<azonenberg>
miek: looks like you also missed registering the DCAEdgeTrigger class
<azonenberg>
The way your driver is written you *probably* can still load save files without it, but it's better if you do it
<azonenberg>
If you fix that it looks good to merge
<azonenberg>
also yay ISI channel test boards should be coming tomorrow
<azonenberg>
they took one heck of a circuituous route, thanks fedex
<azonenberg>
Shenzhen -> Guangzhou -> Anchorage AK -> Oakland CA -> Memphis TN -> Seattle WA
* GenTooMan
thinks of an even crazier route for azonenberg, Shenzhen (CN) -> Ulaanbaatar -> Pyongyang -> Rio de Janeiro -> Machu Picchu -> Mexico City -> Browsville TX -> Memphis TN -> Seattle WA
<azonenberg>
GenTooMan: have you ever actually shipped anything via that route? lol
<azonenberg>
i dont think fedex ships to north korea :p
<azonenberg>
(and you forgot a layover in antarctica)
<GenTooMan>
it takes a year because things get frozen for a bit I've heard.
<azonenberg>
lol
<azonenberg>
anyway... new AKL-AV1 board is assembled and looks good
<azonenberg>
there is one small mechanical fix i need to make, a resistor too close to an IC (by "too close" I mean "partially under it" lol, but there's enough land sticking out to solder it still)
<azonenberg>
about to solder up the tip resistors and give it a test
* GenTooMan
hopes nothing explodes ... well nothing that isn't supposed to explode at least.
<azonenberg>
AV1 response before/after adding equalizer to the output
<miek>
ooh, nice
<azonenberg>
I'm excited lol
<azonenberg>
This is one of my most heavily analog circuits to date
<azonenberg>
like, the other probes are super simple architecturally
<azonenberg>
the PT5 is basically just a resistor, the AD4 is just two PT5s feeding into an amplifier
<azonenberg>
the AV1 may not be super high bandwidth but it's a much more architecturally complex design with more analog stuff going on to get good flatness across the whole band
<azonenberg>
there's three different stages of equalization and flatness tweaking to get good response across the whole operating band
<azonenberg>
you have trimming of the low and high band loops of the BUF802 (lowest frequency), the trimming on the R-C divider (higher frequencies), and then now the fixed setting CTLE at the output
<electronic_eel>
azonenberg: the AV1 results look really good!
<azonenberg>
Yep :) It looks like a winner
<azonenberg>
i have three ECOs planned for the next version
<electronic_eel>
have you done a vna measurement of it targeted at low frequencies too? i'm thinking of the crossover region between the buf802 and the low freq opamp?
<azonenberg>
1) fix the clearance violation where the 0402 is colliding with the sot23
<azonenberg>
2) add the equalizer cap
<electronic_eel>
iirc that is between dc and 1 mhz (or maybe a bit more)
<azonenberg>
3) add an additional ground contact on the other side of the signal allowing left/right handed deployment
<azonenberg>
My VNA bottoms out at 300 kHz
<azonenberg>
I do flatness correction at low frequency with a scope and a ~20 kHz squarewave off a function generator
<azonenberg>
then twiddle R9 (low freq, BUF802/opamp crossover gain matching) and C8 (higher freq, R-C divider matching to BUF802 input capacitance) to get the squarest waveform possible
<azonenberg>
just like compensating a normal r-c scope probe
<azonenberg>
the diff being that since these are internal matching vs matching to an external scope they are considered factory calibration steps the end user will not need to mess with
<electronic_eel>
do you have a regular dds function generator? that could also be used for lower frequency analysis
<azonenberg>
i was using the LFO on my siglent vector siggen
<electronic_eel>
is not a full vna of course, but if you just care about magnitude
<electronic_eel>
ok, so if you have looked at the lower freq output of your siggen and didn't notice anything being off with the AV1 output then i think it is fine
<electronic_eel>
the composite loop mode of the buf802 is a bit more complicated than your other probes. could have been a path for issues to sneak in
<azonenberg>
yeah i did flatness correction with the siggen at around 20 kHz squarewave
<azonenberg>
and manually tweaked the trimmers until it looked nice
<azonenberg>
R9 is much lower freq than C8
<azonenberg>
so you tune R9 first then C* to get the last bit
<azonenberg>
C8 i mean
<azonenberg>
ISI channel test boards arrived. Assembling and testing after work
GenTooMan has quit [Ping timeout: 245 seconds]
GenTooMan has joined #scopehal
Degi_ has joined #scopehal
tiltmesenpai4 has joined #scopehal
balrog_ has joined #scopehal
Degi has quit [*.net *.split]
balrog has quit [*.net *.split]
Degi_ is now known as Degi
tiltmesenpai has quit [Ping timeout: 240 seconds]
tiltmesenpai4 is now known as tiltmesenpai
ericonr- is now known as ericonr
_whitelogger_ has joined #scopehal
GyrosGei1r has joined #scopehal
laintree has joined #scopehal
asy_ has quit [Ping timeout: 240 seconds]
welterde has quit [Ping timeout: 240 seconds]
Yamakaja has quit [Ping timeout: 240 seconds]
GyrosGeier has quit [Ping timeout: 240 seconds]
electronic_eel has quit [Ping timeout: 240 seconds]