bear said:
I am still stumped at what your point is...
Its a different buffer. Not necessarily any better or worse.
Usually a diamond buffer has the driver collectors to the rails as I have done.
Still its an interesting schematic and I don't mind the dialog.
Cheers!
Russ
Russ,
Sorry to intrude... apparently erroneously I thought that the "cross coupled" buffer and the "diamond buffer" were different names for the same thing... all originating with Van Scoyoc in 1948 using tubes. The Van Scoyoc circuit can be found in Radiotron Designer's Handbook 4th Ed., page 663 (fig. 15.43A). 🙂
Guess they're not exactly the same...
_-_-bear
Sorry to intrude... apparently erroneously I thought that the "cross coupled" buffer and the "diamond buffer" were different names for the same thing... all originating with Van Scoyoc in 1948 using tubes. The Van Scoyoc circuit can be found in Radiotron Designer's Handbook 4th Ed., page 663 (fig. 15.43A). 🙂
Guess they're not exactly the same...
_-_-bear
bear said:
Actually Bear, quite the contrary. I am quite glad you joined the discussion. No need to apologize.
I have been playing with your "cross coupled buffer" idea.
I am actually quite pleased with the simulation when paired with a parallel CFP output stage like the attached simulated amp. Forgive the roughness, its not polished yet. 🙂
But the simulation into a 32ohm load is very very promising.
Just to give you an idea (of course this is just a simuation) into the 32 ohm (with 1nf in parallel) load at 20khz we get this:
Code:
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 2.000e+04 1.995e+00 1.000e+00 -2.04° 0.00°
2 4.000e+04 3.546e-07 1.777e-07 129.23° 131.27°
3 6.000e+04 1.904e-06 9.542e-07 79.98° 82.03°
4 8.000e+04 2.497e-09 1.252e-09 -79.41° -77.37°
5 1.000e+05 7.206e-07 3.611e-07 -102.41° -100.37°
6 1.200e+05 9.329e-10 4.675e-10 33.43° 35.47°
7 1.400e+05 2.936e-07 1.472e-07 73.30° 75.35°
8 1.600e+05 2.258e-09 1.131e-09 -59.88° -57.84°
9 1.800e+05 1.200e-07 6.015e-08 -111.86° -109.81°
Total Harmonic Distortion: 0.000105%Ok, I think I like this even better.... 🙂
And thats into 32ohms. 🙂
I think I am going to have to build this puppy....
Here is the cct:
Code:
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 2.000e+04 1.991e+00 1.000e+00 -4.22° 0.00°
2 4.000e+04 3.111e-07 1.563e-07 -98.05° -93.82°
3 6.000e+04 1.418e-06 7.124e-07 86.61° 90.84°
4 8.000e+04 1.352e-07 6.791e-08 -10.54° -6.31°
5 1.000e+05 3.631e-07 1.824e-07 -95.75° -91.53°
6 1.200e+05 3.338e-08 1.677e-08 162.13° 166.36°
7 1.400e+05 9.910e-08 4.978e-08 77.72° 81.94°
8 1.600e+05 8.126e-09 4.082e-09 -28.18° -23.96°
9 1.800e+05 2.661e-08 1.336e-08 -114.38° -110.16°
Total Harmonic Distortion: 0.000076%And thats into 32ohms. 🙂
I think I am going to have to build this puppy....
Here is the cct:
Build is progressing slowly since I have kids running around making a mess 🙂 The resitors are a pain in the butt to mount since the holes are a little bit too close for comfort. They fit, but not as easily as I would have liked. I ordered a TP Low Current Bipolar Power Supply to run this thing today. Hopefully it will be shipped swiftly, nudge nudge Brian and Russ 😉 I will steal the PSU from my Opus to test in the mean time. I have some thoughts on the gain of the Diamante. The BOM I used now will give it a gain of 2, which is probably a bit low for a headphone amp. Is it safe to take it all the way up to 11 in gain?
MrMajestic said:
I would be surprised if 2X gain is not enough for a headphone amplifier. That's 4VRMS from normal consumer sources at max volume. That much output into most headphones would quickly make you deaf. 🙂
I am running mine at 2X, but 11X should work fine too.
Cheers!
Russ
Russ White said:
I will have to build it with 2X gain any way, since thats what I have 🙂 I will socket the resistors so I can play around with it a bit. I would like to have the option of going deaf at least 🙂
MrMajestic said:
Well just for reference 11 x 2VRMS is well past what the amp can do at full volume at any case by more than double. 🙂
The best the amp can do is about 9VRMS so I would say 4X gain is about as much as you can go without clipping from a 2VRMS source.
Cheers!
Russ
Russ White said:
Hehe, ok, I will take your advise and not go past 4X gain then 🙂 Hopefully I will have the amp up and running within a few days. I cant wait.
Russ White said:
Ok, I must have messed up the mathematics. Care to elighten me what would get me closer to 4X gain?
I found an error in the BOM that also crept into my own BOM. Q9 is listed twice with both BC550 and BC560. Which is supposed to be? Im guessing its a BC560 but I would like to be sure.
MrMajestic said:
Q9 is NPN BC550 same as Q10. When in doubt check the schematic.
The gain is figured like this:
G = (RF / RG) +1
Cheers!
Russ
Here are the assembled boards. Looking pretty good if I may say so myself 🙂 I havent powered these up yet, but I will soon.
An externally hosted image should be here but it was not working when we last tested it.
