It seems like this subject became pretty popular recently.
The easiest way would be using DRV134, however, I was thinking about something more challenging and less usual.
Combining some schematics I've seen previously (basicaly Pass balanced GC and Borberly preamps), I came up with this. Would it work well in that application?
There is also a possibility of using regular op amps for conversion, and I will post some examples later.
The easiest way would be using DRV134, however, I was thinking about something more challenging and less usual.
Combining some schematics I've seen previously (basicaly Pass balanced GC and Borberly preamps), I came up with this. Would it work well in that application?
There is also a possibility of using regular op amps for conversion, and I will post some examples later.
Attachments
Seems like the subject is challenging indeed, as I only received one comment so far (by e-mail), mixed advice though, and I'm not sure if I could even consider it, especially when it was trying to correct Mr. Borberly's choices. I'm sure more e-mails will come. 
In a meantime, we move to ICs, as they are always safer to play with. This is my vision of pretty good, op amp preamp. It is in fact based on an actual, very well sounding preamp design, so no funny e-mails this time please.
I guess, for bridging purposes, the circuit could be used without input buffer (gain stage), but if somebody plans to add volume control, the buffer may be actually preferred. I'm using here OPA627, and it sounds not bad. The phase splitting is done using OPA2604, and in this position they also sound fine, but Carlosfm suggested recently AD815 and I wouldn't mind trying those here, especially when they are described as balanced line drivers.
The two 1k resistors in series, between two op amps, were used like that on purpose and one can add another gain stage for truely balanced preamp operation at the point between them. There is also a switch allowing to change the gain setting.
In a meantime, we move to ICs, as they are always safer to play with. This is my vision of pretty good, op amp preamp. It is in fact based on an actual, very well sounding preamp design, so no funny e-mails this time please.
I guess, for bridging purposes, the circuit could be used without input buffer (gain stage), but if somebody plans to add volume control, the buffer may be actually preferred. I'm using here OPA627, and it sounds not bad. The phase splitting is done using OPA2604, and in this position they also sound fine, but Carlosfm suggested recently AD815 and I wouldn't mind trying those here, especially when they are described as balanced line drivers.
The two 1k resistors in series, between two op amps, were used like that on purpose and one can add another gain stage for truely balanced preamp operation at the point between them. There is also a switch allowing to change the gain setting.
Attachments
Hi Peter,
You have a switch on the input buffer for 3x and 7x gain?
7x gain seams a little high...
As for the AD815, it seams to only be available in SMD version, and it needs a heatsink.
It's not so easy to use this chip, although... it pays.
Sonically I prefer the LM6172 or OPA2132 to the OPA2604.
You have a switch on the input buffer for 3x and 7x gain?
7x gain seams a little high...
As for the AD815, it seams to only be available in SMD version, and it needs a heatsink.
It's not so easy to use this chip, although... it pays.
Sonically I prefer the LM6172 or OPA2132 to the OPA2604.
Those values can be always played with, whatever preference. My preamp actualy is using those exact values and the gain isn't too high.
I also got some samples of SMD AD815 and I'm quite curious about its sound. After all we usually agree on sonic signatures of various chips.
So far I didn't replace OPA2604 yet, but might do it one day.
For gain setting it's probably advisable to be adding resistor in parallel with 1K (instead of switching resistors), so there woudn't be any switching noise.
I also got some samples of SMD AD815 and I'm quite curious about its sound. After all we usually agree on sonic signatures of various chips.
So far I didn't replace OPA2604 yet, but might do it one day.
For gain setting it's probably advisable to be adding resistor in parallel with 1K (instead of switching resistors), so there woudn't be any switching noise.
I've heard good things about LM6172 so it is on my list of chips to try.
As to the transformer, well that is probably most elegant approach, yet not everybody wants to go with it.
I'm considering here a simple circuit to add to the LM4780 board and convert it into bridged amp, as many people asked about it. Transformer, apart form the size restrictions, will not always fit in everybody's budget.
Also, if somebody already has the balanced preamp, they don't really need anything extra and can drive two separate channels in a bridged mode, isn't it? This creates also another possibility; instead of adding a bridging board inside the amp, one might build a balanced output preamp and drive stereo amp in a bridged mode. I might go for that myself, actually.
As to the transformer, well that is probably most elegant approach, yet not everybody wants to go with it.
I'm considering here a simple circuit to add to the LM4780 board and convert it into bridged amp, as many people asked about it. Transformer, apart form the size restrictions, will not always fit in everybody's budget.
Also, if somebody already has the balanced preamp, they don't really need anything extra and can drive two separate channels in a bridged mode, isn't it? This creates also another possibility; instead of adding a bridging board inside the amp, one might build a balanced output preamp and drive stereo amp in a bridged mode. I might go for that myself, actually.
Steve Eddy said:
Using transformers?
I'm presently using S&B 102 TVC and as good as they are, there are certain limitations, like softening of highs in some systems/on certain material. Also dynamics are not always comparable to an active preamp. How would it be to combine transformer with an active line driver, like AD815 for instance (or discreet buffer)? I think there is even an example like that in application notes.
Re: Re: Creating a differential signal from a single-ended signal for bridged amps
I think the solution here is not that challenging as I would have to hire a paid consultant.
But, I may be interested in discussing with you licensing your active I/V stage as I like it a lot. I'm also planning on developing my own version of parallel TDA1541 DAC and this circuit of yours fits there well.
Pedja said:For the adequate compensation and % of the $€£(...), I may offer you one such solution.
I think the solution here is not that challenging as I would have to hire a paid consultant.
But, I may be interested in discussing with you licensing your active I/V stage as I like it a lot. I'm also planning on developing my own version of parallel TDA1541 DAC and this circuit of yours fits there well.
Peter Daniel said:
That approach has worked very well for me.
I posted this almost a year ago:
http://www.diyaudio.com/forums/showthread.php?postid=264145#post264145
se
Absolutely true. Keep in mind fixed ratio transformers for some reason sound a lot better than the S&B. And are much cheaper.
I have used Lundahl LL1660 for phase-splitting or step-down duties and can assure you it sounds way better than the S&B.
For line input splitting Lundahl make a very tiny (AD815 size?
) and very cheap transformer - LL1544. I have a set and if you let me know the configuration (input buffer? max input voltage?) will gladly hook them up to a couple of your favourite opamps and let you know how it sounds.
analog_sa said:
Yeah. The thing about the TVCs is it's like having 24 or however many different transformers. A given transformer tends to work best with a specific source/load impedance. A TVC will typically see just one source/load impedance for all of its positions which means that most of its positions will have a less than optimum loading.
se
Really good transformers are big dollar items and come with their own set of problems. The DVR134/5 does a nice job with little extra parts and you may find after all the work on a discrete circuit that it was a learning experience with the same results. You dual FET circuit is an option but who wants more caps in the signal path?
Peter Daniel said:
Steve Eddy said:
This one is fairly respectable... a couple Joes have compared it favorably with much higher priced transformers.
http://www.edcorusa.com/transformers/audio/wsm/wsm10k-10k.htm
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.