John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Engineers arguing the meaning of language seems much like the hifi reviewers using language to argue the meaning of engineering.
I would agree that in this case "musical" is a language issue as opposed to the strict definition of a word. Words used by engineers have a precise meaning, and if they don't, one should be agreed upon before discussion, which I think is what Jan was trying to do. Trying to agree a definition of musical or musicality or music for that matter, is a losing game. However, I think the context should be taken into account, and it's ok to call a piece of equipment musical and is having degrees of musicality if it's high fidelity
 
Administrator
Joined 2004
Paid Member
Hi Dan,
IME cable direction can be a direct controller of 'musicality'.
Ahhhh, no.

What is important is that the outer shield ends up on the source side. That's it. The arrow never had anything to do with which way the AC signal ran. It's determined to thwart your plans to convey this as a truth since the electronics are displaced equally in both directions. The emf is also travelling in equal amounts in both directions.

Now, you may be experiencing a higher noise level with the cable installed the wrong way. How you interpret that (better or worse) is all up to you. Just be aware that the outer shield is normally connected to the source side due to the (normally) lower impedance on that side. Folks may discover that from a passive attenuator / switch box (AKA, passive "preamp") to the target amplifier, the connections need to be reversed for the outer shield. It's all about one connection being able to drain the noise to ground more effectively.

Note that because the outer shield is only connected at one end, the only current that can flow is from capacitive pickup to ground. No signal current flows on this conductor.

-Chris
 
Sounds like an agenda to me, and an offhand attack on other peoples work.
Well, are you aware that there is a member also named scott wurcer who does just this in the Funniest Snake Oil Theories thread?

So engineers like you and I tend to object to the bending, manipulation or even outright breaking of mathematical and scientific principles. That's our nature. Should we apologize for this?

I think an upside challenging things is someone usually learns something from it.

I think the intention of an Analog Devices data sheet is to describe the product and to market the product. This heated debate is about the wording of the AD797 data sheet.

C-C looks like subtraction to me, so what if it fails at high f the results are posted please comment on my points and the actual data and drop the editorializing. Again as a point of discussion, if the buffer is modeled as an exact G = 1 with no frequency limitations it is simply C-C = 0.
The model is not true to the chip. The model assumes an impossible infinity and thereby makes the glib statement that distortion simply cancels and even shows that the effect of the buffer, A, vanishes.
That is a wonderful claim from a marketing point of view but it is simply not true in the chip. If the data sheet qualified this and pointed out that the real circuit cannot achieve this and remain stable but the feedback loop created by Cn helps reduce distortion then that would be more accurate.

Should we say error reduction rather than cancellation? Such a commotion around a circuit where nothing but a simple capacitor added to an already finished circuit had an unforeseen benefit.
Yes we should say that. Agreement! :)

The commotion is not around the real circuit, which is a ground-breaking design by all accounts, just the data sheet explanation.

High time to move on...:cool:

Aside: In general (not talking about the AD797) I think it's trendy to mix up the terminology of error correction, from the digital world, with error reduction in control theory. This seems to happen a lot in the pages of DIYaudio too. Feed forward does do addition and subtraction of signals and so the term cancellation is applicable IMO. Feedback certainly does not as it is a continuously adaptive system and its effectiveness depends on ratios and and stability.
 
Last edited:
My question too; are they perhaps X5R or X7R? These are known to be horribly non-linear.

Jan

MLCCs are available up to at least 47uF at 10V. There may be larger capacitance ones available, but I know the 47uF capacitors are in our system.

The capacitance varies with applied voltage. Therefore they are not linear. No need to measure them as the data sheet tells enough to know they will not couple sound in a linear fashion.
 
Last edited:
Well, are you aware that there is a member also named scott wurcer who does just this in the Funniest Snake Oil Theories thread?

Peace, let's drop it though I do believe attacking fraud and garbage science is different. It does not take much effort to find highly respected scientists getting frustrated over zero point energy, water powered cars, etc.

As a piece of history that I remembered last night, Walt Jung and I at the time were having some fun with playing around with feedback to the null pins of op-amps. You have a similar situation, short the inputs and let the amp be open loop, driving the null terminals via a large resistor has a huge gain to the output easily driving it rail to rail. But connecting the resistor from either null pin to the output is never unstable and even can make the Aol virtually infinite or even reverse its sign. Furthermore doing this has little or no effect on the closed-loop response of the main amplifier. I have published this here before.

I was simply trying to make an analogy to this behavior since the current mirror is where the offset is injected. Breaking the loop and observing the huge gain does not make for the "usual" feedback story.

BTW an ideal current conveyor is not bilateral and I think your simplification goes a little too far and misses some of the behavior of the circuit. I got stranded yesterday with a battery that no longer can take a charge and caught a bad head cold waiting for the service truck, if you can bear with me I am working on a simple circuit that illustrates what I mean and might interest people as a circuit not just an argument. :)
 
AX tech editor
Joined 2002
Paid Member
But connecting the resistor from either null pin to the output is never unstable and even can make the Aol virtually infinite or even reverse its sign. Furthermore doing this has little or no effect on the closed-loop response of the main amplifier. I have published this here before.

Coincidentally, I was reviewing some AP circuitry today and noted an NE5534 with a 100 ohms resistor from comp pin 5 to the output pin 6. I tend to pay attention when Bruce does something like that.

That what you referred to?

Jan
 

TNT

Member
Joined 2003
Paid Member
I'm afraid it could be worse than this. I think he means the extrusion direction. Yes - the way the copper was made into a strand. It plays differently depending on how you connect it. "They" say.

//

Hi Dan,

Ahhhh, no.

What is important is that the outer shield ends up on the source side. That's it. The arrow never had anything to do with which way the AC signal ran. It's determined to thwart your plans to convey this as a truth since the electronics are displaced equally in both directions. The emf is also travelling in equal amounts in both directions.

Now, you may be experiencing a higher noise level with the cable installed the wrong way. How you interpret that (better or worse) is all up to you. Just be aware that the outer shield is normally connected to the source side due to the (normally) lower impedance on that side. Folks may discover that from a passive attenuator / switch box (AKA, passive "preamp") to the target amplifier, the connections need to be reversed for the outer shield. It's all about one connection being able to drain the noise to ground more effectively.

Note that because the outer shield is only connected at one end, the only current that can flow is from capacitive pickup to ground. No signal current flows on this conductor.

-Chris
 
Administrator
Joined 2004
Paid Member
Hi TNT,
Then one has to ask. What direction was the cable made in, because the inner conductor and insulation is one step only at the most. I can see where the stranded wire in the centre would be spooled before winding into a stranded wire. Wound again before it has the insulation covering. and so on.

There is no telling how the wire was drawn.

-Chris
 
Status
Not open for further replies.