[gtp]

I searched and didn't find an answer to my question.

Has anyone tried the AD8397 in a Go-Vibe? If so, did it work out OK? I know there are some offset issues, but I don't know if the GoVibe circuit has the corrective elements. I have a sample AD8397, so if I put it in, I need to measure the output offset voltage and make sure it doesn't rail, correct?

Thanks,
George

 

[MisterX]

1) If you consider the Go-vibe a Cmoy then the answer is yes.
2) It doesn't work unless you change the resistor values.
3) Correct.
Never, ever, ever plug your headphones into an amp after swapping opamps unless you have made sure the DC offset is not going to damage said headphones.

 

[gtp]

Hello, MisterX.

Thanks for your reply.

Too bad it probably won't work in the GoVibe as is. I also have an 8620 to try out, so I still have something to look forward to!

Since there has been so much positive bantor here and over at Headwize about the 8397, I'll start digging around for a good circuit diagram.

Thanks,
George

 

[fewtch]

Quote:

Originally Posted by gtp Hello, MisterX. Thanks for your reply. Too bad it probably won't work in the GoVibe as is. I also have an 8620 to try out, so I still have something to look forward to!

IMO the AD8620 sounds terrific in the go-vibe... really brings it to the next level in terms of refinement, clarity and 'energy'. Then again, I've loved the 8620 in everything I've tried it in, so YMMV

.

P.S. I think the Go-Vibe is best described as a "Cmoy-like" commercial design. So might Ray Samuels' Hornet amp possibly be described, at least from what I've read.

 

[gtp]

IMO the AD8620 sounds terrific in the go-vibe... really brings it to the next level in terms of refinement, clarity and 'energy'. Then again, I've loved the 8620 in everything I've tried it in, so YMMV

.
QUOTE]

Hi Fewtch,

Well I got my 8620 soldered onto the Browndog, set it in the GoVibe, and WOW, I'm hearing things that I didn't hear before. There seems to be more detail than the 8066. It sounds like there is a bit more extension at both ends and the sound is brighter overall to my ears. This chip is definitely not as forgiving as the 8066 - I can hear a lot more faults in my mp3s.

I'll have to spend a couple of days with the 8620 to decide if I like better than the 8066.

Thanks,
George

 

[fewtch]

Quote:

Originally Posted by gtp Hi Fewtch, Well I got my 8620 soldered onto the Browndog, set it in the GoVibe, and WOW, I'm hearing things that I didn't hear before. There seems to be more detail than the 8066. It sounds like there is a bit more extension at both ends and the sound is brighter overall to my ears.

IMO it's not strictly brighter, but more 'energetic' and detailed in a way that makes it seem brighter. Or if it's brighter, it's a neutral sort of brightness (heh, figure that one out

) that exposes other opamps as being a bit on the dark side.

Not sure how I'm going to like it with the DT880 I have on order, since that headphone is supposed to be a bit on the bright side. IMO it's excellent with HD580s though.
Quote:

This chip is definitely not as forgiving as the 8066 - I can hear a lot more faults in my mp3s.

Yep I found that as well... very transparent to the point of being unforgiving of poor sources and/or recordings. Did you notice a certain sense of refinement that other opamps don't have also? IMO it makes the go-vibe less 'raw' sounding (which seems like its primary sonic signature, lacking buffers and all).

 

[gtp]

Did you notice a certain sense of refinement that other opamps don't have also?

It's going to take me (at least) several hours of listening to comment on this. I am new to the amplifier scene, so my ear is not tuned to the nuances of headphone amp sound. Hopefully tomorrow I'll have time to listen more at work!

Thanks,
George

 

[Andrea]

Quote:

Originally Posted by MisterX 1) If you consider the Go-vibe a Cmoy then the answer is yes. 2) It doesn't work unless you change the resistor values. 3) Correct. Never, ever, ever plug your headphones into an amp after swapping opamps unless you have made sure the DC offset is not going to damage said headphones.

May I ask a question? Perhaps it's paranoiac, but, would you say that the proper Browndog SO8 to DIP adapter is adequate for the AD8397's output power and current drain? The copper traces on it seem a bit thin...

 

[cire]

Quote:

Originally Posted by Andrea May I ask a question? Perhaps it's paranoiac, but, would you say that the proper Browndog SO8 to DIP adapter is adequate for the AD8397's output power and current drain? The copper traces on it seem a bit thin...

we're talking tiny amounts of current and voltage here... theres no way it would make a difference....

