--- Log opened Tue Dec 15 00:00:48 2020
00:06 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-igvorzqzoqcebntp] has joined ##hplusroadmap
02:15 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-igvorzqzoqcebntp] has quit [Quit: Connection closed for inactivity]
03:13 < fltrz> lol bought myself a bunch of flexible tubing, and 10 earbuds with mics of one brand, another 7 of another and another one
03:14 < fltrz> forgot to buy the attenuating earpluga...
03:15 < fltrz> acoustics experiments yay
03:15 < fltrz> going to try to measure otoacoustic emissions
03:16 < fltrz> yesterday I had a totally unrelated idea regarding op-amp TIA's...
03:17 < fltrz> not sure if anyone here is into transimpedance amplifiers?
03:20 < fltrz> theres this annoying trade-off between high amplification, but with higher thermal noise or less amplification but more noise... then realized I can replace the feedback resistor with the ADC, then connect nothing else to Vout
03:20 < fltrz> use the input impedance of the ADC as op-amp directly... I'm amazed I haven't seen this yet
03:20 < fltrz> *use the input impedance of ADC as feedback resistor directly,... man I'm tired from the factory
03:21 < fltrz> you remove a component, and remove its thermal noise
03:22 < fltrz> its very weird because my proposal does not connect the ADC analog ground pin to ground but to virtual ground of the op-amp TIA
03:23 < archels> this uh, may or may not work
03:23 < fltrz> archels, should I draw a circuit, are you're pretty sure what I'm suggesting?
03:24 < fltrz> archels, assuming you see the same schematic in your minds eye, I guess you agree that such a topology is extremely unconventional right?
03:24 < fltrz> but it makes perfect sense, the opamp output voltage must be the signal current through the ADC input impedance
03:25 < fltrz> I suspect there will of course be equivalent input voltage noise of the ADC and this is probably identical to its sampling noise?
03:26 < fltrz> if this works, it seems like everybody blindly copy pasta'd the same basic op-amp amplifier schematics from each other
03:26 < archels> ADC ground goes to op-amp inverting input, ADC signal in goes to op-amp output?
03:27 < fltrz> archels, yes
03:27 < archels> what ADC are you using?
03:27 < fltrz> archels, and ADC positive input is connected to normal ground
03:27 < fltrz> archels, it's just an idea, I didn't try it out yet :)
03:28 < fltrz> my mind went tripping balls for at least an hour afterwards, i swear
03:28 < fltrz> archels, do you agree that at least superficially it makes sense and could work?
03:29 < archels> the assumption is that the current is the same between ADC signal in and ADC AGND, right?
03:29 < fltrz> my proposal asks you to effectively ignore the ADC datasheets
03:29 < fltrz> archels, yes, otherwise lots of ADC circuits would be fucked
03:30 < fltrz> and the ADC wouldn't be able to be modeled as a resistor. what does input impedance of an ADC even mean if the incoming and outgoing current don't match?
03:32 < archels> but where does the assumption come from that you can model an ADC as a resistor?
03:33 < archels> I'm not an expert, but you typically have to deal with rapid impedance changes while the ADC is sampling your signal
03:34 < archels> see e.g. the MCP3204 datasheet, figure 4-1 https://ww1.microchip.com/downloads/en/DeviceDoc/21298e.pdf
03:34 < archels> this is why I asked what ADC (or what ADC architecture) you're using, like sigma-delta, SAR or a simple resistor chain
03:34 < fltrz> for example https://www.planetanalog.com/adc-guide-part-13-input-impedance/ its basic essentially
03:36 < fltrz> yes the quantization, samplingm ... noise will be there
03:36 < archels> this article that you linked literally says "Unbuffered ADCs have a switched capacitor-based input", and "the input impedance of an ADC is not purely resistive"
03:37 < archels> so, I hate to ruin your trip here, but...
