<gruetzkopf>
wq: if i want to "ooh - shiny" you into implementing support for TI CC-link programming (and 8051 debug) where do i need to dropship target devices (and where do i buy them so they actually arrive
simukis_ has joined #glasgow
simukis__ has quit [Ping timeout: 240 seconds]
Getorix has quit [Quit: My iMac has gone to sleep. ZZZzzz…]
ma1 has joined #glasgow
Getorix has joined #glasgow
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
ma1 has quit [Quit: ma1]
ma1 has joined #glasgow
futarisIRCcloud has joined #glasgow
Getorix has quit [Read error: Connection reset by peer]
Getorix_ has joined #glasgow
Getorix has joined #glasgow
Getorix_ has quit [Ping timeout: 268 seconds]
parataxis has quit [Ping timeout: 268 seconds]
analprolapse has quit []
analprolapse has joined #glasgow
fibmod has quit [Ping timeout: 240 seconds]
<whitequark>
electronic_eel: the whole point of my exercise was to lower part count
<whitequark>
it's fairly obvious how to do this with a ton of discrete parts
<whitequark>
gruetzkopf: are there docs
<gruetzkopf>
there are official protocol docs, yeah
Stormwind_mobile has quit [Ping timeout: 268 seconds]
Stormwind_mobile has joined #glasgow
Exec1N has joined #glasgow
emilazy has quit []
emilazy has joined #glasgow
Exec1N has quit [Ping timeout: 272 seconds]
Exec1N has joined #glasgow
Stormwind_mobile has quit [Ping timeout: 240 seconds]
Stormwind_mobile has joined #glasgow
bvernoux has joined #glasgow
Getorix_ has joined #glasgow
Getorix has quit [Ping timeout: 240 seconds]
<whitequark>
awygle: so here's something interesting
<whitequark>
i put a glasgow right on top of the antenna and ran benchmarks to stress usb
<whitequark>
it emits surprisingly little noise
<pjustice>
and apparently resists external rf reasonably well
<whitequark>
conversely, if i turn on my T12 soldering iron...
<whitequark>
as far as i can tell it emits broadband interference that raises the noise floor by 5 dB everywhere
<pjustice>
One of these things is cheap Chinese ... stuff
<pjustice>
The other is not.
<awygle>
lol that is interesting
<whitequark>
it's actually in a metal case which makes it doubly surprising
<whitequark>
oh wait
<whitequark>
it's not "everywhere", it interferes with the SDR after the mixer
<whitequark>
because it doesn't matter what my LO is set to, i see these peaks at the same position
<pjustice>
switching ps?
<whitequark>
the iron?
Getorix has joined #glasgow
<pjustice>
yeah
<whitequark>
yeah that's how all t12 irons work
<whitequark>
hm, wonder if ts100 fares any better
<pjustice>
Other than that, do you like the T12?
<whitequark>
hm it does actually, ts100 produces very little rfi
<whitequark>
as far as i can tell
Getorix_ has quit [Ping timeout: 268 seconds]
<whitequark>
pjustice: yes, t12 is awesome
<whitequark>
i wouldn't settle for anything less these days; metcals are better if you don't work with old stuff and can afford them
parataxis has joined #glasgow
<pjustice>
I wouldn't think the tip heater would need to be cycled particularly fast. So the TS80/TS100 which are DC powered should have the advantage over the AC powered T12 in noise.
<whitequark>
oh sorry, I meant I like the cartridge
<whitequark>
that specific T12 controller is ok i guess; it was cheap mostly, and served me well
<awygle>
i'd be interested to see a near field probe on the glasgow usb
<awygle>
it's probably pretty good though, short run by competent engineers
Stormwind_mobile has quit [Ping timeout: 272 seconds]
<tnt>
whitequark: btw, wrt to some other of your questions on twitter, don't forget the LO is not a clean sine, it pretty much has all harmonics present so if you tune to f, you will most likely receive 2*f, 3*f, ...
<whitequark>
ohhhhhh
<whitequark>
oh THAT's why the LPF setting is so important
<whitequark>
or, wait
<tnt>
Well the LPF is only for the baseband signals, that's to avoid the sampling image folding.
<whitequark>
yes
<whitequark>
hm
<tnt>
To deal with the rest ... you need band specific filters in front of the SDR.
<whitequark>
so the only way to deal with LO not being perfect sine is to reject it bef... yeah
<whitequark>
ok
<whitequark>
makes sense
<whitequark>
hm. what could be an absurdly powerful signal at 650 MHz
<whitequark>
wait, that's an artifact again
Stormwind_mobile has joined #glasgow
<awygle>
SDRs: fun and definitely not a trap
<whitequark>
hahah
<whitequark>
currently staring at what i assume is wifi
<whitequark>
20 MHz wide signal at 2.437G
<awygle>
sounds like wifi to me
<pjustice>
channel 6
<hl>
what SDR is this?