 

[BrownDog]

Quote:

Originally Posted by Andrea May I ask a question? Perhaps it's paranoiac, but, would you say that the proper Browndog SO8 to DIP adapter is adequate for the AD8397's output power and current drain? The copper traces on it seem a bit thin...

The traces on the BrownDog 970601S are 8 mils (0.008") wide using 1 oz. copper. That translates into a cross-section of 8 mils x 1.4 mils or about 11 square mils. I checked with Analog Devices to find out the size of the internal bonding wires between the external pins and the pads on the die. For the AD8397, they're using 0.7 mil diameter wire which yields a cross-section of 0.385 sq. mils - I'm not sure what kind of wire they're using, but it would be gold, copper, or aluminum. So the PCB trace on the adapter has a cross-section over 28 times greater than the die bonding wire. Furthermore, they told me the fuse current (i.e., the current required to blow open a bond) is an astonishing 50 amps!

While the AD8397 is capable of delivering over 300 ma into a 32 ohm load - a substantial current for an op-amp - I don't think you have too much to worry about in this regard.

Hope this helps.

 

[Andrea]

Quote:

Originally Posted by BrownDog The traces on the BrownDog 970601S are 8 mils (0.008") wide using 1 oz. copper. That translates into a cross-section of 8 mils x 1.4 mils or about 11 square mils. I checked with Analog Devices to find out the size of the internal bonding wires between the external pins and the pads on the die. For the AD8397, they're using 0.7 mil diameter wire which yields a cross-section of 0.385 sq. mils - I'm not sure what kind of wire they're using, but it would be gold, copper, or aluminum. So the PCB trace on the adapter has a cross-section over 28 times greater than the die bonding wire. Furthermore, they told me the fuse current (i.e., the current required to blow open a bond) is an astonishing 50 amps! While the AD8397 is capable of delivering over 300 ma into a 32 ohm load - a substantial current for an op-amp - I don't think you have too much to worry about in this regard. Hope this helps.

Thanks! This blows away any slightest perplexity I could have dared to conceive. I felt it was a bit paranoiac, and it was.

Then the AD8397 is my next thing to play with

 

[gtp]

I finally got my AD8397 Browndogged and got a chance to test it in my Go Vibe. I measured the DC offset at listening volume level to be ~15 mV. Fully cranked, the DC offset is ~50 mV. I figured 15 mV was safe enough, so I plugged in my headphones. So far it sounds great! I haven't listened enough to compare with my other opamps, though.

I have a couple of questions: Is 15 mV safe for long term listening? Is it safe to assume the DC offset will remain fairly constant?

Thanks!
George

 

[Andrea]

Interesting! Thanks for reporting

I guess you can assume that the DC offset will remain constant (as you measured it at a given volume level). Still, anyway, 15 mV isn't really ideal. You can safely live with it with higher impedance phones (say 250-300 ohm) though.


The alternative (sorry for your wallet) would be to get two OPA551's and mount 'em on a "dual DIP" browndog. That way the offset voltage could never be a problem, and you'd have the advantage of a separate opamp per channel. Output current would be 200 mA, slightly more than the AD8397. Only "?" is how well they'll perform with a 9V battery.

 

[gtp]

Yup, the OPA551 is next on my opamp wish list. I have gone a little crazy with rolling ever since I got my Go Vibe. It came with the 8066, but I have since added AD8620, OPA2227, OPA2132, AD8397, and OPA2107 is on its way...

Quote:

Originally Posted by Andrea Interesting! Thanks for reporting I guess you can assume that the DC offset will remain constant (as you measured it at a given volume level). Still, anyway, 15 mV isn't really ideal. You can safely live with it with higher impedance phones (say 250-300 ohm) though. The alternative (sorry for your wallet) would be to get two OPA551's and mount 'em on a "dual DIP" browndog. That way the offset voltage could never be a problem, and you'd have the advantage of a separate opamp per channel. Output current would be 200 mA, slightly more than the AD8397. Only "?" is how well they'll perform with a 9V battery.

 

[Raymate]

Quote:

Originally Posted by gtp Yup, the OPA551 is next on my opamp wish list. I have gone a little crazy with rolling ever since I got my Go Vibe. It came with the 8066, but I have since added AD8620, OPA2227, OPA2132, AD8397, and OPA2107 is on its way...

I use the 8066 and just placed an order for the 8620, how did you find the 8620 over the 8066, what cans are you using?

Cheers

R