There are better small-signal PNP transistors available:
2sa992 (if you need beta > 400, and linearity)
2sa970 (highly linear)
2sa1015 (inexpensive and highly linear)
bf421/423 (inexpensive, fast and Vceo > 200 V)
mpsa92/93
2n5401
or just plain
bc558 (inexpensive, widely available, and audibly indistinguishable from bc560c)
2sa992 (if you need beta > 400, and linearity)
2sa970 (highly linear)
2sa1015 (inexpensive and highly linear)
bf421/423 (inexpensive, fast and Vceo > 200 V)
mpsa92/93
2n5401
or just plain
bc558 (inexpensive, widely available, and audibly indistinguishable from bc560c)
Thanks for the suggession.
well, I am looking for the low noise version.
Vceo is not an issue to me,its ok even vce is max 50v.
I know there are many similar PNP's but I am very much concerned about the Nfe and max Nfe should be > 2db.
If I use substitute ,I think I need to change the passive components as per its electrical characteristics ,if I am wrong please correct me.
well, I am looking for the low noise version.
Vceo is not an issue to me,its ok even vce is max 50v.
I know there are many similar PNP's but I am very much concerned about the Nfe and max Nfe should be > 2db.
If I use substitute ,I think I need to change the passive components as per its electrical characteristics ,if I am wrong please correct me.
I have BC546C and BC556C if you are interested.
http://www.fairchildsemi.com/ds/BC/BC556.pdf
http://www.fairchildsemi.com/ds/BC/BC546.pdf
Probably has the higher noise figure no influence at all. It depends whether the design is optimized regarding the noise. If you'll get 114 or 112 S/N will not be particulary important.
http://www.fairchildsemi.com/ds/BC/BC556.pdf
http://www.fairchildsemi.com/ds/BC/BC546.pdf
Probably has the higher noise figure no influence at all. It depends whether the design is optimized regarding the noise. If you'll get 114 or 112 S/N will not be particulary important.
> Vceo is not an issue to me,its ok even vce is max 50v.
Then you can't do much better than the 2sa1015, which is inexpensive and widely available in most gain grades, including original Toshiba-manufactured devices.
> I am very much concerned about the Nfe and max Nfe should be > 2db.
The noise figure is only of significance in a preamp for moving-coil pickup or similar low-amplitude input. It has little or no significance for a power-amp, even for an LTP input stage. You will find that you will get better improvements on audible thermal noise by switching to metal-film resistors in place of carbon composition resistors, etc. You won't find an audible difference between a bc558c and bc560c in most audio applications - they're just differently graded bins of the same device.
Moreover, 2sa992 and a970 are comparable to or better than the bc560c on NF, hFE as well as Ic/Ib linearity.
> If I use substitute ,I think I need to change the passive components as per its electrical characteristics ,if I am wrong please correct me.
If it's a well designed circuit, there should be no dependence on any transistor parameter, not even the beta. Any small-signal PNP with the appropriate values for Ic/Vceo should be a drop-in replacement for the bc560c without any change in the passives. The only thing that I can think off is that if the circuit is at the edge of stability, than the compensation may sometimes have to be tweaked to maintain stability if you're using a faster transistor in place of a slower one.
Then you can't do much better than the 2sa1015, which is inexpensive and widely available in most gain grades, including original Toshiba-manufactured devices.
> I am very much concerned about the Nfe and max Nfe should be > 2db.
The noise figure is only of significance in a preamp for moving-coil pickup or similar low-amplitude input. It has little or no significance for a power-amp, even for an LTP input stage. You will find that you will get better improvements on audible thermal noise by switching to metal-film resistors in place of carbon composition resistors, etc. You won't find an audible difference between a bc558c and bc560c in most audio applications - they're just differently graded bins of the same device.
Moreover, 2sa992 and a970 are comparable to or better than the bc560c on NF, hFE as well as Ic/Ib linearity.
> If I use substitute ,I think I need to change the passive components as per its electrical characteristics ,if I am wrong please correct me.
If it's a well designed circuit, there should be no dependence on any transistor parameter, not even the beta. Any small-signal PNP with the appropriate values for Ic/Vceo should be a drop-in replacement for the bc560c without any change in the passives. The only thing that I can think off is that if the circuit is at the edge of stability, than the compensation may sometimes have to be tweaked to maintain stability if you're using a faster transistor in place of a slower one.
Linuxguru,
Your post is impressive.
Good catch on Noise figure, "no significance for a power-amp".
I accept your suggession .
I am planning to go with 2sa1015 can you please tell me what is exact complimentary NPN for 2sa1015 ?.
some where on the net I found the equivalence for bc560c is 2N5087 , what's your suggession keeping indian retailers in mind ?.I know, now again speaking of bc560c makes no sense ,but just a curiosity.
Can you elobarate "circuit is at the edge of stability" from your last post.
Regards,
Haseeb
Your post is impressive.
Good catch on Noise figure, "no significance for a power-amp".
I accept your suggession .
I am planning to go with 2sa1015 can you please tell me what is exact complimentary NPN for 2sa1015 ?.
some where on the net I found the equivalence for bc560c is 2N5087 , what's your suggession keeping indian retailers in mind ?.I know, now again speaking of bc560c makes no sense ,but just a curiosity.
Can you elobarate "circuit is at the edge of stability" from your last post.
Regards,
Haseeb
> I found the equivalence for bc560c is 2N5087 , what's your suggession keeping indian retailers in mind ?
The Fairchild 2n5087 is a close equivalent with lower Icmax and fT, but otherwise very similar. However, I haven't checked for availability of this device in Bangalore or Chennai.
> Can you elobarate "circuit is at the edge of stability" from your last post.
If the amplifier circuit is at the threshold of breaking into oscillations, then substituting a higher gain device or faster device in place of the original lower gain or slower device, can sometimes cause the circuit to become unstable and break into oscillations.
The Fairchild 2n5087 is a close equivalent with lower Icmax and fT, but otherwise very similar. However, I haven't checked for availability of this device in Bangalore or Chennai.
> Can you elobarate "circuit is at the edge of stability" from your last post.
If the amplifier circuit is at the threshold of breaking into oscillations, then substituting a higher gain device or faster device in place of the original lower gain or slower device, can sometimes cause the circuit to become unstable and break into oscillations.
Big Flop
Hi Guys,
At last the project was a big flop.
The amp sounding crap like 60's.
The output gain is far far lesser than what is expected.
Hell lot of distortion high quieset current and what not.
The input stage schematic looks quite promising but a crap in practical.
This project Details are posted in another thread under Solid State section (http://www.diyaudio.com/forums/showthread.php?s=&threadid=120484)
But,I have not gave up, I am now doing RND on input stages using npn transistors.
Keep going ......................!
Hi Guys,
At last the project was a big flop.
The amp sounding crap like 60's.
The output gain is far far lesser than what is expected.
Hell lot of distortion high quieset current and what not.
The input stage schematic looks quite promising but a crap in practical.
This project Details are posted in another thread under Solid State section (http://www.diyaudio.com/forums/showthread.php?s=&threadid=120484)
But,I have not gave up, I am now doing RND on input stages using npn transistors.
Keep going ......................!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.