// #eevblog — 2026-01-17
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
00:58:20 <Rhine_Labs> Yesterdays Tektronix Surplus Haul. https://www.stevenrhine.com/wp-content/uploads/2026/01/Rhine-Labs-Tek-Surplus-Score-2026-Crop.jpg
01:03:08 <K> nice
01:03:20 <K> condition?
01:04:07 <Rhine_Labs> Monitor works but for sure needs service. https://www.stevenrhine.com/wp-content/uploads/2026/01/RHINE-LABS-Tektronix-655HR-1-Monitor.jpg
01:04:23 <Rhine_Labs> Wave For Monitor Works... PSU Unsure.
01:05:43 <K> oh it is color
01:06:14 <K> to my mind the HV psu is most interesting piece there
01:06:53 <K> well, technically - all 3 devices have HV PSU in them eh
01:35:54 <Rhine_Labs> Yea, getting good results from the x-ray machine. https://www.stevenrhine.com/wp-content/uploads/2026/01/Data-IO-Unipack-2B-351B-070B-Adapter-X-ray-By-Rhine-Labs-scaled.jpg
01:36:15 <Rhine_Labs> Another 40KV PSU in that..
01:36:45 <K> ooh niceee
01:36:55 <K> xray machine would be on my ultimate list
01:37:55 <K> well, is, not would be.
01:38:11 <K> but i do not see it happening
02:23:51 Join: Kuschelpony ([email protected]) to #EEVblog
02:34:13 <Kuschelpony> good weekend
02:34:48 <K> hello Kuschelpony
02:34:56 <K> is do cold over there?
02:35:02 <Kuschelpony> and hello K
02:35:19 <Kuschelpony> no absolutely not, today we had spring-like sunny 12°C
02:35:36 <K> QUE?!
02:35:41 <Kuschelpony> today and tomorrow are expected to be the most mild days of this januarx
02:35:51 <Kuschelpony> what do you mean?
02:36:01 <K> wait, you are in Germany, right?
02:36:07 <Kuschelpony> yes i am
02:36:31 <K> +12C is hard to believe for me rn, we had like -12C
02:37:23 <K> ah, indeed, you have positive temperatrues over there
02:37:47 <K> we have snow, ice and missery
02:39:33 <arch> Kuschelpony, are you familiar with hybrid switched-mode + linear stage power supplies? i am trying to understand how this creates the tracked switched rail: https://www.analog.com/en/resources/evaluation-hardware-and-software/evaluation-boards-kits/dc2132a.html
02:40:05 <Kuschelpony> not specifically but im sure i can find it out :) let me check in about ten minutes
02:40:11 <arch> here is the schematic: https://www.analog.com/media/en/technical-documentation/eval-board-schematic/DC2132A-1-SCH.PDF i am looking at the U3 chip
02:41:14 <arch> the feedback pin (24) appears to be "buffering"(?) the actual output voltage from Vout 0-24V, and it must be having the switched output rail be some voltage above (maybe 3V-5V headroom)
02:44:06 <arch> jeeesus...fuck.... what is up with linear ICs being $14/pop. ffs
02:44:16 <arch> this is why china is winning
02:46:33 <Kuschelpony> nice circuit and easy to understand on the first look
02:47:05 <Kuschelpony> its all about
02:47:11 <Kuschelpony> Q2 and the resistors around
02:49:02 <arch> ok, let's say v_out = 1V. U3's output past the L1 inductor is 1V+(some voltage). i looked at its datasheet and it has a feedback voltage of 0.97V
02:49:22 <Kuschelpony> ou understand it best this way to see Q2 as a current source, emitter is at th e input voltage of the linear regulator, base is derived from the output of the linear regulator
02:49:44 <Kuschelpony> now calculate backwards
02:50:18 <Kuschelpony> to get 0,97v at r18 needs slightly less than 200µA
02:51:05 <Kuschelpony> so you need to get 1/5th of this voltage or about 190mV across the emitter resistor of Q2
02:51:52 <Kuschelpony> add about 0,6V base-emitter-voltage and you need about 0,8V at the base relative to the linear regulator input
02:52:42 <Kuschelpony> loo at the divider of R38 and R21 which halv the voltage difference across the linear regulator
02:53:02 <Kuschelpony> so you get about 1,6V across the linear regulator
02:53:19 <arch> ahhh ok, so let's imagine R21 did not exist--it had no feedback. by default, switching would continue until that fb voltage was seen. at that point, Q2 would source current from the output, thus creating voltage drop across R18, which when it reached 0.97V, it would deactivate the switching?
02:53:32 <Kuschelpony> yes
02:54:07 <arch> ahh, ok. that simplifies things for me greatly. that's pretty cool. you explained it very well about focusing on just the Q2 and surrounding resistors. thanks!
02:54:22 <Kuschelpony> think about the output of the linear regulator has a fixed voltage, and the switcher rises the input of the linear regulator as long until there is so much current throught the transistor until the switcher feedback is 0,97v
02:54:29 <Kuschelpony> sure!
