stereo di box -10dbv to +4dbu balanced

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

extraware

Active member
Joined
Jan 18, 2005
Messages
32
Location
Seattle
OK, I am certain that this kind of project has been discussed here, but I have run several searches and come up with instrument DI boxes.

What I need is a handy way to connect single-ended consumer gear to my console (all balanced inputs). Basically so i can easily connect cd players and tt/phono pre's etc........

I understand that it is basically inputs, +12db gain stage with balanced output stage. Any favorite schemes posted here? I am open to transformer or 'less designs and looking for "transparent" high quality not a "color" box.

Sorry if I missed an obvious thread............

Erik
 
you dont have an adjustable gain on your line input to your console?

All you need is a line amp, there are soooo many to chose from, knock your socks off. Ive used api 325s for tha puspose with great success, but you could build something incredibly cheaper than that.

dave
 
[quote author="soundguy"]you dont have an adjustable gain on your line input to your console?
dave[/quote]

Yeah I do. I see your point. I was just thinking that if I went to the trouble to build something I would make it have more flexibility. But maybe I am trying to re-invent the wheel here-and should, as you suggest, just use an existing or diy line amp.

Err.....humility inducing moment.

Erik
 
unless Im entirely misunderstanding what you are trying to do, bringing -10dbV unbalanced sources up to +4dbu operating levels is pretty much why there are trim controls on line inputs. Trimming your stuff up there is prefectly fine of course if you want a different sound than what your console is doing, you can build a line amp of any variety, I used to do this on a console I didnt like too much, I ran all my tape returns into the console through a variety of different line amps and got some great sounds that way, all depends on what you are trying to do.

dave
 
I have wondered about this also. Could it be done making a simple device of no color with a few opamps. Have old reels with the tape machine putting out -10 levels. Could use a 8 channel unit to go into my converters. Do not want to impart the colors of my consoles pre's on the tape. Will do that on mixdown. Buying a new device will set you back $250 about.

Thank you,
Chris
 
Audio & Design Recording used to make the 'Propak'.

There's more to it than line trim in a number of applications. Line trim doesn;t affect monitoring and/or metering, so the meters read way off and buss/tape switching results in a 12dB drop/jump depending on which way the error is.

A few other companies make more affordable versions than the propak, but all you need per channel is an op-amp trimpot gain trim stage, followed by a OPA134 (or similar) buffer. For the return trip, a differential receiving stage followed by a trimpot-set gain adjustment. This is all in manufacturers' data sheets. I've built several for CD players, cassette decks and Tascam crummy semi-pro stuff interfaced with +4 gear over the years.

It's not really a 'DI' at all, which is why you're only turning up instrument input stuff on your search... It's a level match/balancing interface.

Some cassette decks tended to overload their input circuitry before the record level control if you fed them with +4 signal, plus on consoles like Neves without buffered output distribution, unbalancing the signal to feed the cassette from the +4 output also unbalanced all of your 'real' recorders, so that was a no-no in properly installed studios. That and the fact that playback with or without the presence of a trim control) would scarcely tickle the metering at ful tilt, and the noise pickup risk of unbalanced... I always built the step-up/step down interfacing and everything was good.

Keith
 
This started, for me, in thinking about building one of those simple Jensen transformer circuits for a basic single-ended to balanced output box, but decided I wanted to build a transformerless device.

[edit]-ssl tech, good point-I was mis-applying the term "DI".

So I think that the better question is:

What is a favorite line amp circuit that I might DIY for less then $100 per channel that has great performance? there are kits for around $300/pair (borbleyaudio, etc.) Maybe the Jung super buffer followed by....?

Erik
 
-10dBV+4dBu.gif
 
Thanks for clarifying signal phase. And YES, use the 5532, NOT the 5534. And those 10 pFd caps are a good idea. I might even use 47pFd. I like bigger output resistors but that'ts a matter of taste and expected abuse.

But look again at your gain. Whatever U1 puts out, U2 puts out equal but opposite. And the total output is across both U1 and U2. So there is 6dB more gain than you seem to show. Since the "correct" gain is 12dB, my values are correct.
 
