LED as Zener !

[John_BS]

I've just replaced the yellow back-light LED's on a Solatron 7150 DVM. The replacements are 5mm "white" LED's which, when measured at 1mA, exhibited a forward voltage of about 2.7V.

I thought it might be interesting to use these as low-voltage zeners, so I've compared the voltage / current curve with a 400mW 2.7V zener.

Note that the zener plot is the first one!

John
(0.5V/div hor, 1mA/div vert )

[Electronpusher0]

Interesting, certainly a better "knee". How does the slope compare at higher currents?

Peter

[Radio Wrangler]

There was an audio article in WW a long time ago where the designer used LEDs instead of zeners, saying they exhibited lower noise levels.... It might have been Douglas Self...

David

Red leds are great for constant current sources using a bipolar transistor, the temperature coefficients almost cancel, and you get an indicator free.

[turretslug]

That's both encouraging and interesting- from time to time I've chewed over the idea of using "white" LEDs as free illumination in battery applications where something like 5.6V Zeners were used anyway but took it for granted that characteristics such as knee definition and slope resistance would be rather worse. You learn something new every day....

[paulsherwin]

Presumably you could use two or three in series to obtain higher reference voltages.

[Red to black]

Thorn used an LED on the TX10 chassis Tube base as a means of biasing the video output stages back in circa 1980

[TonyDuell]

There was a crazy power supply circuit used in a Zenith MDA monitor all those years ago. I joking said it combined the reliability of a switcher with the efficiency of a linear. It rectified the mains, used that to power a free-running chopper (no regulation applied), the output of the chopper transformer was rectified giving about 18V DC. This was fed into a linear regulator circuit to give the 12V for the rest of the monitor.

The reason I mention it here is that the reference for the linear regulator was the (green) power-on LED.

[barretter]

You can replace capacitor-bypassed resistors in valve cathode circuits with an LED or a series-parallel combination of LEDS depending on the voltage you need and the current involved.

[ionburn]

Originally Posted by Red to black

Thorn used an LED on the TX10 chassis Tube base as a means of biasing the video output stages back in circa 1980

I remember this well! On the training the rep made a big thing of it.

I often use them, remembering the Thorn example which was, in my experience, reliable. The only issue I have is going through my box of led's testing each to get one that is just right although th colours tend to bracket a voltage.

[Alistair D]

Originally Posted by barretter

You can replace capacitor-bypassed resistors in valve cathode circuits with an LED or a series-parallel combination of LEDS depending on the voltage you need and the current involved.

Scroll to the bottom of this page for an example of LED cathode bias

http://www.valvewizard.co.uk/OtherStuff.html

Al

[G6Tanuki]

I recall a circuit in one of the 1980s ARRL handbooks in which a red LED was used as the source-resistor in an AGC-controlled dual-gate-MOSFET IF amp; it was claimed that by holding the source at a fixed voltage the AGC action was more-linear and intermodulation when the MOSFET was 'throttled back' in the face of strong signals was reduced over what could be achieved with a simple resistive source.

[Bazz4CQJ]

I remember once being in a conversation about whether or not to include a LED on a mast-head RF pre-amp simply to verify DC was being supplied, or whether the LED might be a source of noise; I don't recall what conclusion was reached.

Just quickly looking over Google on the subject of zener/LED noise, this thread on another forum may be of interest https://groupdiy.com/index.php?topic...this%20border..

Re the RF pre-amp, can anyone tell me what sort of frequency range a LED could affect?

B

[Mooly]

You might some useful bits of info here:

http://www.dicks-website.eu/noise_in...ces/part1.html

I suspect a lot of this falls into the 'half of a quarter of nothing' region though.

[Radio Wrangler]

Note that his quiet plot of a zener with a constant-current feed is of a 2.7v part. At that voltage it will probably really be a Zener.

Once you get above a few volts, you'll find that what get flogged as "Zeners" are actually avalanche diodes. A somewhat different mode of operation and very different in terms of noise.

David

[cmjones01]

At the back of my bench is a Maplin 12V 3A regulated power supply whose job is powering low-voltage drills and the like. I was given it because it didn't work. Tracing out the circuit, I found that the voltage reference was actually the red front panel LED, which had itself failed short-circuit! A replacement brought the thing back to life.

I think it must have been done that way for cheapness. Temperature compensation makes no sense, given the lack of thermal contact between the front panel indicator and all the hot bits.

Chris

[JohnBG8JMB]

useful dodge is a red led in the earth lead of a 7812 regulator. makes a 13.5v regulator with a free pilot light.

John

[McMurdo]

Conversely, Roland used clear zeners in its (jupiter 8?) synth power supply and they lit up red.

More recently, an amplifier design that comes to the workshop from time to time uses 4 green leds

[evingar]

Back in the day, Quantel used red LEDs as part of a current sources in a video decoder. Complex thing that used loads of MC1495 (4 quadrant multipliers).

[AC/HL]

I don't know how or when light emission was discovered, but as far as I know all semiconductor junctions display the other characteristics to some degree. It's developed further in the dedicated devices of course.