The AM Forum

Yay! The Audiophools don't like little zero bias triodes! I was doing some experiments with one and I like it and maybe should buy a few.


I'm happy with nit.. It's perfect as Borg.

• 300V
• zero bias
• can make 10W

But as all audio gurus and experts know these are not good for hi-fi absolutely no good at all for that, only for modulators and high voltage amplifiers and that is why they are $5-10 on the internet.

But some folks think otherwise if it says "Mullard"
2Pcs. NOS MULLARD NR73/CV1285=EC​C31/6N7G $269

Not sure what audio amps used them. Just some ham gear and industrial stuff?

The 53 is a 2.5V version.

What other smallish zero bias triodes are there?

6A6

1635

6Z7

73DG

For dual triodes the 12BH7 is good but its a bit on the audiophool radar.

The 6N7 and 12BH7 were popular for mobile rig modulators.

Ive used the 53 but there is a HD version that was used in elevator controllers but I forget the number.

Carl

What got me looking more closely at the 6N7, besides zero bias modulator duty, is this old limiter. It is a little different than most and uses a 5814 which is a 12AU7 in a bridge circuit  (The limiter used a 6N7 for this in early units).

This is a different circuit from the variable-u and variable gain and "screen/supressor/pentagrid" schemes. The compression/distortion specs are interesting @1%<15dB and it claims an attack time as fast as 100us.

The 5814 is a 9-pin miniature dual triode like a 12AU7where the 6N7 is much beefier. I don't know why they changed.

Another interesting "feature" is the way the 6N7 is used in the input circuit. Being a zero bias tube it is still given some cathode bias of 2.6, maybe so the line could drive it meaningfully. The plates feed the bridge directly. I'd like to get my hands on one and run through it with a signal generator and a scope.

Other tubes like the 12BH7, 6FQ7/6CG7, etc. are good but I'm thinking of similar sized tubes that are usually run zero bias with medium or rated plate voltages. The 12BH7 will do that at low voltages like 60-80V.

The 6N7 was used as the intermediate tube in the speech amplifier section of the Globe champ 175 (between the 6SJ7 and the PP triode connected 6F6's). That rig always got excellent audio reports, and I have subsequently built two speech amplifiers based off the original design. One was used with a T-368 and was identical to the WRL design using triode 6F6's as the driver tubes. The other is currently being used in my homebrew 254W rig. That one uses 6A3's as the drivers. I believe the 6N7 is used as a phase inverter to take the single ended output from the 6SJ7 to the push pull 6A3 grids. The output of the 6N7 is capacitatively coupled to the 6A3's, no transformer. Again, I get excellent audio reports. See the schematic below. The first schematic is the speech amp, the second is the power supply, and the third pic is the actual speech amp. 6SJ7 and 6N7 at top, then ps components. 6A3's at bottom, S-9 driver transformer, 1uf 5000V cap for modified heising, 811A's and S-22 at top left.

Ron

I know where there's a Collins 26W, no longer in service, just holding a place in a rack. Will check whether early or late for tube complement.  If you want us to power it up and check behavior or other parameters let me know.

Thanks! I guess I wanted to vary the input voltage and observe the AC and DC levels on the input amp electrodes and the limiter bridge tube electrodes as well as across the driver transformer primary to better understand it. Its the only unit I know of that uses a bridge. It would take all day fiddling about. I would not ask you go to to that much trouble.

That's a very good looking modulator there. 6A3's are kinda scarce, I would probably go the 6F6 (or more likely other more common pentode or tetrode) route and use then as triodes. How did you like the 6F6 in triode mode? Did it have good regulation driving the 811's? Can you recall about B+ for them and the bias method or value? Some limiters and compressors have a small power stage and some don't, but I tend to want one. One question, since the 6N7 can make 8-10 watts, was why didn't you use it for a driver to the 811's? Not enough overhead?

Running a class 'B' tube into another 'B' audio-wise is a bad idea.  A good driver for 6N7's was another 6N7 with the elements tied together, forcing it into more of an AB operation.

If you want to use a real triode in driver service without having to resort to neg feedback, try PP 6B4G's.  They are sorta on the audiofool list, but can be had for not a lot of money.  Same as a 6A3 with a octal base.

Russian surplus examples are available NOS as well.

53-6A6-6N7's also makes a good balanced crystal oscillator with the crystal feeding into the two grids.

73DG

Here's a little table of triode connected 2 and 3 grid tubes. Maybe this can be kept up with and added to as more applications are found here and there.

so far: 6L6, KT88, EL34, 6F6, 807

It's interesting that the 6F6 data found was class AB2. I did not find AB1 or A data for the 6F6, but I've only looked on Bunker and on Pete Millett's site.

