John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Two compensations. Zobel for inductance (can be in two parts when (often) not linear, specially for tweeters) and motional (RLC)l for resonance frequency.
I am curious and interested to read your cons arguments.

(1) Zobel.

In an amplifier, a zobel is mainly related with oscillation issue, not for impedance correction, so the capacitor used is relatively low [less than 100nF]. Even at this low value, pure listening in many occasions always bring me to prefer the no zobel. ABX is easy [zobel always brings more "refined" sound] but deciding what to prefer is very difficult here.

At higher capacitance value as in tweeter's impedance correction, OTOH, the audible effect is much bigger. I can't blame speaker designers if they opt to use it.

I have lived long with many Dynaudio tweeters and in many high end speakers, zobel is used. Actually Dynaudio used and endorsed this zobel implementation. So I have many experience where I tried the zobel and decided in the end that no zobel is much better in term of subjective enjoyment and musicality [in objective term, especially in commercial speaker where consumers judge the speaker in short listening, and with poor ears, I think it is a hard decision to leave out the zobel].

So what is the technical argument for my cons beside the subjective listening preference? None. But when I designed a speaker by ears and do measurement at the end, I cannot guarantee that the FR will be the flattest, or the IR will looks perfect, but I can guarantee that it will have good phase behavior! So my preference has a strong correlation with phase behavior and I conclude that phase, however "small", is audible.

(2) RLC [Notch filter].

Interestingly, for me an RLC is a must [for expensive design where part value is nothing expensive].

When I was a beginner, RLC was an evil, especially if you have sensitive ears [Admit it, what did you hear when you add those filters to the tweeter? The vocal becomes unnatural, right?]. But with experience, the RLC is a great tool for me. Its function is not only to notch the impedance peak but to tailor the phase behavior!

So the impedance peak doesn't need to be notched 100% to be flat. Less than 20 Ohm is fine imo. The impedance is not the direct goal, but the phase is.

How long and how (sinus, pink noise, what level) are-you doing the break-in of a speaker before measuring-it ?

Break in will change some T/S and distortion but I don't think the change is critical enough. The change will improve the sound but will not affect the design so much. And especially if you tune the final result by ears [which means differs slightly from the best measurement result] then it should be not important imo.
 
I was wondering if the spider was not the main 'spring' in the assembly, so the surround could have little influence on resonance frequency (apart added or reduced mass of the foam).

You cannot assume that the cone moves in pistonic. The spider holds the cone in small radius and directly on the voice coil, while the surround holds the cone in bigger radius far away from voice coil. Actually, the surround is the main mechanism which change the pistonic behavior.

The cone movement animation here is nice: TMD - Tuned Mass Damper - Focal

I hope i will be able to make a miracle, but such a hard work, as the crossover is at the limit for the size of the membrane. (12", 1500Hz, go figure !)

I think, if the woofer is to be used as a midrange, you can sacrifice on the surround behavior related to xmax. Half roll rubber in no good here.
 
There aren't any 'quacks' here, just people who don't believe in the experience of other people.

Yes there is... the subtext of the messages that have accumulated is that I am.

Yep, they "don't believe in the experience of other people" means they only believe their own?

But they could surely just test it:

An externally hosted image should be here but it was not working when we last tested it.


[Only temporary post here - until suitable place where it will be published permanently - with other similar material]



-
 
Last edited:
Yes, of course. No way to find the original surround and my boomers are no more in catalogs :mad:
I hope i will be able to make a miracle, but such a hard work, as the crossover is at the limit for the size of the membrane. (12", 1500Hz, go figure !)

Not true!

In 1996, when I decided it was time to return my AR94 to its original sealed box state, since the foam surrounds had literally disintegrated, and I had two nice BR openings, I also assumed this would be so. It turned out that the man who was to set things right knew of a German company which manufactures rubber surrounds as replacements for cases just like mine.

Their catalog was about 5 cm thick, with a listing of loudspeakers such as I have never seen before that. AR occupied two whole A4 format pages, and sure enough, my model was there. He ordered them and some two weeks later, he had them.

