AustNet AustNet Archive Project — IRC Logs
« All channels | Contribute a log »

// #eevblog — 2026-06-16

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
00:09:42 <Kuschelpony> would be easier if i would had used them ever
00:55:33 Join: steve30 ([email protected]) to #EEVblog
00:55:35 Mode: (by ChanOP) +o steve30
01:07:23 <MrMobodies> Steve30, what milk do you like with your coffee?
01:07:42 <steve30> whole
01:08:14 <MrMobodies> how about the ones in the glass bottles that haven't been pasterized?
01:08:33 <MrMobodies> I found one in Waitrose I found them the best
01:09:26 <MrMobodies> this was it
01:09:47 <MrMobodies> https://www.waitrose.com/ecom/products/the-estate-dairy-unhomogenised-organic-whole-milk/762999-1-2
01:10:18 <MrMobodies> really complements the cofee
01:20:10 <steve30> its fine if you shake it up
02:42:31 Join: blastsnoop104892acb ([email protected]) to #EEVblog
02:44:11 <blastsnoop104892acb> any EE's on here? there's a public spat between two company EE's which could do with further thoughts. https://www.reddit.com/r/bikewrench/comments/1rkw96c/son_ladelux_dynamo_hub_drag/
02:45:30 <blastsnoop104892acb> It's a German and UK company that make high-end bicycle/usb charging products for dynamo hubs (for bicycles). Lots of heat on charging small li-ions at 2C, reactance and others
02:50:44 <blastsnoop104892acb> high-end light/usb charging products
02:57:23 <Kuschelpony> whats your question??
03:10:46 <rc_garage> he asked if there are any EE's here
03:11:21 <blastsnoop104892acb> not so much a question it's to correlate what's being said on the thread
03:12:42 <blastsnoop104892acb> there are a number of points raised such as whether charging a 1200mA small lithium at 2C is acceptable. Whether correcting inductive reactance without capacitance is possible.
03:13:51 <blastsnoop104892acb> supplying 1A current limit to a USB using BC 1.2 profile (1.5A) and whether the EE is right to suggest a UVLO is needed
03:15:18 <blastsnoop104892acb> the overvoltage handling of one product, whether its a good idea or not
03:19:37 <Kuschelpony> i can comment at least about the reactance thing:
03:19:59 <blastsnoop104892acb> if you know your stuff please do
03:20:26 <Kuschelpony> a capacitor is the most simple thing but can tune the alternator inductivity only at a specific rotation speed, below this speed the power output drops entirely instead of gradually
03:21:35 <Kuschelpony> it woud be possible to engineer an circuit (based on an Class-D amplifier) that presents itself as a negative inductivity that compensates the positive inductivity of the alternator at any speed, and at the same time works as active rectifier
03:22:57 <Kuschelpony> maybe someone who loves tinkering and overengineering has already done this, but i strongly suspect that it gives any meaningful gain over a simple passive rectifier plus maximum power point tracking circuit (easier)
03:23:48 <Kuschelpony> also i thinkpeople will dislike it when the bike draws considerably more than 3W from their pedaing power at higher speed
03:23:50 <blastsnoop104892acb> wouldn't it need an inductor roughly the equivalent size, modulated by the PWM?
03:25:35 <Kuschelpony> no, the inductor could bemuch smaller, lookat the 5*10*10mm cubes on (true) 50W TPA3118 boards, this one inductor only needs to deal with the switching frequency of the Class-D chip but the energy storage for the virtual negative inductivity is the capacitor smoothing the DC bus anyway
03:28:02 <blastsnoop104892acb> what's the theory behind how it works? If an inductor and frequency produce a positive reactance how do you get back to zero phase?
03:29:33 <blastsnoop104892acb> and how would that work with active rectification, one isolated and inverted?
03:31:09 <Kuschelpony> letssee it as a series circuit
03:31:35 <Kuschelpony> there is the EMF inside the alternator and its inductivity in series
03:32:34 <Kuschelpony> the inductivity obviously chokes the current (side comment, this is exactly what makes the classic light bulb system workable at all)
03:33:11 <blastsnoop104892acb> yep get that
03:33:47 <Kuschelpony> now the electronics senses the current that flows and feeds back an voltage that is 90° shiftet relative to this current
03:34:23 <Kuschelpony> this appears as a negative inductivity or a capacity that is magically matching resonance at any speed
03:35:05 <Kuschelpony> because with 90° phase shift the electronics delivers purely reactive power and therefore zero average DC side current
03:35:38 <Kuschelpony> in this case, current would only be limited by the resistance of the alternator wire
03:37:19 <blastsnoop104892acb> I undersatand the theory but not the practical implementation. If the phase is shifting (frequency is changing) how would it know how much negative inductive equiv to produce?
03:40:52 <blastsnoop104892acb> I've been trying to find an example circuit or chip that can do it but so far unable
03:41:34 <Kuschelpony> outside loop is a regulator for the voltage of the DC bus or the tapped off current to charge battery, this is injected as phase shift set point into a inner loop with an PLL controlling an oscilator
03:42:37 <Kuschelpony> mmh i think we can draw up easily an example circuit, but i strongly doubt there is any readymade chip, because as i mentioned above a much simpler approach would give almost the same performance
03:43:02 <blastsnoop104892acb> simpler being capacitors?
03:43:34 <blastsnoop104892acb> but bike speed changes so the frequency is volatile. Here it may have an advantage
03:44:49 <Kuschelpony> simpler as in no tuning at all, just rectifying and running an maximum power point tracker (well known from photovoltaics) to extract the most power from the simple rectifier
03:45:37 <blastsnoop104892acb> simpler but less power, 25-30% less
03:46:22 <Kuschelpony> beyond something 10km/h a bike alternator is a constant current source and you can raise the voltage and therefore extracted power practicaly linearly with speed
03:46:56 <Kuschelpony> btw: if it has no clamping diodes, you can reach beyond 50V at 40 to 50km/h
03:47:03 <blastsnoop104892acb> yep but power factor displacement is still a limitation
03:47:14 <Kuschelpony> right, i do not disagree
03:48:12 <blastsnoop104892acb> I've been scouring google and I can not find a single example of negative impedance with active rectification
03:48:24 <blastsnoop104892acb> not saying it doesn't exist, wondering if I'm searching correctly
03:48:36 <Kuschelpony> active rectification is just a by-product of the class-D amplifier
03:48:50 <Kuschelpony> like a variable frequency drive in regeneration
03:49:51 <Kuschelpony> we are talking about something like a (synchronous machine) variable frequency drive set to maximum regeneration
03:52:17 <blastsnoop104892acb> If it's possible to make a scrap block diagram or something that would be awesome, it's new to me
03:53:11 <Kuschelpony> im happy to do tomorrow (its getting closer to bedtime here)
03:53:34 <Kuschelpony> maybe someone else in the chat has further ideas to explain
03:53:36 <blastsnoop104892acb> very cool, I'll come back on
03:56:34 <blastsnoop104892acb> can you see anything on google which is close?
03:57:24 <Kuschelpony> oh i dont use google, but might do some duckduckgo searches tomorrow
03:57:48 <Kuschelpony> (sorry dont want to kill this conversation but i should be on the train at 5 tomorrow)
04:00:58 <blastsnoop104892acb> no thanks for your help
04:01:07 <blastsnoop104892acb> much appreciated love learning
04:02:06 <Kuschelpony> thats one of the reasons we have the chat here
04:02:32 <Kuschelpony> and sorry that we ive on different continents with different timezones, goodnight
04:02:40 <blastsnoop104892acb> goodnight
04:02:42 Quit: Kuschelpony ([email protected]) Quit: Ex-Chat
05:04:12 Quit: blastsnoop104892acb ([email protected]) Ping timeout: 2 minutes
11:00:38 Quit: DogWomble ([email protected]) Connection closed
11:11:39 Join: DogWomble ([email protected]) to #EEVblog
13:33:28 <MrMobodies> Well that is nice
13:33:29 <MrMobodies> https://news.sky.com/story/trumps-iran-war-is-one-of-the-greatest-strategic-blunders-in-us-history-13554425
13:33:30 <EEVBot> [url] Access Denied | news.sky.com
13:33:44 <MrMobodies> I hope the energy prices can now start falling instead of going up.
13:34:22 <MrMobodies> "After a war that has cost an estimated $30bn, killed thousands and destroyed hundreds of millions of dollars worth of US military hardware"
13:34:42 <MrMobodies> I didn't know it could run up that much in what appears to me a short while.
13:36:17 <MrMobodies> one step to avoiding a world war 3
13:40:44 <MrMobodies> https://www.pulsetoday.co.uk/news/special-investigations/making-complaints-work/patients-using-ai-to-send-legal-letters-to-gp-practices-says-indemnity-provider/
13:40:46 <EEVBot> [url] Patients using AI to send legal letters to GP practices, says indemnity provider - Pulse Today | www.pulsetoday.co.uk
13:41:23 <MrMobodies> Please to see them getting a taste of their own medicine. Talking of the ones that switched to an AI chatbot ONLY...
16:06:44 Quit: ibiza ([email protected]) austnet.org!NickOP (Kill requested by neo)
16:06:50 Join: ibiza ([email protected]) to #EEVblog
16:50:22 Quit: blacklotus89 ([email protected]) Ping timeout: 2 minutes
16:50:29 Join: blacklotus89 ([email protected]) to #EEVblog
17:01:40 Join: blastsnoop104892acb ([email protected]) to #EEVblog
17:37:46 Quit: steve30 ([email protected]) Quit: <Baljem> steve30: by the way, I had the most bizarre dream the other day that someone came in here and seriously flipped their shit over your quit message and why aren't the ops /doing/ something about it
18:53:14 <blastsnoop104892acb> Hallo, waitinf ro Kuschelpony to come back on. I was discussing with him about phase correction in an inductive circuit using negative impedance rather than capacitance. Anyone came across it before?
19:05:03 <Ysjoelfir> Mr. Schmusepferdchen will be back soon I guess, he's here daily, usually. And, he's a cool guy.
20:40:57 <electrohead> !coffee
20:40:58 * EEVBot pours coffee down electrohead's throat
20:42:40 <DogWomble> !coffee electrohead with a side serving of being slapped about a bit with a rainbow trout with Monty Python's Life Of Brian playing in the background
20:42:40 * EEVBot slides a cup of joe to electrohead with a side serving of being slapped about a bit with a rainbow trout with Monty Python's Life Of Brian playing in the background
20:44:12 <electrohead> lol
20:45:11 * DogWomble stickso n an episode of invader zim
20:46:17 <electrohead> Think I found a mole in the chat
20:46:22 Kick: ibiza was kicked by electrohead (electrohead)
20:50:24 <electrohead> > Kill requested by neo
20:50:33 <electrohead> That's the part that stands out like a sore thumb
20:50:35 <electrohead> heh
22:06:18 * Ysjoelfir starts digging to find more moles
23:18:25 Quit: krish2487 ([email protected]) Ping timeout: 2 minutes
23:18:29 <electrohead> I'm just curious if said person rejoins and etc
23:18:45 <electrohead> Anywho, happy tuesday and things!
23:19:22 <electrohead> !caffeine Ysjoelfir
23:19:22 * EEVBot forces caffeine into Ysjoelfir