The 807 data in the Sylvania '59 manual had a column for ICAS but the data was exactly the same as for CCS ?? huh.

"ultralinear" and other "tapped transformer" data was omitted because it is not true Triode operation and we don't generally see that in ham radio modulator drivers and such driver transformers as would be required for it are not off the shelf items.

The "special triode connection" where G1 and G2 are driven together was also omitted because that is not the same thing. What is the general opinion about whether that is appropriate? It would do no harm but the focus of the whole 6N7 topic and the triode connection subtopic seems to be triode connected driver amplifiers that do not themselves require meaningful drive power.

Other beam tubes and pentodes that would be interesting would be the 6AQ5, 6BQ5, 6K6, and certain low voltage ones like the 6Y6

The 6N7 is a strapping little tube. Both sections in parallel transformer coupled to another one with both sections in P/P can be pushed to close to 10w outpoot. One as a P/P class-B modder would be great for a small PW rig.

The circuit that Ron posted is a pretty classic 6N7 fuzz inverter for a twin triode with a common cathode. The common cathode seemed to be a stumbling block for designers in audio service, but is very handy for other applications like mixers, product detectors, etc.

I have always been curious how 2 of them would work in P/P parallel. But I have never seen anyone do that. The only parallel applications I have seen are usually class-A low level stuff.

The 2A5 and 42 should be listed with the 6F6;  the 46 and 59 also has triode ratings.

One tube Ive not remembered seeing listed for audio service is the 837.

I started receiving 6N7's from eBay purchases. It is apparently a VT-96. One of the tubes is an Arcturus brand. I didn't know they made those.

added the 2A5 and 42.
I'll work on the chart some more later.

Maybe I should include single ended triode connections, special class B connections (g1 to g2) and small zero bias triodes. Each implementation should have a separate tab.

Here are a whole bunch of distortion specs for the 2A5/42/6F6

http://tubedata.tigahost.com/tubedata/sheets/127/2/2A5.pdf

The only thing I could find on the 807 is 2% in AB1 under Philips

The 6N7 noise limiter in the post war SP-400 Superpro puts both sections in parallel and is basically a fast low impedance shunt audio switch. Cleverness in the form of emission starvation is used by means of a series resistor off the 6VAC in series with the fils to set the tube to a point where it responds nicely to impulse noise. But this almost always goes out of whack due to capacitor, tube and resistor aging basically guaranteeing that the circuit does not work very well in any living example!

Patrick,

I don't really know how to analyze an audio chain properly. Audio circuits have always been more difficult than RF circuits for me to understand. I find a schematic that I like, I build it, and hopefully it works. If it works well, that's icing on the cake. In this particular case, I noticed that everyone raved about the audio from my Globe Champ 175, so when I needed to build a speech amp, I just duplicated the GC175 circuit and again got a good result. Interestingly, the modulator in the Globe champ 175 also uses triode connected tetrodes. Specifically, it uses push pull parallel 6L6s in triode configuration to modulate a single 812. That was a great little (big) rig. There was one at the W0ZUS auction in like new condition that sold for $160.00. That was a good deal.

In the first speech amp that I built, the GC175 clone, I believe I had about 300 volts on the plates of the 6F6's. Screen and plate tied directly together. Cathode bias with 700 ohm/15 watt resistor (one resistor for both tubes). These were metal envelope tubes and I remember they ran very hot. Perhaps they would have been happier with 250 volts on the plates. On the other hand, the tube data sheet that I have shows that you should run the plates at 350 volts for PP class AB2 service. Using -38V grid bias and 350V on the plates, power output is 13 watts, probably somewhat less with cathode bias I would assume. Plate to plate load resistance is 6000 ohms at those values. I would not ever hesitate to use triode connected 6F6's as driver tubes, the experience I had with them was very good. They will probably always be easier and cheaper to get compared to the 2A3/6A3/6B4 family of tubes.

I had forgotten this until today, but the speech amplifier/driver that I built for my homebrew rig is actually out of the Radio Handbook 15th edition (1959). The circuit is shown on page 652 and is practically identical to the GC175 circuit except that it employs 6B4 triodes instead of triode connected 6F6's. I substituted 6A3's in my unit since that is what I had on hand, but it is the same tube except for the base. Again, I believe that I have about 300 volts on the plates of the 6A3's. I am using cathode bias again, this time with a variable WW 1000 ohm resistor. The 6A3 plate circuit is metered, and the bias resistor is set for 80 mA of static current. Power output should be about 10 watts, and plate to plate load resistance is about 5000 ohms.

