Converting tube oscillator to a preamp?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

syn

Well-known member
Joined
Oct 11, 2005
Messages
653
Hi
                I have an old 1031-A General Radio low distortion tube oscillator that I'd like to convert to a some kind of  preamp for  general studio use...gain make up stage for a passive eq  would do as well...
                As far as  I could get it, audiophiles regard 6y6g as a very good tube and a VERY large output transformer promises a lot. 
                All tubes tested  and are OK.                       

Any ideas  are welcome.

Tubes used: 6SJ7, 6Y6G, OD3, 6W6, 6X5GTG, 6SK7,6SQ7, 6SL7GT.
Output transformer ??:600r
Choke: 500mH
Data:(low quality,  sorry)

Cheers

 
I have six preamps which I made using the parts of H/P 200CD oscillators. I bought all the units on ebay for about 25 bucks each. They use 6CW5 output tubes in a grounded plate circuit, using the cathodes to drive the transformer, which is quite large. Some of the other H/P units, like moder 200, use 6Y6 tubes in a standard push-pull circuit with a different output transformer. It did take a little engineering to come up with the circuit I wanted ahead of the output stage. I wound up using some old Triad mic input transformers with the circuit balanced from input to output. The ones I made sound as good as any tube preamp I've heard, or commercial ones I have in my collection. The iron in these things was pretty good and made to last. Also added a bit of "character" to the sound. Good luck!

Jim
 
Thanks, good to know it is doable... but, i've spent 4-5hours yesterday looking for a schematic using 6y6g in ANY
configuration but nothing came up. Where to look for schematics using 6y6g's? Anyone willing to share?
Thanks
 
> 1031-A General Radio low distortion tube oscillator that I'd like to convert to a some kind of  preamp

I think it is a shame to cut it up. But I'm not going to pay for shipping from Serbia.

The "obvious" thing is to snip the tuner and AGC networks. The 6J7 (or 6SJ7) is a VERY fine voltage amplifier. The 6Y6 is used as a fat dumb cathode follower, which is good. The output attenuator and transformer system needs no change. And you know what the transformer is: the plan shows "5000 ohms" direct, "600 ohms" through the transformer.

Apply a signal from Bias Battery to 6SJ7 grid. Or, since that battery may be rotted and not all signals can float DC, break 6SJ7 cathode and insert 1K with 47uFd to ground for self-bias. You want to trim the 1K so that 6SJ7 plate is still near 100V.

6SJ7 gain is about 100-200, 6Y6 unity, transformer about 3:1 step-down.... overall gain 30-60, 30dB-36dB. As a mike-amp you will use 1:5 or 1:10 mike transformer, total gain 500-2,000 or 54dB-66dB. Maximum input at 6SJ7 is around 1V, so you will need a pad with hot mikes and loud sounds.

Very few people have more tubes in their power supply regulation than in their amplifier.

15g3sqd.gif
 
> looking for a schematic using 6y6g in ANY configuration

gosh.... it is just a power pentode. Optimized for low voltage high current. In fact it is pretty-near a 25L6 except with 6V heater.

"Usually" you use 25L6 when you have LOW supply voltage (such as 105V DC from US 117VAC wall outlets) and need significant audio power. Then you will flow the tube's DC current through the primary of a transformer connected in the tube's plate circuit. This needs a transformer designed for DC, and has high distortion, but it works. Can get several Watts this way. 25L6 can also be used up to 200V DC; 6Y6 is variously rated to 200V 250V, or 300V depending which spec-sheet you look at.

BUT: your oscillator claims "low distortion". So the 6Y6 is used Cathode Follower. And the compromises between bandwidth and DC current are harsh. The oscillator does not need BIG output power, probably only 0.1W. So the 6Y6 is resistance-loaded, DC current through a resistor, then a DC-block cap to attenuator and switch/transformer. This will not drive a loudspeaker LOUD (even if you had a 600 ohm speaker), but is plenty for a bench oscillator. And plenty for any studio "line" level.

I really suggest you keep the 6SJ7-6Y6-OT system as original as possible. It is already a fine amplifier, just buggered to oscillate. It is already built, by beautiful(?) ladies in Patterson NJ who soldered better than almost anyone alive today. You can't buy parts as good as GenRad used routinely.
 
PRR
        Thank you very much.

PRR said:
I really suggest you keep the 6SJ7-6Y6-OT system as original as possible. It is already a fine amplifier, just buggered to oscillate. It is already built, by beautiful(?) ladies in Patterson NJ who soldered better than almost anyone alive today. 

       I will, it is so simple and the parts used promise very interesting results.

PRR said:
You can't buy parts as good as GenRad used routinely.

This thing is extremely well built,  every part is work of art.

I tried to draw what I actually understood. Any mistakes?  I have no experience with tubes, meaning this is my  first tube build. I'm thinking of adding passive eq stage ala Pultec (schematic of  passive eq by Gyraf) as well. There is plenty of make up gain, and thanks to parts/ plan used I think results will be very interesting...Also, I'm thinking of changing to SolidState  PSU.

eq.jpg



 
You need to have the cathode resistor and capacitor in parallel in that first tube stage. In series, like you have them, there is no D.C. path for the tube. You should also think of maybe hanging a resistor off the grid of the 6SJ7 to keep is stable while switching between the mic and the EQ. A gain control off the output of the EQ would also be nice, since without one, you'll probably overload the preamp.

Jim Zuehsow
 
One more thing. I usually like a jack switching arrangement that disconnects the transformer secondary when going from balanced to unblanced. In the EQ section, the unbalanced input still sees the transformer secondary as a load. Not a good engineering practice.
Also, the 47K resistor at the output of the EQ should be eliminated as it really serves no purpose. That capacitor doesn't have a charge on it so it doesn't need a drain. Puting a grid resistor there at the first tube, where one belongs, would have both resistors in parallel. Depending on what transformer you use, you may need a terminating resistor like you show. Some transformers perform better with a Zobel network consisting of a series resistor/capacitor, then again, some transformers require no termination. If I were doing this, I would probably use a fixed pad at the EQ output to drop the line level out of the EQ down to mic level so the amplifier would see something in the range it could handle. You would have to measure the loss through the EQ to determine the pad value, which will just be a simple voltage devider.
I have never liked driving a transformer from a pot. The transformer sees a different source resistance as you change level. I would connect the transformer directly to the D.C. blocking capacitor, and use a 600 ohm balanced "T" attenuator to control level. That keeps the transformer happy, looking into a constant load. I must admit I generally prefer trimming the gain at the input, rather than the output of a mic pre, or by using negative feedback to adjust gain. YMMV.

Jim Zuehsow
 
Jim

         thank you very much. 
         Sorry about eq input confusion,  cleaned it up.
         Last night I played with the oscillator and it sounds that good, that is going to get sampled  . Really BIG.
I hope I'll not loose that character in the process of conversion to a preamp/eq.
         I am aware of the fact that pot driving an output transformer is not the best solution (for the reasons you already explained in your post) however I'll give it a shoot at least until I finish the preamp and then listen to the results.
         At the same time I'm thinking of doing Redd47 alike first stage and negative feedback network (no out pot in that case, but  still ,maybe, a balanced "T" ). I'll have to make up my mind at some point...
         Shall I leave grid 2 and 3 floating (first stage)? Does input to the first stage looks better now? Suggestions please:

eq2.jpg
 
That's looking better. You can't just let those grids float! Tie pin 3 to pin 5 for starters. Then connect pin 6 to B+ using a 2.2 megohm resistor along with a .047 ufd cap from pin 6 to ground. The 6SJ7 doesn't like to see more that 100 volts on grid # 2 (pin 6). You will still need to pad the output of the EQ to mic level in order to not overdrive the preamp, but you can do that once the unit is built.

Jim
 
Jim

      thank you very much, once more. Below is the updated diagram, hope I didn't  miss something.
I'll sort eq out level once i build the pre.
Thank you.
eq3.jpg


 
emrr said:
Cool; we're DC coupled between tubes!


Emrr

        thank you. Good point, but we are looking at the strip down schematic (General Radio is calling it elementary schematic diagram, downloadable from the first post).
Coupling caps are on the actual unit, and I'll have to make a proper schematic showing them,
once i get into the guts. For now mine main concern is the  first stage as it is the one that needs moding...

Cheers

Milos
 
Back
Top