The Opamp thread

[leeperry]

but the OPA2727 cannot work on 12V, can it? it's 12V max swing, so that's ±6V

and can I try the LT1229 on my soundcard? or is it like a VERY bad idea?

the upper spectrum is rather dark on the 1358, but it doesn't sound like grind paper being applied into your inner ear at least

maybe a tad more trebles would be good, I got so many more LT samples pending anyway

I will use OPA211 on adapters, is there any point that I try the OPA2211A as well?

 

[Filburt]

12V is the specified max but you may be able to run it higher. I've run the OPA2727 at up to about 14V or so. I'd personally use a THS4032 for I/V on a PCM1792 like the config of the STX. The 2727 would be more appropriate for the reconstruction filter afterwards, although it wouldn't be my first choice.

 

[12Bass]

Quote:

Originally Posted by Filburt /img/forum/go_quote.gif ... I'd personally use a THS4032 for I/V on a PCM1792 like the config of the STX.

Just curious... when a chip is used for I/V, how does that change its impact on the sound, say, compared to using it in an output section? Just wondering how the sonic artifacts imparted by a device might differ depending on application? For example, should one expect a relatively neutral chip like the AD797 to sound neutral when used for I/V (assuming, hypothetically, that it had a sufficiently high slew rate)? Or should one expect a different sort of response for devices when use in an I/V application?

 

[Filburt]

.

 

[12Bass]

Thanks Filburt,

Perhaps my confusion lies in that I'm unsure about what is going on during I/V conversion and what sort of demands it places on the I/V conversion device (commonly an opamp). Above, it is implied that greater bandwidth is desirable in this application; however, I'm not sure why. It's more of a curiosity anyway, as none of my DACs require I/V... though curiously, I have a Sony ST-S555ES tuner which uses current output (which has been modified to voltage).

 

[leeperry]

ok thanks Andrea for the additional infos, and Filburt for the tip! I will order a bunch of THS4032 samples if they offer good synergy w/ my DAC

I think I'll order the 4031 instead, the crosstalk is pretty bad on the 4032....and a very skilled friend of mine will be soldering them on adapters anyway

http://focus.ti.com/docs/prod/folder...t/ths4031.html

THS4031CD or THS4031ID? the ID is more expensive and has better temp. specs, should I go for this one?

 

[ROBSCIX]

Quote:

Originally Posted by leeperry /img/forum/go_quote.gif ok thanks Andrea for the additional infos, and Filburt for the tip! I will order a bunch of THS4032 samples if they offer good synergy w/ my DAC I think I'll order the 4031 instead, the crosstalk is pretty bad on the 4032....and a very skilled friend of mine will be soldering them on adapters anyway High Speed Amplifiers (Greater than equal to 50MHz) - Voltage Feedback - THS4031 - TI.com THS4031CD or THS4031ID? the ID is more expensive and has better temp. specs, should I go for this one?

THS4031, are single channel soic. You will need adapters and a very steady hand.

 

[ROBSCIX]

Quote:

Originally Posted by Filburt /img/forum/go_quote.gif 12V is the specified max but you may be able to run it higher. I've run the OPA2727 at up to about 14V or so. I'd personally use a THS4032 for I/V on a PCM1792 like the config of the STX. The 2727 would be more appropriate for the reconstruction filter afterwards, although it wouldn't be my first choice.

Sure, you might get away with runnig it 1 volt over. These guys are talking about running this opamps at over twice it's Max recommended voltage.

I cannot see that ending very well.

 

[Pluto2]

Excellent finding on ADA4898-1 Kim : )
Hi Filburt, have you compared this ADA chip with AD797? JW claims its much better than LME49722MA.....

 

[12Bass]

Got a pair of the ADA4898-1 around here somewhere, along with a pair of the AD797. Just haven't had a chance to audition them yet. Not sure if I have the proper adapter for the SOIC ADA parts, however. I'll be giving the LME49720/LT1358/LM4562/OPA211/OPA2211A a listen soon as well.

 

[Pluto2]

.....and nicer than AD8599, with better stability etc. than AD797 too : )

12Bass, please do!!!

 

[Pluto2]

And costs only $5 each too! Wondered why there's been so little reviews......I guess it's skepticism, judging from data charts showing it's not an audio chip and is not a good video chip either : )

 

[ROBSCIX]

Will have to try that one. I have many AD units but haven't heard of that model.
I will need to review the spec sheets.

 

[Filburt]

I don't know why Andrea thinks the ADA4898-1 is more stable than the AD797; it's a 65MHz op-amp with no internal feedback and a single stage high gain VAS. ADI doesn't even specify the audio band performance on it, and it looks like it may just not be as optimized as the 797 for that type of use. It looks more like something you'd maybe use with a high speed ADC than an audio app. The THS4051 is OK. I didn't find it to be superlative.

leeperry - The crosstalk spec on the THS4032 is at 1MHz. If you look at the graph, it's probably better than -90dB in the audio band. I doubt you'll do better using the browndog single-to-dual adaptor.

 

[Filburt]

Andrea, I'm not sure why either you or Jim (at least, so you say) have trouble with the AD797 (although in Jim's case I'm more inclined to think it's something reasonable). There are many factors which may be responsible with a device such as this. After working with it for a few months, I found it isn't particularly easy to use, relatively speaking. However, the performance has been good for me since I got some sense of what to do with it. Since I don't subscribe to innateness mysticism when it comes to electronic devices, it doesn't strike me as particularly strange that there will be some divergence of experience based on usage alone. There's also the issue, of course, that some devices just don't work well in certain circuits (such as AD797 as in integrator I/V).

You're free to like whatever you like with these things, but any serious exercise in using high performance chips probably requires a serious exercise in learning appropriate application. That isn't to say to say a particular design doesn't place constraints such that potential performance is bounded, but it does mean that inference and conjecture based on some sort of organic principle about op-amps is a largely frivolous exercise.