03:37 < fltrz> with low pass (which you want anyway for nyquist limit) the ADC input impedance will be low pass filtered
03:39 < fltrz> when you quote switched capacitor, thats exactly the same as "sampling noise" I mention above
03:43 < archels> I wonder if AGND will stay near 0 V if you tie it to the op-amp input like that
03:43 < archels> I expect it will not, because the connection to the op-amp is high impedance, whereas the AGND pin of course expects to be connected to a low impedance
03:43 < fltrz> sure, for the same reason as with a real resistor
03:43 < fltrz> archels, impedance is not per pin, but across pair of pins
03:48 < fltrz> suppose the amplification is -1000, and suppose initially the virtual ground and outpout voltage are 0V and there is some unknown voltage across the ADC input/ground, now suppose a disturbance causes a 1mV rise in the virtual ground, then this fall in output voltage will bring the virtual ground voltage down and steer it back to 0V
03:51 < fltrz> archels, lol, by the same reasoning we could put earphones instead of the ADC
03:51 < archels> once you're inserting that low-pass filter to counter the sampling noise, aren't you just inserting that resistor that you were trying to eliminate back into the circuit?
03:51 < fltrz> archels, low pass filter can be made with L's and C's
03:53 < fltrz> archels, I take it from your skepticism that you too have never seen such a circuit?
03:53 < fltrz> perhaps we cracked ADC voodoo magic
03:57 < archels> do tell how it works out
03:58 < fltrz> archels, I was thinking of perhaps downloading a bunch of SPICE models for each type of ADC, then simulate what happens with some test signals?
04:41 -!- Codaraxis_ [~Codaraxis@91.193.4.10] has quit [Read error: Connection reset by peer]
05:29 < kanzure> concurrency control and snapshot isolation https://github.com/rxin/db-readings/pull/16/files/f07aa75bc2ac2b5301b86b7d38af27be96fe10c9
07:58 < fenn> eddie hall explains how he accomplished the world record 500kg deadlift https://www.youtube.com/watch?v=mnZ4ftlifik
08:12 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-bgcnvunukkohsgnw] has joined ##hplusroadmap
10:17 -!- Urchin[emacs] [~user@unaffiliated/urchin] has quit [Ping timeout: 256 seconds]
11:15 -!- CryptoDavid [uid14990@gateway/web/irccloud.com/x-bgcnvunukkohsgnw] has quit [Quit: Connection closed for inactivity]
12:00 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds]
12:22 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
12:43 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has quit [Ping timeout: 272 seconds]
13:48 -!- spaceangel [~spaceange@ip-89-177-48-202.net.upcbroadband.cz] has joined ##hplusroadmap
13:55 -!- darsie [~kvirc@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 240 seconds]
14:33 -!- Llamamoe [~Llamagedd@178235180200.dynamic-4-waw-k-4-0-0.vectranet.pl] has quit [Quit: Leaving.]
15:24 -!- spaceangel [~spaceange@ip-89-177-48-202.net.upcbroadband.cz] has quit [Remote host closed the connection]
15:37 < kanzure> fenn: actually 501kg is the record https://www.youtube.com/watch?v=4e8azCO3seg
16:53 -!- shawwwn [uid6132@gateway/web/irccloud.com/x-lsyitowicalkgkjz] has joined ##hplusroadmap
17:35 -!- Urchin[emacs] [~user@unaffiliated/urchin] has joined ##hplusroadmap
18:36 < lsneff> Turns out jlcpcb's smt assembly service has all the parts I need for my boards, so might use that. Nice how easy this stuff is these days
19:56 -!- justanotheruser [~justanoth@unaffiliated/justanotheruser] has joined ##hplusroadmap
20:05 < fenn> lifting half a ton is pushing the limits of humanity; lifting 501kg is one-upping another human
20:33 -!- shawwwn [uid6132@gateway/web/irccloud.com/x-lsyitowicalkgkjz] has quit [Quit: Connection closed for inactivity]
21:18  * fltrz visualizes yet another ant or group of ants dragging yet another joint
22:39 < lsneff> .t https://sci-hub.scihubtw.tw/10.1049/el.2013.3815
22:39 < saxo> Sci-Hub | X7R dielectric multilayer ceramic capacitors show good micro-actuating properties with little hysteresis | 10.1049/el.2013.3815
23:05 -!- darsie [~kvirc@84-113-55-200.cable.dynamic.surfer.at] has joined ##hplusroadmap
23:11 -!- Llamamoe [~Llamagedd@178235180200.dynamic-4-waw-k-4-0-0.vectranet.pl] has joined ##hplusroadmap
--- Log closed Wed Dec 16 00:00:47 2020