The iBasso P3+ Herron has made its way to my door and there are images and more . . .

[theory_87]

Quote:

Originally Posted by HiFlight /img/forum/go_quote.gif So far, I have tested the AD743, OPA350 (single channel version of OPA2350), and OP27 in Class A and the AD744 in Class A + Output bypassed using the AD8616, EL8201 and LTC6241 as buffers. All combinations sound very good and can be used without modification in the P3, P3+ and D10. They vary slightly in soundstage and imaging and all seem to be more resolving of minute nuances of detail. The better the rest of the equipment and recording chain, the better they sound.

coming up with new set of top kit for D10?

 

[hopeless]

Quote:

Originally Posted by HiFlight /img/forum/go_quote.gif So far, I have tested the AD743, OPA350 (single channel version of OPA2350), and OP27 in Class A and the AD744 in Class A + Output bypassed using the AD8616, EL8201 and LTC6241 as buffers. All combinations sound very good and can be used without modification in the P3, P3+ and D10. They vary slightly in soundstage and imaging and all seem to be more resolving of minute nuances of detail. The better the rest of the equipment and recording chain, the better they sound.

How on earth could you use the LTC6241 as a buffer?

Anyway the P3+ is interesting. Perhaps I'll buy me a new portable amp. Will it drive an HD650 decently, with external power supply attached?

 

[HiFlight]

Quote:

Originally Posted by hopeless /img/forum/go_quote.gif How on earth could you use the LTC6241 as a buffer? Anyway the P3+ is interesting. Perhaps I'll buy me a new portable amp. Will it drive an HD650 decently, with external power supply attached?

By inserting each adapter into the socket!
Seriously, try it first, before you make negative comments.

 

[HiFlight]

Quote:

Originally Posted by theory_87 /img/forum/go_quote.gif coming up with new set of top kit for D10?

I'm working on it, but there are several Class A combinations that are just to close for me to try to pick a "best", so more than likely, I will just offer the Class A modules and buffers "alacarte" to those who don't wish to make them up personally.

I do have some "beta" sets out for impartial testing.

The Class A AD743, Class A OPA350, and Class A + bypassed output AD744 are where most of my efforts have been focused to date.

The D10 is not as easy to work with as is the P3 or P3+ because the LR socket is designed for dual-channel opamps only and each buffer socket is also for dual-channel buffers.

For LR Class A, it is necessary to use 2 single channel opamps and solder them directly to a 2>1 adapter, as there is insufficient room for a socketed adapter.

If a soldering mistake is made during the assembly process, it necessitates scrapping one adapter, 2 opamps and 4 SMD resistors.

 

[rhw]

Quote:

Originally Posted by HiFlight /img/forum/go_quote.gif I'm working on it, but there are several Class A combinations that are just to close for me to try to pick a "best", so more than likely, I will just offer the Class A modules and buffers "alacarte" to those who don't wish to make them up personally. I do have some "beta" sets out for impartial testing. The Class A AD743, Class A OPA350, and Class A + bypassed output AD744 are where most of my efforts have been focused to date. The D10 is not as easy to work with as is the P3 or P3+ because the LR socket is designed for dual-channel opamps only and each buffer socket is also for dual-channel buffers. For LR Class A, it is necessary to use 2 single channel opamps and solder them directly to a 2>1 adapter, as there is insufficient room for a socketed adapter. If a soldering mistake is made during the assembly process, it necessitates scrapping one adapter, 2 opamps and 4 SMD resistors.

WOW
Really sounds intriguing.

I think I will not solder soic chips or smd resistors any more,

but I do not understand this:
Quote:

one adapter, 2 opamps and 4 SMD resistors.

compared to this:
Quote:

Originally Posted by HiFlight /img/forum/go_quote.gif One must remove the #6 pin, connect the #5 pin to the #6 solder pad, then use a 2.2k resistor to bridge #6 to #7 pin.

(for those who want to solder ...)

 

[SpudHarris]

At present (thanks to Ron's input) I'm using the 744's in Class A with the Output Bypassed + Ibasso Buffers & AD797's in Ground again in Class A. I only have the standard P3 and am amazed by the resolution, dynamics and general SQ of this amp so I can only imagine that the P3+ must be something else!

On the subject of SMD resistors, they are so tiny that I can't even think about using them. With some clever thought to the placement though I've managed to use .6W metal film caps and they work a treat.

 

[hopeless]

Quote:

Originally Posted by HiFlight /img/forum/go_quote.gif By inserting each adapter into the socket! Seriously, try it first, before you make negative comments.

Seriously, the LTC6241 has 30 mA of short circuit output current. There are a thousand of better opamps to use for buffers. OPA2727, OPA2350, and mostly the LME49726 with 300 mA output current (unfortunately only in MSOP package) come to my mind right now.

 

[hopeless]

Quote:

One must remove the #6 pin, connect the #5 pin to the #6 solder pad, then use a 2.2k resistor to bridge #6 to #7 pin.

That way you bypass the output stage of the opamp, then you bias the bypassed output stage of the said opamp into class A. Kind of a waste of power and time, ain't it?

