What I need is a PCB with 4 or 5 opamp sockets and a selector switch, so I can choose my opamp according to my mood. Remote control optional, but prefered.
In your dreams!
I don't actually like the softer Burr Brown sound, and think the OPA2604 is one of the dullest sounding amps I know. Oddly enough I never notice it in Naim equipment, where it sometimes makes up some of the best kit I know of. Though if there is one in the Nait 5i, that would explain a lot. Maybe it's because I have only used the dual. And duals are not a good idea usually, since they are a hidden worst case of shared ground paths.
My favourite amp at the moment is the LM6172, which is a symmetrical design that goes out to 100MHz with an astonishing slew rate of 3000V/uS. I think it is basically a current feedback amp that's packaged to be used as an ordinary op amp. But I haven't looked into it closely enough since I started to find out what current feedback amps do. See Barrie Gilbert etc. I'd be very interested to hear what you make of it alongside its illustrustious DiL-8 companions. The only problem I have ever found with it is that it tends to fail quite easily, or degrade. I don't treat these parts very well usually, but I was surprised when I found I was down to about two of them in a drawer that must have had 20 in it to begin with.
Incidentally, I would mention that if you want a real leap in performance - and this can be done with TLO72s - then go differential. Particularly as the output of a pre amp. I know one's only putting in an inverter (and you'd think an extra component would make it worse rather than better) but I assure you it's the only way. I have always had differential topologies in all the stuff I've done, yet I was stunned when I just filled in the gap. There was space for this op amp on the board, even had a socket there. And it sat like that for two years with me being perfectly content and thinking it wasn't needed. Boy, was I in for a surprise.
Good luck, Christian
In your dreams!
I don't actually like the softer Burr Brown sound, and think the OPA2604 is one of the dullest sounding amps I know. Oddly enough I never notice it in Naim equipment, where it sometimes makes up some of the best kit I know of. Though if there is one in the Nait 5i, that would explain a lot. Maybe it's because I have only used the dual. And duals are not a good idea usually, since they are a hidden worst case of shared ground paths.
My favourite amp at the moment is the LM6172, which is a symmetrical design that goes out to 100MHz with an astonishing slew rate of 3000V/uS. I think it is basically a current feedback amp that's packaged to be used as an ordinary op amp. But I haven't looked into it closely enough since I started to find out what current feedback amps do. See Barrie Gilbert etc. I'd be very interested to hear what you make of it alongside its illustrustious DiL-8 companions. The only problem I have ever found with it is that it tends to fail quite easily, or degrade. I don't treat these parts very well usually, but I was surprised when I found I was down to about two of them in a drawer that must have had 20 in it to begin with.
Incidentally, I would mention that if you want a real leap in performance - and this can be done with TLO72s - then go differential. Particularly as the output of a pre amp. I know one's only putting in an inverter (and you'd think an extra component would make it worse rather than better) but I assure you it's the only way. I have always had differential topologies in all the stuff I've done, yet I was stunned when I just filled in the gap. There was space for this op amp on the board, even had a socket there. And it sat like that for two years with me being perfectly content and thinking it wasn't needed. Boy, was I in for a surprise.
Good luck, Christian
My electronics knowledge is almost non-existent so much of what you said above is over my head, but I appreciate it nonetheless. After looking at the spec sheet for the LM6172 briefly, my gut was saying "overkill". I'd love to hear your thoughts on how it sounds.
My time last night with the OPA2107 was shorter-lived than I expected. It has distinct midrange which worked for the mostly vocal or three-piece stuff last night and the depth of the soundstage was astonishing at times, however I did grow rather tired of its in-your-face midrange . This morning I put on some mixed hits from the 60's, 70's, 80's and 90's and the OPA2107 wasn't particularly good for that. Within a few songs, I was popping the AD8066 in there. Definitely better.
I used to like the "warm, analog" Burr Brown sound whereas I felt I had to "tolerate" the ADs. As my entire system evolves, though, I find that the 2 ADs are getting much more time in my DAC. I have my eye on a dual-to-single DIP8 adapter so I can start trying pairs of singles, such as AD797, AD825, etc.
Thoughts on how it sounds. Not easy, as there's nothing really special about it. It doesn't grab and say "look how good I am". The LM4562 does, otoh. But while it's lovely and clear, it really doesn't do music that well. It's a bit sharp and somewhat jangly and irritating. the LM6172 is much smoother, marginally less clear, but doesn't become boring and lifeless like a 5532. If anything its got a slightly light presentation, a bit like an Audio Technica cartridge, but not quite that light.
Don't worry about the slew rate or the 100MHz - one doesn't use it out there. It's a good idea to filter the input at somewhere around 100kHz, or perhaps a touch higher.
