A) Why do you care? The way the devices are used has more effect on "tone" than the actual devices. Especially in "chips" which are professionally designed and are very generous with devices.
B) What "chips"? There have been so many. On some (mostly older) the datasheet shows partial guts. LM383 data shows all BJTs (there could be a JFET hidden in the bias but that's probably not of interest). LM3886 is likewise an all-BJT design. You don't generally find JFETs in "power amps"; their special virtues are not essential and BJTs provide more gain per unit of area (=$ in chip making). You do find JFETs in the audio-popular TL072 where JFETs allow "zero" input bias currents and thus low hiss in high impedance inputs.
B) What "chips"? There have been so many. On some (mostly older) the datasheet shows partial guts. LM383 data shows all BJTs (there could be a JFET hidden in the bias but that's probably not of interest). LM3886 is likewise an all-BJT design. You don't generally find JFETs in "power amps"; their special virtues are not essential and BJTs provide more gain per unit of area (=$ in chip making). You do find JFETs in the audio-popular TL072 where JFETs allow "zero" input bias currents and thus low hiss in high impedance inputs.
If you do a parametric search on ti.com you can add the "architecture" column by checking the box on the left hand side.
Precision Op Amps | Products | Operational Amplifiers | TI.com
Precision Op Amps | Products | Operational Amplifiers | TI.com
IC circuit diagrams are often simplified or just plain wrong. Why should the IC company reveal a trade secret?
They also change a lot over a product lifetime. You can be confident that an early TL081 used very different IC fabrication technology to a recent part, with a much smaller die size and probably using ion implantation rather than diffusion.
They also change a lot over a product lifetime. You can be confident that an early TL081 used very different IC fabrication technology to a recent part, with a much smaller die size and probably using ion implantation rather than diffusion.
Thanks for the replies everyone.
PRR - Because in my experience bipolars in the output stage sound better than fets. But the fact that you are telling me the lm3886 is all bjt is messing with me cuz when I heard one it had no balls at all (in a Cambridge Audio amp) compared to an earlier version of the same amp which used a TDA1514a which i thought was all bipolar as well but now I'm not sure!
Thanks Paul. I can't seem to find a parametric search function though. Perhaps I have to register to do that.
JMFahey - What I mean by chip is chip amp. We're in the chip amp forum!
Davidsrsb - I'm not saying they should.
PRR - Because in my experience bipolars in the output stage sound better than fets. But the fact that you are telling me the lm3886 is all bjt is messing with me cuz when I heard one it had no balls at all (in a Cambridge Audio amp) compared to an earlier version of the same amp which used a TDA1514a which i thought was all bipolar as well but now I'm not sure!
Thanks Paul. I can't seem to find a parametric search function though. Perhaps I have to register to do that.
JMFahey - What I mean by chip is chip amp. We're in the chip amp forum!
Davidsrsb - I'm not saying they should.
You can usually tell from the input bias current. If it's in the pA range, you're dealing with a MOSFET input amp, around 1-10 nA is likely a JFET input, 100 nA and beyond would indicate a bipolar/BJT input. It's still a guess, but at least then an informed guess.
If the input is a MOSFET input, the entire chip could be CMOS or BiCMOS (mix of bipolar and CMOS). If the input is JFET or BJT, it's most likely a BiCMOS chip. It could be a mostly bipolar opamp with a few FETs sprinkled in where needed. I've designed a few opamps like that.
Tom
If the input is a MOSFET input, the entire chip could be CMOS or BiCMOS (mix of bipolar and CMOS). If the input is JFET or BJT, it's most likely a BiCMOS chip. It could be a mostly bipolar opamp with a few FETs sprinkled in where needed. I've designed a few opamps like that.
Tom
> ...messing with me cuz when I heard one it had no balls at all ... compared to an earlier version of the same amp which used a TDA1514a which i thought was all bipolar as well but now I'm not sure!
See? Implementation details are more important than devices. Don't get a grudge from limited testing.
See? Implementation details are more important than devices. Don't get a grudge from limited testing.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.