« first day (2519 days earlier)      last day (2713 days later) » 

00:29
@Asmyldof so basically it just lies to you all day every day?
@Skyler It involves firehoses and high pressure pumps
now im picturing you sucking in a birthday cake through a vacuum and spitting it out a hose
 
1 hour later…
02:00
@Asmyldof Now many containers?
02:18
hey there @NickAlexeev
 
2 hours later…
04:23
Go home AVX, you are drunk.
Their "1210" SMT tantalum is actually 138 thou by 110 thou typical
I would call that a 1411, not a 1210. Anyway, that just royally screwed me on a huge board order and I had to vent.
05:11
@NickAlexeev I suspect 2 full pallets
@PeterK On the one End Tantalums are notorious for their liberal sizing. On the other hand 1210 is one of those packages that, granted be it more often in width, regularly is flexible with exact sizes for no reason I know of.
 
2 hours later…
07:23
@Asmyldof woop. I bought a PSU.
 
3 hours later…
10:01
@BartekBanachewicz That's not whoop. Now it'll be much harder to bamboozle you into buying a fancy one
Which one?
10:45
@Asmyldof Korad ka3005p. It's kinda fancy when you think about it.
only for rather crude definitions of fancy...
kinda
I've concluded that I have too many things to possibly extend my workshop with to get just a PSU for the whole budget
like a new multimeter at some point
which reminds me of that I still wanted to buy and test that cheapo meter dave talked about...
I was reading through this post and comments and one said that, lemme
> It’s a pity thermal imaging cameras are so expensive – it’d be really good to have one looking at a circuit as it’s powered up.
I actually found a meter with such a camera recently, now I know what that was for
@Asmyldof also I might still free you from other lab equipment you might not need ;)
be careful, he wants money or favours in exchange for that
10:55
@PlasmaHH money is fine
you have somet hat you dont need? ^^
@PlasmaHH sort of.
11:10
let me compile you a list of things and where to send it to...
lol
@PlasmaHH it's just that I have a pool of imaginary money (earned on the internet) that I spend on random stuff
since it's earned on top of my wage my SO doesn't mind myself buying the strangest things with it :)
imaginary money? so you buy imaginary stuff with it?
@PlasmaHH e.g. games
11:36
@PlasmaHH and now I'm watching a "how to find a $50 scope on ebay" vid which also counts as imaginary :D
Anyone able to recommend a PCI card that does Profibus for a Windows PC and has a nice and straightforward C++ API?
windows 98?
lol no, Windows 7 Pro
@KyranF Can't say that I do, no
What about one that has a shitty API, like the SERCOS CIFX cards by Hilscher?
11:51
@KyranF National Instruments if you have the money
the CIFX cards by hilcher are about US$1k, and support profibus, but the main cost is SERCOS III I think. I'm hoping to find something around $500 that just handles profibus to PLC
damn I've been reading about scopes and now I want one
it seems that living my life without one has been a waste
and wow seems they really dropped in price
not much, in the last couple of years there has appeared a low price market that didn't exist before
a good one still costs as much as years before, its just that for much less money you get a usable one.
12:11
Might have a 500MHz turn of the century Agilent "left over" if I can reset it and calibrate properly * cough *
defective 500mhz scopes sell on ebay here for about the same price whether they are functional or broken :(
@PlasmaHH That's why I tried to win the broken one in the US
Still spensive though
Because Agilent
and shipping
But, I'm hopeful I can double it in value
Well, this is the 400kg lot of stuff, so the one 18kg Agilent wasn't going to be adding much cost
ah ok
12:20
Which is why I pre-bid on the 500MHz, the 1GHz and the 1.5GHz one, hoping to win one of the GHz and the MHz as backup and then none of the auctions pop-corned and I won them all
1GHz has been sold pre-emptively
Well, not guaranteed sold, but likely they'll have it
Keeping the 1.5GHz one
So probably going to have the 500MHz to either keep if the offers aren't high enough (who doesn't want 8 channels of Agilent in his lab?!), or sell (because, who really needs more than 4 channels of Agilent!?)
Hey, when you guys are troubleshooting a system, do you take notes constantly, or just load it all into headspace?
depends
@Giskard42 Depends. At the office there's so much hassle and interruption that I write even the tiniest thing down. Home lab I'm always alone and can follow a plan to the letter, so all the Scope images will also be in accurate order and I can always type the report weeks later without missing a detail
@Asmyldof lol
@PlasmaHH yeah that's what I meant. I've been looking at one of those newer Sigilent ones and it's kinda cool that it doubles as a bus analyzer
they brag I2C, SPI, UART, serial...
I suppose I'd use that much more than super high frequency stuff, but then again I have no idea what I mean
most seem to have it, and in most it doesnt seem to work that well...
e.g. in my rigol its next to unsuable
12:32
hmpfh
I've looked at Saleae before, but they seem to be kinda expensive for what they offer
it's just much less versatile than a scope
oh shit they also accept paypal
Just make an excel sheet, one row for each feature you can think about that might be useful to you and enter stuff for every candidate scope to get an overview
also think about features like a bandpass filter in the math functions ^^
@PlasmaHH well right now I'm thinking that scopes are mostly for analog stuff and digital probes are for digital stuff
this is a huge generalization obviously since they overlap to quite some extent
I would say that the best thing is to buy a scope for analogue and salae for digital stuff if you don't want to buy a big mso from agilent or so
@PlasmaHH IOW, both :D
If I had some spare mmoney I would probably try out if a tektronix mdo is useful
12:44
it's a pity Saleae charges $40 for shipping
but then again
12:58
oh hey I found a shop that ships bus pirates for free
well then
oh ugh no
13:14
@BartekBanachewicz get a fake salaeaeae from aliexpress
They work with the freely available sealaalalaaalaa software
Paid $29 including shipping for two of the simplest way back when
Or some such
13:41
@Asmyldof I think salamieae said they're going to lock down their software soon
Won't work with off-brands
@BartekBanachewicz I think you're going to be pretty disappointed with anything less than a DS1054Z
I had an off-brand Atten scope for a few years and the difference is pretty substantial for like $100
The salamiaeaes are semi-useful, but I haven't found them more useful than a bus pirate and a good Rigol analog scope
13:57
@Asmyldof mm
they seem to be going for around $8
14:11
@Giskard42 I have the sabbellabbedingdongs just to leave at customer sites. And I always keep installers of the last two versions of any software I use, so no worries
Hi there,

