modifying a circuit for trafo-less output

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

jstark

Well-known member
Joined
Oct 2, 2004
Messages
114
Location
Austin, TX
I'm a college student.  I'm poor.  Transformers are the bane of my DIY existence.  I want to build circuits without them.

So, modifying the Country Hick DI, available at : http://www.diyfactory.com/projects/countryhick/countryhick.htm
I have this:

hick.jpg


This should work, right?  I just have to choose R1. 

Here's my R1-choosing-logic, please tell me where it's wrong:  The original circuit is driving a 2.5:1 trafo, which is pushing, say, a 600 ohm load.  So the output transistors were designed to see a 2.5*600 = 1k5 ohm load.  Applying this to the opamp circuit, R1||R1 = 1k5 ohm, thus R1 = 3k ohm. 

Is that right?

Assuming I'll never hook this to unbalanced gear, would the cross-coupled opamp scheme offer any significant advantage over this one?

Thanks in advance,

David
 
move the inverting input to a different node and think about the opamp circuit DC points.

Part of the DI use is hum reduction with the use of a transformer.
 
After giving this slightly more thought, are Q2 and Q3 even necessary anymore?  They're set for unity voltage gain, so they are only providing current to drive a line.  But now, that's what the opamps are doing too.  Could the opamps just get hooked up to the output of the FET?  Thanks,

David
 
You don't even need to use the fet. Use a fet input opamp

The coolness of the circuit is how it works and overloads and sounds.
 
Two thoughts:

First, by losing the transformer you also lose the galvanic isolation of whatever it is you're DI-ing from the preamp/console. That opens you up to ground loops. That may or may not be an issue, but keep it in the back of your mind.

Second, for driving any kind of real-world cables, there needs to be some build-out resistance on the outputs of those opamps to avoid instability. If you're going to be driving transformer-input preamps with these, I'd suggest 100 ohms per side, and match them as closely as possible using a DVM or impedance bridge to maintain common mode rejection.

Peace,
Paul
 
If you are going to skip the transformer and all of its benefits, I would also skip the opamps. Just use and impedance balanced output scheme like fig. 1 in the dself link. This keeps current draw in check with phantom and gives you the option to add the transformer later when you have a groundloop.
 
[quote author="jstark"]Here's my R1-choosing-logic, please tell me where it's wrong: The original circuit is driving a 2.5:1 trafo, which is pushing, say, a 600 ohm load. So the output transistors were designed to see a 2.5*600 = 1k5 ohm load. Applying this to the opamp circuit, R1||R1 = 1k5 ohm, thus R1 = 3k ohm. [/quote]
Impedances reflect as the square of the turns ratio, so the figure from your 600 Ohm example should be 3.75K.

You don't need the R1 to ground on the non-inverting opamp as you already have R1 to (virtual) ground on the inverting opamp.

That said, the comments from the other respondents are more important - the transformer is there for a reason.
 
Alright, so you've all convinced me that I *do* want a transformer-ed DI.  How low should the frequency response of the transformer extend?  The Triad cheapies I'm looking at in Mouser only go to 300Hz.  Low 'E' on a bass guitar is 41Hz, so I'm thinking I pretty much have to get a transformer that extends to 20Hz.  Is that right?  Thank you all for your help,

David
 
[quote author="Gus"]You don't even need to use the fet. Use a fet input opamp.[/quote]

JLM DI
http://www.jlmaudio.com/JLMDI.htm
and
http://www.diyfactory.com/projects/jlmsimpledi/jlmsimpledi.htm

but the Country Hick was more about the
The coolness of the circuit is how it works and overloads and sounds.
as Gus said
 

Latest posts

Back
Top