+/-floating tube PSU for guitar ?! Schematic error ?

[papasteack]

Hi,

I'm about to build a little single ended really classic tube amp based on Blackheart Bh5h schematic found on the web (even if it would cost less to buy one ^^). On every schematic found, audio ground is floating. Chassis is connected to main earth, but after secondary, even input jack ground is not at earth potential, but is connected to earth with a resistor and a cap in serie (and not in parrallel). I'm worried, and confused, should'nt thoses two component should be at least in parrallel as so circuit's ground to be at earth potential while still reducing ground loop risks ? I really don't want to let psu floating, it's not safe. Is it a error, or i am missing something ?

Here a link to the schematic : http://rowbinet.files.wordpress.com/2009/06/bh5_schematic-rev-20070611.pdf

Edit ! found same thing on AX84.com on only one amp, a high gain, but without the resistor :
http://ax84.com/static/ubergain/AX84_UberSEL_091224.pdf


Thanks for help

 

[famousmockingbird]

Looks normally grounded to me. The input jack has two sleeve connections in the schematic........one goes straight to ground the other through RC to ground.

 

[Mooly]

I had a look for curiosity. My thoughts,

So the audio ground isn't connected directly to mains ground. I guess there are many arguments for and against but safety wise... unless its built to, and built using components (mains tranny for one) components that comply to double insulated specs then there is a risk.

You can connect the audio ground direct to mains ground... I'm surprised its not tbh. I guess the cap and resistor in series is meant at some attempt to tie the audio circuitry to mains ground at hf but even that is dubious imo.

I would say connect to ground, and if you experience a ground loop by so doing then look to using a 10 ohm ground lift resistor, and bypass that resistor with a four diode high current bridge connected so that you get two "diode volt drops" each way (so - an + linked on the bridge and AC input terminals used to connect across the resistor). The bridge has no effect other than under fault conditions when it will pass the full fault current to ground.

 

[famousmockingbird]

The two schematics are different 😕

 

[Merlinb]

You don't need to connected the audio ground to chassis, but you can do, if you want. Most people do.

 

[famousmockingbird]

The designer details grounding in construction notes on that ax84 SEL schematic.

 

[papasteack]

Ok, so here are where my reflexions have come now.

Looks normally grounded to me. The input jack has two sleeve connections in the schematic........one goes straight to ground the other through RC to ground.

Ground and earth. Chassis is at eath potential. But not ground as chassis is connected to ground only by a resistor in serie with a cap. And in guitar amp, ground go straight to guitar ground, to strings, to human body. So, wich potential do we want to be connected to ?...I vote for earth. But, there's the ground loop thing...(see bellow)
First, for safety, but from what i've readen, it gives too a path for human body noise to earth when touching strings (human body, from what i've readen, is "noisy" and not a "ground"), and for strings (other antenna) noise too. But it's only what i remember having readen somewhere, maybe on geofex, don't remember...

[...]
I would say connect to ground, and if you experience a ground loop by so doing then look to using a 10 ohm ground lift resistor, and bypass that resistor with a four diode high current bridge connected so that you get two "diode volt drops" each way (so - an + linked on the bridge and AC input terminals used to connect across the resistor). The bridge has no effect other than under fault conditions when it will pass the full fault current to ground.

I thinks i'll do as you said mooly. I just remember now it was well explained here http://www.valvewizard.co.uk/Grounding.pdf as ground loop breaker, but when i was reading this, i didn't have a tube construction project, so i didn't really though about it.
Thanks for confirming, i was doubtful, and since this amp will be for a very good friend, i prefer going for the safer way.

The designer details grounding in construction notes on that ax84 SEL schematic.

In the pdf, it's only the standard ax84 pdf notes, that do not seem to correspond to the schematic, as on it, there's a cap between earth and circuit ground (on jack). If the note was good, there wouldn't be any use of adding this cap in parallel of...a wire ^^. Or maybe have you found other construction notes somewhere else ?

 

[Merlinb]

If the note was good, there wouldn't be any use of adding this cap in parallel of...a wire ^^.

Actually there would: If someone makes the ground-to-chassis connection somewhere other than the input, the cap still ensures a good RF ground connection at the input. That's why the cap is there.

Of course, if the ground-to-chassis connection is made right at the input then the cap does indeed become redundant.

 

[trobbins]

I doubt the amp would pass common safety standards if it was a commercial product. Unless the amp gnd is connected to protective earth (mains earth), then it is regarded as floating and so any part of the circuit is deemed to be at up to B+ from protective earth (ie. 0V ground for everyone outside the amp).

 

[Merlinb]

I doubt the amp would pass common safety standards if it was a commercial product. Unless the amp gnd is connected to protective earth (mains earth), then it is regarded as floating and so any part of the circuit is deemed to be at up to B+ from protective earth (ie. 0V ground for everyone outside the amp).

But can you point to an actualy safety standard that is violated by this? I can't.
The sort of safety bodies that deal with consumer appliances liek audio amps are interested in mains voltage safety. They don't care about or check what happens on the secondary side of the transformer (until you get into the kilovolt range, possibly. I wonder if the secondaries of microwave oven transformers are earthed? I suspect the EHT of CRT televisions isn't).

 

[Mooly]

I suspect the EHT of CRT televisions isn't).

It is referenced to the chassis which may in a really old set have a direct connection to the mains neutral. So in a sense ground referenced when you realise neutral and ground are at the same potential in many installations. The chassis was non touchable though and could be live with reversed polarity of the mains lead. The customer was safe at all times. The only connection to the outside world was an aerial socket which was isolated (I think to around 2 to 3kv from memory).

Modern CRT sets still have the EHT chassis referenced. An SMPS ensures isolation mains isolation and there is no direct grounding. The chassis is user accessible via all the typical connections and can and does carry quite a high (albeit low current level) leakage. Its easily possible to "draw an arc" to any connected item if you deliberately try.

 

[trobbins]

60950-1 has a section on protective earthing (clause 2.6.1) that requires parts likely to carry fault current and assume hazardous voltage level in event of single fault. Hmmm, possibly applicable if B+ faults to PE. Also basic insulation requirement for unearthed hazardous voltage secondary circuit.

We have a generic appliance standard AS/NZS 3100, for use where no other standard is more applicable. The PT would need to be rated for double or reinforced insulation to allow someone to touch say the input circuit (which would be regarded as exposed metal), otherwise all exposed metal parts are to be effectively earthed.

But its Friday night and brain is too numb to work through what is valid and what I am not properly comprehending - sorry.