Pass Discrete Opamp in DIP-8 Package

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Disabled Account
Joined 2002
I think you mean Harris who made class A opamps a long time ago. Good sounding too at that time. But IMO class A alone is not a guarantee for good sound. I heard class AB opamps that were better.

JRC/NJR make cheap silicon that sounds mediocre (to stay on the positive note). Never heard a real good product of them while the audio world seems to like their NJM4580, probably because of the price. That NJM4580 can be replaced by LME49860 with guaranteed good results....
 
Last edited:
after seeing this topic i couldn't resist turning this in a DIP8 opamp:
Discrete Operational Amplifier

as its fully class A making it with smaller transistors limits the output current even worse.
I got a 7mA bias, so 1.5mA output is about the max, but it works, has the same pinout as an 741 and it was fun.
sorry about the soldering skills, had a bad day :)
http://img844.imageshack.us/img844/4522/dsc02197xc.jpg
http://img191.imageshack.us/img191/8581/dsc02199dl.jpg
resistors are 0603 and i used NXP BC857's, 2 transistors in a small sot666 package.
 
Hello Patrick,

here the BF862 version.

Is the LSK489 a good candidate for version 1? I´m starting with layouting and I will try to put all on the top side to simplify the heatsink design.

Or is today another simple discrete OPAMP available for DIY? I want to use it for the DSC1 DSD-DAC redesign.

Jean-Paul
 

Attachments

  • BF862_OPAMP_TOP.jpg
    BF862_OPAMP_TOP.jpg
    224.9 KB · Views: 918
  • BF862_OPAMP_BOT.jpg
    BF862_OPAMP_BOT.jpg
    227.9 KB · Views: 882
If I have to do it again today, I would use 2SK209 / 2SK2145 instead.
BF862 has high gate leakage when Vds is >5V.

I still have no solution for the P-channel device, and also for the DC drift.
I would have gone for a LSJ74, but then they do not have them in SOT23.
Maybe you can try J175, but they are rather noisy.


Patrick
 
2SK2145 has no matching guarantee, though they track thermally better than 2x 2SK209.
Also the package can dissipate half the heat of a 2SK209.

For the first stage, the current is determined by the CRD, so the Idss of the JFET pair does not matter.

Perhaps it is worth trying J175 for 2nd stage.
Third stage current is a load / thermal consideration.
You can also make a CRD with 2SK209 using a suitable source resistor.


Patrick
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.