Light to tubes.

[served]

Hi.

I have a problem.
Finished building a tubeamp. Now I would like to add some colour to it. I was thinking on puting lights under the tubes. It would look cool.

I was tihinking on using heater voltages to supply power to leds. Its 12,6V and was thinking on making half rectification with two diodes.
But I can not figure out if this is such a good idea what kind of voltage will I get from it?
Then I could calculate correct resistor values.

The other thing I just noticed when reading. Tube heater can be powered with DC current, then It would be less noisy. So if I have two Tubes 12AX7 and EL 84. Should I try DC or it doent have that much effect?

[Skruffyhound]

I know next to nothing about tubes but coincidentally in the course of todays reading I found a piece talking about DC on heaters wearing tubes out faster.

In fact it was right here on this forum, just a few threads up 
http://www.diystompboxes.com/smfforum/index.php?topic=84878.0

[maarten]

why should you want to use rectification? The leds will run on AC as well; as they conduct only in one way, they will light only half of the time, but you won't notice as this goes too fast for the eye. It will save you current.

Heating with AC does not give you any noticeable noise if you twist your wires to the heaters and keep them away from signal lines.

Maarten

[amptramp]

Beware of running LEDs on an AC heater supply.  Many LEDs have a reverse voltage rating of 5 volts or less.  You can put an antiparallel diode across the LED to prevent it from seeing more than a diode drop in the reverse direction.  In fact, put a LED under one socket facing one way and another LED under the other socket facing the other way and run them off the same dropping resistor.  The LED never sees a reverse voltage greater than the forward voltage of the companion LED.

Many amplifier designers connect the heater wiring to a DC voltage of about +30 volts to minimize the amount of heater modulation on the signal.  Even with the high work function of pure tungsten, some electrons are emitted from the heater and find their way to the plate, adding an amount of hum that is slight to a lot depending on how much of the filament protrudes from the cathode.  By raising the heater potential, the presence of the grid tends to block the filament emission from getting to the plate.  Note: I use heater and filament interchangeably here although filament is usually directly heated and heater is indirectly heated.

[PRR]

> Heating with AC does not give you any noticeable noise

Correct (to a point), but that's pure smooth AC.

The wire around the half-wave AC-fed LED will radiate big buzz, and if right under the socket will be hard to cancel.

> LEDs have a reverse voltage rating of 5 volts or less

Yes, but if current is limited they are not damaged. In fact on nominal 6VAC power with appropriate forward-limit resistance they won't be damaged. (This changes when the raw AC is much higher: I ran an LED from 120V AC and the reverse breakdown dissipation restricted my safe forward current and brightness.)

The back-back plan is elegant and efficient. Still puts an unbalanced square-wave buzz between all your signal pins.

If you can spare another 10mA from your B+ (or steal it from power tube cathode bias): wire all your LEDs in series with an appropriate resistor. This may waste 3 Watts, but you can run a hundred LEDs in series. You may need more than you think to get "any" light to come up through a tube base. And this DC can be as clean as you want.

> I have two Tubes 12AX7 and EL 84

The EL84 has a 270 ohm cathode resistor, right? Split this to two 470 ohm resistors. Stick two LEDs in series with each resistor (4 total). This gives correct EL84 bias and four max-bright LEDs.

[served]

Thanks to you all! This is really good info!

I thought about this halfrectified power and it seemed a bit off to me also.
The Humm problem is probably from something else. I have never had this kind of problem before. Its not my first amp, but I thought that if I have heard about it, then there must be something that I have to try out to see if DC realy is an improvement.

I should probably redisign my turret, keep the heater lines more futher away from the circuit.

I will try to use EL84 power to supply LED.s it looks like a good idea. I just have to keep the bias right.

And I would like to correct one thing. The Heater voltage is not 12,6, its 6.3V, I some how managed to confuse my self.

I tried to light through some tubes and one led did manage to do the job, so I will use only two leds, on for AX7 and the other for EL84.

So I will try to test all this and will let you know!

[petemoore]

  A cheap/quick way to test heater circuit for noise is simply replace the supply with a battery [temporarily] to see if the heaters AC is actually noisy.

[merlinb]

There are many ways to power LEDs for free in a tube amp http://www.freewebs.com/valvewizard2/ampwithLEDs.jpg

Also, wiring up antiparallel LEDs across the heaters is very unlikely to cause any more buzz than you usually get with AC heaters.

[served]

HEy! Thats mean one. I will go for it! I think I ll use ECC83s anode power. I have a 100K bias resistor there and I will put the LED before .
This was a super good information! Thanks, I will let you know tomorow!

[served]

Hey! Thanks! I got it! Now I just have to bias it again. Light is a little bit low, but I ll go and get better LEDs

[amptramp]

12AX7 / ECC83 anode (and cathode) current is typically 0.5 to 1.2 mA.  Not many LED's will produce a bright light from that.  The EL84 / 6BQ5 has a plate current of 48 mA which is probably too much for an LED but the screen current goes from 5.5 mA at no signal to 10 mA at maximum signal and this should be adequate.