azonenberg changed the topic of #scopehal to: libscopehal, libscopeprotocols, and glscopeclient development and testing | https://github.com/azonenberg/scopehal-cmake, https://github.com/azonenberg/scopehal-apps, https://github.com/azonenberg/scopehal | Logs: https://freenode.irclog.whitequark.org/scopehal
<marcos__> Thread 1 "glscopeclient" received signal SIGSEGV, Segmentation fault.
<marcos__> std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string (__str=<error reading variable: Cannot access memory at address 0x68>, this=0x7fffffffd5f0)
<marcos__> at /usr/include/c++/7/bits/basic_string.h:440
<marcos__> 440 { _M_construct(__str._M_data(), __str._M_data() + __str.length()); }
<marcos__> (gdb) bt
<marcos__> #0 0x00007ffff7b8f002 in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::basic_string(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) (__str=<error reading variable: Cannot access memory at address 0x68>, this=0x7fffffffd5f0) at /usr/include/c++/7/bits/basic_string.h:440
<marcos__> #1 0x00007ffff7b8f002 in OscilloscopeChannel::GetHwname[abi:cxx11]() (this=0x0) at /home/marcos/paltatech/tools/glscopeclient/scopehal-cmake/lib/scopehal/OscilloscopeChannel.cpp:199
<marcos__> #2 0x0000555555581e52 in OscilloscopeWindow::UpdateStatusBar() (this=this@entry=0x5555558e6210) at /home/marcos/paltatech/tools/glscopeclient/scopehal-cmake/src/glscopeclient/OscilloscopeWindow.cpp:736
<marcos__> #3 0x0000555555581ff8 in OscilloscopeWindow::OnWaveformDataReady(Oscilloscope*) (this=this@entry=0x5555558e6210, scope=scope@entry=0x55555582b4d0)
<marcos__> at /home/marcos/paltatech/tools/glscopeclient/scopehal-cmake/src/glscopeclient/OscilloscopeWindow.cpp:707
<marcos__> #4 0x000055555558230b in OscilloscopeWindow::PollScopes() (this=0x5555558e6210) at /home/marcos/paltatech/tools/glscopeclient/scopehal-cmake/src/glscopeclient/OscilloscopeWindow.cpp:664
<marcos__> #5 0x00005555555acbf1 in ScopeApp::run() (this=0x5555557cdf30) at /home/marcos/paltatech/tools/glscopeclient/scopehal-cmake/src/glscopeclient/main.cpp:101
<marcos__> #6 0x000055555557da51 in main(int, char**) (argc=<optimized out>, argv=<optimized out>) at /home/marcos/paltatech/tools/glscopeclient/scopehal-cmake/src/glscopeclient/main.cpp:237
<marcos__> um, sorry, thats probably not the proper way to paste stuf into irc
marcos__ has quit [Ping timeout: 248 seconds]
marcos__ has joined #scopehal
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
<azonenberg> ok there we go, that's something i can actually work with
<azonenberg> marcos__: yeah a pastebin is normally preferred for that many lines
marcos__ has quit [Ping timeout: 272 seconds]
<azonenberg> Welp, got some initial measurement
<azonenberg> LED ground pins (whatever they're made of) are no better than the steel needles
<monochro1a> azonenberg: iirc... more steel
<monochro1a> or something, they stick to magnets :P
<sorear> steel + rf sounds like a bad time, skin effect in iron is bad enough at 60hz?
<azonenberg> let me see if those are magneti
<azonenberg> you gotta be kidding me
<azonenberg> they are
<azonenberg> here i was thinking they were tinned copper
<azonenberg> so this means my ground impedance has been all out of whack too
<azonenberg> not just the signal pin :p
<sorear> so how do you remotely begin to get a calibrated signal in absolute units out of a rig like this?
<azonenberg> The hope is that i can eventually get it flat enough that i just tell the scope it's a 10x probe and i'm good out to a couple GHz
<sorear> is there any way to actually verify the flatness and establish an uncertainty interval around that 10?
