So really its distortion-less, as the sim shows exactly the inherent THD within a AD826. Am I correct, and who wants to start their apology first ?
Cheers / Chris
Cheers / Chris
Dear Chris,
The sheet shows THD harmonics curves for a 2x gain, which is different from this circuit which is unity gain. Secondly, no other specifications about load are given, other than the test frequency of 1MHz. The simulator uses a model that describes the opamp's behaviour and it uses ideal power supplies. That's going to be less distortion than it will be for real. You still conveniently omit the fact that the directly powered opamp has a distortion of < 0.0005%. So, let's say if this directly connected opamp in reality will yield 0.013% @ 1MHz, the transistored version will be way worse off as that mod is the difference between the two opamps in the simulation, resulting in these different results, in this case, worse THD.
I advice you not to increase the base resistor - the spikes will turn progressively worse, pulling the sinewave completely out of shape.
Think of it in a natural fashion: Compare the opamp to a car engine. Here comes this fat intro of "Born in the U.S.A.", demanding output power from the opamp like the engine would demand fuel after stepping it down. Then surely you ain't gonna cut the fuel by pinching it off under various conditions, and with this opamp, cut its life support, it's food lines? If anything, you want to give it a supply with an as low an impedance as possible and that's not what the transistor circuit offers.
The sheet shows THD harmonics curves for a 2x gain, which is different from this circuit which is unity gain. Secondly, no other specifications about load are given, other than the test frequency of 1MHz. The simulator uses a model that describes the opamp's behaviour and it uses ideal power supplies. That's going to be less distortion than it will be for real. You still conveniently omit the fact that the directly powered opamp has a distortion of < 0.0005%. So, let's say if this directly connected opamp in reality will yield 0.013% @ 1MHz, the transistored version will be way worse off as that mod is the difference between the two opamps in the simulation, resulting in these different results, in this case, worse THD.
I advice you not to increase the base resistor - the spikes will turn progressively worse, pulling the sinewave completely out of shape.
Think of it in a natural fashion: Compare the opamp to a car engine. Here comes this fat intro of "Born in the U.S.A.", demanding output power from the opamp like the engine would demand fuel after stepping it down. Then surely you ain't gonna cut the fuel by pinching it off under various conditions, and with this opamp, cut its life support, it's food lines? If anything, you want to give it a supply with an as low an impedance as possible and that's not what the transistor circuit offers.
Last edited:
Magic Box
Exactly the opposite happens, dynamics improve immensely with the transistor connected model, The sim is visually appealing for some, and has I think fooled many. It reliably proves though that is correctly attributing the inherent THD of an AD826 which is .01258 which must apply.🙂 And yes I agree 6ma is best.
Cheers / Chris
Exactly the opposite happens, dynamics improve immensely with the transistor connected model, The sim is visually appealing for some, and has I think fooled many. It reliably proves though that is correctly attributing the inherent THD of an AD826 which is .01258 which must apply.🙂 And yes I agree 6ma is best.
Cheers / Chris
You're free to think what you like. While simulators do not show a perfectly realistic world, but rather, a too perfect world, the difference with reality is minor. Just hook up a scope across the collector and emitter and do a frequency sweep. You'll see the modulation on the opamp supply pin and if you set up your scope properly, even get into view the distortion. Don't you have the urge to back your idea with the actual electronic reason for the subjective improvement?
Now Now Don,
It is the Holy season where is your charitable heart?
Op-Amps garbage? No they are not!
I too love tubes and the tube warm glow and sound.
Remember all the hundreds of Op-Amps the signal has gone through before the DVD, CD or LP got to your house.
DT
No private joust is taking place. I received two strange PMs from madtecchy but decided to ignore them. As far as I am concerned this is a public technical debate, with the usual public non-technical comments taking place too.madtecchy said:
This thread seems to be stuck in the usual 'ears' vs. simulation vs. theory debate. My line on this is that if someone puts forward a theoretically unhelpful circuit which is claimed to improve sound then the onus is on him to demonstrate that it really does improve sound and to offer a plausible explanation of how it does this. So far we have neither of these. The original confusion about what an emitter follower is and how it works does not fill me with confidence.
That in bold is at least something that can be quantified. dB figures for stereo separation before and after the mod would be interesting to see.
I dont take offense, I take pity. And I never said anything about simulation so stop making stuff up. And I have no idea what the rest of the rambling means.
My sentiments exactly, smells a lot like snake oil. Why does asking for a plausible explanation always start the mud slinging? (because thats all they have)
Memory loss
So you also want to attack me on this forum.Just read the first post you aimed at me . That is mud slingin dude . I see you do not like it when it is aimed back at you. And react just like a child . Wind it in dude this is not a place for pitty or mud slingin... And errm i gave you the proof . I think it sounds better and so do others lots of others . Now if you would like to show me a graph or somthing maybe a screen shot off your computer that shows me im hearing wrong i would be very intrested to see how it sounds just like you. DuRRRRRRR
So you also want to attack me on this forum.Just read the first post you aimed at me . That is mud slingin dude . I see you do not like it when it is aimed back at you. And react just like a child . Wind it in dude this is not a place for pitty or mud slingin... And errm i gave you the proof . I think it sounds better and so do others lots of others . Now if you would like to show me a graph or somthing maybe a screen shot off your computer that shows me im hearing wrong i would be very intrested to see how it sounds just like you. DuRRRRRRR
Hi
The circuit I have come up with involving DC with a NPN transistor and PNP transistor, connected to a op amp power supply joined by a resistor at 6ma bias to each polarities base , bears similarity to, a collector to base self biased circuit, However Rc normally the mechanism by which current is drawn from V+, is for the most part absent as it is for the most part direct , other than resistance back to the V+ and V- generation point- ie negligible resistance, however obviously integral for operation .
