azonenberg changed the topic of #scopehal to: libscopehal, libscopeprotocols, and glscopeclient development and testing | https://github.com/azonenberg/scopehal-apps | Logs: https://libera.irclog.whitequark.org/scopehal
<mxshift> azonenberg: a while back we were talking about VNAs w/ large # of ports. Analog has a prelim datasheet for a VNA-on-a-chip: https://www.analog.com/media/en/technical-documentation/data-sheets/adl5960.pdf
<azonenberg> GenTooMan: lol
<azonenberg> mxshift: oooh let me look
<azonenberg> 10 MHz to 20 GHz, wow. this is for a single port i assume
<azonenberg> Might be worth adding a VNA to our instrument-to-build list
<azonenberg> Any word on expected pricing yet?
<azonenberg> Also, general update - AKL-AD3 tip flex finally shipped, a week later than expected
<azonenberg> should be able to assemble and test over the weekend
<d1b2> <Darius> wonders how unobtainable they are
<d1b2> <Darius> oh "preliminary", so very 😄
<d1b2> <Darius> there is an application circuit for 2 port which uses 2 of them
<azonenberg> Yeah. So a 4 port should be very achievable too
<azonenberg> You could probably fit it in 1U or so
<azonenberg> with room to spare
<azonenberg> and i don't mean like a 1U pizza box, i mean like a 6" deep chassis
<d1b2> <Darius> yeah
* GenTooMan looks, "It appears they only have tested it thus far at 1GHz probably a convenient value"
<azonenberg> I definitely would like to have a VNA on the instrument portfolio eventually
<azonenberg> but it's not the highest proirity
<mxshift> I have no other info. That datasheet showed up in another chat
<azonenberg> Well, realistically i am not starting any new hardware design projects for quite a while
<azonenberg> nor am i finishing most of my current ones
<azonenberg> because components are impossible to source
<azonenberg> i picked the perfect year to have a baby lol
<azonenberg> all my projects are on hold anyway :P
<mxshift> Heh
<azonenberg> the probe stuff hasnt been hit too hard because i have plenty of passives, SMAs are reasonably easy to get
<azonenberg> and while the LDOs I use on the differential probe are sold out on digikey mouser had enough i was able to buy them in reasonable bulk (ten units of each)
<azonenberg> as well as several of the diff amplifiers
<azonenberg> So i can do several more hardware iterations with stuff i have on hand
Degi_ has joined #scopehal
Degi has quit [Ping timeout: 252 seconds]
Degi_ is now known as Degi
<azonenberg> So one thing... if i were to build a VNA around this
<azonenberg> I would want to add relays to mux between the AD chip and a low frequency direct RF sampling frontend
<azonenberg> so i can characterize response down to super low frequencies with a DDS synthesizer and ADC running at a couple hundred Msps
<azonenberg> Not having response <10 MHz is an issue if you want to use the measured s-params for stuff like de-embedding of signals with significant low frequency content
<azonenberg> like say a ddr ram bus with down time between read bursts
<d1b2> <Darius> wouldn't you need a separate directional coupler then?
<azonenberg> so i want a DC coupled path that can run down to kHz range
<azonenberg> My thought is to have a SPDT RF relay
<azonenberg> with one path going to the AD chip and one to my own low frequency path
<azonenberg> No directional coupler, direct RF synthesis/sampling
<azonenberg> It might be forward path only for that path (transmitted power, reflected power assumed zero)
<azonenberg> as channel emulation and de-embedding only cares about the forward path and that's my main use for really low freq sparams
<azonenberg> i just need attenuation and propagation delay of each freq as low as i can go
<d1b2> <Darius> OK
<azonenberg> But by measuring the TX waveform vs what i expect to see
<azonenberg> i think i could get the reflected path that way too
<azonenberg> Anyway, the diff probe, power rail probe, and at least one scope are coming first :p
<azonenberg> last thing i need is another project at this point lol
<azonenberg> Speaking of which the AKL-AD3 flex is finally on the way a week behind schedule
<azonenberg> not sure what happened at oshpark
<d1b2> <Darius> heh
<azonenberg> Looks like it should be coming Friday around lunchtime
<azonenberg> So i basically have from now until then to get the EMC shield paint on the prototype enclosure (eventually i'll want to test without it, but for now i'm keeping it)
someone-else has joined #scopehal
bvernoux has joined #scopehal
someone-else has quit [Quit: Connection closed]
someone-else has joined #scopehal
someone-else has quit [Quit: Connection closed]
someone-else has joined #scopehal
CarlosEDP has quit [Write error: Connection reset by peer]
tiltmesenpai[m] has quit [Write error: Connection reset by peer]
CarlosEDP has joined #scopehal
tiltmesenpai[m] has joined #scopehal
bgamari_ is now known as bgamari
bgamari has quit [Ping timeout: 246 seconds]
bgamari has joined #scopehal
someone-else has quit [Quit: Connection closed]
bvernoux has quit [Quit: Leaving]
someone-else has joined #scopehal
<azonenberg> So I think i am going to refactor the Y axis units API slightly
<azonenberg> in particular, i want to remove the restriction that multiple output streams of a channel have the same units
<azonenberg> this would allow you to have things like mag/angle s-parameters be one channel
<azonenberg> with two sub-streams
<azonenberg> That's the immediate use case but I think there are other situations in which it would be helpful to be able to have two logically connected streams, or outputs from a filter, that are not the same units
<azonenberg> but for example if i wanted to be able to add a filter that simulates a channel model and outputs mag/angle s-params, this would be super nice