Swapk_ has quit [Remote host closed the connection]
electronic_eel has quit [Ping timeout: 265 seconds]
electronic_eel_ has joined #glasgow
Getorix has joined #glasgow
Getorix_ has quit [Ping timeout: 256 seconds]
PyroPeter_ has joined #glasgow
PyroPeter has quit [Ping timeout: 265 seconds]
PyroPeter_ is now known as PyroPeter
<_whitenotifier-3>
[glasgow] whitequark closed pull request #198: applet.interface.uart: allow to invert the UART line signals (=idle low) - https://git.io/JJd99
<_whitenotifier-3>
[GlasgowEmbedded/glasgow] whitequark pushed 1 commit to master [+0/-0/±2] https://git.io/JJFZH
<_whitenotifier-3>
[GlasgowEmbedded/glasgow] electroniceel 02b15ba - applet.interface.uart: allow to invert the UART line signals (=idle low)
<_whitenotifier-3>
[glasgow] whitequark commented on pull request #198: applet.interface.uart: allow to invert the UART line signals (=idle low) - https://git.io/JJFZQ
<whitequark>
seems fine, it's even exactly what the vendor says to do
electronic_eel_ is now known as electronic_eel
<electronic_eel>
esden: ah, a different part with dedicated manual reset
<electronic_eel>
on a cursory glance over the datasheet I see no problems with this solution
<electronic_eel>
the 4 pin sot-143 is a bit unusual, I have seen this just used for rf transistors before
<electronic_eel>
but it seems to be common enough for theses reset circuits, I found a few alternatives
<d1b2>
<esden> @whitequark hehe... yeah @DX-MON found the first version of the part
<d1b2>
<esden> @electronic_eel and yeah there are a few alternatives. The first match was a Maxim part, so we searched for a different vendor and settled on this. 😄
<d1b2>
<esden> the first try was to add some pullup/pulldown resistors and other strange stuff. I think the "correct part for the job" seems like the cleanest solution at the end. 🙂
<tnt>
esden: I'm a bit confused by the role of the diode and capacitor with very weak pull ups.
<d1b2>
<esden> It says there in text comment "Keep CY_RESET active for >5 msec after 3.3v switches on"
<electronic_eel>
esden: ...especially if the correct part for the job is about the same price as the alternatives
<tnt>
(and yeah, I can read the comment but that APX chip has a built-in 100 ms delay so the 5 ms to charge cap are irrelevant ?)
<d1b2>
<esden> yeah but that also keeps the 3v3 disabled
<electronic_eel>
the 5ms count after 3v3 has risen
<tnt>
Oh .. I see.
<apo>
oh, I didn't think esden's campaign was actually going to happen in <soon>
<apo>
I guess if that's the case I won't have to bother Attie :)
sorear has quit [Ping timeout: 244 seconds]
_florent_ has quit [Read error: Connection reset by peer]
_florent_ has joined #glasgow
sorear has joined #glasgow
bvernoux has joined #glasgow
<d1b2>
<Attie> @apo ha, I'm wondering how many might drop out because of that, but I'm still keen to go ahead, now I've got the ball rolling
<d1b2>
<Attie> did you contact me via twitter / did I put you on the list?
<apo>
Attie: Yeah, I did - hadn't seen that you're here, so I disabled my usual redirect to nitter and logged in (I'm lvboltzmann)
ali_as has joined #glasgow
<d1b2>
<Attie> ah ok - want me to remove you from my list?
<apo>
I'm unsure =P
<apo>
gotta figure out the diff between C1/C2 - if in doubt I can always donate the C1 to my hackspace
<d1b2>
<Attie> ha ok - let me know your email and I'll keep you informed (twitter dm is fine)
<noopwafel>
C2 will have the INA233, allowing current sensing, usb-c connector, and use parts/footprints which are much less painful/expensive to assemble. also stop button?
<d1b2>
<Attie> stop button sounds neat 🙂
<d1b2>
<Attie> otherwise, i believe so, yes
<apo>
Hmm
<apo>
current sensing and stop button sound good
kbingham has joined #glasgow
SwapK has joined #glasgow
<electronic_eel>
esden: about the new reset ic and the reset button: I suggest to add a 100nF cap to gnd on the mr input on the reset ic
<d1b2>
<esden> @electronic_eel for debounce I guess?
<electronic_eel>
esden: for two reasons: (1) minimum current through the switch. there is no spec for the mr pullup on this part, but when you look at the ds for '811 ics from other manufacturers, they usually have 20k pullup
<electronic_eel>
that is just µa and that is usually not enough to burn away oxidations from the contacts
<whitequark>
TIL you have to design for this
<electronic_eel>
and (2) noise immunity. there is nothing about this in the ds from diodes. but look for example at the one for the max811 or the stm811
<electronic_eel>
whitequark: switches usually have a _minimum_ current rating.
<electronic_eel>
if you don't take this into account, the switches become unreliable over time
<whitequark>
electronic_eel: yea it makes total sense
<whitequark>
i just didn't know it
<electronic_eel>
esden: debounce isn't necessary for these reset ics, they usually have an internal circuit for this
<d1b2>
<esden> @electronic_eel ahhh ok, that makes a lot of sense... TIL 😄
<d1b2>
<esden> @whitequark we are both learning together 😄
<d1b2>
<esden> @electronic_eel ok I will add the cap 🙂
<electronic_eel>
and I got a lot to learn about HDL from both of you, so in the end we all come out wiser
<d1b2>
<Attie> in things like relays that are energised over long periods of time, you can also find the varnish migrating from the coil onto the contacts
<ebb>
So the cap would allow a little inrush to shake off the oxide?
<d1b2>
<Attie> can cause nasty issues - e.g: crackling in audio - anywhere between intermittent operation and full on failure to operate
<d1b2>
<Attie> that's the plan, yeah
<d1b2>
<esden> @Attie ohh "wetting current" is a nice term!
<electronic_eel>
is that term from the very old times when you had wet mercury on the contacts?
<electronic_eel>
at least for high quality reed relais wet mercury was used for quite some time to reduce contact bounce and contact degradation due to oxidization
<d1b2>
<Attie> i'm not sure where it comes from, but that sounds plausible
<d1b2>
<Attie> it also "feels" somewhat appropriate - dry / crusty contacts won't work properly, but wet ones will...?
<d1b2>
<Kia> i imagine theres probably a regime in terms of switched voltage/current where you can get by with just inert gas / gold on the contacts because oxidation isnt enough of an issue and not need to wet the contacts with mercury and not need to "wet" the contacts with voltage/current
<electronic_eel>
Kia: as far as I understood it, you can just build reliable contacts for a current range, always min/max. there is always a minimum current needed for the wetting process. but with inert gasses and the right contact metallurgy you can extend this range quite a lot
<electronic_eel>
but if you go beyond a certain point it becomes quite expensive to further improve. so adding just a cap on the input is a lot cheaper than a more refined button
<d1b2>
<Attie> i think it's also important to remember that contact pressure is a big part of the equation - physically pressing the button "quite hard" will often suffice, but a gentle user / lightly actuated sensor / relay with gentle springs / etc... maybe not
<d1b2>
<Attie> as electronic_eel says, a simple cap will do wonders for the result
<electronic_eel>
but you also have to be careful not to overdo it. if you'd put a 1000µF low-esr polymer cap over the button, the high discharge current will burn away the contacts