// #eevblog — 2021-06-14
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
00:00:37 <rust_collector> just snip the other end, lol
00:00:42 <rust_collector> leave it in there
00:01:12 Join: TheCranston ([email protected]) to #EEVblog
00:01:43 <Lemon> well sometimes its not possible
00:01:48 <Lemon> also, heres the linesman I like https://www.amazon.fr/02-225-universelle-d%C3%A9multiplication-atramentis%C3%A9e/dp/B000CDSQEW/ref=sr_1_1?__mk_fr_FR=%C3%85M%C3%85%C5%BD%C3%95%C3%91&dchild=1&keywords=KNIPEX+Pince+universelle+%C3%A0+forte+d%C3%A9multiplication+%28225+mm%29+02+01+225&qid=1623592880&sr=8-1
00:01:50 <EEVBot> [Title] -
00:01:54 <Lemon> its quite stronk
00:03:18 Quit: Kuschelpony ([email protected]) Connection closed
00:05:53 <rust_collector> I find it is very rare that I pick up one of those to do anything. I wish they removed the cutting part of those, and made the jaws a little bit shorter
00:07:39 Join: modrobert ([email protected]) to #EEVblog
00:09:09 <TheCranston> Greetings. I'm trying to block DC bias into an audio isolation transformer. I've got a capacitor in-line with the signal and then a resistor to the other input effectively creating a high pass filter. Is there an equation or how does one account for the impedance of the transformer itself in that filter?
00:10:47 <hsn> https://visiontoreality.theredsea.sa/sub_domain_pages/sheybarah_en
00:10:50 <EEVBot> [Title] - Breaking New Ground
00:10:52 <hsn> there is money in this
00:10:53 <hsn> :D
00:11:07 <T3sl4> TheCranston: yes, the transformer's inductance acts in parallel with the resistor, in fact you don't need a resistor because of this
00:11:23 <coppice> Breaking New Wind
00:11:32 <rust_collector> hsn that is a killa design, lol
00:11:58 <hsn> rust_collector, yeah that saudi ppl are crazy
00:12:07 <rust_collector> god I hate sites like that
00:12:13 <T3sl4> set C = L / Rsrc^2, give or take
00:12:20 <TheCranston> T3sl4: so just use the nominal impedance from the transformer in the filter equation to figure out the right capacitor?
00:12:26 <coppice> hsn: looks like the set for a dystopian sci-fi movie
00:12:36 <T3sl4> at least one of source or load impedance must be a resistance, otherwise the filter response will be wrong
00:12:50 <coppice> resistance is futile
00:13:01 <T3sl4> the transformer's impedance is not its inductance, no
00:13:14 <T3sl4> src as in source
00:13:47 <TheCranston> what's L in the above equation, then?
00:13:59 <hsn> coppice, another big one would be https://www.neom.com/en-us
00:14:01 <EEVBot> [Title] - NEOM: an accelerator of human progress
00:14:02 <T3sl4> magnetizing inductance
00:14:30 <TheCranston> oh boy.. where do I get that value? this is the transformer I'm using: https://www.hammfg.com/electronics/transformers/audio/101.pdf
00:14:50 <hsn> shybara is part of neom
00:15:38 <T3sl4> measure it, or email them i guess
00:16:22 <coppice> hsn: most of things are clearly someone's idea of cool, but end up looking like the set for a dystopian movie like Elysium
00:17:09 <hsn> :D
00:17:47 <T3sl4> it should be safe to use a larger value i think, say C = 5 / (2 pi Fc Rsrc)
00:17:54 <coppice> hsn: many of them are never finished, or fall apart quickly
00:18:16 <hsn> coppice, despite that .. there is alot of money in that :D
00:18:21 <hsn> its like milking a cow ..
00:18:37 <hsn> milk saudi ppl .. take money .. bye bye
00:18:39 <hsn> :D
00:18:44 * T3sl4 milks hsn
00:18:57 * hsn gives T3sl4 a coffee
00:19:08 <hsn> full of fat
00:19:29 <coppice> hsn: of course. the goal is not to build, but to grab cash.
00:19:35 <TheCranston> ok. the Fc being the cutoff and Rsrc the resistance from the transformer? so 600 ohms?
