G.Kleinschmidt said:
As being a bipolar small signal fan and designer
I am constantly surprised what can be done
by only using
standard mass produced very low price TO92 like the BC550 / BC560.
Compare Bipolar to most JFET,
and the over price ratio can be as high as 25:1, 50:1
to achieve just about nothing more, by Buying and Using JFET:s.
Add the number of devices you have to buy, to get one MATCHED JFET input pair
compared to BC550C .......
And I wouldn't be surprised to see PRICE ratio 100:1, in favour of standard low noise BJT.
----------------
I must say G.Kleinschmidt circuit is a very good example of this.
And of course I have downloaded one copy of your circuit!
Thanks, G.Klein
Low cost bipolar designs are certainly not out of picture.
Not even at 50-200uA.
And when using 500uA-1.000uA that can't be beat by anything
at same price level.
Lineup Audio Lab
chief designer & audio semiconductor article writer
lineup
G.Kleinschmidt said:
Okay... Here's the logical conclusion of elvee's topology (adding a current mirror).
Iq = 85uA.
Gain = 10dB
THD = 0.02% (6Vpp out, 1KHz, 10K load)
THD = 0.0005% (200mVpp out, 1KHz, 10K load)
Zin > 2M
An externally hosted image should be here but it was not working when we last tested it.
I dunno why people are aghast at a design like this. I mean, people use opamps without batting an eyelid, and this is nowhere near as complex as most opamps.
I think this is what I'll build. SST404s are only a dollar or two, and MMBT06/56s are a dime a dozen. If I want more distortion, I can always lower the quiescent current
Cheers,
Suzy
suzyj said:
That's about double the THD I was getting!
I think this is what I'll build. SST404s are only a dollar or two, and MMBT06/56s are a dime a dozen. If I want more distortion, I can always lower the quiescent current
OK...looks good. A couple of refinements and pointers though:
You can get rid of R11 and R12. DC feedback is provided through R13 (Rf) so R11 and R12 are superfluous. Removing R11 and R12 will reduce your gain, but you can reduce R14 (Ri) in value to compensate.
C4 won’t need to be anywhere near as high as 1uF. As a general rule, the -3dB frequency pole formed by C4 and R14 (Ri) should be equal to or lower than the -3dB frequency pole formed by the input coupling cap and the input resistance. Complying with this rule will eradicate a stepped frequency response in the low frequency roll off.
For frequency stabilisation, C4 between the collector and base of the VAS transistor must form a dominant pole for the whole amplifier. I can almost guarantee that you will need a value much higher than 4p7 for a practical implementation of this circuit with acceptable frequency stability (maybe 15-100pF). Don't be fooled by the phase margin LTspice is giving you. As well as some inherent complexities in the operation of semiconductor devices, LTspice doesn't take into account stray capacitances and other parameters associated with a circuit’s physical construction. Bandwidth approximations in spice simulations for circuits such as these are usually overly optimistic.
Another thing you’ll most likely need to do to ensure unconditional stability (Elvee mentioned this already) is to isolate the load capacitance from the output with a series output resistor of a hundred ohms or so (NFB not taken from the load side of the resistor!).
All up you've got a good circuit there that, with a bit of fudging and experimentation, should work well. Building it will also give you a good handle of how well spice smiulations conform to reality. Have fun!
Cheers,
Glen
suzyj said:
I am not against any good amplifier.
And if you can easily and cheap build it, and you want it that way:
Why not! Who can stop you?
I am for making amplifier simple as possible, like some other.
Few and easy to get parts and n trouble understanding, because is not complex.
A good reason for KEEP IT SIMPLE, is that the more you add into a circuit
the more chance you get some problems.
It is like when you are sick. You take some medicine.
And not so many side effects. Now you take some stronger medicine to get well faster.
Now you can get more bad side effects, which will make feel bad and maybe even sick.
So you take some extra medicine, against that dc-offset
and you want absolutely symmetrical circuit.
And this can be done.
But there is a backside to everything. side effects and other problems by what you introduce.
most any new feature would want some voltage and current to work.
Will cause new currents in rails and ground plane.
Simple but good circuit = few possible causes for trouble.
And easy to find out where problem or unwanted harmonics, distortion comes from.
Complex circuit with many components = more chances for some small unwanted distortion signals.
More rails in between everything.
Maybe even dual supply! Which add another rail from supply.
V+ and Ground is only 2 rails.
I guess you know now, my opinion and how I figure it.
This is why many so simple circuits, at times can outperform complex.
Even though not having dual supply and DC-servo
and error correction this and that .. and you name it ..
lineup
Wavebourn said:
I didn't need to. Look back through the thread, and you'll see plenty of comments (many of which are yours) which in essence distill down to "use as few devices as possible". I find these comments are rarely constructive.
Gaincard hype is a good example of this. I've read reviews of this amplifier, which go on about the 30mm or whatever signal length, and nine components. So what! There's perhaps a hundred components on a bit of silicon inside the poweramp chip.
