Bob Cordell Interview: Error Correction

lumanauw said:
Hi, Toino,

It really works? That's good news 😀

What is your impression of the sound before and after EC is implemented. Is it better sounding?


Hi lumanauw

I have made that modification some time ago. At the time I have made some simulations on pspice V8 before test the circuit in the amplifier.
I must search for the files in my hold backups and then I will post it here. Just give me some days…
The amplifier is not Hi-Fi. It is a 4 kilowatt beast for concert loudspeakers and I have never listened it with the EC.
But the THD was so low that the A1-Neutrik I have at the time could not measure it. (< 0.003%)
I remember that I have constructed a separated working prototype based on the simulations… mosfet drivers and bipolar output devices… will see the files when I found then.
 
lumanauw said:
Hi, Toino,

Traderbam found out that the modified EC (which is driven from the emitors of the VBE multiplier) is not working.

In here
http://www.diyaudio.com/forums/showthread.php?postid=1061815#post1061815
You measure (with Neutrik A1?) and it works to lower distortion ?

Lumanauw
Don’t worry about the circuit. Surely it works… just give me some time to search for the files.
Driving it from the emitters forms a kind of “current feedback” amplifier
 
lumanauw said:
Hi, EDL,

The Etalon Integra is not using global feedback. Is this amp low biased classAB or full classA? How many mA bias in final transistors? I try to figure out whether the CCT used in Integra is HEC or not.

What is the value of VR in bases and in emitors of BD139-BD140 (2 pots marked with "just")?

Etalon Integra-1 is HEC cct, output bias is ~300mA , VR on the bases is 20k, VR on the emitters of BD's is 10k.
 
Untried and untested Yokoyama-Hawksford mods.

Hi Andy_C.

Continuing with our discussion from here, here and here.

Some obvious mods, the most notable of which is the increase in voltage efficiency, and the change of drivers to BJTs: i am generally uncomfortable with MOSFET drivers. 😉

Whether the whole thing makes sense in practice, or even SPICE, remains to be seen. 😀
 

Attachments

  • hyec.png
    hyec.png
    16.7 KB · Views: 412
Hi Mike,
I've been working on a similar idea with a bjt output amplifier. Right down to the current sources. My error reference is right from the Vas. Still playing with it, but the technique is effective.

I've been delayed over a year. 😡 One day I'll get back to it. You guys are giving me a stronger urge to complete it.

-Chris
 
Hi Mike,
It's interesting. I get lost in the math of this and it comes right around to something I've put together. Cool! 😎 I did find that BJT outputs sounded better than mosfet types in my work, and not by a little bit either.

You now have my complete attention sir! Use small words so I can follow you. 😉

-Chris
 
Hi Mike,
Yes, and my error detector is reasonably high impedance. The Vas has enough current so far.

I don't know that this would translate into an amplifier with 80 V rails though. I'm working around the 35 ~ 50 V mark now.

I built my correcton circuit on a small PCB. This is so I could move the connections around to test it. One day I swapped the inputs around by mistake. It took me a bit to figure out why the performance was worse! Stop laughing!

-Chris