Moreover the emitter, normally in the collector to base self biased circuit the opposite polarity
may I think be termed in this case as Re which is unconventional. I determine that the op amp itself presently tested with AD825, opa627 and opa1611 interacts favorably to improve power supply connection with transistors connected in this manner.
Cheers / Chris
The circuit I have come up with involving DC with a NPN transistor and PNP transistor, connected to a op amp power supply joined by a resistor at 6ma bias to each polarities base , bears similarity to, a collector to base self biased circuit, However Rc normally the mechanism by which current is drawn from V+, is for the most part absent as it is for the most part direct , other than resistance back to the V+ and V- generation point- ie negligible resistance, however obviously integral for operation .
Moreover the emitter, normally in the collector to base self biased circuit the opposite polarity
may I think be termed in this case as Re which is unconventional. I determine that the op amp itself presently tested with AD825, opa627 and opa1611 interacts favorably to improve power supply connection with transistors connected in this manner.
Cheers / Chris
Last edited:
As we are surely looking at a specific induced distortion and the psycho-acoustic impression of stereo image, I don't think we'll be seeing measurable separation differences any time soon.
Chris, these explanations are incomprehensible in English or electronic-speak. I don't think they serve your cause, though I'm sure your idea works fine. Sorry, but there it is.
Last edited:
Hi Ian
Sorry its not in a form that is easy, but thats how I understand it. To solve this language issue quite easily- breadboard it and try yourself, then you will hear the difference, which will start to make better sense as it applies to an op amp. I recommend Analog Devices AD825
Cheers / Chris
Sorry its not in a form that is easy, but thats how I understand it. To solve this language issue quite easily- breadboard it and try yourself, then you will hear the difference, which will start to make better sense as it applies to an op amp. I recommend Analog Devices AD825
Cheers / Chris
It looks like technical English, it uses technical English words, but I can't understand the sentences. Obviously my PhD in electronics is not sufficiently advanced to follow Chris's thinking here.Chris Daly said:
I don't get where the 6mA is supposed to come from. If 15V is applied across a 2.2K resistor, then a current of about 6mA would flow through it.
However, in the given circuit, the voltage across the resistors is very much lower than that. The current through each resistor will be equal to the current drawn by the opamp, divided by the opamp's current gain.
Or was there a misprint in the circuit? If it is supposed to be as shown below, then there will be about 6mA flowing through each resistor.
😕
Attachments
Oh I wholeheartedly agree, but an "increase in separation" is the one technically measurable claim I've seen, as opposed to all the nontechnical phrases used saying how this is an improvement.
But of course a measurement that showed no change in separation would only convince the claimant that the measurement is faulty. There may be no way of convincing him otherwise.
I thought of posting earlier in this thread that "a little knowledge is a dangerous thing."
Hi Godfrey
How about we ask Chris to tick the right box:
1) The transistors are biased such that an idle current of 6 mA flows to the Opamp supply pins.
2) The resistors are chosen such that 6mA flows through them. (your surmise)
3)The resistors are chosen such that 6mV, or some other potential, appears across them.
4)This is a calculated value that may be in error.
Any more or can we start with that? 😉
How about we ask Chris to tick the right box:
1) The transistors are biased such that an idle current of 6 mA flows to the Opamp supply pins.
2) The resistors are chosen such that 6mA flows through them. (your surmise)
3)The resistors are chosen such that 6mV, or some other potential, appears across them.
4)This is a calculated value that may be in error.
Any more or can we start with that? 😉
Sadly there is no misprint in the circuit, but there is a whole lot of misunderstanding how it works. There is a lot of misunderstanding regarding the difference between WHAT it does, HOW it's manifested, and HOW it does it.
I would suggest that people who think it's enough to 'build it and you will hear the difference' are really on the wrong site, and would be better off at audio asylum etc, where the WHY of things is not that important, neither is the question weather some objective measure of the performance of a circuit can be established, or some plausible technical explanation of it's operation can be given. All three are the very point of THIS site. At least that was my impression.
I would suggest that people who think it's enough to 'build it and you will hear the difference' are really on the wrong site, and would be better off at audio asylum etc, where the WHY of things is not that important, neither is the question weather some objective measure of the performance of a circuit can be established, or some plausible technical explanation of it's operation can be given. All three are the very point of THIS site. At least that was my impression.
Interestingly, I wonder how Chris manages to bias the current at all - the quiescent current of the opamp is a fixed given, in the case of an AD826 it's around 7ish mA. The only way to affect this current is to increase the base resistor so much that the transistor starts to pinch the current, i.e. enough drop over the resistor, pushing the emitter of the transistor 'against' the outputstage voltage wise, leading to a reduction in quiescent current. But when you reach that condition, you're effectively running the opamp at a supply voltage that's way below the minimum operating voltage. All the voltage will be across the transistor's collector/emitter channel 😉
- Status
- Not open for further replies.