If you 'output bypass' the AD744 by using pin 5 as the output, then class A biasing the output stage (pin 6 with the resistor to V+) doesn't make sense anymore at all. Either you bias pin 6 to make the output stage work in class A, or you just use pin 5 so as to bypass the output stage completely.

 

[rhw]

Hopeless,
thanks for the explanation.
I was irritated by the 4 SMD resistors.

SpudHarris,
maybe I should built my own.
Well, have to look for all the parts.
For some BrownDogs and opamps, shipping will become more expensive than the parts.
Hmmmmmmmm

 

[HiFlight]

Quote:

Originally Posted by rhw /img/forum/go_quote.gif WOW Really sounds intriguing. I think I will not solder soic chips or smd resistors any more, but I do not understand this: compared to this: (for those who want to solder ...)

Sorry for any confusion, but I am using the SOIC version of the AD744 instead of the DIP style, removing the (tiny) #6 pin, then soldering it onto an adapter from which I have removed the #5 pin and previously installed an SMD 2.2K resistor bridging the 6-7 solder pads on the adapter and an SMD 0 ohm resistor between solder pads 5-6 on the adapter. This makes for a very small module and since I do the mods to the adapter first, I don't run the risk of having to scrap an entire module if I make a mistake soldering the SMD devices.

I tried a number of different methods to achieve the OB+CA operation, and this one seems to be the quickest for me with fewer mistakes. Keep in mind though, that I have worked with SMD devices for a number of years and despite their small size, they are about the easiest devices to use for this project.

Once the 2 modules are complete, I then solder them both into a Browndog 2-1 adapter to allow their use in the D10.

For P3 and P3+, the 2-1 adapter is not required. One can also use Class A opamps on a 2-1 adapter for the ground channel, if desired (again, only in P3 & P3+)

 

[HiFlight]

Quote:

Originally Posted by hopeless /img/forum/go_quote.gif That way you bypass the output stage of the opamp, then you bias the bypassed output stage of the said opamp into class A. Kind of a waste of power and time, ain't it? Yes, it would be. If you 'output bypass' the AD744 by using pin 5 as the output, then class A biasing the output stage (pin 6 with the resistor to V+) doesn't make sense anymore at all. Either you bias pin 6 to make the output stage work in class A, or you just use pin 5 so as to bypass the output stage completely.

You misunderstand me...Pin 6 (output) is removed and is not connected in any way to the socket. It is only the compensation pin (#5) that is biased.

Regarding the LTC6241HV, It swings to within 30mv of the supply rails which is very good, and the DC offset is extremely low, which is also very good. Although I also usually favor a buffer with a higher output current, the 6241 actually sounds excellent as a buffer and has plenty of drive, very close to that of the 2727 and 49720, while drawing much less quiescent current. I read the datasheet as the DIP LME49720 having an output of about 23 ma, very similar to that of the 6241.

This is one of those opamps that looks great on the datasheet, but, IMO, does not sound that good in actual operation.

If output current is the only consideration, it is difficult to consider anything but the AD8397.

It is not my inclination to quibble about the characteristics of various opamps but rather to provide information to those who might be interested in improving the sound of their amps.

Please feel free to either accept or decline my suggestions.

 

[denging]

@Ron,

I really like the idea to put the LTC6241HV on P3's Buffer socket... unfortunatelly, the Buffer socket in P3 (and perhaps the P3+) is designed for 2x single opamp ... do you use some sort of a 'dual to single' adapter (which i hardly find it anywhere) ? btw, is it possible that 'HV' on LTC6241HV is for 'High Voltage'...

thanks

 

[hopeless]

Quote:

Originally Posted by HiFlight /img/forum/go_quote.gif You misunderstand me...Pin 6 (output) is removed and is not connected in any way to the socket. It is only the compensation pin (#5) that is biased. Regarding the LTC6241HV, It swings to within 30mv of the supply rails which is very good, and the DC offset is extremely low, which is also very good. Although I also usually favor a buffer with a higher output current, the 6241 actually sounds excellent as a buffer and has plenty of drive, very close to that of the 2727 and 49720, while drawing much less quiescent current. I read the datasheet as the DIP LME49720 having an output of about 23 ma, very similar to that of the 6241. This is one of those opamps that looks great on the datasheet, but, IMO, does not sound that good in actual operation. If output current is the only consideration, it is difficult to consider anything but the AD8397. It is not my inclination to quibble about the characteristics of various opamps but rather to provide information to those who might be interested in improving the sound of their amps. Please feel free to either accept or decline my suggestions.

For buffer the LT6234 is a better choice than the LTC6241, and draws less current too. It has more output current and can drive heavy loads.


That said, regarding the class A biasing... I just commented the quote I quoted, as it was written.

Anyway biasing pin 5 of the AD744 is probably no good to do, since on that pin the opamp will have veery little output current, and that way you choke it even before considering the load represented by the input of the buffer stage...

 

[hopeless]

Consider the rail-to-rail and very nice LT1678 and LT1632 too. I want to try them (especially the former) in my DAC.

 

[jamato8]

I am back to the stock configuration. I am using the HD650's right now and the sound is very fine. The bass is well controlled and there is waaay more than enough volume if needed. Very open and transparent. Can't wait to hear the JH13 Pro IEM's with this and the other amps.

Wow, this is really driving the 650's. The bass. . . great stuff.