I might have the same issue with the soundstage and dist. I was blaiming my preamp the whole time. I'll do the modifications and report back.
/Kari
/Kari
Kari,
After additional checking, if you wish for convenience sake, you can make R9 and R10 500 ohm resistors too. It is less "optimal" but I don't beliieve that it will make an audible difference.
Also, for everyone else, I have realized that the values for the resistors are very dependant on the voltage of the power supply rails. The values I posted are valid in the 29-31 volte range. The good news is that I can now determine the optimum values relatively easily for any range of supply voltage. If you need values for a different set, let me know.
This circuit is very much a delicate act of balancing parameters. The stability limits on the feedback, and the biasing issue for the differential pair are tightly intertwined, and changing one changes the other, so that you have to find the right combination that doesn't compromise either side.
Damn you Nelson Pass!!! Before I found passdiy.com I was content with my ignorance and mediocre sounding stereo system. No one warned me that this hobby is an infectious disease causing a slow spiral into insanity. I spend way too much time now giggling uncontrollably while surrounded by electronics paraphenalia, murmering about the search for the ultimate circuit.😎
After additional checking, if you wish for convenience sake, you can make R9 and R10 500 ohm resistors too. It is less "optimal" but I don't beliieve that it will make an audible difference.
Also, for everyone else, I have realized that the values for the resistors are very dependant on the voltage of the power supply rails. The values I posted are valid in the 29-31 volte range. The good news is that I can now determine the optimum values relatively easily for any range of supply voltage. If you need values for a different set, let me know.
This circuit is very much a delicate act of balancing parameters. The stability limits on the feedback, and the biasing issue for the differential pair are tightly intertwined, and changing one changes the other, so that you have to find the right combination that doesn't compromise either side.
Damn you Nelson Pass!!! Before I found passdiy.com I was content with my ignorance and mediocre sounding stereo system. No one warned me that this hobby is an infectious disease causing a slow spiral into insanity. I spend way too much time now giggling uncontrollably while surrounded by electronics paraphenalia, murmering about the search for the ultimate circuit.😎
metalman said:
Hello Terry,
I follow this tread with high interest. I would be nice to you to calculed the resistors value for a range of 25-28V rails supply. I will use 300va 2x25V through high reagulated supply.
Marc
metalman said:
Terry,
Would it be possible to describe the issues and
the calculations involved?
metalman said:
So true...Are there any DIYaudio self-help groups? 🙂
Dennis
Terry,
Did you recieved my mail? If not:
I would be nice if you could calculate the resistor values for 36-37VDC.
Cheers,
Vit
Did you recieved my mail? If not:
I would be nice if you could calculate the resistor values for 36-37VDC.
Cheers,
Vit
Sorry for my absense guys, between work and my misbehaving aleph-X I haven't had any time to spare.
I had your numbers crunched and somehow lost them. I'll get them to you ASAP.
The short answer on the calc's is that I don't actually calculate them, but I have a Spice model for the circuit that I can try different parameters and see where clipping might occur at different circuit nodes, as well as see when the circuit goes unstable at high frequencies. I use an iterative process of tweaking the values and checking the results, and have developed a feel for what tweaks have what effects. Also, I have tried many iterations of changes in the real functioning circuit to corroborate that what I see in the sims does correspond with what happens in the real circuit.
The long answer is being drafted, where I will try to descibe the cause and effects of the circuit behaviour. When I post that discussion, please realize that while I have 8 years of university education in engineering and 12 years engineering experience, none of it has anything to do with electrical design, so what I will besaying comes strictly from my experience as a hobbiest. Don't take anything I say as gospel.
No, I didn't get your e-mail (Spam filter probably ate it!), but I'm happy to figure out values for you too.
Cheers, Terry
I had your numbers crunched and somehow lost them. I'll get them to you ASAP.
