• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

bias arrangements for P-P amps.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Disabled Account
Joined 2004
Wow…
I have missed a lot those two days offline!!

Hey the problem over here is about biasing output stages and not with the Greek letters ì and Ù, which are European letters, too!
But…
I’m using R because I prefer to use R than “ohm” and not because I’m European. I prefer to write 0R5 than 0.5 Ohm, or 120R than 120ohm. It’s something like “kilo”. You use only the letter “K”. Like 1K5 or 120K or 47K or whatever. You don’t write 1.5 kilo-ohm. Not even 1.5 kilo. Omega also is not a common letter to find on keyboards.

@SY
Bravo with your idea!! Keep us informed!
 
OT but looks like we got two going on here!
In the US it is/was? example: 1.2k for resistors, but I prefer to write the European way 1k2... For caps 0.01µf -> 10n or 10nF. Wasn't all that hard to get used to for caps! Get rid of that darn dot! Sorry, couldn't resist resident! :D

Wayne ;)
 
Sure, let's talk about mhos instead. Whoops.... I meant Siemens. :D

Or how about mF and MF in place of the more obvious uF for micro-Farads? There's almost no excuse for that one! :rolleyes:

Such debate is potentially endless, but FWIW, I personally prefer using the European style resistance numbers; 8R2 and 4K7 take up about the least space of the various alternatives, and this can be very important on densely packed schematics. Heck, many schematic editors don't even have a way to put in the omega symbol. Sometimes I'll even omit the unit entirely, since a schematic indicates what the part is anyway.

Anyway, I think LED biasing is a great idea, expecially for preamp and lower-current stages, or outputs which require little bias voltage. For most output stages, I think I'd stick to fixed bias, but please keep us informed of your results, SY. It could be very interesting to compare the noise and impedance of various colours - blue vs. standard red vs. high-intensity red etc.
 
frugal-phile™
Joined 2001
Paid Member
hifiZen said:
Anyway, I think LED biasing is a great idea, expecially for preamp and lower-current stages...It could be very interesting to compare the noise and impedance of various colours - blue vs. standard red vs. high-intensity red etc.

This info from an olf JoeList thread might shed some light.

> > Is there a good reference on LED bias out there on the web?
> > I'm looking at some method to calculate the LED to use but most of what
> > i've found its more or less trial/error.
> >
> > I've tried some blue led's on my 12B4 amp so far with good sound
> > resulting, unsure if its better than cap/resistor though. Need to listen
> > to it more.
>
> steven,
>
> i've heard that the voltage drop across the LED is proportional to
> wavelength emitted... thus the infrared diodes have the lowest, then in
> order of ROYGBIV.
>
> the orange diode i'm using in the cathode of a 6s45pi develops about 1.9vdc,
>as a frame of reference. i would think that the blue is a bit more than
>that... how much Vf do you measure in circuit?
>
>ken

It is worth bearing in mind that an LED is not perfect - although better
than a resistor I think - the reason being that they have a slope
resistance, thus a small amount of feedback will be developed.

I understand that the red LED has the lowest slope resistance and therefore,
is the best - if 1.6V suits you. Avoid super-bright or special reds; buy
the cheapest you can find - I am told that these are the best.

Not often you that in this mad hobby!

Paul

dave
 
I don't think noise is an issue- remember that this is a high-level stage and (most important) it's differential. The noise is common mode, which means that it will be knocked down another 40-50 dB below its already-low value. Anyway, I've got the first iteration of the EL84 version wired up and we'll give it a run tonight or tomorrow.
 
SY said:
I don't think noise is an issue- remember that this is a high-level stage and (most important) it's differential. The noise is common mode, which means that it will be knocked down another 40-50 dB below its already-low value. Anyway, I've got the first iteration of the EL84 version wired up and we'll give it a run tonight or tomorrow.


SY,

I think the interest in the subject of LED's for developing cathode bias expressed here may have developed beyond just the scope of biasing an output stage where as you point out, noise will be a insignificant issue. I know that I for one would like to know more about their application in line level or lo-noise preamplifier stages. Perhaps a new thread should be established to discuss this issue.

As far as o/p bias. If one wanted to use LEDs in cathode circuits where say 80 mA was flowing, rather more than the ~20 mA that a LED can safely handle, has anyone tried parallell LED arrangements? Would small equalizing ....um, you know, be required in series with each LED as one would definitely employ when parallelling power diodes?
 
The current equalizing question is one I'm wondering about myself. Experiment will tell.

FWIW, I've been running some tests on 5692s configured as voltage amplifiers (similar setup as Morgan Jones's 6SN7 measurements) with red LEDs as cathode bias and a LED-referenced current source as plate load. The S/N of the whole stage is better than -100dB ref 2.5V, so the LEDs can't be contributing much.
 
LEDS as references

When talking about semiconductor devices for references there is an easy way to distiguish what devices will be "low noise" and what devices you need to be careful of noise.

Reverse conduction (controlled breakdown) devices like zeners are always noisy - and its largely avalanche noise which is damn near impossible to entirely get rid of once generated.

Forward conduction devices like LEDS and various "band-gap" devices are Low Noise.

If you need a low noise "zener" then wire a transistor as a Vbe multiplier.

If you need higher voltage low noise "zener" then put one or more LEDS in series with the emitter of the transistor and turn it into a Vbe plus LED volts multiplier.

Noise is the reason for use of Blackgate bypass capacitors as well. They do sound good BUT their main advantage is their truely outstanding noise performance.

On the "R" issue, one poster above got it right. Its purely a way of getting rid of the dot "." before the photocopier/laser printer etc. does it for you and you are left wondering if the value was 4.7 Ohm or 47 Ohm, 4R7 says it all.

R is used as a multiplier (like K or M) not as an Ohms symbol - yeh its restricted to use for resistors BUT its "Industry Standard".

Really guys - someone should find you a horse, a long stick and a windmill.

All the best for Xmas and the new year to you all.
Cheers,
Ian
 
Oh - on the topic

What do I use for output tube bias in my EL84 PP Ultralinear Amp?

An MJE340 + BC547B "ring of two" current source in each cathode with a 470uF/50V Blackgate standard bypass.

For "newbies" the circuit description is (draw it up from the description):
1) 0V connected to emitter of BC547B and one end of 180R resistor
2) Other end of 180R resistor to emitter of MJE340 and base of BC547B
3)Collector of BC547B to base of MJE340 plus one end of 6K8 resistor
4) other end of 6K8 to MJE340 collector and to the EL84 cathode.
5)bypass cap 470uF/50V Blackgate across the lot from EL84 cathode to 0V.

I found this sounded MUCH better than a simple resitor (270R)plus bypass capacitor in each EL84 cathode.

My V+ is 315V, This arrangement gives about 37mA per tube at a developed bias voltage of 10.8 Volts.

I am going to stop by the electronics shop today to grab a batch of LEDs to give SY's scheme a whirl.

Cheers,
Ian
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.