X-Scale has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Would you like to know more?]
X-Scale has joined ##openfpga
user10032 has joined ##openfpga
m_t has joined ##openfpga
eduardo_ has joined ##openfpga
eduardo__ has quit [Ping timeout: 248 seconds]
Bike has joined ##openfpga
rohitksingh_wor1 has quit [Read error: Connection reset by peer]
rohitksingh has joined ##openfpga
DrLuke__ has joined ##openfpga
DrLuke has quit [Ping timeout: 256 seconds]
noobineer has joined ##openfpga
<pointfree>
rqou: how's everything on Zonenberg Island
<pointfree>
OT: Can anyone recommend an (air) CO2 scrubber? I'm converting the 60sqft storage locker next to my apartment into labspace. I've got a power outlet in there but no windows or ventilation and I don't want hypoxia.
<reportingsjr>
that sounds really sketchy
<pointfree>
right now it's filled to the top with Forth programming language books and magazines but I'm moving them to another storage locker which is less full.
<reportingsjr>
That would also be very energy intensive
<reportingsjr>
unless you have the growth tanks outside which.. at that point just ventilate the air straight outside.
<Zorix>
i thought scrubbers used lime or something
<Zorix>
id go with what the international space station uses.. that is proven
<pointfree>
Zorix: hm interesting.
<pointfree>
I guess I'll also be writing a driver for an air quality sensor + alarm to my forth system.
<Bike>
can you not just leave the door open? or maybe cut a vent?
<pointfree>
Bike: I just checked and there is a rubber draft stopper filling the gap between the door and the floor. Maybe I could run a tube under the door and use a fan.
<Zorix>
scary
<whitequark>
awygle: god damn it
rohitksingh has quit [Quit: Leaving.]
<whitequark>
so let's say I have an external device swithcing 16 channels simultaneously
<whitequark>
such as a GBC displaying a sprite
<whitequark>
the FXMAs will draw, what, 35*8=560mA?
<whitequark>
anyway this actually causes enough noise that on the opposite side of the board, the FPGA misreads the flag coming from the FX2
<whitequark>
it's not even in the same IO bank!
<reportingsjr>
hello filtering components
<reportingsjr>
whitequark: Is the noise fairly high frequency?
<whitequark>
dunno, I have no test points there
<whitequark>
I think slew rate limiting resistors will fix it
<reportingsjr>
fair enough
<whitequark>
awygle: actually come to think of it.
<whitequark>
can you open a case with ON Semi about slew rate limiting and FXMA108?
<whitequark>
this has to be troublesome enough they'd know it
<whitequark>
this chip has been nothing but a nightmare
<openfpga-github>
Glasgow/master a89f5a2 whitequark: Rename FIFO parameter early_in to streaming and fully disable pktend.
<openfpga-github>
Glasgow/master 8f572d8 whitequark: Fix data corruption writing last byte available in an FX2 IN FIFO.
<openfpga-github>
Glasgow/master e6c3390 whitequark: Add early_in parameter to get_in_fifo to disable automatic OUTPKTEND....
rohitksingh has joined ##openfpga
rohitksingh has quit [Quit: Leaving.]
<Bike>
"every step seemed locally optimal at the time" is an eight word horror story
<whitequark>
it's THE archetypical horror sotry
genii has joined ##openfpga
Lord_Nightmare has quit [Ping timeout: 252 seconds]
Lord_Nightmare2 has joined ##openfpga
Lord_Nightmare2 is now known as Lord_Nightmare
<reportingsjr>
^^
noobineer has quit [Remote host closed the connection]
<qu1j0t3>
whitequark+++++++++
<reportingsjr>
argghh, does anyone else seriously struggle with SMPS designs?
<Zorix>
just usually smps repairs
<reportingsjr>
I decided to use an IC adafruit uses in some of their lipo battery boards. Sounded like a decent chip and I'd be running it at like 60% of its max spec. Well, the internal switch pins kept shorting to ground. Searched and found other people with this issue.. I put the caps way too far away from the switch pin which was causing spikes on that pin and killing them. :(
<reportingsjr>
now that I compare my design to adafruits mine looks like hot garbage.
<reportingsjr>
Out of the like 10-15 boards I've designed with SMPS only about 5 of those have worked reasonably well.
<rqou>
i struggled with smps design a lot while taking power electronics :P
<Ultrasauce>
inductance in the wrong place is a killer, same with motor controllers
<reportingsjr>
yeah, it feels like I just haven't done enough intense design to really understand the current loops in SMPSes and so I keep making bad designs.
<reportingsjr>
Unfortunately I am cheap and when this happens I get discouraged from spenidng so much on respins/parts and drop projects before I find the correct solutions.. << dumb thinking on my part I suppose.
<rqou>
hmm usually I've never heard of this type of problem
<rqou>
people usually add clamp circuits for this
<rqou>
what type of smps are you building
<tnt>
most chips actually have a recommended layout in their datasheet. If you follow that, it'll work.
<rqou>
unless you pick a shitty inductor and saturate it
<tnt>
well most chips I ever used also have a 'design guide' that will tell you exactly how to pick a part :p
<tnt>
I mean if it's an IC that adafruit uses, it's not going to be a 50A 1V core supply for a i7 ...
<rqou>
well, at least they're not legally prohibited(?) from doing so unlike SFE :P
<rqou>
but yeah, multi-phase smps design is even more difficult
<reportingsjr>
tnt: true, but some times you want things to be arranged differently from te recommended layout.
<tnt>
reportingsjr: why ? Usually having my design to work is pretty high on the priority list.
<tnt>
tbh, I'd really be curious one of your failing layout.