Simple unbalancing and balancing line amp

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
I was under the impression that this was for a small home setup. The only time this arrangement is a problem is when the negative side of the signal line is shorted to ground. That scenario can accidentally happen in a patchbay. If there is a patchbay it’s a bad idea. If the only way it will ever be connected is balanced it’s arguably a better solution than cross coupled .
I use that DBX118 in a pro setup and with a patchbay ...
 

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
Another possibility is finding a second-hand SCV Universal Buffer. It's a card system. There are dual input cards (electronically balanced or xfmr-balanced), dual output cards (xfmr balanced) and combined input/output cards.
I sold tons of them in the 80' and 90's. Almost unused these days, you can find them for the price of iron.
They look just like what I would need, but they seem pretty rare on the 2nd market 😅
I’ll try to find the schematics and study them
 

Jen

Member
Joined
Sep 26, 2018
Messages
18
Hello
I have been looking around but haven't found much. I am looking to build a simple stereo unbalancing line amp and a simple stereo balancing line amp to use before and after my DBX118 in a balanced environnement. The idea is also to be able to play with the gain a bit, like this : input amp would mostly need attenuation or very little level boost, and output amp would mostly need amplification to provide makeup for extreme comp settings.
Since the comp is already pretty colored soundwise, I'm looking for something clean.
Anybody know some schematics ?
I could also use a board like this and tap some gain pots ?
THAT Corp makes excellent balanced line receivers and transmitters at the 'chip' level. For over 10 years now I have used only the THAT Corp chips, or high nickel core transformers for balanced interfaces. The THAT output device is configured to allow either node to be open, or shorted to ground, without upsetting performance.



My preference in line level transformers is Cine-Mag, though from time to time I also use Sowter and Jensen.
 

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
For some reason I cannot access the thatcorp website, Chrome Browser tells me the connection is not secured. I've been googling around and it sure looks interesting and quite cheap. Do you have a schemo of a typical output stage ? Is there a similar chip for a input stage ?


---------OR---------


At this point my other option I guess is the following : I went through my stash of parts and PCBs and found some API "Bel All" PCBs, based on the 312 but with many options. What do you guys think of using a board like this with no input transformer, a cheap DOA (or even better a 5532 if it can be done), then a simple Edcor output xfrmr ? I have a bunch of 10K:10K and 10K:600R
Here's the schematic :
 

Attachments

  • API_BE_ALL.png
    API_BE_ALL.png
    21.7 KB · Views: 1

ccaudle

Well-known member
Joined
Mar 18, 2010
Messages
498
Location
Houston
Chrome Browser tells me the connection is not secured

They apparently did not configure their certificates correctly, the certificate is for "web-hosting.com" which is their hosting service provider.
You can click the advanced button, view the certificate information to verify it looks like a simple configuration error, and allow the connection.
 

ccaudle

Well-known member
Joined
Mar 18, 2010
Messages
498
Location
Houston
See post #3

OK, we can quibble about whether not including a potentially redundant buffer amp makes it "wrong" or not. You also have to drive the V+ and V- pins with a low impedance power supply, does that make it "wrong" for not including regulators on board?
The device datasheet in the "Applications" section states that it "must be driven from a low-impedance source, preferably directly from opamp outputs, to maintain the specified performance."
 

abbey road d enfer

Well-known member
Staff member
GDIY Supporter
Moderator
Joined
Jan 22, 2008
Messages
15,000
Location
Marcelland
OK, we can quibble about whether not including a potentially redundant buffer amp makes it "wrong" or not.
I used the word "wrong" originally because, in the context of the OP's application, he wants to have a level control, so a buffer is a necessity, not an option.
You also have to drive the V+ and V- pins with a low impedance power supply, does that make it "wrong" for not including regulators on board?
Poor example, because the THAT IC's have excellent PSRR, so even with a half-decent PSU, the on-board 0.1uF capacitors guarantee correct operation.
The device datasheet in the "Applications" section states that it "must be driven from a low-impedance source, preferably directly from opamp outputs, to maintain the specified performance."
Yes, it is in THAT litterature, but not in Sparkfun's. The fact that Sparkfun have not included a buffer shows that either they have not read the specs or they have not considered the typical operation (where the signal comes from a source of moderate impedance like 100-1kohms or even more). It is the duty of a proper designer to consider the whole picture, not just a series of simplified diagrams, each dedicated to one particular aspect of operation. If they just had looked at the demo board, perhaps that would have caught their attention.
 