00:21:06 <T3sl4> at least one of source or load resistance, or both
00:21:36 <T3sl4> Fc is the signal cutoff frequency, probably lower than the 300Hz on the transformer
00:21:49 <T3sl4> which is only -0.5dB, Fc is measured at -3
00:23:18 <TheCranston> got it. sorry for the all the baby step questions.
00:24:40 <hsn> lol what a good point as an ad https://snipboard.io/QWXRf1.jpg
00:25:40 <TheCranston> I'd like to hit Fc at 100Hz given the signal coming thru is I/Q from an SDR radio... Really just trying to stop DC from flowing thru the transformer. had no idea that it was DC biased when I started on this.
00:26:29 <T3sl4> that's ok. filter circuits are nontrivial. they aren't at all impossible, but it takes quite some experience to look at a thing and say, this component has to be such-and-such; or to measure unknown elements and design with them (like the transformer)
00:27:01 <hsn> I/Q = ?
00:27:07 <TheCranston> non-trivial is the fact..
00:27:42 <T3sl4> hsn, that's a complex question
00:28:10 <hsn> T3sl4, IQ modulation in radio stuff?
00:28:30 <T3sl4> i would assume so
00:28:38 <TheCranston> hsn: the quadrature signal. https://www.pe0sat.vgnet.nl/sdr/iq-data-explained/
00:28:40 <EEVBot> [Title] -
00:28:41 <TheCranston> yup, radio
00:30:55 <TheCranston> SDR radio provides access to the I/Q signals so I can use them for grater dynamic range on an external waterfall display. Radio only displays about 40KHz of the bandwidth but the radio is doing about 150KHz internally. My gear tops out at 96KHz for sample rate.
00:31:43 <TheCranston> but need to isolate the two parts (radio and interface) and rid ourselves of the the DC bias from the radio into the transformer.
00:32:44 <TheCranston> Technically I need to do this on both sides of the transformer as the input device on the computer assumes it's a microphone and provides 2.7v DC bias... argh.
00:34:10 <T3sl4> so uh
00:34:17 <T3sl4> is this SDR not actually on the same ground as the PC
00:35:42 <TheCranston> correct
00:36:10 <TheCranston> found it... 288mH inductance.. https://www.hammfg.com/files/parts/pdf/101F.pdf
00:38:05 <T3sl4> yay
00:40:13 <coppice> Blocking DC in an IQ signal is such a sad thing to do. One of the nice things about a system designed for an IQ receiver is being about to abandon the need for the coding to DC balance
00:40:53 <T3sl4> it does rather defeat the purpose of direct detection
00:41:39 <TheCranston> mfg. put a pair of pull-up resistors on the outputs giving me the bias.. not sure what else to do..
00:43:52 <TheCranston> so.. if I did the maths correctly, I'm looking at a .764uF cap to set the filter Fc at 100Hz. no additional resistor, just relying on the transformer. that sound about right? C = .288 / ( 2 pi 100 600)
00:45:50 <T3sl4> .288?
00:46:10 <electrohead> 3.14?
00:46:20 * electrohead yawns and runs
00:46:24 * T3sl4 banana cream pies electro in the head
00:46:34 <TheCranston> .288 is 288mH, right? that was the inductance from the spec sheet
00:47:36 <T3sl4> 5 is not L?
00:47:59 <TheCranston> oh. oops.
00:48:49 <electrohead> T3sl4: I'm just waking the hell up >.<
00:49:25 <T3sl4> EEVBot: give coffee to electrohead
00:49:25 * EEVBot pours coffee down electrohead's throat
00:49:42 <T3sl4> https://i.imgur.com/oY5qzMs.jpg
00:49:47 * TheCranston getting a crash course in filters.. still working on first cup 'o coffee
00:59:46 <TheCranston> so the earlier equation, then C = .288 / 600^2 gives 0.8uF.. with the 5 you were just giving me a safe adjustment of the 1/(2 pi Fc Rsrc) equation, correct?
00:59:55 * T3sl4 Brians up TheCranston a bit
01:00:16 <T3sl4> yes
01:00:33 <TheCranston> yes... thank you! :)
01:00:55 <T3sl4> which is 4uF... the other way says 13uF
01:01:03 <T3sl4> somewhere in there should be fine
01:02:32 * TheCranston ponders my order of magnitude math error...