I am searching for a vertical RJ45 connector for a 1000Base-T application, like this one:
http://www.mouser.com/ds/2/706/fastjack-vertical-gigabit-520346.pdf
Circuit E, where all 4 center taps are seperated and with 0.01µF or 0.1µF connected to ground.
But this one is rather expensive. An additional transformer stage is not necessary.
Can you help me here?
I've seen a few from Wuerth Electronics but horizontal, are the vertical ones the rare ones?
@Asmyldof why would you leave them anywhere?
anyway, I just bought a $5 clone. We'll see how useful it is in a month
But I do use them myself, also @BartekBanachewicz, because they can record MANY more points than a Rigol and that's extremely useful for statistics
@BartekBanachewicz Often embed them into test equipment as acquisition boards
well if the chinese clone is good enough then I'll just need a scope without all those fancy decoding features
@Giskard42 oh I see
14:13
@BartekBanachewicz When you're a Freelancer and you leave a driver board you know works and it doesn't work with whatever a customer has made, you leave them some tools and say in a more friendly manner "figure it out, or pay me to"
99/100 times in my case it turns out I'm fully to their spec, but their spec isn't correctly written, or they actually didn't design to their own spec
Which is very much not a me-problem
@BartekBanachewicz Seriously, DS1054Z. It's worth it.
@Giskard42 €400, huh.
@Giskard42 Only when you need it for what you want to figure out
14:18
why would I need 4 channels?
@Asmyldof Wut
I worked the first 3 years of getting paid as a freelancer with a good and calibrated Fluke, second hand and a 20MHz analogue scope
@BartekBanachewicz You don't, but rigol doesn't make a 2 channel
Perfectly fine
Yeah, but how much did that run you?
14:19
yeah I've figured that the PSU is the most important part to get actually
I bought my 25mhz atten for $250
The 20MHz scope I found in a dumpster? I found it in a dumpster
2
@BartekBanachewicz I'd scour ebay for the psu
so that's what I ordered. I could use a new multimeter, a new soldering station, stuff
@Giskard42 that's ordered already. I went for the Korad.
Ah, good man
14:19
Second hand low-end Fluke back then was 80 euros or so
But one of those Fluke 17 ones from China is, what 110 now?
choosing multimeters is just a nightmare for a beginner btw
Okay, I guess I'm wrong, I thought digital scopes were more expensive
guess I'm behind the times
well if you get rid of USB then it's sorta ok
the functionality overlap between all of those is also pretty confusing
chosing anything is just tedious if you follow the excel sheet approach, nothing more
@Giskard42 What I'm saying is you do not need digital when you're starting. That's right nonsense
14:21
@PlasmaHH that requires you to know what you want
@BartekBanachewicz You can get GDM-8251 bench multimeters really cheap on ebay right now, there was a big liquidation
@BartekBanachewicz you better make your mind up about that then
@Giskard42 oooh, shiny. That's a sexy looking meter alright.
Any old analogue anyone is doing away with that works or that you can probably fix will do for the first year to even decade
@Asmyldof Maaaan, whaat?
14:22
@Giskard42 oh nice, 180 bucks and 220 bucks shipping... yeah
@PlasmaHH I found one for 275+20 so far
Snagged mine for like $150 +20
oh there you go 190+35
is that thing really more convenient than a "handheld" one?
I suppose being heavy and not moving around could be a feature
@BartekBanachewicz Hella precise, really high resolution, and RS-232
@BartekBanachewicz Depends on what you feel is convenient and what you expect to be doing in the next X months
14:24
@BartekBanachewicz customs and taxes included?
@PlasmaHH hmpfh. That's a bit random here.
I have never found something in the US that is worth it on ebay...
@Asmyldof so far I'd like a one that beeps
usually the shipping prices are ricioulous
mine doesn't beep in the continuity mode
14:26
This does, but the latency is atrocious
It's about a half-second on continuity mode before it beeps
lets all buy daves meter then ^^
@Giskard42 interesting, so it would seem that this one isn't that great after all :)
@BartekBanachewicz It's not perfect
@Giskard42 a huge amount of the "slow" continuity mode meters I found are only that because they have shitty probes. my cheap 3 buck ebay probes work much better than those that came with my agilent meter
@PlasmaHH Don't need to have 2 Keysights and 8 Flukes
14:32
@Asmyldof I mean the one he might one day sell
@PlasmaHH Ah yes, got myself some genuinely golden probes off China
@PlasmaHH I'm not that confident in the actual down under the hood quality of it all, if I may be so blunt
It would be intresting to see, since feature wise it has some noticeable things (15V diode mode, low burden voltage) but I would not be so sure if the software side of things is ok, and the long term stability...