I have always wondered what prompted the engineers at WRL to use the triode 6F6 as a driver. I would guess that it had something to do with costs and mass production, but in 1948, I would think that 6A3's, 6B4's and 2A3's were all plentiful and cheap. Not true today of course. By the way, 6A3's are still commonly seen on Ebay. New ones typically go for $40.00-50.00. If you watch close, you can sometimes get a better deal. I got a group of four used 6A3's for $30.00 about a year ago. They are definitely out there if you are willing to pay the price. If you know how to do hi fi tube audio, they are probably worth it. For the rest of us who don't know how to use oxygen free copper wire properly, it's just for braggin' rights. ;)

Ron

Thanks Ron for that info. You did right to not re-invent the wheel, and that info's great and I'll look it up and keep it. For my part RF has always been the challenge, and even some audio stuff is tricky. I don't design tube audio as a profession, just have repaired many tube amplifiers over the years and built a lot of them.

The reason for all the interest in triode connecting high-grid-count tubes was initially for possible replacements for tubes like the 6N7 that I thought were scarce and costly (but it seems they are not, today). Nonetheless it's worth collecting as many examples as possible of triode connection operating conditions for reference. I guess the specs for triode connecting larger tubes like the 6CA7 or 6L6 have always been around but until I got started in RF I didn't realize the value of triodes for drivers.

]]]]

About those distortion figures, those are without feedback, so a great deal of improvement is possible. 2% isn't bad but vintage "storebought" ham gear seems to fall short of the mark.

Patrick,
            Reading your post the other day stuck with me as I attended an AWA swapmeet on Saturday. Just before leaving, I thought I'd browse through the tailgate section and ran across a fellow selling tubes for a buck apiece. Remembering the tube number, I picked up half a dozen NIB GE 6N7's. A PW project might be in order now.

Phil 

Just FWIW, The older Super Pros (SP-200, 210, Bc-779, and 1004) that everyone loved the audio from used a triode connected 6F6 as the driver and triode connected 6F6s for the outpoots.

I think the true pentodes lend themself much better to Class-A triode cnnected applications then the beam tetrodes. I'm sure there is someone out there that knows more than me that can explain why. There are also triode connected specs published for all or most of the "beam power amp" tubes like 6V6, 6L6, etc
But the triode connected 6F6, 42, and a few other pentodes were more popular.

Just another interesting sub note. Class-A and other non grid current classes of operation had the screen and supressor tied to the plate. Class-B and other classes that sucked up a little grid current always had all of the grids tied together as one. I guess having the grids tied to the plate gives a higher Mu which is more suitable for class-A operation.

The Super Pro triode phase started with PP 42's in the SP-10, 6F6's from the SP-100 to early SP-400 and 6V6's in the later 400's.

Carl

I made a 6N7 Jones oscillator transmitter a couple of years ago and it is a slick little unit.

Mike WU2D

I've done all I really want to for now. I did't fill out the sheet too well for the special class B triode connection. There is little info for it except in the modulator sections of the ham mags. Nothing a ruler and some plate curves wouldn't fix though.

Good! $1 is cheap for that much tube.



Are the gimmicks under the socket or are they the capacitances of the tube and wiring?

That is the very circuit I was referring to in a previous comment.

Great minds having one thought..... ;)

73DG

I can see who the real leaders of this operation are. That transmitter sort of has SBE written on it.. granted the crystal may not pull much.. How is the keying profile on CW? My friend Dennis W5FRS says that just a tiny amount of chirp sometimes makes it easier to copy.

The whole story is in all the Jones handbooks from the mid-30's.

Frank J. used it as the basis of many VFO designs before ECO circuits came into vogue.

73DG

There are some Hammond interstage transformers that might work OK with the 6N7 as a driver to another tube (p-p class A). Some specify 40-60mA in the primary, to drive the next stage grids. The only gripe with Hammond is the transformers spec from 150Hz up ad its hard to say whether that is with all that current flowing or not. So could be good or bad. I'm guessing that at 0.2W from an interstage application the frequency response will be better.

124A   TRANSFORMER TUBE DRIVER CLASSIC   124   
Hammond Manufacturing   
Tube Driver
150Hz - 15kHz
5 Watts
45mA   10K Primary
90KCT Secondary   
insertion loss ±1dB

they sound quite good. I have a 124D in my 4X1 rig, no one ever complains about the audio. I think the usable response goes well below 150 cycles. Don't forget that you would be using it for voice and not music. They are a pretty good little tranny for the price.