Unfortunately, I do not remember their name, so you will have to search for them (key words possibly "loudspeaker surrounds"). That was like 10-12 years after I had purchased them, so I cannot swera to this, but after the change, they sounded pretty original to me - and are still going strong. "Satisfied" ia the least I can say.
 
I finally found an old plot I posted elsewhere showing statistical distribution of change in hearing thresholds versus frequency as a function of age between ages of 18 and 55, based on ISO standard 7029.

If you think this is disturbing, don't bother looking at what happens over age 55...........really.

Then, based on the claims of audibility on this thread, I can be statistically confident that most of you are probably 18 year old females ............;) You know who you are.........!

This is the hard fact of how hearing changes as we age, but onset and degradation starts at a pretty early age.
 

Attachments

  • hearing%20thresholds%2055yrs%20versus%2018%20yrs.jpg
    hearing%20thresholds%2055yrs%20versus%2018%20yrs.jpg
    83.3 KB · Views: 166
Lucky,

The last octave may be a lost octave, but this only relates marginally to critical listening imo. Training and experience make up for much. But for evaluating tweeters, it might be a good idea you ask the young ones.

Other than that, only an 18 year old fool would rely on his ears to develop an audio product anyways.
 
Last edited:
Training and experience make up for much.

I think I am a better listener today than I have ever been. The biggest problem is just being totally honest to yourself about it.

And the worst is stupid shoot-outs, I just tune out. Submitted an amp for a club shoot-out, slept right through it and it won! I thought it was a hoot, something to put on the website and forget it.

-
 
Out of curiosity, I checked the mosquito thing you mentioned, didn't even require to adjust the volume for the 17KHz sample. (usb dac/amp)

Never realised I'm a virgin with perky boobs.
I've given serious thought to ways of making women think I'm way under 30, but so far I've only been succesful with roofies.

Looking at an 18-55 graph, I can not but wonder what the average exposure is/was.
Most audio types I've bumped into, were not the types for loud ambient noise level environments, (cd) walkmans, mp3 headsets, or concerts that make your ears bleed.
Who knows, a large percentage of the elder audiophiles could still be highly shaggable.
 
I think I am a better listener today than I have ever been.
When i was a young sound engineer, i had a lot of problems with cymbals. Wondering some old colleague's mix. Trying to add bright to them at very high frequencies, getting "distortion". Too much informations in my ears.
With age, losing treble sensitivity, i was better and better. Able to distinguish the weight of the 'metal', the hit of the stick. Older, I was always asking musicians, during mixes about the trebles, and they usually found my work ok for them.
Too, I was unable to enjoy violins. The friction of the archer was unbearable.

If I suffer from my lost of vision, I don't have the feeling to have lost audition. Treble less aggressive, for sure, but discrimination, ability to separate and analyse the sounds, better.
What about you ?
 
(In an amplifier, a zobel is mainly related with oscillation issue, not for impedance correction, so the capacitor used is relatively low [less than 100nF]. Even at this low value, pure listening in many occasions always bring me to prefer the no zobel.
May-be you simply love to get more trebles ? Not me. On my side, i don't like any soft dome tweeter. Feel like they are modulating some kind of white noise instead of reproducing the frequencies... I use one horn for medium and treble. No tweeter.
For my amps, you know, i care about high slew rate (CFA), stability margin, no overshoot on square waves and little signals. With high stability margin, my amp makes no difference to me with or without Zobel in it.

But compensating the impedance curve of my driver (Fc=1500, 6 Ohms from 0 to 40 000) makes, for sure, a real positive difference. Everything sound more fluid and natural. less "hifi". There is a main difference between the zobel in an amp, and the one you can use to flatten the impedance: the second is the other side of the cable.
Of course, you have to tune frequency response to be flat with it. If they make-you lose some needed treble, you will prefer without ;-)
 
Last edited:
Status
Not open for further replies.