@PlasmaHH He's still a bit... clunky in his path of assumptions and inferences in that really low noise or really high stability, I'd want to see the boards before I told anyone to go out and buy at the price driven by the claimed specs
@Asmyldof yeah, I often hope its just his wanting to be quick on camera and that when he sits down with enough time he gets it right. But what I think will happen is that many people will take that one apart and do suggestions of all kinds so that either the 1st revision can be "hacked" to be better, or the 2nd revision will be useful
@PlasmaHH He is not doing that just for the camera
14:46
@Asmyldof I mean he seems to want to get everything in one take and rushes through the stuff. I hope he takes more time off camera. If he rushes through the stuff in that same way off camera, then well... have you ever had a look at any of his products?
@PlasmaHH I have not. But I mean he does it in all his videos where he starts with assumption A, works through process B, gets to assumption C, which invalidates assumption A, and then goes through Process D on basis of assumption C, etc
Well, not all his videos
But the ones where he is fixing or designing something
Yeah, when he is fixing stuff he reminds me of some coworkers I had that had degrees in physics or chemistry or something like that and never really learned structured debugging and jumped from here to there without collecting data first and only then drawing conclusions
Or keeping an overview of the information gathered when making assumptions or decisions
And if you do that in such videos, you do that with everything
hm, maybe. I find it at times difficult to explain my approaches to people, I should do my own youtube channel to figure out if I would be the same on videos ^^
14:58
The comment section would be fun, people screaming all over the things that I get wrong
@Asmyldof Have you ever encountered a problem in which a SAM4S will enable the clock, waits for the rc oscillator to start, switches to the RC oscillator, but then hangs waiting for the clock source to start?
@Giskard42 Not on SAM4S no, there's an erratum which can cause something similar if you use the accurate RC into the PLL, but that only applies to the M0+ devices
Not related to M3 or M4 types at all
Huh. It's definitely a hardware problem, because one board revision prior works fine with the same code.
So weird
switch chips?
@PlasmaHH You mean I might have a bad batch of chips or something?
I've tested like five new boards and they all have the same problem
15:06
@Giskard42 who knows? at least you can rule them out that way.
@Giskard42 It's not unheard of in lower price assembly houses that reflow profiles are "guesstimated" rather than controlled
which package do you use? qfp or bga?
that shouldnt be too hard then to switch the chips, just for having it ruled out
Will do, brb
Nope
15:21
sot he new chip works fine on the old board (and/or vv?)
Old chip works fine on the old board, old chip doesn't start on the new board
good point to start from there.
I guess I would start at square 0 and check every pin for what its doing and what is the difference.
gnd vs. floating etc.
actually measure it
Yeah, probably a good idea.
stability of the power rails and stuff like that
thing is, like you might have goofed up somewhere and a pin is not conneted to where you thought it is
have seen that with quite a few gnd pins that went nowhere
anyways, have to catch a train to go home
@PlasmaHH Okay, thanks for the help
And asmyldof
15:39
they aren't even different boards
this is doing my gorram head in
They're the SAME revision
16:21
@Giskard42 Some other part has the wrong PN installed, maybe.
@ThePhoton ...It's a good idea, but A. I've checked the BOM, nothing's been changed, and B.
There's nothing I can think of that could possibly have this effect
The only external component involved is a watch crystal, but if I remove it on the old boards I get a totally different response to what I'm seeing
@Giskard42 Don't check the BOM, check the actual boards. Assembly shops do make mistakes. The machine might drop a part and they pick the wrong one up to replace it (because you didn't budget them spares to allow for this kind of loss). I've also seen distributors (well-known ones) ship mis-labeled parts.
@Giskard42 Wrong value R or C in your RC circuit...or cracked capacitor.
@ThePhoton It's an internal rc
in the chip
I'm powering just the chip, bypassing other components
Putting 3.3v on the rail
All that's in there is decoupling and the watch crystal
I've replaced the watch crystal
@Giskard42 mmm
@Giskard42 No floating reset/NMI/JTAG pins?
16:46
@ThePhoton Nope, I'm stepping through
(with JTAG)
loads code fine
just halts on wait_for_osc()
 