The short answer on the calc's is that I don't actually calculate them, but I have a Spice model for the circuit that I can try different parameters and see where clipping might occur at different circuit nodes, as well as see when the circuit goes unstable at high frequencies. I use an iterative process of tweaking the values and checking the results, and have developed a feel for what tweaks have what effects. Also, I have tried many iterations of changes in the real functioning circuit to corroborate that what I see in the sims does correspond with what happens in the real circuit.
The long answer is being drafted, where I will try to descibe the cause and effects of the circuit behaviour. When I post that discussion, please realize that while I have 8 years of university education in engineering and 12 years engineering experience, none of it has anything to do with electrical design, so what I will besaying comes strictly from my experience as a hobbiest. Don't take anything I say as gospel.
No, I didn't get your e-mail (Spam filter probably ate it!), but I'm happy to figure out values for you too.
Cheers, Terry
metalman said:
I'm wating for you values before i fire on first time XGC Blocks
Between I'll progresse a little in casing
An externally hosted image should be here but it was not working when we last tested it.
Marc
Marc,
For rails in the +/- 25 to 28V range, use the values in my last posted schematic but with the following changes:
R15 --> 125 ohm 1/2W (or higher watt rating)
R16 --> 12.5 ohm
C3 --> 470uF
R16 sets the bias for the differential pair and thus the voltage drop across R1 and R2, which needs to be ~10.4V to avoid clipping of the input differential. R15 is lowered to increase the amount of voltage drop from the rail to R1 and R2, increasing the voltage headroom for the differential pair. Unfortunately, reducing R15 also decreases the circuits rejection of ripple from the rails more than I like too see, so I suggest the increased value of C3 to bring the ripple rejection back up.
Don't worry about getting precisely those values. The closest available value to those number in whatever series of parts you are using will suffice.
I agree with Mark, you chassis is coming along beautifully.
Vit,
Your numbers are coming next!
For rails in the +/- 25 to 28V range, use the values in my last posted schematic but with the following changes:
R15 --> 125 ohm 1/2W (or higher watt rating)
R16 --> 12.5 ohm
C3 --> 470uF
R16 sets the bias for the differential pair and thus the voltage drop across R1 and R2, which needs to be ~10.4V to avoid clipping of the input differential. R15 is lowered to increase the amount of voltage drop from the rail to R1 and R2, increasing the voltage headroom for the differential pair. Unfortunately, reducing R15 also decreases the circuits rejection of ripple from the rails more than I like too see, so I suggest the increased value of C3 to bring the ripple rejection back up.
Don't worry about getting precisely those values. The closest available value to those number in whatever series of parts you are using will suffice.
I agree with Mark, you chassis is coming along beautifully.
Vit,
Your numbers are coming next!
Vit,
For 36 - 37 V rails (i.e. +36 positive rail, and -36 negative rail)
R15 --> 360 ohm 1W
R16 --> 13 ohm
C3 --> 220uF minimum (470uF is OK too!).
Cheers, Terry
For 36 - 37 V rails (i.e. +36 positive rail, and -36 negative rail)
R15 --> 360 ohm 1W
R16 --> 13 ohm
C3 --> 220uF minimum (470uF is OK too!).
Cheers, Terry
metalman said:
Thanks a lot
To resume for 25-28V :
R1/R2 : 500R
R9/R10 : 625R
R11/R12 : 500R
C3 : 470µf
Marc
Marc,
Exactly Correct!
Kari,
Received last night the XGC-boards you sent me. They are very well done, I'm quite impressed! What layout software did you use?
Cheers, Terry
Exactly Correct!
Kari,
Received last night the XGC-boards you sent me. They are very well done, I'm quite impressed! What layout software did you use?
Cheers, Terry
metalman said:
Those where made with Eagle. However i'm searching for a package with a spice frontend. I have copies of Protel DXP and Orcad but they seem way to complicated to use, and expensive. What i'm looking at now is Target3001!, comes with spice, can import and export Eagle schematics and pcbdesigns. It can also import Eagle and Protel libraries which is a big thumbsup.
Btw, what cad/sim software are you using?
/Kari