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
<Degi> Huh hydrogen annealed Fe can get 200k
juli964 has quit [Quit: Nettalk6 - www.ntalk.de]
juli964 has joined #scopehal
Degi has quit [Ping timeout: 260 seconds]
Degi has joined #scopehal
<_whitenotifier-9> [starshipraider] tpterovtt forked the repository - https://git.io/Jfc8f
futarisIRCcloud has quit [Quit: Connection closed for inactivity]
noopwafel has quit [Ping timeout: 256 seconds]
noopwafel has joined #scopehal
apo has quit [Ping timeout: 256 seconds]
apo has joined #scopehal
bvernoux has joined #scopehal
<azonenberg> bvernoux: my splitter came in, and the resistors
<bvernoux> ha great
<azonenberg> going to rework the frontend after work and see how the gain looks
<bvernoux> MiniCircuits splitter ?
<azonenberg> yeah
<azonenberg> its surprisingly small
<bvernoux> and the resistors are 75Ohm RF ?
<bvernoux> yes it is small
<azonenberg> no the resistors are 93.1 ohm to replace R13 and R14 in the frontend
<azonenberg> we had 100's there but there's some output impedance in the amp we didnt take into account that is causing gain error
<bvernoux> its a RF resistor too ?
<azonenberg> Nothing fancy, this is for the 100 MHz AFE
<bvernoux> ha ok
<bvernoux> I was thinking it was for the Probe
<bvernoux> anyway you are lucky as I suspect you receive things in 1 day in USA especially from DigiKey/Mouser which are in USA
<bvernoux> In France I prefer to buy to Farnell now as things come from EU most of the time so faster ...
<bvernoux> But anyway some things can be found only @DigiKey ...
<bvernoux> azonenberg, also I have feedback from Darrell about my measurement ;)
<bvernoux> and He confirm that OSHPark FR408HR are not so good for RF
<bvernoux> especially for GCPW as it seems really better for microstrip
<bvernoux> so far I have not done any microstrip test as it often take too much space vs GCPW ...
<azonenberg> well is it fr408 or the enig thats the problem?
<azonenberg> i would imagine there's lots of losses in the nickel layer too
<bvernoux> it seems the most problematic is ENIG ;)
<bvernoux> hmm the KC908 2nd batch will be even better than 1st one ;)
<bvernoux> they are redisigning the RF board to go up to 22GHz ;)
<bvernoux> screenshot of work in progress in 2nd batch https://www.kechuang.org/r/312174
<bvernoux> there is 2 channels in fact which act as dual receiver or transmitter
<bvernoux> and the new batch will have a CPU working at 125°C ;) (previous one was 100°C)
<Degi> azonenberg: https://i.imgur.com/NuH8B8g.png is this good
<Degi> I got an oven now
<Degi> The korona one
<azonenberg> Degi: looks a little noisy up toward the top, but should work
<Degi> The noise is the thermostat
<Degi> And maybe the fan
<Degi> Hmm how about I do a trial run
<bvernoux> critical part is when exceeding 180°C(SnPb)/217°C(Pb-free) ...
<bvernoux> which shall be between 60s to 150s
<bvernoux> with peak to 230°C/260°C during 10s to 30s
<bvernoux> on your curve the peak duration is about 200s
<bvernoux> I hope it is SnPb as it will not work for Pb-free
<Degi> Hm pb free
<Degi> Meh ill mod the furnace to remove the thermostat
<Degi> Whats the best way to plug syringe tips
<bvernoux> it seems your top temperature will be not very good for pb-free
<bvernoux> especially peak
<bvernoux> but it could work but for sensible parts like BGA ...
<bvernoux> also preheat is not respected
<bvernoux> and ramp down is clearly too slow
<bvernoux> you shall open the door with fan ;)
<Degi> oki
<bvernoux> it shall be very fast
<bvernoux> check NXP pdf it is quite clear
<Degi> Datasheet says 30 sec between 249 and 217
<Degi> It says 6 °C maximum per sec
<Degi> Is the ramp up ok or should i preheat longer
<Degi> meh should be fine
<bvernoux> the worst is probably it is too long at 240°C
<bvernoux> 200s
<bvernoux> is clearly too long
<Degi> Yes ill just turn it off once it reaches 249 °C
<bvernoux> I will recommend SnPb reflow ;)
<bvernoux> easier and less dangerous as less heat
<bvernoux> even if the curve is not very good results are often not too bad for prototypes
<bvernoux> with my plancha i do it basically ;)
<bvernoux> I preheat
<bvernoux> then check temperature up to 240°C (wait 10s) and I remove the power ;)
<bvernoux> Anyway results are not consistent but sometime good enough for simple prototypes
<bvernoux> with 1 board out of 8 with soldering defect ;)
<bvernoux> I do not speak about some qfn as they often fail (solder bridges...) when not done correctly
<bvernoux> It also depends on the solder paste used as it is very important
<bvernoux> with bad solder paste some component jump ;)
<bvernoux> anyway Pb-free is tricky even if we should all use that as it is not toxic vs SnPb it is hard to work with when you do not have good oven ...
bvernoux has quit [Quit: Leaving]
* monochroma whispers the dreaded phrase "Tin Whiskers" and slinks back into the darkness
<Degi> Not with 3% Ag
smkz is now known as __builtin_trap
futarisIRCcloud has joined #scopehal
<Degi> Huh
<Degi> Apparenthy the temperature reading is a bit below
<Degi> Judging by the smell of the pcb
<Degi> https://i.imgur.com/PYkdDi5.png Too hot, pcb got kinda smoky, tin got dark
<Degi> I hope too rapid cooling isnt bad