:00:58 -0000 (UTC), piglet wrote:
>
Cursitor Doom <cd999666@notformail.com> wrote:
On Sat, 1 Jun 2024 15:44:17 +0200, Jeroen Belleman wrote:
>
On 6/1/24 14:07, Cursitor Doom wrote:
>
I've taken a shot of the waveform into the 50 ohm input. It's
around 850mV peak-peak. Hopefully the slight distortion I
spoke about is visible; the slightly more leisurely
negative-going excursions WRT their positive-going
counterparts. So it's not a pure sine wave as one would
expect. Does it matter? I don't know!
>
<https://disk.yandex.com/i/7cuuBimDbOIBZw>
>
>
And <https://disk.yandex.com/i/z6fYbeVfPRK7aA>
>
>
The shape looks perfectly acceptable to me. This is +3dBm into
50 Ohms.
Is that what it's supposed to be? Canned reference oscillators
most often deliver +13dBm, sometimes +10dBm.
>
Is it? I only make it about half your figure: +1.65dBm.
I admit I'm frequently prone to careless errors, so stand to be
corrected,
but here's my method:
850mV peak to peak is 425mV peak voltage. Average of that is
0.425x0.636 =
0.27V. Average power is average volts squared divided by the
load impedance of 50 ohms = 1.46mW = +1.65dBm.
>
I shall consult the manual to see what it ought to be - if I
can find it, that is, as PDF manuals are a nightmare to
navigate IME.
>
>
>
Use 0.71 for RMS instead of 0.636 ! I make that about 1.8mW or
+2.6dBm ?
>
Thanks, Erich. But there's no such thing as "RMS power" strictly
speaking IIRC, so that's why I took the average figure; not that
it makes much difference in practice. it does seem a bit on the
low side, but despite reading through the most likely sources
(the service manual and the trouble-shooting/repair manual) I can
find nothing stated for what that signal level should be! This
may be due to the user-unfriendliness of very large PDF manuals;
I just don't know. Anyway, not very satisfactory! Later today I
plan to do a direct power meter measurement of the ref osc (since
none of us here seem to agree on what 850mV vs 50 ohms equates
to!!)
>
>
>
>
>
Since you have a power meter, a signal source, and an oscilloscope
why not measure the peak to peak voltage on the scope and power on
the power meter and see which calculation 0.636 vs 0.707 gives the
closest agreement?
>
It wouldn't prove anything one way or ther other, though, since that
power meter hasn't been calibrated for "quite a while" so to speak.
:)
It'll give a 'good enough' reading for my purposes, but won't be
accurate enough to meaningfully test your otherwise fine suggestion.
>
>
The 0 to +10 dBm range I mentioned came from the service manual.
>
Looking at your scope picture, it looks like a 3 Vpp signal, which
is +13 dBm, a very common distribution level, but one that exceeds
the analyzer's allowed range. All that's needed to fix this is a
3dB inline attenuator. Here is one for SMA connectors:
>
.<https://www.amazon.com/MWRF-Source-Male-Female-Attenuator/dp/
B07MP9D9GC?th=1>
>
Just buying a few of these and doing some experiments will be far
cheaper and faster than the various alternatives discussed.t
>
Joe Gwinn
>
I think you're looking at the first picture with the signal into the
scope's 1 Meg input. The 50 ohm trace is only 850mV peak-to-peak or
thereabouts and when I measured it with an actual power meter, came
out at about +2.5dBm so within the range you stated; no attenuation
needed (thanks for the range, by the way; I needed to know that).
>
What we don't know is exactly how you made the various measurements.
If you are observing the signal from the 10 MHz reference where it
enters the analyzer, I would expect that there is a T-connector with
the scope (set to 1 Mohm) listening in to passing signals.
>
You did ask me this before and did post an answer. See Message-ID:
<v3fsbp$2u0a6$1@dont-email.me>
>
You also still appear to think that the 10Mhz signal is going into the
analyzer. It isn't. It's coming out. Again, see Message-ID:
<v3fsbp$2u0a6$1@dont-email.me>
>
I did read that, but didn't know what to make of it. I think an
annotated drawing is required.
>
On this drawing, where do the various scope traces mentioned up-thread
come from?
>
Joe Gwinn
>
>
>
>
>
>
In this case, the load seen by the incoming reference is that provided
by the input on the analyzer. Which input is +10 dBm max. If you set
the observing scope input to 50 ohm, the reference will see a 25 ohm
load, cutting the signal seen by the analyzer by 3 dB. Which will
take +13 dBm down to +10 dBm, which is in range.
>
A 3dB attenuator in line will drop the signal to 10 dBm as well.
>
I've built lots of systems like that. The 10 MHz reference is
delivered to everybody at +13 dBm, and it is the receivers'
responsibility to attenuate it to whatever they need.
>
>
I've now measured the 100Mhz oscillator and that seems fine, although
I only saw 0.61V p-p into 50 ohms, so somewhat less than the 10Mhz
oscillator's output.
So far, I've not measured anything which screams "the fault's here!"
as all the expected signals are present - although admittedly I have
many more to test. But certainly all the *major* signals within this
complex beast are present. It's looking like it could be an issue with
one of the phase detectors or LPFs. Sigh....
>
To my eye, it does scream.
>
Joe Gwinn
Joe, I appreciate you're only trying to help, but don't worry about it. I
don't believe the oscillator this thread relates to is causing the PLL
unlock error so we've all gone down a bit of a rabbit hole with this one.
I really need to look elsewhere for the culprit.
For what it's worth, I found a manual on the web saying that the
10MHz output should deliver +5dBm into 50 Ohms. It's a bit low,
then, but I doubt that this is your problem.
Jeroen Belleman