Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

LED tube biasing, pros and cons
LED tube biasing, pros and cons
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Thread Tools Search this Thread
Old 7th July 2018, 03:03 PM   #11
GoatGuy is offline GoatGuy  United States
diyAudio Member
GoatGuy's Avatar
Join Date: Dec 2012
Location: SF Bay Area
Default LEDs

Originally Posted by jackinnj View Post
Just because an LED has a certain forward voltage doesn’t mean that it is low impedance - and a high Z LED militates against its use.
You make a fair point. I counter that tho' with this: it would take an exceptional LED to show a significant ΔV / ΔI (≡ “Z”) slope around its nominal VF forward recombination voltage.

One significant possibility is that an exceptional LED has an integral series "ballast" resistor inside its package. Or with more sophistication, an integral constant-current regulator (which isn't terribly hard to make: a lightly doped JFET is a compelling constant current regulator.)

But I've only seen one of these in my entire life so far. I've been “doing microelectronics” for 55 years now. From the dusk of vacuum tubes… LOL…

No. We just need to think of LEDs as diodes with a relatively high VF recombination voltage¹ and an also — entirely serendipitous — nearly temperature invariant VF as well. Before LEDs were available in the shorter wavelengths (i.e. blue-green, blue, deep blue, UV and 'white') … the VF ranged from 1.6 to 2.2 volts nominally. RED and YELLOW, and yellow-green.

While perhaps OK (onesies) for preamp or front-end input stage biasing, they didn't have the current carrying to make good biasing on downwind stages. At the core too was manufacturing variance. You could buy “a bag” of 100 to 1000 inexpensive LEDs, set up a test jig with a voltmeter and a whetstone bridge, and while watching the TV, test-and-bin all the LEDs in a few hours. Then become vexed to realize that the clustering of VF relative to IQ was so variable. Manufacturing variance.

Today however, because LEDs are being made with exactly the chip-making technologies used for our fantastically complex smartphone chips, the precision and repeatability of manufacturing big wafers of LEDs all-at-the-same-time has peaked.

I bought a “1,000 bag” of tiny bright blue LEDs for $3.50 on EBay; unlike the 1970s, I now have a 4½ digit multimeter and an older HP metrology 10 PPM reference variable voltage supply… so “testing and binning” the LEDs doesn't even require setting up a whetstone bridge for high accuracy.

Quite to my bemusement, there were only TWO LEDs that didn't “bin” to the same ±0.1% “bin” of VF as the others. Being only 2 of them, I just chucked 'em in the rubbish bin. T'heck with them.

Moreover, the "1,000 bag" was apparently weighed-to-count, and there 1,060 LEDs in the bag. (All were “double tested” for both VF and apparent brightness, non-quantitatively, at 1, 2, 5, 10, and 20 milliamps by rotary switch, and being inserted on a plugboard between two LEDs used as output-brightness references.) It took about 2 hours to test 'em all.

SEPARATELY I have advocated using simple 1N4148 or 1N4001 type ubiquitous budget rectifier diodes in series as simple-to-tune VF cathode bias solutions. No pretty green or blue glow, but solid-as-a-rock performers. Thing is, that when a few are tied end-to-end in a series chain to get say “4 volts”, within the ±30oC tube chassis operating temperature range (or more!) the VF for non-LED diodes varies by a few percent. Enough to mung the original biasing purpose? No… not really.

Before LED biasing became popular, I was using string-of–1N4001 diodes to create the raised cathode bias point all the time. All those 30+ year old amplifiers are still in use. It works. And frankly, in the days when “good” resistors were ±10% accurate devices, and we all used them without binning for cathode biasing … and with the ±25% variation of tube operating points, ±40% “life-curve” operating point drift of tubes in service, and the ±10% line voltage variation, having a 10% binned bunch of resistors was easy to work around.

A chain-of–1N4001's having ±3% VF thermal drift was the least of my problems.

So, just saying: it would be mighty rare to find a bunch of LEDs that exhibited “high Z” impedance, without being engineered to do so as a side-effect of some other purpose-and-function ideal.


¹ recombination potential — for the curious, all solid-state diodes work by having a "P" (shortage of electrons) and "N" (surplus of them) pair of layers molecularly bonded into a so-called PN junction. Wires are attached, and simplistically, there you are. A diode.

When current passes thru in the forward direction, electrons jump from the N side to the P side, and “recombine” with “holes” left in the crystal lattice valence electron cloud on the P side. Because quantum physics requires it, every last recombination is accompanied by a photon emission in “recombination”.

For 1N4001 thru 1N4007 silicon, the recombination energy is low enough that the photons aren't visible to the human eye. (But they are to infrared cameras!) For LEDs, the materials have higher “bandgap energy” differences between the P and N sides.

