Gilmore, Szekeres and CHA47 Amps -- A Brief Comparison for Fellow Newbies

[tangent]

Quote:

are you refering to the dual 823 or the Sing 843

The 843s. I wouldn't call the 823s "detailed". I like their sound, but they're not detailed on the same level as the 843.

Quote:

even with a buffer, many newer chips, esp from AD, offer similar performance (though very different sound) for much less...

The OPA627 is, without question, overpriced, even in the "A" grade. That's one reason I'm happy to have discovered the AD843 for myself recently. I'm on the verge of deciding never to use another 627 again.

 

[BoyElroy]

Hi Tangent,

Thanks for your comments re: the 2134. I'd like to try a higher quality opamp with the same pinouts as the 2134. Do you have any suggestions? There are so many different opamps being talked about on the Head* sites that I'm not sure where to begin.

 

[jarthel]

OPA2228, AD826, OPA2604, OPA2132

I tried those four on my "CD player" and I love the OPA2228 the best.

Jayel

 

[tangent]

BoyElroy,

I find that all Burr-Brown op-amps have a similar sound. Sure, there are differences, but it seems to me more useful to try entirely different op-amps if you're not happy with the sound you're getting with a Burr-Brown op-amp. If, instead, you like the sound but just want a higher-quality version of it, go ahead and try other Burr-Brown chips. There's enough variation on the BB theme that you might find something that's balanced a little better for your tastes.

I second the recommendation of the 2228, so long as you can get it to be stable. I had a bit of trouble with it. The 2604 is even worse -- very picky, and not all that different from the 2132, IMHO.

As for the AD826, that's a bipolar input op-amp, which means it's difficult to make it work right. At some point if you keep at this hobby you'll want to tackle bipolars, but I think you should stick with FET-input op-amps for your first few experiments.

 

[BoyElroy]

Hi Whit,

Just want to let you know that I put up the Gilmore power supply pcbs today. Its a revision of the pattern that I used with my Gilmore that does away with all extra wiring except one. Its up at the head of this thread, in case you missed it.

Hope it works okay for you--

 

[jerikl]

BoyElroy, thanks for giving me an idea of how much the Gilmore costs. Probably going to try and have it done sometime soon... well, as soon as I get a job and can fund the power supply for sure!

 

[BoyElroy]

Hi Jerikl--

To tell you the truth, the power supply shouldn't be all that much if you build just the LM317/LM337 section initially and then add the REF02/OPA548 later...


--happy building and good luck with the job

 

[ppl]

I have used the AD-843 ever since it came out and i like it alot while the AD-9610 is Close and in some respects Better than the 843 it is still not as Dynamic as the 843 The 843 is not for portable amps it draws as mutch Current as an EL-2008 1 Amp Buffer and wiill only swing to within 4 wolts of the supply rails. Search for posts regarding My AC powered Headphoneamp/Preamp on these Fourms for more info

 

[BoyElroy]

Hi PPL--

I haven't tried either opamp but I'd like to make a portable opamp/buffer unit sometime soon. What is your personal favorite as far as battery powered (efficient!) opamps go? Is the OPA627/637 considered a bit long in the tooth now? I have to say that I've tried both the OPA2132 and the OPA2134 and I was hoping there'd be a better sounding, equally efficient and affordable opamp out there. Does the AD-9610 fit the bill? And how would it match up with OPA634 buffers?

-Thanks and sorry about all the questions...

 

[tangent]

I don't consider the OPA627 "long in the tooth", but it is expensive and requires a lot of voltage (~12V) to sound good. Not good for portables.

For a single-9V amp, my choices would be, in increasing order of desirability: LM6172, OPA2228, AD845, AD823. I really like the AD823 for portable use, and if you buffer it, its only major weakness (low current drive) is unimportant. The others do well with a single 9V battery as well, just not as well as the AD823.

I'm sure there are other worthy opamps in this class I haven't tried yet, so do a search and see what you can dig up...

 

[BoyElroy]

Hi PPL--

Doh!...I just read your AD-8512 article and realized most of the info I'm looking for is already on that thread...



Tangent--

Thanks for the advice--I'm going to try the AD823 with the OPA634 and see how it runs off of a 12Vdc battery pack.

I think I'll get a couple of OPA627's as well. Do you know what the difference is between the OPA627 and the OPA637?

 

[tangent]

Quote:

Do you know what the difference is between the OPA627 and the OPA637?

Wouldn't it have been quicker to just read the datasheet than wait for me to answer your question?

The OPA637 is "uncompensated" which trades unity-gain stability for speed. The OPA627 is unity-gain stable, and is a 16 MHz part. The OPA637 is only stable at gains of 5 or over, and it's 80 MHz at G=10. This makes the OPA637 harder to use, but it may give a sound you prefer, once you get it stable.

 

[BoyElroy]

Hi Tangent--

Yes, I could and did read the datasheet--but most of the tech. spec data was beyond my ability to interpret...hence my req. for your help.

So...thanks.

Specifically, I was curious about how they sounded...does whatever tech. differences they have affect their sound quality...

 

[aos]

Those are the same chips, except that one is "raw" or "wild" if you will, and can only be used if it's set to gain 5 or more (by its feedback resistors) otherwise it will oscillate. The other has been compensated (e.g. by a capacitor between out and inverse input that has been added internally by the manufacturer) and thus can be used in any gain configuration but the compensation process has reduced its bandwidth (and slew rate) - basically a tradeoff between stability and bandwidth. Faster opamps (i.e. uncompensated) like OPA637 might sound better on transients but you can only use them if you've set gain to 5 or more.

You can convert uncompensated opamp to a compensated one by adding external parts by yourself. However that requires more board space, external components and "hassle" of calculating the right compensation network. By buying compensated amp you buy hassle-free part and trade convenience for some performance that is in many cases not noticeable.

Basically, unless you know what you're doing, or you're building based on somebody else's tested design, use compensated amps.

 

[BoyElroy]

Hi AOS--

Thanks for the explanation--seems like the 627 it is, then!