Hi
The below curve shows why many constructors turn to BJT's for good linearity in the output stage, and not Mosfet.
The Linear region of operation starts when the output impedance becomes roughly independent of the load current. In this region the output stage doesn't contribute with any significant distortion.
On the curve this corresponds to a flat horizontall line.
As you can see you only have to run the Ic current at a certain level (depending on the transistor type 70 - 500 mA) to get a Ic independent output impedance.
However in the MOSFET the curve only gets completely flat at something like 5-10A, but it gets closer and closer to horizontal, the higher the Id.
This is why MOSFET's are good for Class A amplifiers, providing you keep the idle current above 5-10 Amps. But MOSFET will never get a low THD in Class A/B amplifiers. I have made bunches of MOSFET amplifiers, (like the ZETA), the good thing about MOSFET is that they are very easy to work with, don't require a lot of gate drive, are rugged, easy to come by and dirt cheap. But they will never be the ultimate output device for high quality amplifiers.
(Well maybe except for Class D of course
)
On a more academic level, there is nothing that prevents you from building a Pass amplifier using BJT's instead of MOSFET. Of course it's no longer a Pass Amp, but it should work with no significant complications. I recommend a good BJT from SANKEN 2SC2922. It should fill in nicely for the IRF's, and for idle currents below 5 Ampere is bound to give you a better result than MOSFET.
The below curve shows why many constructors turn to BJT's for good linearity in the output stage, and not Mosfet.
The Linear region of operation starts when the output impedance becomes roughly independent of the load current. In this region the output stage doesn't contribute with any significant distortion.
On the curve this corresponds to a flat horizontall line.
As you can see you only have to run the Ic current at a certain level (depending on the transistor type 70 - 500 mA) to get a Ic independent output impedance.
However in the MOSFET the curve only gets completely flat at something like 5-10A, but it gets closer and closer to horizontal, the higher the Id.
This is why MOSFET's are good for Class A amplifiers, providing you keep the idle current above 5-10 Amps. But MOSFET will never get a low THD in Class A/B amplifiers. I have made bunches of MOSFET amplifiers, (like the ZETA), the good thing about MOSFET is that they are very easy to work with, don't require a lot of gate drive, are rugged, easy to come by and dirt cheap. But they will never be the ultimate output device for high quality amplifiers.
(Well maybe except for Class D of course
)On a more academic level, there is nothing that prevents you from building a Pass amplifier using BJT's instead of MOSFET. Of course it's no longer a Pass Amp, but it should work with no significant complications. I recommend a good BJT from SANKEN 2SC2922. It should fill in nicely for the IRF's, and for idle currents below 5 Ampere is bound to give you a better result than MOSFET.
Attachments
pass with bjt's
I understand from previous discussions that just replicating the Pass circuit with BJT's (take for example the mini-aleph, a 2-stage design) is not so easy. One small transistor will not drive a bjt power stage to reasonable levels. Would it be possible to have the circuit here on the forum with some values and numbers, better still with some measured or simulated data? I guess a great deal of diy'ers are interested!
I understand from previous discussions that just replicating the Pass circuit with BJT's (take for example the mini-aleph, a 2-stage design) is not so easy. One small transistor will not drive a bjt power stage to reasonable levels. Would it be possible to have the circuit here on the forum with some values and numbers, better still with some measured or simulated data? I guess a great deal of diy'ers are interested!
What we need is power stage with BJT. with two transistor i get load current of 500 ma rms with max input (300mv).
How did you get 5 A on load with just BJT transistors?
Anyway , you all find it . it seems like you all have built so many amplifiers.
I am waiting..... meanwhile who has best simulation program?.
Belated Merry christmas and Happy new year to all.
ALexk
Toronto
How did you get 5 A on load with just BJT transistors?
Anyway , you all find it . it seems like you all have built so many amplifiers.
I am waiting..... meanwhile who has best simulation program?.
Belated Merry christmas and Happy new year to all.
ALexk
Toronto
Schematic
see my post above
with this picture attachment:
http://www.diyaudio.com/forums/attachment.php?s=&postid=797165&stamp=1135293820
Output here is a combination of 2 transistors, because like rmgvs says, BJT power transistors are not so easy to drive.
This 2T output device gives an easier load for the voltage amp stage.
Also Q2 shouldnt have too low current.
Like 15-30 mA, depending of how much power amp is to produce.
As said before (R2/R1 + 1) is the voltage gain.
And the resistor R8 should be replaced with a current source.
Could be 1-1.5 A
Voltage supply could be 24-30 V.
see my post above
with this picture attachment:
http://www.diyaudio.com/forums/attachment.php?s=&postid=797165&stamp=1135293820
Output here is a combination of 2 transistors, because like rmgvs says, BJT power transistors are not so easy to drive.
This 2T output device gives an easier load for the voltage amp stage.
Also Q2 shouldnt have too low current.
Like 15-30 mA, depending of how much power amp is to produce.
As said before (R2/R1 + 1) is the voltage gain.
And the resistor R8 should be replaced with a current source.
Could be 1-1.5 A
Voltage supply could be 24-30 V.
S 30:
snip Lars
"It's not very special or unconventional..."
Hello Lars
indeed, I meet something in the early 70ths.....
may be I am too old!
snip rmqvs
"Why do you put the speaker to the + volage rail?"
