Bob Pease on the New LM4562

[gaetan8888]

Hi

Anyone do have the Bob Pease article on the LM4562 and could send it to me by email ?

Thank

Gaetan

 

[DSP_Geek]

Yes, a part of that low level thd+n are the noise floor, but with a typical input noise of 2.7 nv at 1 khz the raise of thd+n whould not start at 5 volt but at a lower output voltage less than 5 volt ?

Look closer at the noise number: it's 2.7 nV/sqrt(Hz), where sqrt is square root. That means you must multiply the raw number by the square root of the bandwidth to get the real noise value.

For example, if you have (to make things easy) 10 kHz bandwidth, then take the square root of 10,000 to get 100 sqrt(Hz) , multiply by 2.7 nV/sqrt(Hz), and get 270 nV. That calculation only works if the bandwidth past 10 kHz drops like a stone, which it wouldn't. Typical op-amp circuits roll off at 6 dB/octave, so the spectrum beyond rolloff would also contribute to the noise figure. I haven't run the numbers, but another 3 dB of noise due to post-rolloff spectrum sounds about right. For this example you'd multiply 270 nV by 1.414 (3 dB) and get 382 nV of noise.

So, going back from 1 uV, we see the factor is about 370. Reduce that by 1.414 to account for post-rolloff, and we get a multiplier of about 262, which squared yields a corner frequency for the test of about 70 kHz, which seems to be a bit high for typical opamp testing.

Let's now look at the IM distortion curve. It increases only by a factor of 10 between outputs of 1 volt and 100 millivolts, whereas THD+N increases by a factor of 100 between those two points. All else being equal, one might expect IM to track THD if there were crossover distortion. Since it doesn't, we must conclude the THD+N is dominated by the noise component instead of harmonic distortion.

 

[peranders]

Hi

Anyone do have the Bob Pease article on the LM4562 and could send it to me by email ?

Thank

Gaetan

Google words
LM4652 bob pease

Tested that?
http://www.edn.com/index.asp?layout=blog&blog_id=1700000170&blog_post_id=780007278
http://www.edn.com/blog/1700000170/post/790011279.html

 

[gaetan8888]

Hello peranders

I did ask that question because wen I did try to read this article;
http://www.elecdesign.com/Articles/...3166/13166.html

There was a never-ending advertising and I can not read the article.

Thank

Gaetan

 

[peranders]

Could you fix the link please?

You can't copy the text you see, you must copy the real link.

 

[AndrewT]

the first link gives me a search and adverts.
What are we supposed to change?

 

[gaetan8888]

Could you fix the link please?

You can't copy the text you see, you must copy the real link.

Hello

Oups.. here it is;

http://electronicdesign.com/Articles/ArticleID/13166/13166.html

It is the link in the first post of this tread.

Thank

Gaetan

 

[col]

LM4562 in Rod Elliot ESP P88 and P09 crossover

Hi,

Recently received 3 sample LM4562 from natsemi and have put 2 of them in a P88 preamp which has 6db of gain on the first stage and 7.86db on the second stage. This is the default gain structure when building the preamp from the AI.

I replaced 2 OPA2134 with the LM4562. The P88 preamp is powered by the ESP P05 PSU which uses variable regs (LM317/LM337). It is very clean and has virtually no noise.

The P88 preamp feeds a 3 way stereo active crossover which is assembled from 2 x ESP P09 24db linkwitz riley crossover boards and is also powered from the P05. I am currently using one LM4562 as the input buffer to the crossover, 4 x OPA2134 as the output buffers, 4 x OPA2134 in the high-mid filters and 4 x NE3522 in the low-mid filters.

It is sounding very nice

There is some pictures of the project here:

http://minirig.org.au/2008/06/19/minirig-interm-esp-crossoverpreamp/

and I wrote it up in diyaudio as I built it:

http://www.diyaudio.com/forums/showthread.php?s=&threadid=123409

I read this entire thread before posting. It is very interesting to hear the connoisseurs of opamps discussing their latest meal

The thing I found most interesting is the combinations of opamps depending on there task.


So I am wondering if I have it correct for my setup?

My preamp/crossover has:

8 dual opamps for crossover filters (High and high-mid OPA2134, low and low-mid NE3522).
4 as crossover output buffers (OPA2134)
1 crossover input buffer (LM4562)
2 preamp (LM4562)

Can anyone suggest a better configuration of opamps? I'm wondering wether to take out the LM4562 from the second stage of the preamp and put back the OPA2134. I guess I can always try it .

col.