01:03:07 <TheCranston> ok. so split the difference and 8uF is *right* there.. I have some of those in the lab.
01:23:43 <Lemon> hey Ampera, do you know how much could I get with 4 x 8gb ddr4 2133Mhz ecc sticks ?
01:23:51 <Ampera> for, money?
01:23:52 <Lemon> prices on ebay seems variable
01:23:55 <Lemon> yeah
01:24:21 <Ampera> well, as you said, the price is quite variable
01:24:33 <Ampera> do research, find what you think to be a fair price, and counter that with what you want for it
01:24:46 <Ampera> you're in an entirely different market than me, so I have no clue
01:24:55 <Dude> buh
01:25:01 <Ampera> personally I'd pay maybe 150 - 200USD or so
01:25:03 <Dude> matlab's equation solver is trash
01:25:07 <Ampera> but that's an out of my ass number
01:26:06 <Lemon> thats also the price I thought so you confirm it
01:26:08 <Lemon> thanks buddy
02:03:16 Join: Kuschelpony ([email protected]) to #EEVblog
02:13:24 Join: chivesncheese ([email protected]) to #EEVblog
02:13:32 Quit: chivesncheese ([email protected]) Quit: Leaving
02:23:37 Part: TheCranston ([email protected]) from #EEVblog
02:32:17 <Kuschelpony> good evening
02:49:30 <Kuschelpony> hmm, i wonder if some chip designer was also a player of "snakes" :) https://www.richis-lab.de/Bipolar11.htm
02:49:30 <EEVBot> [Title] - TIP2955 Leistungstransistor
03:34:55 <T3sl4> dont think they had that back then lol
03:36:39 <Kuschelpony> hmm, i think that game is available since the 80s, was definitv part of an 486SX-25 PC package and not new then
03:40:59 <T3sl4> 3055/2955 go back to the 60s, though those epitaxials are likely 80s
03:41:27 <T3sl4> i don't remember how old that game is, maybe there was a terminal version in the 70s?
03:41:41 <Kuschelpony> these specific ones from ST with the snake pattern are more like from the 2000s
03:42:11 <Kuschelpony> i can say for sure that this game wqritten in BASIC running under DOS was not new when that 486 was bought
03:44:02 <coppice> the 3055 goes back to the day the first silicon was diffused
03:44:09 <DHess_US> terminal version?
03:44:29 <DHess_US> Oh, computer terminal, not wire terminal
03:44:39 <Kuschelpony> looool
03:45:27 <DHess_US> 3055 goes back to single diffusion mesa with 200 kHz Ft and a square SOA.
03:46:29 <Kuschelpony> im too lazy now, but i might look back to find my databook from i think 1965
03:46:41 <DHess_US> Tektronix had some instruments with three different versions of the 2n3055 inside.
03:50:07 <T3sl4> DHess_US: to be fair, it wouldn't be hard to wire from relay logic
03:53:45 <Kuschelpony> you gave me a mental image, playing snakes in relay logic might work even with closed eyes :)
03:59:50 <DHess_US> Bummer, Tektronix doesn't give vendor names for the transistors they used, only part numbers.
04:01:01 <hsn> Using different vendor on your own risk?
04:01:21 <Kuschelpony> dhess, i totally forgot to say i played last sleepless night a bit with discretze I2L (plus 10 Ohms ballast resistors), did work not bad
04:01:29 <DHess_US> Tektronix qualified parts by vendor, which makes sense because I've had to do the same sometimes.
04:02:19 <DHess_US> So they had a Tektronix part number for a given IC, and a list of suitable Vendors and their part numbers for that part.
04:10:51 <DHess_US> So for instance the 156-0511-00 could be the RC4531NB from Raytheon, or NE531N from Signetics. But the -01 had to be the NE531V from Signetics, and the -02 had to be the NE531N-IIB from Signetics.
04:11:28 <Kuschelpony> ok, interesting to learn
04:19:29 Join: Guest69 ([email protected]) to #EEVblog
04:32:05 Quit: Guest69 ([email protected]) Quit: My MacBook has gone to sleep. ZZZzzz…
Search this day's buffer: