silly power supply

Until recently, you could get the very pretty, very quiet LM329DZ for way under a buck. Then TI discontinued them, and now AD's second-source version is $5.

(Yes, I stocked up before that happened, but the 329 was an important part of my bag of tricks for getting high performance at very low cost.)

Cheers

Phil Hobbs

DP270 is an apropos adhesive. My best guess - and it's only a guess - is nitric acid will soften it. Unfortunately, my experience with nitric acid is limited to college chemistry classes. Danke,

Supertex DN2530 is a sot-89 higher current depletion fet, Idss 200 ma min, and it has an LTSpice model.

I don't understand what you are doing. Got a schematic?

What controls the LED currents?

We used TALEMA SMJ-140B but it's expensive and 39 weeks lead time now. It was most cool but it's time to design it out. We do have 7000 in stock! Given that, I could still use them on the current project.

We have got equivalents from Amgis too.

Pink plastic antistat bags. My life is centered around Uline blue plastic bins. Each project spins off a bin, so project-specific protos go there, with one more bin for generic prototypes. Most are small. A folder on a server has a sub-folder for each proto, with schematic, photos, and notes. Some protos are just little bench tests, some are dremeled, some are actual PC board layouts.

We (in theory) aren't allowed in the stock room so have to formally request parts, so we have a lot of sample kits for engineering, and I always request extras when I do the official request, and I keep them in coin envelopes, again in blue bins.

I like the coin envelopes because I can scribble notes on them.

---------/R <-> Z 10V9\-------

+12V-G <-> Z 10V8---- GND

---------\B <-> Z 9V1/-------------

in PCB 1.1" long (11 pins edge connector) and 0.5" wide (or as small as possible).

The LEDs will blow up.

How? I got some running for a few days now.

Actually, the zeners should blow first, but positive temp. coeff. helps. As it heat up, voltage increases. So far, i don't have too many blowing, except for wrong wirings.

On a sunny day (Wed, 28 Dec 2022 07:53:05 -0800) it happened John Larkin snipped-for-privacy@highlandSNIPMEtechnology.com wrote in snipped-for-privacy@4ax.com:

Oh I see!

Those are nice bins and give some protection to those proto boards indeed.

We had a sign in the huge spare parts department in broadcasting, something like: "We have everything, the impossible takes a little longer" Since it was almost 24/7 (worked in shifts) we made notes in a book about what we did for the next shift to keep things running...

In the design times I stored things and data per product. Sometimes I find a little box I build here and wonder: "What was this?" OTOH I have thrown a lot of big stuff I desigend and build away, especially when moving house (several times), still many boxes full of electronics are everywhere. Wonder where it all will go if I am no more, some market perhaps :-) I have also donated stuff.

Nothing limits the currents. Crank the voltage up a bit.

And there's no reason to buy precision zeners since the LED forward drops are in the threshold equation.

The 12V is shunt regulated with a 5W zener shunt, or 50W TL432 shunt temporary. If the zeners blow, then likely have to replace the 12V shunt. 5W is good enough when the battery packs are balanced. 50W if seriously unbalanced, until the other catch up.

It just runs 50% duty, RC osc style. So only deviations in RDSon from the high to low side results in asymetric drive. But that doesn't matter, since the coupling capacitor takes care of that.

I do too. But in the present IC, there is no regulation, only feedforward. So no stability issues to worry about

For the half bridge you have 2 transistors. Which coupling are you talking about with 1?

In the old days they always used 2. An offline DCDC transformer (push pull style, Royer converter)

Zeners are sometimes bad, sometimes good :-)

On a sunny day (Wed, 28 Dec 2022 22:17:46 +0100) it happened Klaus Vestergaard Kragelund snipped-for-privacy@hotmail.com wrote in <toibpo$11qi$ snipped-for-privacy@gioia.aioe.org>:

The 20uF I think I just would go about it totally different, but would need more info like load current and other requirements, Microchip PICs are my standard tool:

Maybe mine are older?

Sometimes I use LEDs as zener added to a real zener to get the right voltage, like in my PIC programmer...

PIC18F14K22 has build in reference, ADC multiple analog inputs, 2 hardware comparators, PWM out, serial UART, logical outputs, DAC, 64 MHz PLL, internal oscillator or external xtal, EEPROM for calibration if you need more precision etc etc.

Can make anything I like with it.. Seen my frequency counter?

scope? Of course I have a RTL_SDR sticks too for the RF work: Bit of software, no bloat, few parts...

Oldest PC I still use every day runs Linux, has a PCI DVB-S satellite card, programs PICs via the par port, plays movies, does internet, now almost 20 years old. Was a top mobo those days, AMD. New is now Raspberry PI 4 8GB ARM. Quiet, cheap, uses hardly any power, has a nice IO GPIO. This laptop I use to type this is a Samsung from 2011 running Slackware Linux with of course my own Usenet reader:

and a lot of other programs I wrote.

Its all too easy, do not understand that Microsoft bloat that has nothing extra to give except requiring you to buy ever more expensive hardware to run their crap. Is so insecure and crap written that security is impossible to guarantee and forces people to 'updates' on a regular basis. They grab every bit of your emails on their servers not even encrypted for US to spy on you!!! Don't get me started! Snake oil it is!!

Here's the current version.

Seems fine. The DQ127 dual-winding inductor is running cool at above expected power out and at 100 KHz.