By gaincard reasoning if I gooped up the nine transistors that make up this amplifier with black plastic, then I could claim there was only one component...
Regards,
Suzy
suzyj said:
It means, you don't want to understand why some people don't like some designs.
No, I did not mean that "as few as possible" is like a Mantra, regardless of putpose. Audio design is always about finding satisfactory compromises.
You may find the proof of my words 2 postings earlier when I confirmed that one resistor more is a good idea. However, if you remove 2 odd transistors this resistors will not be needed...
Wavebourn said:
We are on the same thinking, Wavebourn, many times .. but not always
Neither do I mean with
Keep It Simple!
that we should over simplify.
Because this can mean very bad audio.
But we should think before adding this and that.
We might add unwanted 'features' along with the good things.
My amplifier designing can be seen as having 3 phases:
============================================
1. I have have a good and often simple audio solution.
2. I put together circuit, adding components, one after the other.
Creating voltage working points and current sources and transistors.
Until simulations / real life testings shows very good data.
3. The BIG CLEAN UP.
I remove as many of the added 'features' as I can.
Trying more simple ways, like resistors instead of current sources
and resistor dividers instead of active voltage references.
============================================
Hopefully I end up with a Simple practical implementation
of my good and simple idea.
At least as simple and good as can be.
As Wavebourn just said:
Audio design is always about finding satisfactory compromises.
I agree with this.
I would not say always ... but I would say many times.
--------------------
There are 2 words, I am careful not to use too often:
- always
- never
I know some people at this forum, has not learned this so well, yet
There are more colours in real life, than
- all white and all black.
And likewise in audio designs and matters.
And here, I do not mainly think of Wavebourn.
I say for him, it is more his own way of expressing himself.
Regards, as always
lineup
I don't play acoustic. I saw the prices of nice Taylor/Martins and decided electric was the way to go. Had a Yamaha acoustic, but traded it on my first Fender Strat. Now, have cheap Ibanez Les Paul knockoff.
Distortion on an acoustic? Ewwwwwww.... now that's just silly.
Diode fuzz would of course require much more gain and be more complex than the preamp itself.
I've never seen a DIY guitar effect done without opamps and without tubes. It would be a novelty. Maybe one that doesnt need to exist given the nice THD and Vox stuff that's out there.
Distortion on an acoustic? Ewwwwwww.... now that's just silly.
Diode fuzz would of course require much more gain and be more complex than the preamp itself.
I've never seen a DIY guitar effect done without opamps and without tubes. It would be a novelty. Maybe one that doesnt need to exist given the nice THD and Vox stuff that's out there.
suzyj said:
Hey, don't get discouraged by the Golden Eared loonies that have a penchant for sticking pentodes into microphones and can hear .0005% THD, subjectively quantify an amplifiers ‘stamp’, ‘panache’, ‘mistiness’ and ‘semiconductorness’ and perceive the magical aura of vacuum tubes!
I wanna see a progress report and some lab test results for this little amp.
Cheers,
Glen
G.Kleinschmidt said:
Glen;
you forgot to add: "Don't trust your own ears, trust my beliefs".
Out perception (sight, hearing, everything) reacts better on dynamics than on static. It means, absolute level of harmonics means less than their change. In the nature, the louder is the sound, the richer it is with harmonics. When you kick a drum that suppose to produce clear sinusoidal signals it at the beginning sounds rich with harmonics, during decay their level is less and less. The same about guitar, piano, brass, strings, any musical instrument, any reverberation in any environment. No exceptions, except audio amplifiers that are developed with controversial requirements: fair audio reproduction and effectiveness. In order to save electricity dynamic of sound amplified by such amplifiers is changed, level of harmonics higher when strings and reverberation decay, exponential law of decay is compromised, which is audible better than just constant level of distortions, and even better than distortions of much higher level that are higher on higher level of signal.
Wavebourn said:
No, that should read: "Don't avoid commonsense in favour of delusion and fantasy”.
If you think that the low THD designs presented here produce a level of harmonic distortion that will have any appreciable impact on the sound of a guitar then you obviously don’t know what you are talking about. Either that or you’ve been eating Kryptonite and have developed a sense of perception akin to Super Man.
In fact, thus far, I haven’t seen you make a single meaningful comment or contribution to any of the discrete op-amp circuit design ideas put forward here. The topic of the thread obviously disinterests you and I have no idea why you persist in hanging around, continually posting irrelevant, off topic, nonsensical waffle, other than for the purpose of being a pest.
This is going to be my last response to you in this thread. I’m not going to encourage you to post any more BS.
G'bye
G.Kleinschmidt said:
Naah, I generally just ignore them. I'm developing a good discriminator to sort knowledge from uninformed (or misinformed) opinion.
G.Kleinschmidt said:
I knocked together a little board last night in Protel - 15x32mm, with all parts surface mount. I figure once it's put together I can just bung some heatshrink over the whole thing and hide it amongst the volume and tome pot.
Cheers,
Suzy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.