Tube grounding diagram

[rp]

I am building a two-channel tube preamp with an external PSU. I've been reading up on grounding, and I've made this diagram to think it through and see if I understand it correctly. Does this look right?

Here are the "rules" I'm following:

- Safety/earth ground connected to PSU chassis at mains inlet
- B+ and heater circuits connected to PSU chassis near the cable.
- Pin 1 of input and output xlrs connected to preamp chassis
- Phantom power ground connected to preamp chassis at xlr input pin 1
- Both chassis connected to each other via cable.

In the diagram, TX = input and output transformers.

Thanks in advance for any input! Sorry the image is a little fuzzy.

 

[Tubetec]

Ive done up a few pre's similar to what you have in the drawing ,I generally go with  chassis of the pre  connected to screen and  back to the psu, and then provide seperate ground wires from say the neg side of a b+ decoupleing cap , I also completely isolate the signal ground from chassis in the preamp itself ,and again provide a seperate wire back to the psu ,doing it like that means that signal ground is completely screened and doesnt have to carry any induced rf or magnetic hum voltages that chassis/screen might pick up . With the heaters mostly I tend to provide the path to ground through a pair of resistors centre tapped to ground at the output socket of the psu ,if you happen to have a valve with a helical heater,like say an Ef 86 this arrangement will  give a usefull reduction in heater induced noise ,as you get a certain amount of noise cancellation with the 'balanced to ground' set up ,works good with both ac and dc heaters also ,generally though in say mic pre's your probably going to want Dc for lowest noise.
Heres an interesting article on the subject ,
http://studylib.net/doc/18350724/an-article-on-hum-in-valve-amplifiers

 

[Tubetec]

Oh one more thing I forgot to mention ,I usually keep my input transformer seperate from the pre itself ,theres a couple of reasons for this ,one it allows you to orientate its position for minimum induced hum ,and secondly you can sub in different ratios depending on the requirements .I set the grid resistor as high as possible for the valve in question then use a phono plug with a socket soldered to the back end of it with lower value grid resistor inside ,to suit the transformer in use . On balanced outputs I also leave it unterminated in the preamp itself ,that way when I have a very high level sound source I can match into a higher impedence ,say 20 kohm instead of 2 kohm and transformer core distortion is less of an issue ,depending on your own circuit arrangements and if your using nfb or not you may need more stable impedences for things to work as expected though.One of the downsides of Nfb is that it can effect the overload characteristics, triodes with no feedback tend to have a very gracefull  overload ,with high feedback the abillity to recover quickly from transient overloads isnt as good, also higher order harmonics are created when feedback amplifiers approach maximum output .

 

[ruffrecords]

Your scheme is basically sound. CHeck this out:

https://drive.google.com/open?id=0B_n67A1hN3qtSzNkQUhhdHR1TkE

Cheers

Ian

 

[abbey road d enfer]

- B+ and heater circuits connected to PSU chassis near the cable.

Are you using DC heaters?

 

[rp]

Tubetec - Thanks for the advice and the reading material! I look forward to digging into it when I get a few minutes. It's Interesting to hear your approach. I'm going to keep the transformers on the chassis, as I'm going for compactness, but it makes sense why you would keep them separate.

Ian - Thanks for the link! I used your document as a reference. If I'm correct, the difference between mine and yours is that mine lacks star grounding to the safety earth connection point.

Abbey - I was planning on elevated AC heaters. What do you think? This is my first attempt at such a project. I was wondering if dc heaters were necessary, and I took a gamble in design/parts sourcing and decided to go with AC. The circuit I'm doing (the nyd 6sn7 one-bottle) doesn't require elevation, but the hope is that I can use the PSU for future projects that may or may not require heater elevation. I was also referencing aronaut's PSU in this thread: https://groupdiy.com/index.php?topic=45990.0

 

[abbey road d enfer]

Abbey - I was planning on elevated AC heaters. What do you think? This is my first attempt at such a project. I was wondering if dc heaters were necessary, and I took a gamble in design/parts sourcing and decided to go with AC.

So you shouldn't ground it. You should balance it, either with two resistors of 68-150 ohm, or with a potentiometer (check any Fender amp schemo). You should do that also with elevated heaters.

 

[supersonic]

Articles like these are so darn important. And its simple. Whats hard is letting the people know...

Ive done up a few pre's similar to what you have in the drawing ,I generally go with  chassis of the pre  connected to screen and  back to the psu, and then provide seperate ground wires from say the neg side of a b+ decoupleing cap , I also completely isolate the signal ground from chassis in the preamp itself ,and again provide a seperate wire back to the psu ,doing it like that means that signal ground is completely screened and doesnt have to carry any induced rf or magnetic hum voltages that chassis/screen might pick up . With the heaters mostly I tend to provide the path to ground through a pair of resistors centre tapped to ground at the output socket of the psu ,if you happen to have a valve with a helical heater,like say an Ef 86 this arrangement will  give a usefull reduction in heater induced noise ,as you get a certain amount of noise cancellation with the 'balanced to ground' set up ,works good with both ac and dc heaters also ,generally though in say mic pre's your probably going to want Dc for lowest noise.
Heres an interesting article on the subject ,
http://studylib.net/doc/18350724/an-article-on-hum-in-valve-amplifiers

 

[rp]

Aha. I made an error in the diagram. I intended to show that the heaters were referenced to ground in some way, but did so by drawing one of the heater legs wired to ground. My intended circuit follows your suggestions, I believe. The heaters are connected to a voltage divider off the B+ supply via 100 ohm resistors.