ShinOBIWAN said:
If your DAC has opamps in the output section the easiest and probably cheapest would be to put an opamp at the entrance of your power amplifier for balanced input
All other changes would only pay off if the output of your DAC is discrete (but that is my opinion only).
Only a Pooger would use an input tranny....... I'm sure the real balanced thing is possible since this is a differential imput. I wonder if Stuart or Jacco are out there and see this??????
Mark
Mark
jacco vermeulen said:
I'm pretty sure its discrete, take a look at the photo:
An externally hosted image should be here but it was not working when we last tested it.
Its kinda small but you can make out the DAC's and then what I presume to be the opamps which are fairly small and toward the center of the board.
Sorry to be a pain but could you perhaps post a diagram/schematic of the bits I'd need to assemble and just where they'd connect to the KC board.
Any help would be greatly appreciated.
Thanks,
Ant
Mark A. Gulbrandsen said:
Maybe you should do a little investigating yourself instead of slinging. An electrically balanced input takes very extreme matching to come even close to the CMRR of a transformer, and then only if it is precisely matched to the particular source component. Not only that, you get natural RF rejection that isn't affected by the source impedance like an input RC filter that adds resistor noise. Change the source component, and you have to rematch with an electronic balance. No worries for a transformer, and you get DC isolation as well. It's true that a transformer costs a lot more, but works as advertised. If you want to be cheap, you get what you pay for.
Poogie,
I am hardly singing any song....... I have extensive experience with audio type transformers of all types from both my broadcast background, recording equipment maintainance, and home audio projects.
The biggest problem with transformers is that even the best of the best are nowhere near perfect. The assiciated loss, ringing, phase shift, and LF rolloff that you will get is just not something that any of those building the KSA-50 would want, especially the LF roll off problem........ The KSA-50 has quite remarkable deep LF response! The phase shift problem is then the second worst problem a tranny will cause. I not going to get into specifics on other effects but if you want to clog up your home system with transformers you certainly have the right to do so. Transformers do have their applications in many places, but certainly not in limiting the performance in my home system.
Mark
I went ahead and ordered prototypes made of these boards as well and they will be in my hands by Thursday. I felt that with ordering 200 of them it would be the wiser choice to get protos and be sure all is still good. Hole chganges thay may be but mistales can still happen and its happened to me before at work!
I just didn't feel that I'd want to ship out close to 200 boards that would need to have traces cut or changes made. The full prototype cost will be credited against the production run which will only take 2 weeks, not the 4 weeks since Advanced is still running the 4 week for 2 week deal at no extra cost. I will absorb the cost of having the original protos made up as well as making up the difference to get us to the 200 board count.
Will keep everyone posted on this...... and I will also keep the board order open until at least Thursday the 28th.
Mark
I just didn't feel that I'd want to ship out close to 200 boards that would need to have traces cut or changes made. The full prototype cost will be credited against the production run which will only take 2 weeks, not the 4 weeks since Advanced is still running the 4 week for 2 week deal at no extra cost. I will absorb the cost of having the original protos made up as well as making up the difference to get us to the 200 board count.
Will keep everyone posted on this...... and I will also keep the board order open until at least Thursday the 28th.
Mark
Now, now Gentlemen....
This is not the thread for transformer discussion except for pwr supplies in...
Your brand new Krell KSA-50 clone!...
Did you hear the DJ and announcer voice on that one?
Back to the regularly scheduled thread....
Sorry to step into Al's shoes for a second.... Man they were big.. 😉
This is not the thread for transformer discussion except for pwr supplies in...
Your brand new Krell KSA-50 clone!...
Did you hear the DJ and announcer voice on that one?
Back to the regularly scheduled thread....
Sorry to step into Al's shoes for a second.... Man they were big.. 😉
Alot of commercial amps made today use input chips for the differential becuase the internal components are closley matched. Bit its really pretty easy to hand match the necessary components anyway if you just order extra parts. A few places will supply precision matched resistors at slight additionsl charge if you don't want the hassle. But then its going to be somewhat of a hassle to make a balanced input work and matching up critical parts won't be a bother anyway.
Mark
Mark
rabstg said:
Err, excuse me but I think that my question regarding balanced inputs for the KSA50 was completely valid, despite any inferred implementation details.
By dismissing any discussion on implementation, including counterpoints, of a potential balanced input for the KSA50 you also dismiss stuff which I really would like to read and find interesting.
Do you often dismiss others questions as offtrack if it doesn't benefit, interest or serve a purpose to you?
After reading virtually all of this thread, there hasn't been any suggestion of a balanced input section. On the other hand I see countless reiterations of details throughout this thread that keep cycling almost indefinitely. I suspect that this exact same thing will undoubtedly cause this thread to explode once everyone has the boards and needs advice on construction. But nobody seems to mind that, no diluting of the thread there.
