John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Member
Joined 2016
Paid Member
My Tek 475 is still one of my favourite scopes...

Working on some low noise audio preamp cards many years back, they were misbehaving oddly. We spent ages trying to figure out where the noise was coming from. It got late so we got to the "right, just cut tracks until the noise stops" mentality to try and isolate it. We got down to a single unconnected pin with noise still measurable! Impossible..but... turned out the assembly / defluxing process had left a thin, invisible film of conductive goo on the boards!! A good scrub in strong solvent (that you couldn't use today!) and bingo -- working!
 
diyAudio Member RIP
Joined 2005
My Tek 475 is still one of my favourite scopes...

Working on some low noise audio preamp cards many years back, they were misbehaving oddly. We spent ages trying to figure out where the noise was coming from. It got late so we got to the "right, just cut tracks until the noise stops" mentality to try and isolate it. We got down to a single unconnected pin with noise still measurable! Impossible..but... turned out the assembly / defluxing process had left a thin, invisible film of conductive goo on the boards!! A good scrub in strong solvent (that you couldn't use today!) and bingo -- working!
Flux contamination is underrated.

Although I left UCLA in 1985, a tech who was brought in by someone who hated my guts had managed to break portions of it. His motto ought to have been Leave no knob unturned. He loved to use every possible feature (this delayed by that, intensified by another) despite a simple triggered sweep being adequate in almost every case. Things began to fail...
 
I will freely admit my opinion was that there was a knee-jerk aspect to his initial approach i.e. "wire just can't matter".

Which isn´t really an answer to my question, but of course it was somehow a strange move from Randi. Although the contention that there should be (and can be) no audible difference between _reasonable_ choosen cables was afair always noted....

The initial parameters were poorly negotiated, too bad it didn't go forward. In the end someone like E. Brad Meyer and his bank of parametric equalizers in the chain would probably have had Fremer calling foul.

Yes, too bad it didn´t go forward. But it is quite funny that in fact Randi weaseled out, but your assertion still is, that otherwise Fremer of course would have .......

But anyway billshurv and SY had "heard different" and of course true believers will never change their mind :)
 
Member
Joined 2014
Paid Member
But anyway billshurv and SY had "heard different" and of course true believers will never change their mind :)

Not sure what I have said to mark me as a true believer. Given how downright offensive Fremer was being, I would personally have told him to sod off. This might be why I don't do this sort of thing. Magician vs stand up comic. I mean what could go wrong.
 
Not sure what I have said to mark me as a true believer. Given how downright offensive Fremer was being, I would personally have told him to sod off. This might be why I don't do this sort of thing. Magician vs stand up comic. I mean what could go wrong.

Although imho Randi started the flame a couple of years before, i don´t see really a difference between both actors wrt in this regard.

But anyway, i provided two examples responding to Scott´s remark which illustrate why these meetings between true believers did not work.

And if it is even in the Fremer/Randi event impossible to reach consensus on the "who weaseled out question" (based on the published information including emails, under the assumption that this information was correct) it is imo unlikely to do better in more difficult matters.
 
Disabled Account
Joined 2010
Was doing some late night work on my power amp and could not get a clean - i.e noise free signal from my new Tektronix scope while monitoring the output of my amp. I've got ~72 kHz at c. 100mV pk-pk coming out of my power amp - both channels. I'm sure its not a stability issue - the square wave plots are well behaved and the bias current and offset are stable. When I unplug the input, its quiet. The cable from my sig gen to my amp is about 500mm long and not of a particularly good quality.

Anyway, with the Tek plugged in and showing a less than ideal trace, I powered up the Rigol. Same thing - so it must be the amp.

I power down the Tektronix for some reason - the problem disappears. That cannot be. I power it up again. 72 kHz showing on the Rigol.

Turns out, the LCD screen, or the Artesyn power supply that sits behind it (I can see the name through the ventilation slots) is pumping this dirt out. I wound a few turns of wire into a coil and connected it to the rigol. Its definitely coming from the Tektronix.

The low level trace (5mV FS) on the Rigol is clean, but not the Tek. I've been chasing my tail for hours.

Anyone have a similar experience and do I have grounds to tell Tek its not acceptable? Its a 4 channel 200 MHz BW 1GS scope.

Oh how I miss my Philips analog scope . . . .

Yessir. I like to work with the TEK TDS3052 - but I know it emits loads of common mode noise.
 
AX tech editor
Joined 2002
Paid Member
Was doing some late night work on my power amp and could not get a clean - i.e noise free signal from my new Tektronix scope while monitoring the output of my amp. I've got ~72 kHz at c. 100mV pk-pk coming out of my power amp - both channels. I'm sure its not a stability issue - the square wave plots are well behaved and the bias current and offset are stable. When I unplug the input, its quiet. The cable from my sig gen to my amp is about 500mm long and not of a particularly good quality.