<monochro1a> check against a reference source
<azonenberg> sorear: i'm testing my probe against a direct SMA feed-through
<azonenberg> so i have a good reference to compare to
futarisIRCcloud has joined #scopehal
monochro1a is now known as monochroma
azonenberg has quit [Quit: Leaving.]
azonenberg has joined #scopehal
<azonenberg> anyway i have some 510 bronze wire, 0.635mm diameter, inbound from McMaster-Carr and hopefully arriving tomorrow
<azonenberg> So we'll see how that works
<azonenberg> phosphor bronze is apparently used pretty routinely as an electrical contact material instead of BeCu
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
m4ssi has joined #scopehal
marcos__ has joined #scopehal
marcos__ has quit [Quit: Konversation terminated!]
marcos__ has joined #scopehal
marcos__ has quit [Ping timeout: 272 seconds]
marcos__ has joined #scopehal
bvernoux has joined #scopehal
<bvernoux> hello
m4ssi has quit [Remote host closed the connection]
<azonenberg> new probe tip materials are in
<azonenberg> have to leave for work in a few mins but trying to bang one up quick
<monochroma> :D
<bvernoux> ha nice
<bvernoux> I'm impatient to know if all issues was due to tip only
<bvernoux> evne if it seems it was the case ...
<bvernoux> even
<azonenberg> preliminary results: improved but still more stuff to tweak
<monochroma> azonenberg: significant improvement?
<azonenberg> will compare this to the direct SMA feedthrough when i get home
<azonenberg> dont have time for a full analysis
marcos__ has quit [Ping timeout: 268 seconds]
azonenberg_work has joined #scopehal
<azonenberg_work> So, overall signal quality with the bronze tip isnt much better than I had with the LED pins i think
<azonenberg_work> I still like the bronze from a mechanical perspective, it's my favorite tip material so far
<azonenberg_work> has just a little springiness to it
<azonenberg_work> and its certainly no worse than the other materials
<azonenberg_work> i think the next step is to play with the geometry of the tip
<bvernoux> it is bad after 1GHZ ?
<bvernoux> or before ?
<bvernoux> as from 0 to 200MHz any material shall be good I think
<bvernoux> it shall be sensitive after 500MHz ...
<azonenberg_work> Distortion looks similar to this, but i think slightly less bad
<azonenberg_work> (That test was with the steel tip)
<azonenberg_work> still that step in the edges
<azonenberg_work> by comparison https://twitter.com/azonenberg/status/1131601434481385472 is with the old probe configured as a pass-through with a direct sma connection and an edge launch sma bodged on instead of the tip
<azonenberg_work> Near perfect
<bvernoux> yep nice
<bvernoux> the signal is Gigabit Ethernet ?
<azonenberg_work> So i am fairly confident the tip is the problem
<azonenberg_work> But the material seems to not be the issue
<bvernoux> I could test up to 3Ghz with my signal generator ;)
<azonenberg_work> Which leads me to believe it's the tip geometry
<azonenberg_work> The test signal is a SGMII / 1000base-X idle from an artix-7 GTP
<bvernoux> I have validated it synchronized with GPSDO and my Freq Counter with Power metter
<azonenberg_work> so K28.5 D28.5 basically
<bvernoux> ha ok
<sorear> NRZ clearly
<azonenberg_work> I have also tested on PRBSes
<bvernoux> if you have ref for the bronze tip it could be interesting for other probes
<bvernoux> especially if they are not too expensive
<azonenberg_work> (brb)
<bvernoux> in the quest of cheap GHz probe I have found that too http://jahonen.kapsi.fi/Electronics/DIY%201k%20probe/
<bvernoux> it is quite interesting
<bvernoux> compared to Agilent 7 GHz Infiniimax
<bvernoux> the most funny is the DIY probe has faster rising edge vs Agilent ;)=
azonenberg_work has quit [Ping timeout: 258 seconds]
azonenberg_work has joined #scopehal
azonenberg_work has quit [Ping timeout: 245 seconds]
bvernoux has quit [Quit: Leaving]
azonenberg_work has joined #scopehal
marcos__ has joined #scopehal
azonenberg_work has quit [Ping timeout: 248 seconds]
azonenberg_work has joined #scopehal
futarisIRCcloud has joined #scopehal
marcos__ has quit [Ping timeout: 252 seconds]
azonenberg_work has quit [Ping timeout: 258 seconds]