Binary and tertiary semiconductors such as gallium arsenide GaAs, indium telluride InTe, indium aluminum phosphide InAlP, also known as III-V binaries are used to get bandgap energies of 2 to 4 volts, generating in turn respectively, yellow-to-UV photons in our modern high efficiency LEDs. Since there is almost no color variation with temperature, it is a pretty good bet that the VF also doesn't vary much for LEDs, by temperature.
John Curl's Golden Rule…: 100 kHz bandwidth, 3 μs risetime, 100 W mean output, 100 V/μs slew rate, 2 Ω dynamic load, 20 amp min current source/sink
  Reply With Quote
Old 7th July 2018, 03:25 PM   #12
rongon is offline rongon  United States
diyAudio Member
rongon's Avatar
Join Date: Jul 2009
Location: Across the river from Rip's big old tree...
LED tube biasing, pros and cons
Originally Posted by artosalo View Post
No it does not, because it is not connected as a CCS. It is a CVS and keeps the grid to cathode voltage fixed as long as some (less than) 0.5 mA current flows thru it (and the tube).
Thanks for the info! Yes, that is a very attractive part for this application. I could have used a couple of those for the headphone amp I just built, to get +2V at the cathode with plate current of 55mA. Better yet, I could have experimented with slightly different grid-cathode voltages instead of accepting the +2.2V I got from paralleled green LEDs. Next time...
  Reply With Quote
Old 7th July 2018, 04:20 PM   #13
zlab is offline zlab  Hong Kong
diyAudio Member
Join Date: Jul 2009
TLV431 provide regulated voltage from 1.24v to 6v may be a better choice for low voltage application.
  Reply With Quote
Old 7th July 2018, 04:31 PM   #14
artosalo is offline artosalo  Finland
diyAudio Member
artosalo's Avatar
Join Date: Feb 2010
But max. recommended cathode current is lower too, 15 mA.
  Reply With Quote
Old 7th July 2018, 04:53 PM   #15
DAK808 is offline DAK808  United States
diyAudio Member
Join Date: May 2012
LED tube biasing, pros and cons
Thanks for all the replies! I have read about LEDs here and there, but a more in depth discussion was just what i was interested in. I will give it a go when i get my son who is over 6 foot to move my 60lb amp to the shop . Cheers.
Attached Images
File Type: jpg DSC05655.jpg (443.8 KB, 527 views)
File Type: jpg DSC05661.jpg (391.6 KB, 528 views)

Last edited by DAK808; 7th July 2018 at 04:56 PM. Reason: adding text
  Reply With Quote
Old 7th July 2018, 05:25 PM   #16
jackinnj is online now jackinnj  United States
diyAudio Member
jackinnj's Avatar
Join Date: Apr 2002
Location: Llanddewi Brefi, NJ
LED tube biasing, pros and cons
Originally Posted by rongon View Post
Example of a high Z LED?

According to Morgan Jones (in his book "Valve Amplifiers") a blue LED he measured gave Vdrop of 3.7V with internal resistance of 30 ohms. Is that considered 'high'?
See the “Red Light District” thread. I think thats where I posted some measurements. I am about 3.5k-miles away from the data!
  Reply With Quote
Old 8th July 2018, 05:19 AM   #17
kokoriantz is offline kokoriantz  Lebanon
diyAudio Member
Join Date: Dec 2015
Location: south east asia
I have tried several times tl431 on cathode, it sounds horrible.
  Reply With Quote
Old 8th July 2018, 05:41 AM   #18
trobbins is offline trobbins  Australia
diyAudio Member
trobbins's Avatar
Join Date: Feb 2009
Location: Melbourne, Oz
I have a large batch of surplus 5mm amber LEDs which I use for restorations where the original circuit does not 'need' to be followed. They measure 1.70 to 1.80V over a 0.1 to 1mA range, with an incremental resistance that varies from 500 to 60 ohm respectively. So equivalent to about 20uF bypass at 100Hz, but of course a LED is not frequency selective.

When using a central spigot noval or 7-pin valve holder, it is very convenient to solder directly from spigot to pin, and so avoid the 'space' and wiring for a bypass cap.

The cathode signal current will modulate the idle current, and hence the incremental resistance presented by the LED. Not an issue, and perhaps a blessing, for guitar amp use.
  Reply With Quote
Old 8th July 2018, 05:55 AM   #19
artosalo is offline artosalo  Finland
diyAudio Member
artosalo's Avatar
Join Date: Feb 2010
I have tried several times tl431 on cathode, it sounds horrible.
I have used it too and it performs extremely well.

Please, show your circuit for analysis.

This was mine, 6N6P PP (AB2):
Attached Images
File Type: gif 6N6P_PP_op_skema.GIF (18.2 KB, 509 views)
  Reply With Quote
Old 8th July 2018, 08:21 AM   #20
dreamth is offline dreamth  United Kingdom
diyAudio Member
Join Date: Nov 2010
TL431 has a dynamic impedance of 0.2...0.5 ohms at 1khz so there's no negative feedback, that is why it might sound horrible in some circuits.
  Reply With Quote


LED tube biasing, pros and consHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Seen this ? Pros & cons. R Dijk Solid State 0 15th June 2018 10:01 PM
What are the pros and the cons of those 3 vas ? tessier Solid State 93 30th January 2017 02:53 PM
Cons / Pros Borat Tubes / Valves 34 1st September 2009 01:52 PM
pros/cons of a chipamp VS a tube amp jarthel Everything Else 2 22nd June 2007 02:18 PM
pros/cons of a 845 VS 211 jarthel Tubes / Valves 2 18th June 2007 01:57 PM

New To Site? Need Help?

All times are GMT. The time now is 12:15 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio