Audio Asylum Thread Printer

Get a view of an entire thread on one page

FAQ / News / Events

Tube DIY Asylum

This circuit is balanced but not differential.

75.168.143.89

In Reply to: RE: Balanced input - is this correct? posted by Oldbean2 on February 04, 2024 at 19:07:56

So it needs an input transformer, otherwise you'll have no Common Mode Rejection Ratio (CMRR). Jensen Transformers makes good transformers for this sort of thing.

I'd use a 49.9K plate load at a minimum! The 100K is out to lunch.

If you set up a B- supply, you can run a proper differential circuit with good CMRR values without an input transformer.

Alternatively you could use solid state Constant Current Sources (CCS) replacing the resistor/capacitor values in the cathode circuits right now. But they would work better if you rectified and filtered the 12V supply. You could tie one side to ground and use the minus output for a B- for the CCS circuits. This would increase the effectiveness the CCS circuits to improve the differential effect.

Do not use a single-stage CCS- they hardly work at all. A 2 stage device or one that uses an opamp works much better.

No need at all for the rectifier tube. You can use some HEXFREDs and get the same low noise.

Follow Ups:

RE: This circuit is balanced but not differential. - rage 07:28:58 02/09/24 (1)
Ralph,

Will you elaborate a bit more on the need for an input transformer in absence of a CCS in the tail?

I was under the recent impression with a real balanced input it wouldn't be necessary.

Is there no CMR with a resistor tail load? (or just less)

How would a choke work instead of a CCS in the tail? (again looking at CMR)

Lastly - a single CCS vs a cascode - are we just talking better SNR or a difference in CMR as well? From what I've seen going to a cascode mostly improves the SNR from the -60db range to closer to -100db.. (and that may be generous)

I like simple circuits, less sand, no negative rail when I can help it...so I'd give up some SNR and CMR if I can get away from the CCS... just not sure of the actual performance difference.

Thank you!
- This circuit does not have a 'tail'. - Ralph 08:33:48 02/09/24 (0)
  Will you elaborate a bit more on the need for an input transformer in absence of a CCS in the tail?
  
  Of course it helps to have a balanced input. But tubes are not perfect and neither is the power supply. If you really want the best performance, you have to give it everything you've got. That includes regulating the power supply.
  
  There is a cathode bypass cap (100uf) used with the common cathode resistor 1800 Ohms). This creates a virtual ground and sets the operating point of the tubes. It provides for no differential effect at all.
  
  So if you want a good CMRR value you need an input transformer or a very good CCS. Otherwise the input has almost no CMRR at all.
  
  So that is why replacing the cathode circuit with a good CCS can sort that out and eliminate the need for a transformer. You can execute a good CCS with a couple of tube sections but you'll need a substantial B- supply, equal at least to the B+. If you design it right, if the AC line voltage were to vary, despite no regulation in the power supplies the input stage would see little change in performance.
  
  The tube CCS would need to be a cascode circuit (2-stage) so it can reject power supply noise. A single stage helps but has poor performance.
  
  But you can actually get semiconductors to do the job better than a good tube CCS, with about 2 orders magnitude better performance without the need for such a large B- value. Again a 2-stage circuit is what you need (or an opamp to sense current).
  
  A choke will not work as a CCS, since what you are looking for is wide bandwidth and the ability to regulate the current in the cathode circuit. A choke does neither.
  
  A good 2 stage CCS will improve the differential effect of the gain stage quite a lot over a simple resistor to B-, resulting in more gain, lower distortion, wider bandwidth and lower noise all at the same time. So its worth it to get it right.
RE: Input transformer - Oldbean2 17:42:17 02/07/24 (5)
Example schematic is from the Jensen website; is it as simple as replacing the 12SN7 100k grid resistors with 5k for each tube, to properly load the transformer's 10k secondary? (Plus the r and c off the xlr pin 1)

On the source side, the output impedance of my DAC is 188 ohms. Low enough?

I am using the Jensen as an example, but might use the Cinemag CMLI-15/15B2, which is 15k:15k vs the Jensen's 10k:10k. So I presume it would be 7.5k load resistors.

Plate resistors - I'll end up 30-40k, will have to see what's in the junk box. I'll keep the rectifier, that ugly potato masher fits the TV tube vibe.

Thank you VERY much for the help.

Back for a bit again. Ignore me if you like.
- take a look at this application document - Ralph 08:23:16 02/08/24 (4)
  The values in it suggest that the values you are asking about are incorrect.
  - JT-11P-1 (Open in New Window)
  - RE: take a look at this application document - Oldbean2 09:28:55 02/08/24 (3)
    I'm not sure that the application document you linked is the application I'm using; it appears to take a balanced line feed and convert it to a single unbalanced signal.
    
    The example I copied in my post (linked below from Jensen's site) keeps the balanced signal, with the + side going to one triode and the - to the other - as in my proposed amp. It shows the grid resistor for each triode at 4.99k to ground, for a total of (almost) 10k load on the transformer, matching the 10k secondary of the transformer.
    
    It seems intuitive to me. Or am I completely missing something?
    
    Back for a bit again. Ignore me if you like.
    - JT-11P-1 Balanced Tube Line Input Stage (Open in New Window)
    - My mistake - Ralph 10:18:41 02/08/24 (2)
      You got it right.
      
      The transformer presents about 13K load to your DAC with the dual 4.99K resistors. I don't know what I was thinking!
      - RE: Whew! Thought I had lost it for a minute - Oldbean2 12:02:19 02/08/24 (1)
        
        In Reply to: RE: My mistake posted by Ralph on February 08, 2024 at 10:18:41
        
        Thank you so much for helping out, I wouldn't have a successful build without it. I'll order a pair of input transformers.
        
        Kinda takes away from the junk box character of the amp, but that's ok, it's something new for me.
        
        Back for a bit again. Ignore me if you like.
        
        Yeah, sorry about that. - Ralph 12:09:04 02/08/24 (0)
        
        In Reply to: RE: Whew! Thought I had lost it for a minute posted by Oldbean2 on February 08, 2024 at 12:02:19
        
        Your circuit is similar to the one at the link. You might want to take a look. There is a kit for it IIRC
        http://www.recproaudio.com/diy_pro_audio/diy_files/eqp_1a/eq (Open in New Window)

Post a Followup:

FAQ

Post a Message!

Forgot Password?

Moniker (Username):
Password (Optional):
	Remember my Moniker & Password (What's this?) Eat Me
E-Mail (Optional):
Subject:

Message:	(Posts are subject to Content Rules)

Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:	Automagically notify you when someone responds.

To view your new posting or follow-up, click on the RELOAD or REFRESH button on your browser.

This post is made possible by the generous support of people like you and our sponsors:

Signature Sound

[ Signature Sound Lounge ]