<whitequark>
limesdr mini
<hl>
interesting, you're using an SDR to do emissions testing?
<awygle>
i was about to bitch about people using channel 6
<awygle>
but wq is in russia, so probably not as bad
<whitequark>
hl: no, i'm using an SDR to fuck around
<hl>
ah
<hl>
bet you could save a lot of money on emissions testing equipment though
<hl>
hmm
<whitequark>
awygle: i'm extremely surprised to see absolutely nothing on channel 1
<whitequark>
or 11
<whitequark>
oh
<whitequark>
that's because it's my own wifi router on channel 6
<awygle>
i got channels 5 and 6 mixed up, to my great shame
<awygle>
6 is fine
<whitequark>
i assume everyone else's just attenuated to oblivion
<whitequark>
which is actually strange because the folk wisdom is that 2.4 is unusably crowded here
<whitequark>
except... it actually seems pretty empty
<awygle>
hl: you could do a pretest, but no regulatory body would accept "my limesdr says i meet the emissions mask", of course
<hl>
awygle: of course, but it would be nice to know before sending it to them
<whitequark>
(my own devices use 5G wifi, 2.4G is for people without 5G-capable devices)
<hl>
>2.4 is too crowded
<awygle>
an SDR would be helpful for finding issues you already know are bad enough to need fixing
<hl>
>everyone starts saying "2.4 is too crowded"
<hl>
>everyone goes to 5
<hl>
>2.4 is empty
<whitequark>
to be fair, that's only the second reason i'm using 5G
<awygle>
but you can't confidently say "this is going to be a problem", without doing at least some calibration (which is not that hard, tbf)
<hl>
ah, hmm
<hl>
honestly I know nothing about SDR, etc.
<whitequark>
the first reason is that if i switch to 2.4G, my laptop's wifi card interferes with its own thunderbolt controller so badly, the thunderbolt link drops
<whitequark>
and the wifi link drops, too, but that doesn't cause my kernel to get hung in a mutex
<hl>
hahahahahaha
<awygle>
i wonder if i could run a fat32 driver on a boneless cpu...
<whitequark>
i wrapped the entire wifi card into a kilometer of foil and now it's quite a bit less bad
<whitequark>
but i still use 5G
<electronic_eel>
hl: using a sdr for pre-compliance testing usually isn't worth the hassle. as wq posted, there are a lot of artifacts, spurs and so on. you'd have to check if each of these is a real signal or some crap. spectrum analyzers are a lot cheaper now than a few years ago
<hl>
aah
<awygle>
yeah, grab a 2k siglent, it's worth it if it saves you 30m of chamber time
<electronic_eel>
also a sdr gives no guarantee that the signal level shown is somehow linear over the whole spectrum. it can vary a lot in funny non-linear ways
<gruetzkopf>
in europe you could technically run 1-5-9-13
<electronic_eel>
get the cheapes siglent and hack it up to full specs. its quite easy to do (see eevblog forum)
<awygle>
1 and 5 overlap
<awygle>
slightly
<gruetzkopf>
yeah, slightly
<gruetzkopf>
same for any of those channel pairs
<awygle>
electronic_eel: yes, although they're not calibrated out over the full range
<gruetzkopf>
but sometime running 4-channel is worth that
<awygle>
there's a noticeable non-flatness out past the rated range
<awygle>
at least on mine
<hl>
I know rigol is the modern canonical "so you want to buy a cheapo oscilloscope" vendor. So I take it for analysers Siglent is preferred over Rigol?
<awygle>
dunno about that. rigol didn't have one i liked when i went looking, but they've come out with some nice ones since.
<hl>
ah
<electronic_eel>
the spectrum analyzers of siglent are much better than the rigols, at least for now
<hl>
noted
<awygle>
siglent and rigol are pretty comparable quality across the board ime
<electronic_eel>
for scopes I prefer rigols
<awygle>
siglent has better screens :p
<electronic_eel>
I did a very short test of a siglent scope like 2 years ago on a fair, could get it to reliably show crap after a few minutes
bvernoux has quit [Quit: Leaving]
fibmod has joined #glasgow
Stormwind_mobile has quit [Ping timeout: 240 seconds]
Stormwind_mobile has joined #glasgow
Exec1N has quit [Ping timeout: 265 seconds]
<TD-Linux>
if a motherboard doesn't have a tpm header, what's the next best place to grab onto the lpc bus? superio chip?
<gruetzkopf>
yeah, and if it's antique enough lpc/fwh flash
<TD-Linux>
was thinking of making a flex pcb that solders onto a nuvoton superio chip. and then maybe a ffc on glasgisa
<TD-Linux>
hl, even for scopes there is some rivalry now. rigol scopes have unusable fft for example. so depending on what you are doing the siglent ones might be better