input transformer: center tap of primary to ground

[neilium]

Hey guys, how's things? You're looking well.

But anyway, I'm lashing this circuit together this weekend:

http://www.larryseyer.com/rca.htm

I'm curious about the center tap (pins 3-4) going to ground on the input transformer. I haven't seen any other circuits, solid or hollow state, that have that arrangement. If anything, the center tap is where phantom power is applied.

Any thoughts on this? What about applying phantom power? If I have the center tap to ground and try to apply phantom power to the mic via 2 resistors, wouldn't the quickest path to ground be to ground be via the transformer's center tap, leaving the poor mic unfed?

Thanks in advance. Hope everyone's having a good weekend.

-neil

 

[rafafredd]

And probably killing the input transformer. Just disconnect the center tap from gorund and leave it open.

 

[ebartlet]

I don't see why you couldn't break that CT to GND connection and apply your phantom power to the CT.

 

[BYacey]

Grounding the center tap will not affect operation other than no longer having a floating input. Applying phantom at this point should be accompanied by a bypass cap to ground.

 

[rafafredd]

Just my opinion, but I would prefer not to apply phantom at this CT. Just break the connection of the CT to ground and apply phantom throught 2x6.8k as usual...

 

[NewYorkDave]

I agree with Rafael; just disconnect the CT from ground and apply phantom through two closely-matched 6.8K resistors.

 

[CJ]

Sometimes transformers will sound different if that center tap is grounded.
What this does is change the capacitance of the coils with respect to each other and the core.
I imagine that old RCA circuit used a transformer that benifited somehow by having the center tap grounded.
I doubt you are using the original iron, so just hook the centertap togeather.
cj

 

[volki]

if you do apply phantom throught the center, be sure to use a 3k4 from 48v to the center :wink: :green:

 

[FredForssell]

I'm curious about the center tap (pins 3-4) going to ground on the input transformer.

Ah... Ever wonder where the term "balanced input" came from orginally? A grounded center-tap input is balanced on both phases with respect to ground. If you lift the center-tap (which you can do) then the input is a "floating" input.

 

[neilium]

Fred,

As you could probably guess, I'm from guitar amp land. We're not too balanced here.

-Neil

thanks to all for the replies. Very helpful as always. :grin:

 

[Script]

Old thread, I know, but good info here. I came across the attached configuration which made me wonder. It links the primary CTs of two input transformers for L and R. The CTs here are NOT connected to ground -- but to one another. What's the purpose of this ? Does that change the capacitive load of the TXs wrt one another and still keep them floating ? Disconnecting the link does nothing obvious and measuring AC and DC on the CTs shows nothing - but maybe it does change something? Some program materials sounds maybe 0.1% different, but maybe I'm just imagining. Anyone has an idea what this link does ? Or maybe any ideas what else could be done with it?

 

[emrr]

You came across it?  DIY I assume?  Could it have been grounded at one time?

There are some 1930’s RCA inputs I’ve posted about here that misbehave if the CT to ground is disconnected.  Have not found a specific explanation.  Otherwise, every other old CT to ground I’ve encountered can be disconnected and left floating with almost no change in measurements. 

 

[Script]

Yes, some DIY gear. I think CJ above mentioned that a reference to ground from both CTs does something with the capacitive load (and I assume it probably also injects ground noise ?). And in that other thread about that  enigma transformer which missed CT to ground CJ wrote that such connection can slightly change the frequency response. Hmm, if so, then what would linking CT_L and CT_R do or affect ? Also frequency response, ringing ? One thing is for sure : it's subtle if anything at all. --///-- Hmm, speculating here but maybe linking CT_IN_L with CT_IN_R  'aligns'  the zero crossing points between TX_L and TX_R ? Does that even make sense ??