Read the article. Then note the sales of Bruno's designs.
I have read the article (I buy every issue of Linear Audio), Bruno´s "fashion" is high
feedback, that´s ok with me, but where is the evidence that this will improve the perceived sound ?
Does applying tons of feedback to lower THD to sub ppm numbers make more sense
than extreme slew rate numbers ? The article is very detailed on "how" but not on "why".
Low or no feedback might be a "fashion" and high feedback is ? The "< 0.0001% THD fetish" ?
Sales and quality do correlate ? That´s new to me.
Last edited:
The evidence is that it will cause an amplifier to simply amplify. Indeed, if one is after an effects box, the "perceived" sound may not be as good. But in high fidelity, the amp's output should replicate the input, and for those who want an amplifier to be neutral (John claims that's his aim), the data and analysis are clear.
The last comment was for John, who sneers at anyone who doesn't design and build amplifiers to be sold in his market niche.
No, but did one at the Queen of the South in Hasselt years ago
Thanks for the tip, Corsendonkse Hoeve is ok.
jAN
about that, the thread was closed before I could reply, my apologies for my ignorance of European language traditions/etiquette jan. 'twas a harmless jibe.
Like this?
EMI Rejection Ratio (EMIRR)
The electromagnetic interference (EMI) rejection ratio, or EMIRR, describes the EMI immunity of operational amplifiers. An adverse effect that is common to many op amps is a change in the offset voltage as a result of RF signal rectification. An op amp that is more efficient at rejecting this change in offset as a result of EMI has a higher EMIRR and is quantified by a decibel value.
View attachment EMIRR OPA627 sboz016a.pdf
EMIRR vs frequency
EMIRR frequencies of Interest
It is decisive for opamp 'sound quality'. The higher the dv/dt at Vdiff, the higher the issue and then we separate "slow" designs from "fast" ones and exponential transfer functions from another, less curved. This should be discussed, and not THD of 0.000003% as per datasheets.
I think I could make a pretty good argument that it just wastes output current though. The argument would read: R65 is upstream of the 330 Ohm build-out resistor; there's plenty of loop feedback from this point to control level, so R65's effect is going to be hard to measure; there's no compelling reason to match levels of polarities in balanced lines that aren't run inside the same shield with another signal (common mode rejection is independent of signal levels' balance).
Of course there's always lots that I can't see or figure out, too.
Thanks, as always,
Chris
Last edited:
OK, back to basic talk. This particular design configuration in the JC-2, JC-80, etc, has a unique problem, because it is a single gain block with balanced inputs and balanced outputs. What we have to have, is a small attenuator of about 1dB on one of the outputs, when and only when the gain block is driven by a SINGLE ENDED drive signal. This is a secondary property of the dual negative feedback loops, rather than the topology itself, as the CTC Blowtorch gain stage does NOT have this problem.
The amount of attenuation needed is inversely proportional to the gain of the block. So if you have a gain of 30, then you want to have an attenuation to get 29/30 or so on one side. I am not going back to do the basic math at the moment. Perhaps PMA has the exact equations. This presents a problem with a low overall gain version of this design. It generates a greater gain offset, and therefore requires a more obtrusive output attenuator.
What is interesting is that IF you balance drive this sort of gain block, ALWAYS, then you don't need the correction. In the case of the JC-2, you need the attenuator, sometimes, so we have to enable it with a relay, depending on drive settings, either balanced in or single ended in. (Not so obvious, is it) '-)
The amount of attenuation needed is inversely proportional to the gain of the block. So if you have a gain of 30, then you want to have an attenuation to get 29/30 or so on one side. I am not going back to do the basic math at the moment. Perhaps PMA has the exact equations. This presents a problem with a low overall gain version of this design. It generates a greater gain offset, and therefore requires a more obtrusive output attenuator.
What is interesting is that IF you balance drive this sort of gain block, ALWAYS, then you don't need the correction. In the case of the JC-2, you need the attenuator, sometimes, so we have to enable it with a relay, depending on drive settings, either balanced in or single ended in. (Not so obvious, is it) '-)
maybe because you dont have the original input of that polarity to add to the gain of the stage? as in the case of calculating the gain, you will calculate it as the input plus the gain, while with this conversion/inversion you dont have the neg phase input, so you will use some of the gain inverting the input first for the negative phase. am I on the right track there?
btw I havent studied the schematic aside from the first glance when posted, planned to have a look this weekend. I have quite a lot of the parts so may have a tinker at some stage with the line amp section for fun, with much less gain, what is minimum for stability? i'm not a sim junky, dont have the math, though i'm going back to rectify that.
btw I havent studied the schematic aside from the first glance when posted, planned to have a look this weekend. I have quite a lot of the parts so may have a tinker at some stage with the line amp section for fun, with much less gain, what is minimum for stability? i'm not a sim junky, dont have the math, though i'm going back to rectify that.
Last edited:
- Status
- Not open for further replies.