02:54:38 <Kuschelpony> as said its a quite easy circuit to spot
02:55:16 <arch> analog devices have some really excellent compilation smps appnotes
02:55:48 <Kuschelpony> tobe honest i never looked at them, i design my circuits from first principles
02:56:30 <arch> \o/
02:56:30 <EEVBot> |
02:56:30 <EEVBot> / }
03:04:39 <arch> i think i will build something simple just so i can see its performance on the scope. i was thinking of using a fake ebay lm2596 (which switches at ~50kHz, rather than actually 150kHz like real one). have its output to a LM7805, and see how it performs.
03:04:56 <arch> err not 7805, i mean the adjustable version
03:06:09 <arch> it would really hammer in that concept building it for real and allow me to play with the compensation (?) network they have there with R46, C30, C13, and R19
03:06:22 <Kuschelpony> :)
03:06:37 <Kuschelpony> you mean the adjustable LM317
03:06:51 <arch> yup, i have plenty of LM317 in TO-220 package
03:07:19 <Kuschelpony> a well behaved and forgiving chip
03:07:45 <Kuschelpony> just make sure you always drain at least 5mA or the output voltage will rise uncontrolled
03:08:32 <arch> ahh, good point. i did wonder why many of these cb radio power supplies with lm723, etc, have a load resistor on the output
03:09:45 <Kuschelpony> hmmm with the LM723 the resistor should not be needed, its a special detail of adjustable 3-pin regulators
03:10:37 <arch> i have a lot of projects going on right now, including a honda gx120 engine that needs some TLC, and an old leaf blower carburetor that needs TLC, and an old generator that needs a replacement needle and seat on the carburetor...etc etc... so it might be a while until i get to it
03:10:54 <Kuschelpony> (in the adjustables like the LM317 the internal supply current flows out of the output pin and the adjust pin has high resistance, in fixed regulators like the 7805 the internal supply current flows out of the ground pin and there is no minimum load on the output
03:11:09 <Kuschelpony> no problem, im here almost any evening
03:11:14 <arch> but i noted down your output Kuschelpony so i can reread and reanalyze if needed
03:11:20 <Kuschelpony> my next project, a switcher for a CRT
03:11:32 <Kuschelpony> no problem, its also fine to ask agin
03:11:35 <arch> crt! well, at least free heat in the winter!
03:12:58 <Kuschelpony> ah no, the scope CRT need very little power
03:13:32 <Kuschelpony> 2W filament, some ten uA cathode current at 1,5KV
03:14:16 <Kuschelpony> deflection amplifiers about 4* 1mA idling current at 250V (this one needs little bandwidth)
03:16:16 <arch> is it for a DIY scope? or it's a TV or monitor?
03:18:19 <Kuschelpony> none of all, but monitor probably the closest, a receiver for Feldhell transmissions
03:19:45 <Kuschelpony> basically something in between TV and Fax
03:20:55 <arch> oo
03:23:57 Join: xstevia ([email protected]) to #EEVblog
03:58:24 <T3sl4> https://www.reddit.com/r/ExtraFabulousComics/comments/1qehu2l/give_a_man_a_fish/
03:58:48 <arch> T3sl4, https://www.reddit.com/media?url=https%3A%2F%2Fpreview.redd.it%2Fgive-a-man-a-fish-v0-0wz0s478iqdg1.jpeg%3Fwidth%3D1200%26format%3Dpjpg%26auto%3Dwebp%26s%3D33244fc5f8867a55c13273fd776393ff1ea3c431
04:01:47 <T3sl4> the comments are the real treasure
04:56:32 Join: watcher ([email protected]) to #EEVblog
04:59:39 Quit: watcher ([email protected]) Quit: KVIrc 5.2.6 Quasar http://www.kvirc.net/
05:13:37 Quit: Kuschelpony ([email protected]) Connection closed
05:35:38 Quit: xstevia ([email protected]) Quit: Your SaaS (Steve as a Service) subscription has expired.
06:06:01 <arch> https://i.imgflip.com/ahkg5q.jpg
06:12:03 <K> i was asked to prove my humanity
06:12:17 <K> typical americans.
06:18:03 <arch> typical highly volatile metal
06:35:24 <K> typical!
10:00:17 <Halcyon> Mornings
10:32:50 <K> oy.
10:40:30 * arch flicks a crusty booger at K
11:36:01 <Ampera> hai electrohead
11:36:03 <Ampera> how ya doin
11:36:16 <T3sl4> evenin Ampera
11:43:03 * electrohead touches himself
11:43:16 * T3sl4 swats at electrohead
19:21:07 <K> better living with chemistry!
19:33:28 Quit: watcher ([email protected]) Client exited
19:56:10 <K> https://creepylink.com/
19:56:11 <EEVBot> [url] CreepyLink | creepylink.com
20:21:14 * electrohead potats
20:24:25 <K> no card?
20:29:12 <electrohead> not yet, my man
20:29:14 <electrohead> :/
20:32:38 <K> *angry noises*
22:45:40 * electrohead adds to the angry noises
Search this day's buffer: