1940's GE YGA-4 Audio Oscillator, My Frankenstein Preamp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

millzners

Well-known member
Joined
May 7, 2010
Messages
219
I've had in my possession for some time this old relic.  It's a GE Audio Oscillator, complete with manual and a sales receipt from 1949.  Today I turned it on for the first time, and found that it kind of works.  It has an output, but it's just a constant 120Hz ugly looking sine wave.  The frequency knob does not affect the signal in any way.  The gain knob does work, and you can get from 0 to ~8 VAC output.

So, if you were the proud owner of such an ancient and somewhat functional piece of equipment, what would you do with it?

Some options:

  • Fix it, and use it as it was originally intended.
  • Take the working components out of it and use them in something.  It does have 2 (assumably working) GE 6SN7-GT tubes which I have to assume are the originals, plus 2 6J5, and 1 5Y3-GT tubes, and a couple different xfo's.
  • Tap into the audio amplifier and use it as a preamp.
Any other ideas/thoughts/suggestions?
 

Attachments

  • YGA-4.jpg
    YGA-4.jpg
    163.4 KB · Views: 146
It's a heterodyne oscillator, where the AF is produced by the beating between two RF oscillators, which outputs are "mixed" (but in fact the AF signal is the result of intermodulation). I'm not sure it can qualify for low distortion. The only advantage I see in this design is the short settling time.
If I was in your shoes, I'd check if it has any appeal to collectors; then if I could resell it for 400-500 bucks I would fix it, sell it and treat myself.
If it's not that valuable, maybe I would use it to make a frankenstein preamp. The output xfmr is probably good enough since this piece is supposed to output up to 18kHz.
 
I'm on the side of the frankenstein preamp, but I'm torn.  On one hand, there's a sense of destroying something that's survived intact for 62 years including the original freight and sales receipts...  For me to just scrap it seems disrespectful to the person who took such good care of it for 40+ years.  On the other hand, to create a preamp that's totally unique and use it to create new music is kind of cool...
 
Plenty of ways to create something totally unique without scrapping this thing, lots of beater carcasses on ebay and at flea markets every day that could be used.  I'd say the extra documentation is "added value" to any collectors.  Should be easy to repair, if you want.
 
After some more research on tube design (I'm new to tubes) I've decided to work my way through the process of making this a frankenstein preamp.  Ultimately there's enough tubes in there to make a stereo preamp, although it's pretty much the same tube over and over again...  In any case, that's probably my end goal.  Oh, and to preserve the thing so it looks the same b/c that's part of the appeal.  I'd love to keep the tube eye working.

So first order of business will be disconnecting the oscillator portion of the circuit and the LP "T" filter, and then solder in a 1/4" pigtail on the 240K R11, plug in a guitar and see how it sounds...  Maybe R11 should be replaced with a 1M, we'll see.  If it works then I'll pull out the "Zero Adjust" and replace it with R14 level control so the front panel will have gain and attenuate controls.  Maybe I'll be able to push the thing and get some of that illustrious tube distortion, or maybe not.

If I get that far, the next step will be to get a microphone plugged into it, which will eventually require a transformer, although I think I can get away with using my Jensen-designed reamp box to start...  I'm not trying to go all in until I know this thing is worth the time and money.

The reason for this post is to provoke some thoughts on what to do with the 3 other tubes (see full schematic above) if I get this far without any trouble.  The 6SN7 mixer tube can be repurposed for another channel.  And the oscillator tubes are just half a 6SN7, right?  So maybe the next step will be getting them biased and working as a 2nd stage after the 6SN7...  Any other ideas?

Oh and I'll be careful with the high voltages and discharge the caps and all that.  I've read mountains of safety advice and tips for working with tubes, its time to put them to use.
 

Attachments

  • 1st step.pdf
    13.2 KB · Views: 24
From the attached schematic it looks like you're just a few steps away from what would essentially be one of NYDave's one bottle amps. (The simple SN7 version)

1) remove R20

2) Add a bypass cap from the top of R12 to ground - 25V or > rating,  22 to 470uF

3) Add an input transformer.  Since you don't have one at the moment, you could use a Hi-Z mic/converter to plug a mic directly into the grid to test it.  If it's convenient, replace R11 with a 1M.

Check for any DC voltage on the grid side of C16.  If it's 1V or more you may want to replace that cap.

 
lassoharp said:
From the attached schematic it looks like you're just a few steps away from what would essentially be one of NYDave's one bottle amps. (The simple SN7 version)

1) remove R20

2) Add a bypass cap from the top of R12 to ground - 25V or > rating,  22 to 470uF

3) Add an input transformer.  Since you don't have one at the moment, you could use a Hi-Z mic/converter to plug a mic directly into the grid to test it.  If it's convenient, replace R11 with a 1M.

Check for any DC voltage on the grid side of C16.  If it's 1V or more you may want to replace that cap.

I attached the full schematic above if you want to take a look.

I'm a little concerned about why the schematic doesn't show a plate resistor connected to my B+ on the 2nd stage of the 6SN7 (pin 5).  I'm wondering if it's a typo or if you can power the tube with B+ on pin 2 alone?  That doesn't make much sense to me so I'll definitely be investigating that before I do anything else.
 
millzners said:
lassoharp said:
From the attached schematic it looks like you're just a few steps away from what would essentially be one of NYDave's one bottle amps. (The simple SN7 version)

1) remove R20

2) Add a bypass cap from the top of R12 to ground - 25V or > rating,  22 to 470uF

3) Add an input transformer.  Since you don't have one at the moment, you could use a Hi-Z mic/converter to plug a mic directly into the grid to test it.  If it's convenient, replace R11 with a 1M.

Check for any DC voltage on the grid side of C16.  If it's 1V or more you may want to replace that cap.

I attached the full schematic above if you want to take a look.

I'm a little concerned about why the schematic doesn't show a plate resistor connected to my B+ on the 2nd stage of the 6SN7 (pin 5).  I'm wondering if it's a typo or if you can power the tube with B+ on pin 2 alone?  That doesn't make much sense to me so I'll definitely be investigating that before I do anything else.

That's a transformer coupled feed - the plate doesn't use a resistor here, the B+ is fed directly to the plate via the transformer primary, which acts as the load.

Safety first - make sure there is a correct value fuse and that it has not been bypassed.  I would install a 3 prong AC cord , safety ground to chassis.  Make sure the fuse is reinstalled in the hot side of the incoming AC.  If the caps or rectifier shorts the fuse should catch it but be mindful of the type things that often happen in this condition.  Keep a hand on the PT and filter cap can noting whether they start getting hot really quick.  Tube rectifiers can make a 'cracking' sound (like ice) when they are about to give out.  These things may precede the smells and smoke.  All of these bad things shouldn't happen with proper value fuse installed.

Ok, I see you've already powered it up without issue.  Not having a 3 prong AC can still be risky. Definitely verify the fuse rating.
 
I isolated the "1 bottle pre" portion of the circuit input and output going and plugged in a bass guitar.  I have a small output signal from the bass which is overpowered by the 120Hz 2-3Vp-p signal which I'm thinking is b/c the 60 year old power filter caps coming off the rectifier tube are just not up to the task anymore.  So I'll definitely need the 1M on the input to help out, but getting new caps is a bigger priority.

I'm going to put a 3 prong power cord on there before I do anything else.  The new earth ground just gets soldered to the chassis, right? 

What's interesting to me, having never worked on an old piece of equipment like this, is that there's two externally mounted metal tubes that look like a transformer, except each one houses 2 or 3 capacitors.  You can see on the schematic there's C11A, C11B, C11C, C11D, that's all one 4-terminal device, and each terminal is a separate cap.  I actually found an unused section of C12 (50uF 25V) which I used to put in parallel with R12 for the bypass cap in the 1st stage.  

Anyway I wonder if those transformers coming off the oscillators would be good for anything.  I can't find much information on them, but it would be cool if I could use them in the circuit for color.

So I guess I'll dig around to find some 450V 10uF caps and see if I can get some clean DC to the tubes.  I uploaded a picture of the rats nest I'm working on.  The tube I'm concentrating on is the top left most one.
 

Attachments

  • IMG_5085.JPG
    IMG_5085.JPG
    328.4 KB · Views: 82
The new earth ground just gets soldered to the chassis, right? 


Yes.  If you can't find space you can use a spade connector and attach it where the PT is bolted to the chassis.  Remove any old paint if needed and add a lock washer if there's not one.


Does sound like it may be old filter caps going bad.  May be a good idea to up the value on C11C to 22, 33, or 47uF.  That cap feeds the 1st stage which needs to be really clean. 


 
millzners said:
What's interesting to me, having never worked on an old piece of equipment like this, is that there's two externally mounted metal tubes that look like a transformer, except each one houses 2 or 3 capacitors.  You can see on the schematic there's C11A, C11B, C11C, C11D, that's all one 4-terminal device, and each terminal is a separate cap.  I actually found an unused section of C12 (50uF 25V) which I used to put in parallel with R12 for the bypass cap in the 1st stage.  
Very common in tube equipment, multiple electrolytic.
Anyway I wonder if those transformers coming off the oscillators would be good for anything.  I can't find much information on them, but it would be cool if I could use them in the circuit for color.
These are not audio transformers, they are RF types, destined for about 250kHz operation. I don't think you can use them in an audio project, except for decoration.
 
some oscillators built for lab duties etc can have some decent audio preamps in there. stick with it u could be on to something.

i know a guy who uses an old military decibel meter as a mic pre and it sounds good!!!
 
leadbreath said:
some oscillators built for lab duties etc can have some decent audio preamps in there. stick with it u could be on to something.

i know a guy who uses an old military decibel meter as a mic pre and it sounds good!!!

There's only one way to be sure:  plug 'em in and see what happens.  There's a remote possibility they will be awesome.  I'll follow these test setups tomorrow:

http://www.crystal-radio.eu/entrafometing.htm

Unfortunately I don't have any 450V caps for the PS so I'll have to hit mouser up.
 
The xfo's don't seem to like audio.  Although the transformer is wired in a really strange way.  Internally pins 1 and 2 are wired together and then to the core itself for a single pin on the outside.  On the schematic it looks like pin 4 is the input, 3 is the output, while 5 is AC coupled to ground.  Anyway I ran 5V 1KHz into pin 4 and pin 5 with similar results on pin 3, despite a 1 Meg load or a 1K load I only saw a few mVpp output signal.  As I cranked up the frequency upwards of 500kHz the amplitude grew and grew...  There's still the issue of the signal generator being 600ohm output and the input Z of the transformer being totally unknown, as well as the output Z.  So maybe there's a chance I didn't load the thing correctly, but still, as frequency climbs into the 100's of kHz so does the output voltage, that seems to tell me all I need to know.

I think its a lost cause, maybe I'll come back to it but I didn't get the impression this is worth pursuing further.


Edit:
I ordered all the parts to turn this into a stereo NYD 1 Bottle 6SN7 pre, so I'll use that as a starting point and work from there.  If that works my next question is how should I incorporate the other two tubes, the 6J5's.  Pre or post 6SN7 stage?  Maybe I could use the Gates Radio 7x design http://www.groupdiy.com/index.php?topic=43151.0 for inspiration.
 
Seeing that it is tuned with a 2500pF cap at 248kHz, the inductance is about 160 microhenry.
At the highest audio frequency, 20kHz, its impedance would be 250 ohms, at 20 Hz 0.25 ohm. These are definitely RF.
 
I ordered all the parts to turn this into a stereo NYD 1 Bottle 6SN7 pre, so I'll use that as a starting point and work from there.  If that works my next question is how should I incorporate the other two tubes, the 6J5's.  Pre or post 6SN7 stage?  Maybe I could use the Gates Radio 7x design http://www.groupdiy.com/index.php?topic=43151.0 for inspiration.


I see some definite possibilities with the 6SN7 common plate mixer that's already wired in.  As a line mixer for use in parallel compression apps or FX blending.  It should only take one more OT (Edcor 10 or 15K:600) and a few resistors and pots to do it.  I could draw something up if you're interested.
 
Back
Top