1 transistor negative resistance circuit Now working Attention Winfield Hill :-)

No fair, that's two active parts! I mean, hey, I'm a big fan of dual transistors in tiny packages, can those count as one part? They're often used as if they were.

On a sunny day (8 Jul 2019 14:32:33 -0700) it happened Winfield Hill wrote in :

OK, maybe I should have written 'one amplifying part', it is just ONE transistor with Ib controlled by light, not 'trickety.

Na, that is 2 amplifying parts :-)

SUPER! Rather inventive, i should say. One active part. ..now.. If i cheat like all hell, Zero active parts and only one part total,,,,an Esaki diode.

There are those things they use in light dimmers, trigistors and something else.

DIACs. I have some, after one of the profs came, said he wanted to repair his dimmer, asked if we had any. Embarrassed, I had to say no, but I quickly ordered some.

It appears the LED in the opto is wired backwards, so it will never light up.

Jon

No. For low voltages R6 and R2 supply the LED current and R7 is switched in the circuit, loading V1. For higher voltages D2 and D1 draw the current away from the LED, U1 goes off and R7 'disappears' from the circuit.

For even higher voltages the reverse voltage over the LED will destroy it, which is also a negative effect.

Arie

On a sunny day (Tue, 9 Jul 2019 21:02:39 +0200) it happened Arie de Muynck wrote in :

Indeed, and for even higher voltages:

They're used in CFL ballasts too, to generate the initial trigger pulse that starts the self-resonant oscillator running

Sort of reminds me of this:

Yeah, and OTAs are just as simple as other (transistor, tube) items with transconductance. It's just that, like a Darlington with two transistors, an OTA has...twelve.

You can make a macroscopic UJT by wiring an LED to a voltage divider made with CdS photocells, and pointing the LED at the bottom one. (Latching time in the 10us range.)

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website:

On a sunny day (Tue, 09 Jul 2019 12:59:43 -0700) it happened John Larkin wrote in :

Yes, the idea of a fuses occurred to me, those that automatically go back to normal forgot what those are called.

For a one time effect any fuse will do :-)

Fuses are slow, and your circuit has many more transistors.

LDR and light bulb and 2 zeners may perhaps work too, also slow.

On a sunny day (Tue, 9 Jul 2019 19:28:51 -0500) it happened "Tim Williams" wrote in :

Ah yes, I remember seeing that circuit here long ago. I still have some real UJTs from ebay around.

Fusing is a chronic problem for us. Many of our products run off a DC wall-wart, and a user can potentially give us the wrong voltage or even polarity.

A real fuse needs to be replaced. Polyfuses are horribly inaccurate. e-fuses are IMHO flakey. Polyzens are available in limited ranges. A polyfuse and a transzorb can be teased.

We generally use a poly and a transzorb and go for 24 volts in, assuming that warts over 24 volts are rare.

I'm doing some new gadgets that need 3.3 volts internally. I'll run them from a USB supply, with a USB connector, and assume the input will never exceed 5 volts, and never get reversed. Maybe have a poly just in case.

On 7/9/19 1:25 PM, Winfield Hill wrote:> John Larkin wrote... >> >> There are those things they use in light >> dimmers, trigistors and something else. > > DIACs. I have some, after one of the profs > came, said he wanted to repair his dimmer, > asked if we had any. Embarrassed, I had to > say no, but I quickly ordered some. > > Programmable unijunctions too. (Last time I used one of those was in about 1990, in the interlock circuit of a diode laser controller.)

Cheers

Phil Hobbs

On a sunny day (Wed, 10 Jul 2019 06:50:52 -0700) it happened John Larkin wrote in :

Ah! That was the word, 'polyfuse', never used one.

I use a diode against reverse voltage and a LM2596-3.3 for 3.3V and add a resistor for higher output voltages, that LM2596 because I got a lot of those from ebay for little.

Most transformer based wallwarts are short circuit protected with a thermal fuse, switcher types usually will also somehow switch off and / or keep trying periodically in case of overload.

PC USB will switch off power above about 600 mA IIRC, I do have some Raspberry Pi USB wallwarts that can do a lot more...

I did blow up a WiFi booster by plugging in a 12V wallwart while it only wanted 5V.. Most stuff I have labeled for voltage, not that one..

Just careless.. nothing helps against that :-)

The old thyristor and zener crowbar setup should work too?

When somebody plugs in the wrong supply, you have no idea what is on the other end of that cord. I sure wouldn't short it to ground unless there was a fuse in between--it might start a fire someplace.

The SCR crowbar thing really needs something better than a zener to drive the gate, because as it turns on it steals its own gate drive, which tends to cause hot spots and reduce the dI/dt capability fairly badly.

Cheers

Phil Hobbs

On a sunny day (Wed, 10 Jul 2019 10:28:17 -0400) it happened Phil Hobbs wrote in :

Sure, add a fuse for if they connect it to a 12 V 200 Ah battery :-)

I have used that circuit and it was standard in some TV sets after the switcher to protect against the series switcher failing.

Again, NOTHING can protect against carelessness.

Google "SCR crowbar circuit"

It is a very reliable way of protecting your equipment, Never mind the rest of the universe...^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H