Tube DIY Asylum

Do It Yourself (DIY) paradise for tube and SET project builders.

Return to Tube DIY Asylum


Message Sort: Post Order or Asylum Reverse Threaded

LED cathode bias vs diode 1N4148 in series, Are they the same effect

182.239.88.3

Posted on July 27, 2023 at 04:58:11
Eric Chan
Audiophile

Posts: 285
Joined: October 31, 2001
LED cathode bais act like a bypass cap parallel with cathode R.

Is the normal 1N4148 work the same effect ?

Saying if I want to bias 6922 without adding resistance to its cathode.

Thanks for your sharing

 

Hide full thread outline!
    ...
Yep, they are both diodes with forward voltage drop, posted on July 27, 2023 at 06:11:01
Chip647
Audiophile

Posts: 2578
Location: The South
Joined: December 24, 2012

Measure the current thought the tube plate load to calculate the operating point. The 1N4148 will have a forward voltage drop around 0.7v. If you wanted 1.5 volt bias you would need 2 in series.

People say that different LEDs sound different (in addition to having differences in the forward voltage drop) and they also sound different than diode rectifiers.

 

RE: Yep, they are both diodes with forward voltage drop, posted on July 27, 2023 at 06:23:15
Eric Chan
Audiophile

Posts: 285
Joined: October 31, 2001
Hi Chip,
The LED seems to have constant Voltage whilev4148 may not ad I was told, that is why I want to know their differences in yerm of sound wise and tech aspect

 

Ok..., posted on July 27, 2023 at 07:01:06
Chip647
Audiophile

Posts: 2578
Location: The South
Joined: December 24, 2012
The tech aspect usually revolves around the internal resistance of the part. The 20 to 100 ohms resistance in different LEDs might be audible as the lower resistance will create lower local feedback making the gain very slightly higher. The silicon diode has about the same internal resistance as LEDs, some are a little higher in this regard. For the "constant current" aspect, all diodes have a positive slope for current to voltage, none are completely flat. To get a constant current source you need a constant current circuit.

Once you account for tiny differences in gain due to the internal resistance, the sonic difference falls into the "audiophile ear" world.

You can spend the $1 to plug in a couple of different combinations to see what you like. Just don't be "biased" by the pretty colors.

 

RE: Ok..., posted on July 27, 2023 at 07:19:55
Paul Joppa
Industry Professional

Posts: 7271
Location: Seattle, WA
Joined: April 23, 2001
For what it's worth, some LEDs are noisier than others, though not nearly as noisy as Zeners. It matters in a phono preamp.

 

RE: Ok..., posted on July 27, 2023 at 16:32:09
Stephen R
Audiophile

Posts: 1424
Joined: January 11, 2002
Try led, diodes and whatever else you fancy and see what you like.

Here's some measurements in the link.

 

RE: Ok..., posted on July 28, 2023 at 01:49:07
Eric Chan
Audiophile

Posts: 285
Joined: October 31, 2001
The 1N4007 impedance has even lower, so why use LED then?

 

RE: Ok..., posted on July 28, 2023 at 03:16:45
Stephen R
Audiophile

Posts: 1424
Joined: January 11, 2002
That's a graph of voltage drop. Z of red LED a little lower than 1N4007.

 

RE: Ok..., posted on July 28, 2023 at 07:38:26
Paul Joppa
Industry Professional

Posts: 7271
Location: Seattle, WA
Joined: April 23, 2001
To get the same voltage as a red LED, you need two 1N4007s, which doubles the dynamic resistance.

 

RE: Ok..., posted on July 29, 2023 at 04:18:35
Eric Chan
Audiophile

Posts: 285
Joined: October 31, 2001
The Conrad Johnson ART preamp use 1N4148 2 in series instead of using single RED LED to achieve 1.4V bias

The recovery time for 1N4148 is 4ns which is very fast, is that part of the reason?

 

RE: Ok..., posted on July 31, 2023 at 01:11:09
Stephen R
Audiophile

Posts: 1424
Joined: January 11, 2002
It's all about impedance. There is no switching going on in this application.

 

Additional info you might enjoy, posted on August 8, 2023 at 18:54:34
Ian L
Audiophile

Posts: 1319
Joined: July 5, 2002
Thread with some noise measurements

 

Page processed in 0.028 seconds.