2 hours later…
19:07
@Giskard42 any idea if the pll successfully goes to 12mhz or if its still trying or so?
@PlasmaHH It just constantly waits for the pll to come up
I've filed a support ticket with atmel-cum-microchip, guess we'll see if they have any insight
does the datasheet contain details about how the pll functions exactly? you might be able to assert a few conditions
19:33
start with the cleanliness of the pll voltage ;)
19:58
welp, I incorporated the regulator into my test circuit, and sure enough it draws steady 5mA
but it's cool, I have my own dual voltage setup with the tiny running at 3V3 and the transistor fed straight from the power battery
at the same time I've discovered that the MCU on the damaged board is actually perfectly functional, so it must be something else that fried
I've actually emailed the manufacturer and asked for the schematic, but I don't hope for much
I could theoretically perform the tests with the power LED now but I think I might want to wait for the PSU
with some luck it should actually ship tomorrow
and arrive in 6 weeks
oh by "ship" I actually meant arrive, it's sent locally
and I ordered it to work which is in the middle of the city and a guaranteed pickup
anyway, I should probably (finally!) order stuff for PCBs as well
20:14
hm, in the 80s, have there been diodes (zeners possibly?) that use some body material that can either rust (brownish) or corrode (whitish)? I have on this corroded board a component that has a diode symbol under it but doesnt look like a diode at all
20:50
I'm confused. I programmed a microconroller to communicate over USB as a CDC device and then am listening to it on a hyper terminal. The code is not able to handle inputs yet, so if I right click into the terminal with my mouse, my CDC device will receive the click codes and decide to transmit an entire function from my code in plain text, even formatted in my style from the programming IDE.
I'm confused how this could even exist in the code and how it is transmitted in plain text to my terminal. Only thing I can guess is that somehow the compiler didn't compile the function but instead packed it into the machine code as plain text. Anyone ever see this before?
I also connected the USB code to the stdio printf
not sure how it could get into the printf buffer either
@EwokNightmares have you perhaps copied your source before?
Try doing Paste in some text editor.
yes that was it haha
thank you
That's a big relief and makes me feel silly lol
happens to the best of us :)
:40121628 I thought the terminal was transmitting the ASCII combinations of windows to represent mouse clicks
21:14
hm I think I'll do some analog measurements today as well, since I got the output stage more or less ready
 
