John Curl's Blowtorch preamplifier part II

[john curl]

We are talking about engineering problems and if they are completely addressed.

 

[1audio]

I doubt any current production opamps use masks or processes from the '70s or '80s. Those processes (and wafer sizes) have all been obsoleted.

However the external issues (e.g. the thermal effect on resistors shown on pg 3 of the cited app note) could well contribute to "thermal tails". In real audio circuits there should be a way to measure these transient effects if they are affecting the audio. Any suggestions?

 

[john curl]

Yes, just get one of those TEK adaptors.

 

[Bonsai]

Except that MOST IC's designed in the 70's AND used even in today's products, were designed improperly.

In the right hands, the 5532/5534 is a very fine amplifying device.

 

[Bonsai]

I doubt any current production opamps use masks or processes from the '70s or '80s. Those processes (and wafer sizes) have all been obsoleted.

Correct

 

[fas42]

Just did some listening tests where the 5532 trounced the latest and greatest - it's all about implementation ...

 

[Bonsai]

We are talking about engineering problems and if they are completely addressed.

They were addressed and solved 40 years ago . . .

 

[hitsware]

> 5532 trounced the latest and greatest

I think its the current capability ...

In the same way that a 100wpc amp will sound better
than a 10wpc amp (both @ 1 Watt)

 

[john curl]

Wow! All we need is the 5532/34. Yet, I can improve it, and have done it dozens of times WITH THE SAME FOOTPRINT. Of course, you have to be able to hear the difference. Dave Wilson bought this 'improvement' from us decades ago for his WAMM system.
Perhaps we should discuss what we did. And even YOU HERE can try it for yourself. '-)

 

[Barleywater]

Just did some listening tests where the 5532 trounced the latest and greatest - it's all about implementation ...

Are you referring to Mooly's off the cuff plug and play swap with TLE2072?

 

[fas42]

Yes, but another round was done with the LM4562, in different configurations. Theoretically that should have done the job, but in that implementation it did somewhat poorly - in fact, quite obvious distortion was occurring on peak levels ...

 

[Barleywater]

And where are details of Mooly's LM4562 implementation and test setup?

 

[Johnloudb]

And where are details of Mooly's LM4562 implementation and test setup?

Yeah, I second that. What kind of test? A listening test? Or measurement?

I'm using an LME49720 in a crossover (gain of 5), with a JFET buffer at the non inverting input. I only measured a THD of 0.00067% all 2nd and 3rd harmonic distortion at 1 volt peak, most like due to the JFET buffer. When you start talking about listening tests you get into fuzzy territory. They don't really prove anything, they just offer one persons opinion versus another. Nothing wrong with offering an opinion though!

 

[canyoncruz]

,why does my mesa boogie sound better in class a/b than in class a?

 

[canyoncruz]

Has anyone designed a discrete mixed feedback circuit before? Is there an example out there? Else I am going to have to use opamps

 

[fas42]

And where are details of Mooly's LM4562 implementation and test setup?

Need to speak to Mooly about what was done - it was organised via PM ...

 

[1audio]

The LM4562 (AKA LME49720) is really good when driven from a low impedance source. In noninverting with a 600 Ohm or higher source impedance its distortion rises as much as 20 dB. But that is true of many opamps (and discrete circuits). As said before its all about implementation. The LME49880 dual JFET opamp is significantly better in this parameter.

 

[Jay]

,why does my mesa boogie sound better in class a/b than in class a?

Drive? Try using small crossover less full range speaker to make THD more determinant.

 

[1audio]

,why does my mesa boogie sound better in class a/b than in class a?

Perhaps the distortion sounds good? There are companies that have made fortunes selling distortion generating boxes. Aphex Exciter | Aphex Electronics is a prime example.

 

[1audio]

Yes, just get one of those TEK adaptors.

how would I apply that to an assembled system? It doesn't take a lot to hack together the basics of the Tek, but that was designed to work with the opamp open loop.