I am so free to answer:
Because to save 30 cent for a R and a C! .....
...for that You get a gruesome powersupplyrejection!
Regards
Heinz!
snip Lars
"It's not very special or unconventional..."
Hello Lars
indeed, I meet something in the early 70ths.....
may be I am too old!
snip rmqvs
"Why do you put the speaker to the + volage rail?"
I am so free to answer:
Because to save 30 cent for a R and a C!
........for that You get a gruesome powersupplyrejection!
Regards
Heinz!
Heinz: Right on. But as you must have guessed already it's not about the 30c but about making the simplest solution with the fewest possible parts. (As in the header of the tread).
But like i said before, this is not the same as the best solution........
In fact i can only advise everyone against building the oversimplified amplifier circuits, you will never get a very good result from that, so you are really wasting your time.
But like i said before, this is not the same as the best solution........
In fact i can only advise everyone against building the oversimplified amplifier circuits, you will never get a very good result from that, so you are really wasting your time.
mini-aleph
To Lars as asked in post #104:
http://www.kk-pcb.com/mini-a.html
Would be nice to have such a simple circuit with only BJT's. Two stages, single-ended, moderate amount of overall negative feedback, no output cap betweeen amp and speakers, fairly efficient (like push-pull class A that is), easy to build, measures ok, reasonable amp to drive from normal preamp etc.
To Lars as asked in post #104:
http://www.kk-pcb.com/mini-a.html
Would be nice to have such a simple circuit with only BJT's. Two stages, single-ended, moderate amount of overall negative feedback, no output cap betweeen amp and speakers, fairly efficient (like push-pull class A that is), easy to build, measures ok, reasonable amp to drive from normal preamp etc.
Re: mini-aleph
Lars Clausen, Nelson Pass & me, Lineup + others have here posted some 'simplest bjt' circuit suggestions.
*** Is it possible to build as simple as Mini-Aleph using bipolar only ??
I would say: NO and I would say: YES.
To build a simple working Power Amp is maybe by some opinions easier with MOSFET.
If we would take a bit increased distortion (probably a bit of 2nd harmonics).
If we want a 'better hifi' (lower distortion), it is just as difficult/easy with either MOSFET or BJT.
If we can take a bit of distortion ( so called 'lowfi'), I would say it no more difficult to build a 'simplest amplifier' with bipolar.
May I add,
that to build a good amplifier, in any respect, always takes a great deal of work & knowledge.
Mainly, I conside it good To Know your transistors & Components well.
And when you have used 'your' similar devices in like 20-30 applications,
you will know better how to use them .... and how not to
---------------------------------
For example,
when I use my olden golden op-amp NE5534 .. I know how to set it up.
So it works very good
This does not mean it is no better than modern opamps.
Just that it takes time to learn new opamps and try them in 20-30 applications.
If I took me the time, I would get better results with them.
My Bottomline:
- Better to use a device/component that you know
than
- Use some 'better' device, you can not handle properly.
or
- You may have to spend some time, trial & errors to Get More Knowledge.
Lineup audio regars
_______________________________________________
PS. Nelson Pass knows IRF9610 and IRF240
.. like his own backyard ....
He has used them not 20-30 times .. but more like 200-300 times.
And can tell you right away what is good and not so good way using IRF9610 / IRFP240
if you wake him up any time in his night sleep !!!!
DS.
rmgvs said:
Lars Clausen, Nelson Pass & me, Lineup + others have here posted some 'simplest bjt' circuit suggestions.
*** Is it possible to build as simple as Mini-Aleph using bipolar only ??
I would say: NO and I would say: YES.
To build a simple working Power Amp is maybe by some opinions easier with MOSFET.
If we would take a bit increased distortion (probably a bit of 2nd harmonics).
If we want a 'better hifi' (lower distortion), it is just as difficult/easy with either MOSFET or BJT.
If we can take a bit of distortion ( so called 'lowfi'), I would say it no more difficult to build a 'simplest amplifier' with bipolar.
May I add,
that to build a good amplifier, in any respect, always takes a great deal of work & knowledge.
Mainly, I conside it good To Know your transistors & Components well.
And when you have used 'your' similar devices in like 20-30 applications,
you will know better how to use them .... and how not to
---------------------------------
For example,
when I use my olden golden op-amp NE5534 .. I know how to set it up.
So it works very good
This does not mean it is no better than modern opamps.
Just that it takes time to learn new opamps and try them in 20-30 applications.
If I took me the time, I would get better results with them.
My Bottomline:
- Better to use a device/component that you know
than
- Use some 'better' device, you can not handle properly.
or
- You may have to spend some time, trial & errors to Get More Knowledge.
Lineup
audio regars_______________________________________________
PS. Nelson Pass knows IRF9610 and IRF240
.. like his own backyard ....
He has used them not 20-30 times .. but more like 200-300 times.
And can tell you right away what is good and not so good way using IRF9610 / IRFP240
if you wake him up any time in his night sleep !!!!DS.
Geezzz, can you perhaps lay off the old topics? Your reply brings absolutely nothing to the table that would make it even worthwhile to read, its just a personal opinion really.
Cheers,
Sander Sassen
http://www.hardwareanalysis.com
Cheers,
Sander Sassen
http://www.hardwareanalysis.com
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.