John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
As you have probably gathered, I distinguish distortion/faults/evils into "detectable in Double Blind Listening Tests bla bla" and "not detectable bla bla".
So do I, but I'm not at all convinced that those tests you cite (and for which you have no data) have much, if any, audible consequence unless the results are grossly poor. Just because someone says, "Hirata Test!" doesn't mean they can correlate it with controlled listening results.
Stuart, you will also note I've been trying to get JC to cooperate and help us do some Double Blind Tests bla bla.

I can't be certain about Blowtorch.

But Double Blind Tests were an important part of my life for more than 2 decades. We started them in the late 70's with Peter Fryer's test on audiblity of speaker distortions. As part of this, we listened to OPAs over a very long period starting from 741 (Don't laugh!) and have a list of both music & test signals which showed up the audible faults of OPAs. 5534 was the first OPA we were unable to fault in the late 70s. We actually had pre-prod samples from Mullard when it was called TDA1034.

For Blowtorch, I'm doing what JC likes to do and predict performance from the label on the box/chip/bits. ie sighted tests (prejudices). But I'm doing it with 'calculations' on THD including 7th and its performance on Hirata & Quan from his schematic.

I'm confident that eg Blowtorch will do badly on Hirata & Quan and be worse than AD797 on 7th. On audibility, I'm on shakier ground. My true golden pinnae like John Atkinson & Tony Faulkner are all my age and older.

I'm disappointed no one has nominated new younger persons who may be true golden pinnae. Will they all die out and be replaced with pseudo Golden Pinnae? :mad:

But you when you do enough Double Blind bla bla, you get an instinct about things. eg I can sense if certain subjects are easily influenced by others which is why in the Wharfedale tests, each subject is tested separately, and the 'tester' is NEVER in the same room.

I have this 'sense', about Blowtorch. Its THD performance is veering on the levels known to be audible in our own Blind Tests and other research too. JC isn't the only person to have played with FETs for small signal.

So to clarify, I contend Blowtorch will sound worse than a properly implemented 4558 device. Quan & Hirata, though poor for Blowtorch, may or may not correlate with its poor sound. They are a smoke screen to :wave2s: JC's cage and persuade HIM to provide evidence. 7th is in a similar vein but, though Blowtorch will have more than AD797, it's unlikely to be responsible for its poor sound.

With JC2 Mk1, I'm confident even on audibility. The decoupling & earthing WILL result in poorer distortion and marginal stability. This is audible even on music let alone the poor noise performance. Scott's AD797 datasheet confirms mine isn't the only experience of this.

Just my $0.02
 
PMA,
I guess some money counter got in the mix with that connector choice. Can't imagine anyone selecting that connector for a high end product, but perhaps they were trying to save money to pay for all that carved aluminum in the chassis? If this makes waves in the audio press then you will see that connector changed in an instant.
 
I am confused, who should I believe?
You, of course, trying by yourself, with the same value as my feelings, and in your context.
Looking to your links, i found some noticed differences between NE55xx brands, as well as : "The 5532 has more deep bass also, but has some distorsion in treble".
That seems to correlate what i pretended: Sharper basses sound less 'deep', indeed.
Since 2 decades now i use current feedback AOPs, when i can, and can't differentiate obvious differences with strait wire.
 
Last edited:
I should point out that 5532/4 essentially has C2 built into its compensation with the same beneficial effects on distortion, particularly high order stuff like JC's 7th. It's one reason why its performance still rivals the new uber OPAs in real life.
Waly, as always, its horses for courses. Choose the right device for the application.

For some perspective ..

Cyril Bateman in his seminal series(s) on capacitors, (far more definitive than Jung. Sorry Mr. Marsh :) ) had to make a really ultra supa dupa low distortion oscillator.

He recommended 2 OPAs for this application. One of them was 5534. IIRC, this is after the intro of the NI uber OPAs with advertised 1pp zillion THD.

But for the ultimate, supa ultra dupa plus performance, he of course used [drum roll & fanfare] ... AD797 :D
 
Last edited:
Sy,
Strike one for customer service on that preamp for your boss! Basically take it or leave it seems to be the answer. But at the same time I don't see how an XLR connector with a locking tab would let the connection disconnect from the male plug?

It won't- the XLR supplied with this amp (it's a power amp) doesn't have a locking tab. So the cables have to be re-seated every few hours. He's tried three different types of cables to see if that's the issue- it's not. He is not a dumb guy, he was a highly accomplished engineer before he moved into management. So it's not a stupid-user issue, it's truly a design flaw that even cheap Chinese equipment manages to avoid.

The dismissive attitude is not the way we would EVER treat a customer. When people give us money for our products, we stand behind them, and when they don't perform to expectation, we make it right. The luxury audio industry is apparently different.
 
Kindhornman, the A21 is NOT a super fi amp. It is a MID-FI amp. I will NOT use it in my own audio system. I did not know that the connectors were not Neutrik, but it goes with the territory. For $2500 retail, you get a solid 250W/channel, made to a reasonable standard by the opinion of just about anyone here. We have to produce it in Taiwan at, at least 1/4 the selling price. Try to do this, yourself! In fact, give it to everybody here, giving everybody a credit for at least $5 per hour for their effort. Go for it!
 
Halo A21 is a power amplifier and really has the cheapest non-latching XLR connectors
Hum...let solve your job problem. Can't you block the connectors with ' handcrafted' shims ? Or add a little weld drop on the cable connectors contacts to harden the contact ?
We do not move so often them, in a hifi home gear, time to find a new boss ?
 
Last edited:
The important point in this story is that because I was physically doing the action of switching the polarity (thus I was “certain” that an acoustical different effect will take place) I was genuinely fooled by my senses/brain that I was perceiving acoustic differences.
I had unwillingly turned this test into a blind test, on which of course I failed.
My hearing acuity (very degraded) is irrelevant. Were it relevant, I should have decided that I did not hear any difference.
There's a very powerful tool in the arsenal for picking up what's going on. And that's to use the "right" side of the brain, the intuitive, to do the "measurements" ...

Unfortunately, the "left" side, the consciously thinking part is very easy to lead astray, it has ego, etc, attached. The more you "focus", try and force the intellectual area in the brain to come to a "correct" decision the worse it often gets. This is obviously what ABX, etc, has evolved to be, to try and get around this problem ... but it has to be extremely well implemented to come up with a meaningful result.

I personally have "learnt" to deliberately switch to intuitive evaluation, this is like listening out of the corner of your ear: Clayton's listening we would call it down under - listening when you're not listening. The intuitive doesn't come up with figures, it just knows when something is right or wrong; and that's how I assess system sound, and one thing the right side of the brain doesn't like is typical hifi sound - it knows that this is really, really wrong ...

Frank
 
Last edited:
Status
Not open for further replies.