<Degi>
I wrote a python FFT program for the rigol scope and it can measure 20 attowatts at 350 MHz with no preamplifier (I don't have an actual 20 attowatt source, might try with SDR)
<Degi>
(Well, the noise floor is -137 dBm
<Degi>
at a RBW of 40 Hz)
<azonenberg>
lol nice
<Degi>
Even though it has a noise floor of -147 dBm at 2.4 GHz, holding my phone with iperf3 and 2.4 GHz wifi near it shows nothing
<azonenberg>
is that with an antenna or no though?
<Degi>
Hm, 3-4 cm long piece of solder
<Degi>
It can be used to measure the IMD and image rejection of SDRs though
<Degi>
-14 dB image rejection at 700 kHz... oof
<Degi>
Is the MSO5000 actually a 600 MHz scope?
<Degi>
Or does my SDR output more power at 600 MHz, I measured -40 dBm at 300 as well as at 600 MHz
<Degi>
With 40 Hz RBW I can even measure a 2 GHz signal at -138 dBm
<Degi>
It has approximately -10 dB loss per 100 MHz in the 1.8-2 GHz region
<Degi>
azonenberg, any idea if SERDES can be used to generate frequencies at constant amplitude?
<azonenberg>
Degi: Depends on your definition of "frequencies"
<azonenberg>
you can emit a squarewave at the data rate, or an integer divisor of the data rate, pretty easily
<Degi>
The frequency itself isn't as important as that it is a single-tone frequency and the amplitude is constant
<azonenberg>
then use analog filtering to remove harmonics
<Degi>
Yes, does the amplitude stay roughly the same? I don't really care about the harmonics
<azonenberg>
most serdes do have well controlled swing that's bitstream programmable
<azonenberg>
as long as the load is ~100 ohm differential
<azonenberg>
it's open loop so if you vary the load the swing will change
<Degi>
Hm, I have 50 ohm, but now I remember its quite a shitty termination, since its a BNC T piece with a stub 50 ohm and the 20 pF of the scope... I guess this won't yield spectrographic quality results, maybe if I add a 10-20 dB attenuator before
<azonenberg>
as long as the load is fixed the amplitude will be fixed
<azonenberg>
it just might not be what you set it to
<azonenberg>
Anyway, for the sake of discussion, say you have a serdes clocked at 2.5 Gbps
<azonenberg>
if you feed it an 0x5555 bit pattern, you'll get a 1.25 GHz squarewave
<azonenberg>
So you'll have a fundamental at 1.25, a harmonic at 2.5, a harmonic at 3.75, etc
<Degi>
Hm, if the load is reactive, the amplitude should be frequency-dependent...
<Degi>
Hm yes, I only care about the amplitude of the fundamental, so I won't even need a filter... Might try it out in a bit
someone--else has quit [Quit: Connection closed]
bvernoux has quit [Quit: Leaving]
<azonenberg>
In other news, i'm still going back and forth w/ Sonnet support to tune some of the SMA connector launches
<azonenberg>
Greg has been very helpful at designing reasonably close models of the connectors that still work in a 2.5D solver without needing something fancy like HFSS
someone--else has joined #scopehal
someone--else has quit [Quit: Connection closed]
<_whitenotifier-a>
[scopehal-apps] azonenberg pushed 3 commits to master [+0/-0/±7] https://git.io/JnaeY
<_whitenotifier-a>
[scopehal-apps] azonenberg 63cdd9e - Allow waveform generation filters to be added to an empty session via the "add" menu. See #379.
<_whitenotifier-a>
[scopehal-apps] azonenberg 66113fd - OscilloscopeWindow: proper handling of sessions with no instruments at all. Fixes #379.
<_whitenotifier-a>
[scopehal-apps] azonenberg 90779bd - Fixed "continue" instead of "return" causing Windows build failures
<_whitenotifier-a>
[scopehal-apps] azonenberg pushed 3 commits to master [+0/-0/±5] https://git.io/JnaYK
<_whitenotifier-a>
[scopehal-apps] MegabytePhreak 975d539 - glscopeclient: Find shared yaml-cpp on mingw This was updated for MacOS a while ago but that broke mingw/windows
<_whitenotifier-a>
[scopehal-apps] MegabytePhreak 1bed53a - Fix some windows compile errors These were missed due to already failing CI
<_whitenotifier-a>
[scopehal-apps] azonenberg 264a518 - Merge pull request #380 from MegabytePhreak/windows-fixes Windows fixes