1 hour later…
user228700
22:24
Hello, everyone :-) I was wondering if any one of you has the time to help me with an incredibly dumb question about electric power that has gone unanswered inside my mind for years now.
@Kaumudi.H Shoot!
* kills @Giskard42 *
user228700
Oh, thanks! Right, well, I am learning (for the umpteenth time) about electric power and have come across, of course, problems regarding the calculation of energy in kilo Watt hours.
user228700
My question is this: how is it that simply multiplying kW by h gives me kWh?
Energy flow over time multiplied by time gives you energy
If I understand your question :>
user228700
22:28
Yes, right, however, 1 kW of power signifies that 1000 Joules are being generated/consumed in one second, no?
user228700
1 second. How does it work, then, when we simply multiply this by h?
Ah, I see
user228700
Yes? :-)
One sec, just trying to work this out myself :>
user228700
22:30
:-P Sure.
Okay, just wanted to check weather my understanding was correct
user228700
Wokay.
One "kilowatt hour" is just the amount of energy that corresponds to 1 kilowatt/second for 3600 seconds
if that's a slightly better phrasing
user228700
Oh, crap, one second.
I find it very helpful to play around with WolframAlpha's conversion system
It does this kind of dimensional conversion very well
user228700
22:35
I'm back! Sorry about that. Oh, WolframAlpha, hmm.
user228700
@Giskard42 Where does this number 3600 disappear to when we simply multiply by h to get kWh? O_O
@Kaumudi.H h=3600
I think.
user228700
Hmm, under this definition, does this make sense:
user228700
user228700
To me, it doesn't :-/ That conversion unit, 3600 is nowhere to be found!
user228700
22:41
This is what I found in another source:
user228700
user228700
So...
I fell into the same trap as you, I think
user228700
:-(
user228700
I'm sorry, lol.
22:44
Ah, I think I got it
There are zero dimensions of time in 3.12kw
not /s not /h.
It's just instantaneous power draw.
So when we multiply it by hours, that's our unit of time
not seconds
user228700
@Giskard42 Huh, how is that?
kWh has unit equivalency to Joules
W = J/s
kWh = kW * h = 1000W * 3600s = 3600000J
@Asmyldof Dammit, now you've confused me even more
Hi, can all Karnaugh map simplifications be expressed as boolean algebra simplifications?
user228700
Ah, right, I think I understand. While converting W to kWh by multiplying by the number of hours for which the equipment in question consumed/generated $x$ kWs, the unit of kWh remains Joules $\times$ 1 hr/1 sec.
user228700
22:52
Is that correct?
user228700
I'm not particularly bothered about this unit when multiplying this number by, say, the cost of 1 kWh of electricity to pay my bills is all, yes?
Mainly for my own understanding:
1kw is 1 joule/second
1kw/s is 1 joule/s/s
1kw/h is 3.6x10^6 joules
user228700
@Giskard42 Shouldn't that be 1 $kW\times h$?
@Kaumudi.H AAh, yes
I'm an idiot
1kw/h is joules /s/s
user228700
@Giskard42 As am I, don't worry :-P
user228700
23:00
In any case, thanks, y'all! :-)

« first day (2519 days earlier)      last day (2713 days later) »