Anyway, with the Tek plugged in and showing a less than ideal trace, I powered up the Rigol. Same thing - so it must be the amp.

I power down the Tektronix for some reason - the problem disappears. That cannot be. I power it up again. 72 kHz showing on the Rigol.

Turns out, the LCD screen, or the Artesyn power supply that sits behind it (I can see the name through the ventilation slots) is pumping this dirt out. I wound a few turns of wire into a coil and connected it to the rigol. Its definitely coming from the Tektronix.

The low level trace (5mV FS) on the Rigol is clean, but not the Tek. I've been chasing my tail for hours.

Anyone have a similar experience and do I have grounds to tell Tek its not acceptable? Its a 4 channel 200 MHz BW 1GS scope.

Oh how I miss my Philips analog scope . . . .

If it is an Infinivision 1000 (or 2000, can't remember) I have the same experience. But not always, seems to come and go. Mine is 4 x 70 MHz, haven't gotten around to hack it to make it 200MHz yet....

Jan
 
Yes, too bad it didn´t go forward. But it is quite funny that in fact Randi weaseled out, but your assertion still is, that otherwise Fremer of course would have .......

Well what to you think Mr. Fremer's reaction would be to 60 eight legged pieces of dirty silicon added to each channel?

I far prefer the hypothesis that an arbitrary swap of speaker cables should not be expected to alter the sound in any way outside of frequency response variations due to choice of particular amplifier and speaker. Not eliminating well known issues from a test would be dishonest. The Pear cables in question are outliers in terms of L and C and they even show typical simple frequency response differences of almost 3dB in the audio band.
 
Last edited:
Anyone have a similar experience
Yes.

and do I have grounds to tell Tek its not acceptable? Its a 4 channel 200 MHz BW 1GS scope.
No.


1. Get an iso transformer.
or
2. twist amp and scope power cords tightly to the same duplex outlet, and twist the scope probe around the scope PC to the amp PC, twist it around that, then try the measurement again.

If the coupling is the elements of the boxes themselves, this will not work. This will only work if coupling is via the loop;

John

John
 
The Pear cables in question are outliers in terms of L and C and they even show typical simple frequency response differences of almost 3dB in the audio band.

Normal people unfamiliar with fashion audio are often surprised by the remarkably poor performance of many ultra-expensive products. They mistakenly assume that an expensive wire will act as a competent wire and don't expect its performance to be orders of magnitude worse than an extension cord from Home Depot.
 
Normal people unfamiliar with fashion audio are often surprised by the remarkably poor performance of many ultra-expensive products. They mistakenly assume that an expensive wire will act as a competent wire and don't expect its performance to be orders of magnitude worse than an extension cord from Home Depot.

To be fair they were showing less roll-off from theirs and they showed remarkable restraint in not making outright absurd claims. I don't know what they used to get -3db at 20K from an ordinary speaker I don't remember 6' of even #30 being that bad.

My experience has been that checking even for outright amplifier oscillation is just not done in certain crowds.
 
Last edited:
www.hifisonix.com
Joined 2003
Paid Member
Yes.


No.


1. Get an iso transformer.
or
2. twist amp and scope power cords tightly to the same duplex outlet, and twist the scope probe around the scope PC to the amp PC, twist it around that, then try the measurement again.

If the coupling is the elements of the boxes themselves, this will not work. This will only work if coupling is via the loop;

John

John

John,

They were coming from outlets about 1 m apart (workshop is brand new - fresh wiring etc). Thing is, the Rigol is plugged into the socket right next to the Tek. Amplifier is isolated thru the transformer - assume this is enough?

However, let me try your trick. I'll post some picks up.

Damn. A $600 Rigol is quieter than a $3k TeK machine.
 
Member
Joined 2004
Paid Member
Normal people unfamiliar with fashion audio are often surprised by the remarkably poor performance of many ultra-expensive products. They mistakenly assume that an expensive wire will act as a competent wire and don't expect its performance to be orders of magnitude worse than an extension cord from Home Depot.

The eye patterns and losses on the ultra premium USB and network cables I have looked at would not meet minimum USB/UL cert requirements and cost around 100X what cables that do meet those requirements cost. But lack the magic appellation. It may be the poor performance electrically that provides the better perceived performance. ?less radiation, perhaps??
 
all the way to Europe

What do you expect folks in a nation as small and flat as a postage stamp to do, stare at a picture of the royal family all day ?
Everything preferable to the 7/7 days of Geraldo, Letterman, Leno, and Oprah gossip, even if 90 percent of all news items is online copy&paste material nowadays.

50 bucks on Airforce One renamed D's Duck before NY's eve.
 
Status
Not open for further replies.