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
So I took a few days to think and realised to power supply I have in the DBX is +/- 12VDC and I'd love to stick to that to keep things simple :) Or I could use a +24VDC for the output, I suppose.

2520 DOA can work on +/-12V but that's their low limit. I can use the API Be All Boards, or get one like this : JLM Audio Shop but I can't find the schematics to understand how it's designed
 

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
I guess not, but I'm a little lost and I've found a few circuit designs with 2520 and they seem to do want I want : unity gain or more gain for GR compensation, and balancing with or without an output xfmr, works with +/- 12VDC
If I find an alternative with that uses TL072, or NE5532 or whatever other solution I am totally willing to follow that route :) But I haven't yet !

Edit : I have a bunch of DOAs, some PCBs for them, some transformers lying around ... so it's cheap. I realize it may be a little be over the top to put that after a crappy machine like the 118, and delivering only 20dB max when it gould be a great mic preamp or whatever.
 
Last edited:

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
I went for simplicity : I took out my "Small Signal Audio Design" book by Douglas Self, and got interested in these : what's your opinion ?


Input :

IMG_4808.jpeg


Output, 2 options :
IMG_4807.jpeg IMG_4809.jpeg
In both option, I'll see how I can make gain variation, i'll study the 5532 datasheet. 2nd option is almost too simple to be true, it def has my vote for the moment :)
 
Last edited:

JohnRoberts

Well-known member
Staff member
GDIY Supporter
Moderator
Joined
Nov 30, 2006
Messages
24,168
Location
Hickory, MS
It's still just an op-amp at the end of the day. A 2520 is probably less different from a 5534 than a 5534 is from a TL072.
not even close...
I went for simplicity : I took out my "Small Signal Audio Design" book by Douglas Self, and go interested in these : what's your opinion ?


Input :

View attachment 97341


Output, 2 options :
View attachment 97340 View attachment 97342
In both option, I'll see how I can make gain variation, i'll study the 5532 datasheet. 2 option is almost too simple to be true, it def has my vote for the moment :)
different approaches have different pros and cons...

You suggest simplicity is your primary criteria, transformers are simple (but not cheap, my primary criteria).

JR
 

thomasdf

Well-known member
GDIY Supporter
Joined
Aug 10, 2011
Messages
639
Location
Paris, France
I have mentioned earlier that I have a bunch of transformers lying around, waiting to be used : Edcor, Altec, Lundahl, no name salvaged from old BBC gear ... I was planning on using a pair of Edcors 10K:10K and call it a day.
 

ccaudle

Well-known member
Joined
Mar 18, 2010
Messages
498
Location
Houston
got interested in these : what's your opinion

If you are interested in simplicity, then that "super balanced" input stage is overly complicated and offers no benefit over a standard 4-resistor diff-amp configuration.
And if you are going to use that, might as well get an op-amp with the resistors on package and trimmed to match, like an INA137 or INA134.
And if you are going to do that, might as well get one with buffers so that the CMRR is better, like INA1650 or INA1651.
Hard to have more simplicity than one device.

2nd option is almost too simple to be true

Definitely a good choice if you already have the transformers. The big disadvantages of transformers are space taken, but doesn't seem to be a concern for you, and cost, which doesn't apply if you already have transformers that you consider "free."

Of course transformers are applicable on input as well as output if you have the right devices. If you get the right winding ratio you could have a completely passive input. Otherwise probably INA1650 with a trim pot after for level attenuation if you want is best performance.

not even close... [responding to 2520 vs. 5534 comparison]

OK, maybe a bit hyperbolic, but a 2520 isn't some kind of magic device. It can drive crazy high output current compared to most integrated op-amps, but do you need that when driving modern studio gear? It can run from 20V, but has a lot of output drop compared to a 5534. The 2520 was spec'ed for 15V peak output on 20V supplies, and a 5534 can drive 16V out from 18V supplies. The 2520 was spec'd to drive into a 75 Ohm load, but who does that? That just seems crazy these days.

But my original point still stands, it's not like most circuits using a 2520 can't be adapted for an integrated op-amp, the original post made it seem like it was some sort of impossible task to find a 5534 equivalent to the 2520 circuits.
 

Latest posts

Top