wiring a pentode as a diode...

[kissack101]

Right,

I don't know a great deal about tubes, but i've picked up some 6AJ5 tubes (Datasheet: http://www.mif.pg.gda.pl/homepages/frank/sheets/137/6/6AJ5.pdf) with the view to fooling around with them as clipping diodes. Now 1) I didn't check this before I bought them (in haste, on eBay) but can they be used as diodes? And if so 2) how do you wire them up?

I've been scouring google and the search function here but I can't quite get my head around how to wire the socket. As I understand so far, pins 3 and 4 are the heater (6v to pin 3, pin 4 to ground?), I'm pretty sire 5 is the plate, but I can't figure out which of the remaining pins (some of which are marked 'K', some 'G' - ground as well?) goes where?

Any help would be much appreciated.

Adam.

[Seljer]

as far as using it as a diode goes, I think you'd just use the cathode (pin 7) and the plate (pin 5) and ignore the rest of the stuff (hook up the heater though, 3 and 4, to 6.3V)

[dirk]

I'm not an expert, so I might be completely wrong.

Tubes don't have a voltage drop like normal silicon diodes (0,6 volts) or germanium diodes (0,3 volts). So they can't be used in clipping sections.

Better would be to use these penthodes as normal gain stages to obtain real tube distortion. You can wire them as triodes or penthodes, these give different distortion characters.

[kissack101]

as far as using it as a diode goes, I think you'd just use the cathode (pin 7) and the plate (pin 5) and ignore the rest of the stuff (hook up the heater though, 3 and 4, to 6.3V)

So you hook up 3 AND 4 to 6.3v? What gets connected to ground? I tried using a 6AL5 in a similar role, it did clip but sounded awful, there 3 went to 6.3v, 4 went to ground, as did 6 which is the internal shield (is there one here? I thought maybe pin6, Gs?) and the rest of the pins were the plates and cathodes. Should I just ground all the other pins on a 6AJ5? Will that minimise noise? I'm tempting to just try it but I only have a few and don't want to blow them to pieces.

Many thanks,

Adam.

[R.G.]

Seljer is correct. When the filament is hot, then the anode (plate) and cathode act like a diode. Ignore all the other pins. Just like a semiconductor diode, the anode and cathode may be grounded or not. To avoid injecting hum, the 6.3V filament voltage MUST be referenced to ground by either being tied to ground at one end, having a center tap tied to ground, or tied to a fixed DC voltage.

The only thing that will blow them to pieces is to put too much power (that is, voltage times current) through them. There is a current limit as well, probably about 2-3 ma for small pentodes, but the tube itself will make it very hard for you to exceed this unless you have a lot of volts available.

Dirk's comment is kind of accurate. They don't have an offset voltage that must be exceeded before they turn on. But diode clippers have been built with tubes as diodes. The knee of the turn-on at 0V is very soft. in practice this means that they usually have to be referenced to an external voltage to get clipping like you normally think of. But try it both ways.

[chillhuman]

Does referencing the tube w/ a dc voltage give a sharper knee to the AC passed through it ?

Also, how *does* a tube act as a diode ? Or does it act like a pin diode ?

[Seljer]

Does referencing the tube w/ a dc voltage give a sharper knee to the AC passed through it ?

Also, how *does* a tube act as a diode ? Or does it act like a pin diode ?

The first vacuum tubes that were invented were only diodes, they figured out they could add all that other stuff later

the filament heats up the cathode, which allows it to release electrons into the vacuum (thermionic emission), and if the plate is positively charged in comparison to the cathode, the electrons will then fly towards it, if it isn't then they wont, then plate can't release electrons like the cathode, so altogether, current can only flow in one direction = you have a diode.
http://www.triodeel.com/5ar4_p2.gif

[alextheian-alex]

Does referencing the tube w/ a dc voltage give a sharper knee to the AC passed through it ?

no just the threshold.  You can do the same thing with a SS diode.

Just as a hint... many pedals and even some hybrid amps just wire tubes up as clipping diodes under the guise of 'real tube distortion'.  In #1, the threshold is switchble between two clipping levels, and in #2 it is set at +2v and -2v.  you could easily rig that up to work off a 9v battery.  Also, ignore that filament supply circuit.  If you want to preserve battery life, oxide coated indirectly heated cathode tubes will not be hurt by undervoltage, their emission will just drop... which is not a big deal if you are using them as diodes, so you could knock the heater voltage down a bit with a simple 3 terminal regulator and save some current.

[alextheian-alex]

Actually... that is a pretty good find with that tube... i have not seen them before.  The data sheet has it running at 28v, so you could probably drop that down to 18 with a pair of series connected 9v, or bump a single 9v up to 18v with a simple switchmode voltage doubler and run it as an ACTUAL tube clipping stage.  Or run 3, 9v in series and run it just like the data sheet circuit recommends.  With the 1v of bias that you need, you could just use a rectifier diode or LED in the cathode for bias, and swap out plate and grid resistors until you are happy with the results.

[alextheian-alex]

my browser keeps crashing when i try to post this, but I'll try again.

Here is a simulation that i did of that diode clipper:

[dirk]

Always nice to learn stuff.

I see that the tubes in diode configuration have a nice round clipping behaviour.
How does that differ from normal (schottky/silicon/germanium/led/zener) diodes?