• 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.

6J4 Experiment!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello,
I just sat down to sketch up a circuit to prototype for fun. The project will be a two stage SET amplifier. There will most likely be several iterations. The output tube will come later.
Not that it is the best tube or worst I have a shoe box full of 8532’s (6J4). So this time the driver will be a pair of 6J4’s. The plan is to use two per side in a series configuration much like a SRPP but the output will be tied to the plate of the bottom tube as opposed to the cathode of the top tube.
As I sit to sketch I notice that pins #1, #5 and pin #6 of the tube tie to the grid. Here are a couple of questions.
Does anyone know this tube? Do pins #1, #5 and #6 need to be used? There is a similar thing with the 12B4 both grid pins need to be used to avoid oscillation I read. There are a lot of Darling amplifiers made with this tube and I have not read about any issues.
Any stories to share?
DT
All just for fun!
 
Last edited:
Does anyone know this tube? Do pins #1, #5 and #6 need to be used?

The 6J4 is a UHF triode designed for grounded grid, small signal, RF amplification. It includes all those extra grid connections since these were meant to be connected to the ground plane to minimize impedance, so that it wouldn't oscillate.

For normal audio use, this isn't a consideration (you probably won't be using it as a GG amp anyway either).
 
Any 8532/6J4 stories to share?

I have about 1000 of them, all used. They are in various conditions from dead to new. I tested a few several years ago and decided that they were worth saving. I found a few references that said that they made good input stages. My notes said "lower distortion than a 12AX7". That doesn't tell me much, and I was testing tubes quickly to decide which ones to keep.

I have over 100 6BQ6's, so let us know what you cook up. I'm thinking P-P though. The 6BQ6 works OK in SE. I have melted a few in the Simple SE, but they absolutely rock in P-P. Note the pinout is compatible to the popular audio tubes, just wire the plate cap to pin 3. They like screen drive too as long as you are reasonable with the power requests. Thermonuclear meltdown occurs at 110 watts output on 550 volts B+.
 

Attachments

  • 6BQ6GA.jpg
    6BQ6GA.jpg
    300.7 KB · Views: 938
  • 6BQ6GA_cranked.jpg
    6BQ6GA_cranked.jpg
    279.8 KB · Views: 853
Member
Joined 2006
Paid Member
I tried 8532's in my Darling build a few years back. It was my first tube project. The 8532's I had sounded shrill and are the most microphonic tubes I have used to date. I abandoned the 8532's and moved on to the Hitachi 6GK5 with much better results. Shrill sound from 8532's may have been from high frequency instability. The 8532's were so microphonic that I didn't bother to scope for instability. You might want to plan to use grid stoppers with your 8532/6J4’s.
 
Hello,
See the attached sketch below; This is the starting point of the 6J4 experiment.

The series input stage is lifted from Volume 18 of the 1946 MIT series post WWII. The top tube acts as a resistor, as a tube it is nonlinear in the opposite direction of the bottom tube. Gain is equal to ½ the mu.
drj759
I will try the grid stoppers if the shrill happens thank you.
DT
All Just for Fun!
 

Attachments

  • 6J4 Experiment1.pdf
    19.9 KB · Views: 491
Last edited:
That 6J4 seems more than a bit like the 5842 and 6C45П. Tubes of that nature can oscillate if you look at them cross eyed. :( Put large value Carbon comp. resistors on all grid socket connections and use the junction point of the multiple stoppers as the I/P location. Ferrite beads on the heater wires close to the socket are also a good idea.
 
But it's not. Look at where you're sampling the signal. Output impedance is higher, and when the cathode resistors are matched, the linearity is better. However, it needs to be buffered to realize the linearity; Broskie uses a White cathode follower with a hum cancellation gimmick, Landee shows a cathode coupled amp.
 
Hello,
Yes it looks very much like a SRPP. It also looks much like a Mu-Follower. Both SRPP and the Mu-Follower are discussed on other pages (i will scan and post if there is an interest).

When I ran it in B2 spice it was not constant current. The top tube acted very much like as a grounded cathode amplifier with a plate resistor that has a voltage drop of half the B+ voltage, as described in the attached PDF above. It acted very much single ended but with less harmonic distortion. Push-Pull will deliver more current at the output. The maximum current swing for single end is plus minus the idle current. Push-pull is greater, in theory double the idle of one of the valves of the pair.

SY, much of this is vanity in The Old Testament sort of way, much has been done before. This is old school stuff. The plan is to avoid driving the power SET into A2 and keep the input impedance steady to retain as much linearity at the driver as possible.
I have a sheet of aluminum laid out to start drilling holes for the tube sockets.
DT
All just for fun!
 
Problem is, the output stage is a reactive load and that, along with the loads R5 and R9 (what does R5 do?) will compromise the linearity of that stage. A buffer (you might consider MOSFET if metal is already cut) will clean things up significantly and let you realize the potential high performance of the stacked triodes.
 
Hello SY,
Story be told, metal cut or not, this is a breadboard toy. This will be the first set of real experiments with a PC / sound card based audio test system on my bench. I plan to look at this stacked triode, power supplies, noise and distortion.
The kid in me wants to beat this with a 4 oz ball peen hammer and watch what happens on the computer output. Put more academically, how much will be compromised, without the buffer, I want to see?
The thought regarding R5; R5 is a high value bleeder to drain the coupling capacitor. I will put my feigner in there without thinking.
Please tell me more of the MOSFET buffer how would it be implemented?
DT
All just for fun!
 
Please tell me more of the MOSFET buffer how would it be implemented?

Topologically, this is the easiest way and will work fine, but with a negative supply available, you could put the follower on the other side of the coupling cap and have the option of A2 operation as well as better overload recovery. If you go the easy route (recommended for your first go-round), you could use something like an IRF820, a gate stopper of 1k or so, and a source resistor of B2+/2I, where I is the desired mosfet current (I'd use 10mA or so to start).
 

Attachments

  • mosfet follower.png
    mosfet follower.png
    6.2 KB · Views: 447
I am not so sure that Broskie's Aikido output stage is a White CF.

Hello,
peter t,
Being an old grump I was still born too late to put my name on any of these tube circuits. There seem only to be a few basics and variations on a theme.
Looking back with most of the patents expired perhaps there should be a new naming system sorts the circuits by performance characteristics; input impedance, output impedance, (oops) miller capacitance, gain or buffer. Blah blah blah.
The Aikido Cathode Follower seems to be a unique variation of the cathode follower. If I recall correctly the WCF is also push pull.

Sy,
Thank you for the MOSFET buffer implementation. It will be interesting to test the grounded cathode, stacked MIT triode thing, with and without buffer.
A continuation of a related thought; in a two stage SET amplifier the stages are 180 degrees out of phase. The transfer functions curve in opposite directions and tend to cancel each other out and will create a composite transfer function.
This will be tweeking without the drugs.
DT
All just for fun!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.