Mark A. Gulbrandsen said:
Hi Mark,
So you wouldn't recommend using line level transformers to allow balanced operation?
The link that Jacco gave from Rowland was interesting, it seems the Jensen J1-1P 1 transformers are only -3dB down at 3hz. The graphs from the data sheets certainly look OK. I agree that they will be far more expensive than the opamp route - also thank to LGreen for the schematic which also leads to another question...
What characteristics are needed for the opamp? Unity Gain is essential I presume?
Or would it be best to match the opamps to those that are in the source or DAC in this case?
The only downside I can see is that I'll need a regulated supply to power the opamp stage in the amp.
Actually if you are patient and give this a few days I think you may find that someone just might chime in with the proper way to do a balanced input whilst preserving the qualities of the circuit itself. Using a transformer hardly preserves the overall qualities of this fine amplifier. I think thats the basic reason why Poogie was "jumped". Of course if you find the sound of a certain transformer to be of some merit to your system then by all means persue it.
Mark
Mark
The question about input transformers was not the issue, the ensuing debate that caused people to become defensive is.
We don't need arguments over whether transformers are good or bad, but the different options to do balanced are welcome.
I am not a moderator(nor want to be) but this thread has been remarkably consistent and friendly... I would hate to see it change from an innocent comment or belief.
We don't need arguments over whether transformers are good or bad, but the different options to do balanced are welcome.
I am not a moderator(nor want to be) but this thread has been remarkably consistent and friendly... I would hate to see it change from an innocent comment or belief.
I like traffos, but then I used to work in R'n'R where true isolation was a good thing. Cheap traffos are very nasty. Good Jensens or Sowters are not bad.
The INA134 is a good diff. input device, but not as good at isolation obviously as a traffo. But a lot cheaper.
For domestic use, the INA134 will be fine, unless you have a particularly nasty domestic electrical installatio, that is. 🙂
If you want the actual circuit modded, give me a month or so, and I'll have time to look at it, I'll use the protoboards to experiment. Or if someone else comes up with a mod, I'll try it out, (that may well be quicker!).
The INA134 is a good diff. input device, but not as good at isolation obviously as a traffo. But a lot cheaper.
For domestic use, the INA134 will be fine, unless you have a particularly nasty domestic electrical installatio, that is. 🙂
If you want the actual circuit modded, give me a month or so, and I'll have time to look at it, I'll use the protoboards to experiment. Or if someone else comes up with a mod, I'll try it out, (that may well be quicker!).
balanced in
Hi Guys,
There is no doubt in my mind that this circuit can be modified to have an op-amp 'style' balanced input. In a technical sense it already does, just that the inverting input isn't brought out to the world...adding an op-amp would not be necessary, a couple more resistors perhaps...for a 'working' example, you can see NPs solution for the Aleph's that had balanced inputs.
The impedances and transconductances of the two input legs would need to be equalized (they mostly already are), and possibly other changes might follow from that. I'll happily take a look at this as time permits, but anyone in a hurry will need to find a solution elsewhere. I just have too many things on my plate to think I'll get to this soon.
From reading other threads on the board it seems the 'right' input transformer can be quite transparent and might not diminish the overall quality of the amp, but that transformer could end up costing as much as the rest of the components for the amp...good signal transformers are quite expensive, as much and more than a good power supply transformer...
Stuart
Hi Guys,
There is no doubt in my mind that this circuit can be modified to have an op-amp 'style' balanced input. In a technical sense it already does, just that the inverting input isn't brought out to the world...adding an op-amp would not be necessary, a couple more resistors perhaps...for a 'working' example, you can see NPs solution for the Aleph's that had balanced inputs.
The impedances and transconductances of the two input legs would need to be equalized (they mostly already are), and possibly other changes might follow from that. I'll happily take a look at this as time permits, but anyone in a hurry will need to find a solution elsewhere. I just have too many things on my plate to think I'll get to this soon.
From reading other threads on the board it seems the 'right' input transformer can be quite transparent and might not diminish the overall quality of the amp, but that transformer could end up costing as much as the rest of the components for the amp...good signal transformers are quite expensive, as much and more than a good power supply transformer...
Stuart
pinkmouse said:
Did the originals have ballanced inputs? If not I don't think they should be included in the GB design. It can easily be an add on ciruit without holding up the GB. I say build the protos to make sure everything is right and then lets get these things ordered.
Blessings, Terry
Re: balanced in
I'm just off out in the next few minutes but would really like to discuss this further.
No way! I want the boards at least as much as next guys so after Mark gets the 2nd proto up and confirmed... well I can't wait myself.
pinkmouse said:
Stuart Easson said:
I'm just off out in the next few minutes but would really like to discuss this further.
pinkmouse said:
No way! I want the boards at least as much as next guys so after Mark gets the 2nd proto up and confirmed... well I can't wait myself.