Can you say more about the application, what is the driver and what is being driven? That will help me.

Patrick,
           If you are asking me, I have it turned around for a 1.5:1 stepdown with a 12AU7 with both sections in parallel (class-A) driving a pair of P/P 2E26 AB2 grids.

I needed a modulator capable of around 20w and I didn't want to waste good audio tubes when I literally had a 5-gal bucket full of 2E26s. With close to 500v on the plates in AB2 those littls mothers can make some serious audio power for their size. (Like capable of in excess of 50w) Which is transformer coupled to the screen of the 4X1 with an ARC-5 mod transformer. (No stinkin controlled carrier)

I originally used one of the standard 3:1 step up drivers and it just didn't have enough BAs to push the modders into the grid current range. This made the audio kinda thin and crappy. When I changed it to the step-down configuration, the audio just came to life. I can make 125% without pushing it hard now. I just had to be able to push the little modders into grid current.

Since this skizmatic was drawn I have increased some of the parameters for dramatically improved performance.

Designer snippet:

2 X 2E26 = 1 X 815 (the R2D2 tube).

73DG

I was not originally but I am glad to have as many interesting answers as there are!  Turning the transformer around was something i was going to bring up as drivers are sometimes step down, and the Hammond transformers are listed as step up.

p.s. the 2E26 is a good audio tube according to its success in your application!


It is good news also about the Hammond frequency response sounding good because the alternative is Lundahl and very expensive for that response and quality level. unfortunately there, the spec sheets he has are also a bit confusing to me.



The "evil robot" tube..

I built a stero hi-fi power amp with a pair of 815's years (14?) ago, sounded pretty good, didn't get hot, and used Hammond iron all the way thru.  It was biased for true AB2 operation.

I had a set of OPT's wound by them with 10% voltage tertiary windings and fed that back into the driver trans (One of the potted broadcast jobs).  The interstage tube was of all things a 6N7G, paralled P+G.

Made way over the 35WPC my target was in the design, looked way cool with all those plate caps sticking up on those little squatty tubes. :D

I drove a pair of Altec A7's out in the garage, and at the time I lived in a subdivision so running it hard for any length of time got the neighbors riled up. :o
 
It disintegrated back into the parts bins, but I'm sure all that stuff is still here somewhere.

73DG

The  Gimmicks came from somewhere - probably an old article. I will have to look to see if I actually installed them. The Paraset 6V6 oscillator uses a feedback gimmick like that, plate to grid.

Believe it or not, I was thinking about using an 815 just to be different. Like you said it is a pair of 2E26s in one bottle. But................... 11 years ago when I built that thing I had 2 815s on the shelf and a 5-gal bucket of 2E26s. Not much thought process there........

<snip> ''I was not originally but I am glad to have as many interesting answers as there are!  Turning the transformer around was something i was going to bring up as drivers are sometimes step down, and the Hammond transformers are listed as step up." <snip>

Patrick, the 124D that I have on the shelf was listed to work either way. IIRC, it had the same current handling capacity on both windings. So by chosing the right combination of winding and whether or not to use the center tap and half of a winding for the single ended side, it gives you a lot of flexibility in how you use it.

IIRC, the older AES catalogs from a few years back actually listed it an useable either way. That is where I got the idea from.

Ok, been a while but this is the right place:

Summary Line up of zero bias common cathode dual power Triodes.

56 -  Medium 7 pin base (use same socket as 1625), 2.5V heater.  Pd 3W per unit or 5.5W.
6A6 - Medium 7 pin base (use same socket as 1625), 6.3V heater. Pd 3W per unit or 5.5W.
6N7 - Octal base, 6.3V heater, Pd 3W per unit or 5.5W.
6Z7 - Octal base, 6.3V, 180V B+, 8W Pd.
1635 - Octal base, 6.3V, 10W into 12K p-p, Pd 3W per unit or 5.5W. Curves go to 500V.
5608 - Medium 7 pin base (use same socket as 1625), 2.5V heater.  Pd 3W per unit or 5.5W. Reliability version.

That style of tube can be a lot of fun however real men use 01As and 45s
don

Frank Jones used the 6A6 in his famous (1936) PP xtal oscillator circuit....makes a great little breadboard xmtr for playing with CW and QRP.  I have made many contacts with this rig and variations thereof.  I even used a pair of 955 acorn tubes and a pair of 50L6s in the circuit. 6N7 will work fine, as well.

73,  Jack, W9GT

Needs varnish or at least a nice stain!
No worries about a fire?