Like all op amps it really responds to the cleanness of the power supply and the good bypassing. I personally don't use 100nf ceramics around my op amps - or at least not without a resistor and tend to go for cheap tantalums of 1uF+. Interestingly a number of manufacturers are starting to twig on t6his and are recommending 10uF or more around their op amps. You'll on the datasheet that it suggest 2u2, IIRC. And it was one of the first to do so.
Single op amps are a better idea, incidentally. Amazingly I have never tried a AD797, which I think still tops the list for low noise - though there are some very close competitors (which are probably also better op amps). But I'd be interested in how you think it performs. Especially if you do an A/B with the 6171.
Don't worry about the slew rate or the 100MHz - one doesn't use it out there. It's a good idea to filter the input at somewhere around 100kHz, or perhaps a touch higher.
Like all op amps it really responds to the cleanness of the power supply and the good bypassing. I personally don't use 100nf ceramics around my op amps - or at least not without a resistor and tend to go for cheap tantalums of 1uF+. Interestingly a number of manufacturers are starting to twig on t6his and are recommending 10uF or more around their op amps. You'll on the datasheet that it suggest 2u2, IIRC. And it was one of the first to do so.
Single op amps are a better idea, incidentally. Amazingly I have never tried a AD797, which I think still tops the list for low noise - though there are some very close competitors (which are probably also better op amps). But I'd be interested in how you think it performs. Especially if you do an A/B with the 6171.
I ordered:
- two LME49710 (HA) (will be buying a dual-can to DIP8 adapter)
- LME49720 (HA)
- LM6172
-OPA2132P (already have one, but from e-bay, so not sure it is genuine)
-AD823
-AD8620
-AD8599 (already have one, but from e-bay, so not sure it is genuine)
-AD8066 (already have one, but from e-bay, so not sure it is genuine)
I'll be ordering the 797s from Mouser, along with a bunch of other caps and things I need.
RE: the AD797, I have read some favourable opinions that put it in the same class as 8599 and 8066, but it seems to be fussy. The main reason I want to order them is because I have a pre-amp on the way that uses them and I think it might be interesting to run 797s in both my DAC and my pre-amp. Just for sh!ts n' giggles, of course.
- two LME49710 (HA) (will be buying a dual-can to DIP8 adapter)
- LME49720 (HA)
- LM6172
-OPA2132P (already have one, but from e-bay, so not sure it is genuine)
-AD823
-AD8620
-AD8599 (already have one, but from e-bay, so not sure it is genuine)
-AD8066 (already have one, but from e-bay, so not sure it is genuine)
I'll be ordering the 797s from Mouser, along with a bunch of other caps and things I need.
RE: the AD797, I have read some favourable opinions that put it in the same class as 8599 and 8066, but it seems to be fussy. The main reason I want to order them is because I have a pre-amp on the way that uses them and I think it might be interesting to run 797s in both my DAC and my pre-amp. Just for sh!ts n' giggles, of course.
Last edited:
Are you surprised that it is that good, or that bad?
IMHO, LM6172 is excellent sounding and has always excelled in micro-detail and dynamics. However, its does have stability issues if not adequately bypassed, and sometimes needs a few pF from the output to the inverting input for optimal stability.
When everything is correctly implemented, it sits in pretty rarefied territory and only a handful of discrete opamps are audibly superior.
Last edited:
LM6172 is shockingly good. Differences between other opamps I have tried have been subtle in comparison to the change I experienced last night when I plugged in the LM6172. I really don't care at this point to try any of the other samples I have received from TI (or on the way from AD). I am sure I will try them in time, but for now I am completely captivated.
As linuxguru said, the dynamics and micro-detail are simply mesmerizing. Imaging and soundstage are beyond what I thought my system was capable of. Last night I ended up in delirious laughter. If my DAC circuit is sub-optimal for this opamp, I certainly cannot tell.
As linuxguru said, the dynamics and micro-detail are simply mesmerizing. Imaging and soundstage are beyond what I thought my system was capable of. Last night I ended up in delirious laughter. If my DAC circuit is sub-optimal for this opamp, I certainly cannot tell.
That's impressive, isn't it?
He's certainly keeping us in suspense, though.
I think it's an excellent op amp - and thanks to LG; it's always nice to have someone agreeing with you.
So very glad that I listened to your advice. Thank you!
I still have not ordered the AD797s, as I have a bunch of other parts (still compiling the list) I need to order from Mouser and I'd rather pay for shipping once.
No. Get used to and enjoy the LM6172. I can certainly wait.
If you want to do anything then think about the power supplies and your bypassing. It is worth reading the datasheet in detail, though from here on in a clean power supply is one of the few routes to improvement.
Having said that, if it is working that well then it may not be a good idea to change things, unless you have a good reason. Often one can fiddle about and get something "better", but in fact it's only more to your taste initially, rather than actually better. I try very hard to avoid "my" taste, and that way you get nice surprises that go beyond your desires.