If you have more DIY-plans in you mind right now then an option might be to have a look at the Beh. Ultralink models (MX662 / MX882 if I'm right) and point your soldering iron at that other DIY-topic (that likely will be more the interesting one :wink:

The suggested circuits here will sound better though, since the suggested Beh.-boxes might do fine but could have an opamp upgrade nevertheless.

Edit, oops, no the '662 is only balanced in, unbalanced out, so please skip that - leaving the '882.
 
[quote author="PRR"]
-10dBV+4dBu.gif
[/quote]

Variant on PRR's design: connect a 10pF cap across the feedback resistor of each opamp (for stability). Insert a 100 ohm resistor in series with the opamp output, before the feedback resistor takeoff point (more stability). Use OPA604 amps or two halves of an OPA2604, and now you can ditch the input cap.

Peace,
Paul
 
> connect a 10pF cap across the feedback resistor of each opamp (for stability).

I agree with you and MC. (And I just corrected the "1-pFd" typo in my reply.)

> Use OPA604 amps or two halves of an OPA2604, and now you can ditch the input cap.

Disagree. 5532 offset is low enough it "should" not be a problem. That input cap is for protection against leaky consumer outputs. They are now rare, and I agree it isn't usually necessary to DC-block the input. I just think it is good practice to defend against the unexpected. Sure, for quick/cheap work, leave out the cap and see if it is a problem.

Agree that the '604 is "the new 5534/5532": low THD, over +/-20V rails, good output current. Also the 5532 is a bit noisy with the resistors I show, the '604 may be a hair quieter (either way, quieter than the average -10dBV source). Also there is some dissatisfaction with current-production 553x chips; maybe it has been in production so long that the mold is wearing out (actually: the masks may be copies of copies of copies, and run on very different steppers and wafers than they were taped-up for.)

> Insert a 100 ohm resistor in series with the opamp output, before the feedback resistor takeoff point (more stability).

I don't understand. Resistance between opamp output and feedback takeoff usually degrades stability. If you mean to take the 10pFd cap from opamp pin, resistor feedback after the 100Ω, that's not clear. And using couple-hundred ohm output resistors, keeping all else (including good supply bypassing) tight to the opamp, is usually plenty stable.

I just spotted this:
Yet another paper on opamp stability with focus on practical aspects.
 
Hi Pete,
I've been busting to post that link for a couple of days now but get the feeling that people are getting sick of our OLD web sites.
Mine could well have a blancing box also ... or even the old Group DIY pages.

I think there have been a few balancing units here at the LAB and at least one that combined an FET and a cheap transformer to suit many things from Guitar, Synth and other unbalanced decks.

search ??
you may find that
Pete, Joe, Fabio, Keef, Kev, Elliot, TK and other locals here ... even Eddie,
come up in the search results on a few of the search engines
 
BTW: if loads are always =>10K, and cables are under 50 feet, the TL07x series won't suck, and a TL074 gives you a complete stereo-pair in one $0.39 chip.
 
[quote author="PRR"]BTW: if loads are always =>10K, and cables are under 50 feet, the TL07x series won't suck, and a TL074 gives you a complete stereo-pair in one $0.39 chip.[/quote]

Isn't the fact that loads may vary a good reason to use an input stage/buffer before this arrangement?


Thanks all for all the insights and schemo's! I'll play around with it and post the results. (I know most of you already know what the results will be, but I am just on the beginning of my DIY journey...:)

Erik
 
[quote author="extraware"]Isn't the fact that loads may vary a good reason to use an input stage/buffer before this arrangement?[/quote]
The load won't affect the front end, so no. The load will affect the back end.

Keith
 
I've been considering balancing all my unbalanced gear internally--by adding XLRs where possible or swapping out existing TS 1/4" jacks with TRS ones, adding a balanced driver powered off the existing bipolar supply. In some cases, there's not a lot of physical space to add many components. Any experiences with the DRV-134 chips?

--Bob

BTW: Any idea if (and how) a DRV-134 could be phantom powered? Sort of an odd brainstorm idea: a balanced instrument cable--XLR on one end and phantom-powered active balancing inside a 1/4" TS plug on the other end!
 
[quote author="SSLtech"]The load won't affect the front end, so no. The load will affect the back end.
Keith[/quote]

Oops. Now I see the point being made-again posting b4 seeing!
 
Back
Top