Transformer on Input or Output? Pros, Cons?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

ruckus328

Well-known member
Joined
Apr 14, 2010
Messages
906
Location
Philadelphia, USA
Thought I'd see what some thoughts are on this, I realize this might not be so black and white and subjective, but looking for any opinions on the matter.  I'm fiddling with a couple compressor designs, and the dilemma has come up whether to electronically balance or use transformers on the inputs and outputs.  It's modern in/out impedances I'm dealing with.  I do want some color, and would prefer to put iron on only either the the input or output and electronically balance the other (via THAT 124X/164X or similar) just for cost/simplicity purposes, but the question then is - iron on the input or output?  Are there generally any sonic benfits to one or the other?  Any other pros/cons?  Thanks.
 
Electronically balanced inputs have virtually no cost. A good input xfmr is very costly, because it must be capable of dealing with almost any source impedance, generally uses a high nickel content core.
An electronically balanced output is slightly more costly than unbalanced.
An output transformer can use standard M6 laminations, and deliver excellent performance if the output circuitry is properly designed. The cost increase between EBOS and TBOS can be minimal.
 
Thanks for the input (hur hur hur) guys, just the kind of stuff I was looking for.  I'm not trying to squeeze pennies out at the cost of sound (which is most important to me), but being that electronically balancing is so cheap and easy, I didn't want to deal with iron on both ends if there's not going to be any real sonic benefits.  I've always been under the impression the input xfmr was where it's at as far as mojo, but that's because it's just what I've always heard.  That all being said I am kind of leaning towards going the input route.
 
I think iron (or nickel) will "act" differently in different applications. I'd say test a couple of different things and see what sounds better.

btw, here is what Bill Whitlock had to say about transformer balancing a GSSL.

http://www.groupdiy.com/index.php?topic=8174.msg97033#msg97033
 
if you have three boxes, run them all together and use only input on boz A and output on Box C

that way you save the output on box A, in and out on Box B and input on box C.

when you split the mic pre from the eq and then the compressor, you need to guard angainst ground loops and noise,

but with them all in the same box, you do not need all that iron,

unless you Like iron,

in that case, all 6 or 7 transformers contribute to the sound.

 
From purely an integration standpoint, my perfect world at a minimum would be all inputs balanced.  The home studio revolution was a killer for dynamic range because people normalled all that semi-pro unbalanced equipment through patchbays, and there was hum all over the place.  Only with reducing the patchbay normalling and the selective balancing of certain equipment inputs could you achieve a noise floor close to a professional studio.
Like any project in life, you have to make a list of prioritized objectives and then ignore anything else.  Integration is usually low on the list!

I offer the Culture Vulture as an example.  It can be troublesome to integrate because of the unbalanced I/O.  For that $$$ they could have balanced the input at least, but that was obviously low or off the list of prioritized objectives.  Get it to work quietly and you have a great effect.
Mike
 
What is often overlooked is that only transformers offer galvanic isolation. Electronically balanced inputs and outputs will mitigate against hum loops but they won't prevent them. Only galvanic isolation can do that and only transformers provide it.

Cheers

Ian
 
The only time I haven't liked a transformer so far is in a 57!  Hahaha

No seriously, I do prefer Transformers specially good ones. I have heard a number of well designed projects with no transformers on the output that are great as well.  the THAT chips sound great along with a handful of others. 

 
Thanks guys.  I'm a little green when it comes to transformers I'll admit, starting to get the jist of it but still trying to wrap my head around some things, mainly impedance and loading issues.  For this thing I think I'm going to go with iron on the input, originally was just going to use a 1:1 (cinemag CMLI-15/15 or similar), but what I'd really like to do though is use the input xfmr on the la3a for it's particular color.  Thing is, it's a 1:10 step up, so assuming a +4dbu input signal I'd pretty much be be clipping the rails on the secondary right off the bat, so if I'm right I'd pretty much have to have a 20db pad on the input (which is already present on the La3a).  Don't know if this is a very practical thing to do - (knock the signal down 20db just to bring it back up 20db)  Other thing I'm still fuzzy on is, la3a has a 600 ohm input, but loading resistor is 39K, wouldn't that put it at more like 390R?  What am I missing here?  Don't know if going with a 600 ohm input impedance with a modern design is a good idea or not, I know 10K+ is the modern standard, but then everyone always favors the vintage 600 ohm stuff.

Urei_LA-3A.JPG
 
One other, terribly minor point (feel free to totally disregard or disagree) is that, assuming that your input transformer adds some 'color' of it's own, this will surely shape the compression to some degree.

If you want the envelope of your compression to more accurately track the signal coming into it from the mic pre (or converter), omitting the input iron would get you closer to this, while keeping a transformer on the output would add the ironic mojo after the compressor has done it's work.

Then again, you may like the additional shaping that the input transformer imparts. Who's to know which is better? It comes back to auditioning again - and one transformer sounds different to another, after all.

:)
 

Latest posts

Back
Top