And the fewer messes one makes on a board , the better.
My pleasure, incidentally. CT
If you want to do anything then think about the power supplies and your bypassing. It is worth reading the datasheet in detail, though from here on in a clean power supply is one of the few routes to improvement.
Having said that, if it is working that well then it may not be a good idea to change things, unless you have a good reason. Often one can fiddle about and get something "better", but in fact it's only more to your taste initially, rather than actually better. I try very hard to avoid "my" taste, and that way you get nice surprises that go beyond your desires.
And the fewer messes one makes on a board , the better.
My pleasure, incidentally. CT
Well, I still have not yet ordered the AD797s because I have been busy with other things. However, in the meantime I have tried:
- 2x LME49710HA in dual-can-to-single-DIP8 adapter
- 2x OPA627AP in dual-to-single DIP8 adapter
The LME49710HAs are ridiculously transparent, almost to the point of being "too transparent" I think. Any background noise, hiss, distortion, etc. in the recording equipment or instruments is fully revealed. I found this to be a really interesting experience but I did grow tired of it eventually. There really is a "holy crap" experience with this opamp, but it is not something I would just leave in my DAC and live with forever. Several times I actually thought that my driver cones were malfunctioning (buzzing, rattling, etc.) but then after investigation I discovered that the flaws were in the recording itself. Cool, but I don't necessarily always want to hear that.
The OPA627APs are wonderful. The detail is all there, but the background is nice and dark and the typical BB sound is there. Similar to the OPA2132, but just better in every way I'd say. I have only listened a bit so far, so I have to spend more time with them.
I have made significant changes to my pre-amp recently (capacitor upgrade and swapped CS3310 for PGA2311) which allows me to better discern the differences in my DAC opamps, so I need to cycle through all my favourites and re-evaluate them. The pre-amp, by the way, shipped with dual 49710s in it, rather than the advertised AD797s. Initially I was disappointed, but I think now that this is a good thing to have such transparent, "colourless" opamps in my pre. Perhaps this is why I find the 49710s in my DAC "all too revealing".
I have a long weekend coming up and hope to solder my AD8620 onto a DIP-8 adapter and try it out as well. I will spend some time rolling between LM6172, LME49710HA, OPA627AP and AD8620. Perhaps AD8599 will once again get some listening time, too. I would like to boil it all down to 3 favourite opamps and then sell the rest as a grab-bag.
- 2x LME49710HA in dual-can-to-single-DIP8 adapter
- 2x OPA627AP in dual-to-single DIP8 adapter
The LME49710HAs are ridiculously transparent, almost to the point of being "too transparent" I think. Any background noise, hiss, distortion, etc. in the recording equipment or instruments is fully revealed. I found this to be a really interesting experience but I did grow tired of it eventually. There really is a "holy crap" experience with this opamp, but it is not something I would just leave in my DAC and live with forever. Several times I actually thought that my driver cones were malfunctioning (buzzing, rattling, etc.) but then after investigation I discovered that the flaws were in the recording itself. Cool, but I don't necessarily always want to hear that.
The OPA627APs are wonderful. The detail is all there, but the background is nice and dark and the typical BB sound is there. Similar to the OPA2132, but just better in every way I'd say. I have only listened a bit so far, so I have to spend more time with them.
I have made significant changes to my pre-amp recently (capacitor upgrade and swapped CS3310 for PGA2311) which allows me to better discern the differences in my DAC opamps, so I need to cycle through all my favourites and re-evaluate them. The pre-amp, by the way, shipped with dual 49710s in it, rather than the advertised AD797s. Initially I was disappointed, but I think now that this is a good thing to have such transparent, "colourless" opamps in my pre. Perhaps this is why I find the 49710s in my DAC "all too revealing".
I have a long weekend coming up and hope to solder my AD8620 onto a DIP-8 adapter and try it out as well. I will spend some time rolling between LM6172, LME49710HA, OPA627AP and AD8620. Perhaps AD8599 will once again get some listening time, too. I would like to boil it all down to 3 favourite opamps and then sell the rest as a grab-bag.
Last edited:
Hello again.
No, I do not have an oscilloscope, but I am confused why you suspect it is oscillating since I think it sounds very, very good.
I know it is a ridiculously fast opamp, which may be prone to oscillation. In fact, maybe it is oscillating. Nevertheless, I like the way it sounds in my DAC.
But I think the OPA627AP might be my new favourite. I have to do more listening.
Ideally, I would like to get the schematic for my DAC and tweak the circuit to perform best with the "flavour" of opamp that I like best.
Oscillation at extreme hf isn't always obvious by listening alone...
http://www.diyaudio.com/forums/anal...u-have-checked-see-its-stable-havent-you.html