[Andrea]

in an unbuffered headphone amp?

The specifications say: 70 mA linear output current, FET input, 8 MHz, 20 V/us (& 0.0005% THD+N although not on a headphone load of course

).

It looks perfect for the application, all the more so when it should drive a 300 ohm headphone like in my case.



Does anyone have some experience with these chips? Thanks for sharing

 

[ppl]

this topic has been talked about bfore i might sugest using the search however i did that for you and what you want to know is hear

 

[Andrea]

ppl, thank you! I read through that thread and as a result I think I'll drop the idea

I've other op-amps at hand, which are likely better for my purpose.

One of them being the LT1363 / LT1364. My only concern with this chip (besides it not being FET, which I can deal with) is that high frequency peaking, especially with capacitive loads. My (incoming) amp has a resistor (don't know what value, perhaps 33 or 47 ohm) to isolate the load, but still I'm unsure if that can lead to instability. Oh, and there are tantalum bypass caps on each power rail which should help.

 

[ozshadow]

Wonder what kinda sparks this one would make - is it even useable in an amp ?

http://www.national.com/pf/LM/LM7171.html

 

[Andrea]

Aparently yes, since that (LM7171) is just what the Creek OBH21 and OBH21SE amps have inside (one per channel, without buffers).

But some pretty special care is required to tame the DC offset at the output, besides careful power supply bypassing & board layout due to the chip's very high speed. With a 10 uA input bias current, I doubt that simply balancing the impedances on both inputs of the op-amp can suffice

As for my particular purpose, the LM6172 is more likely to work OK.

 

[ATAT]

The 6172s are great.. 7171s are a beast.. they'll oscillate like hell and cause all sorts of issues unless the circuit is designed for high speed opamps..

Also - Andrew, enjoy the 6172s, they have a kind of warm yet authoritative sound sig that I like more than any of my other opamps..

 

[Andrea]

Mmmm... ATAT, thanks for the encouragement!

But I don't have the 6172 right now. I must order a few samples, which will cost me $10. I think I'll first give the LT1364 a try. I liked the LT1361 (the less powerful brother) a lot before. Oh, and I'm also curious to hear the AD45048.

 

[Andrea]

Quote:

Originally Posted by Andrea One of them being the LT1363 / LT1364. My only concern with this chip (besides it not being FET, which I can deal with) is that high frequency peaking, especially with capacitive loads. My (incoming) amp has a resistor (don't know what value, perhaps 33 or 47 ohm) to isolate the load, but still I'm unsure if that can lead to instability. Oh, and there are tantalum bypass caps on each power rail which should help.

bump...

 

[Andrea]

buump...

 

[Andrea]

buumpp...

 

[ATAT]

So much bumping

Since I'm not as convincing as the engineers at LT, i'll just quote the datasheet =p

Quote:

Capacitive Loading The LT1364/LT1365 are stable with any capacitive load. This is accomplished by sensing the load induced output pole and adding compensation at the amplifier gain node. As the capacitive load increases, both the bandwidth and phase margin decrease so there will be peaking in the frequency domain and in the transient response as shown in the typical performance curves. The photo of the small signal response with 200pF load shows 62% peaking. The large signal response shows the output slew rate being limited to 10V/ms by the short-circuit current. Coaxial cable can be driven directly, but for best pulse fidelity a resistor of value equal to the characteristic impedance of the cable (i.e., 75W) should be placed in series with the output. The other end of the cable should be terminated with the same value resistor to ground.

Basically, your opamp will be stable at any capacitative loading because of the compesatnion at the gain.. this means you can load it however the hell you want. THe only problem is the "high freq peaking" as you say, but this wont change stability, it'll just magnify the FR above 1MHZ, this *might* magnify the RF noise injected into your system, so you could make a low pass filter with a -3 db point at 8 Mhz or so.

Of course this doesnt fix overshoot, which is another problem, but I have no clue how you would solve that one.

Basically it'll work, but you'll have to see how the RF noise / overshoot sounds

 

[Andrea]

Great answer, thanks.


Yeah, overshoot looks a bit of an issue with the LT1364. Less so with the LT1361. Then again, it looks so even with the AD8397. Perhaps I'll order that LM6172

 

[Andrea]

In fact there are other op-amps I'll try first: (2x) OPA551/552, AD8397/45048, (2x) AD8655. Just for the record, the amp will be the Practical Devices XM3. The designer himself (I think?) replied to my inquiry that bipolar input chips like the AD8397 and LT1361 will work OK in it.