John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
The easy test is to listen to the noise made while tracking. Then deliberately play the same music back through the cartridge! When the level is the same you may just find that the drive level is hundreds of millivolts!

More likely a toasted cartridge, the noise made during tracking is virtually 100% mechanical coupling. You should be able to do an experiment with a microscope, an LED "strobe", and a signal generator. Simply image the stylus tip and drive the coils and freeze the motion with the strobe.
 
Hi,

Bohumil Sykora developed a circuit with active damping called "Actidamp" and several phono preamps like Actidamp MkII - MkIV types were produced.

First, Actidamp does not damp the cartridge.

It's operation is interesting, the concept useful in other applications but not MM Cartridges (it does not apply to MC).

Second, it has two considerable disadvantages, one the improvements fall below my personal 3dB rule and secondly, the "Actidamp" applications I know in practice sound MUCH better with it disabled...

Ciao T
 
Member
Joined 2004
Paid Member
The math (I had a grad student working as an intern do this) is based on the compliance (published numbers are a good start) as energy in (Watts) vs. mV across a resistor out (Watts). The ratio is quite large. The device must be reciprocal (a strain gauge cartridge is not for example) for this to have meaning.

If the energy transfer is .1 % the most effect you can have on the mechanicals is .1% (unless I'm missing something about conservation of energy here).

The balanced vs. single ended issue doesn't change this unless Ohm's law doesn't apply (in some audio circles it seems Ohm's law is to be repealed, its too confining).

While I'm throwing bombs here the next myth is cartridge demagnitizing. Keith Johnson and I had a good laugh when this was first circulating. What it does is ensure that the magnetic component of the armature is magnetized. When the AC field is relaxed around the magnetic material in the armature it will pick up and retain the DC field of the external magnet. (The same process as recording on magnetic tape with ac bias.) If you really demagnitized the cartridge there would be no output. . . Whether aligning the magnetic properties of the armature to the external magnet is good and how best to do it may be worth exploring, but only with a good understanding of what really is happening.
 
John,

Looks like 'summing' type input to me. Negative Z would be interesting.

It's inverse bootstrapping. Instead of making a low value resistor look like having a high actual value here a high impedance resistor is made to have a low apparent impedance (the reverse of traditional bootstrapping).

Referring to:

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


which is close to the schematic of a commercial Phono Stage, the greatest improvement in sound quality of this circuit is attained by removing R6 and shorting C2, at least for the specific product (names shall be omitted) I had repeat chances to modify.

Ciao T
 
If we imagine the microgroove phono system scaled up so that the groove were 3 feet deep and we could sit on the edge and dangle our feet in it, the smallest stylus made would be 20 feet tall and tower above us on a tube the length of a football field. WRT retrieving the information in the groove, reciprocity is iffy in the mechanical part too.

Thanks,
Chris
 
Dear John,

John Atkinson
Editor, Stereophile

I have just read a piece of your writing soon to be published... It makes a most tantalising reference to this lecture:

John Atkinson to give Heyser Memorial Lecture at the 131st AES Convention | Stereophile.com

The note linked above mentions that the Lecture would be video taped. I personally consider that the contents would likely be of interest to the posters and readers here as elsewhere.

Is the video available? If not, would you consider making the handouts/slides etc. that you have available to the community?

Ciao T
 
Member
Joined 2002
Paid Member
I'm a little confused by the plots. They look a little different than I would expect. usually there is a flat noise floor to some lower frequency wher the 1/f curve start and it rises below. A HF rolloff is usually indicative of the amp circuit running out of gain, not the noise falling.

Good catch there is a second order 1K filter. I built my measuring boards with some different options that are jumper selectable for filters, gain and reference resistors. Just trying to make relative measurements. The Interfet stuff is a bit pricey at $12.50 @ 1K but hey you can buy Muse op amps for $50 if you want. Quite good sounding really but not what we do for our products.
 
Last edited by a moderator:
John,



It's inverse bootstrapping. Instead of making a low value resistor look like having a high actual value here a high impedance resistor is made to have a low apparent impedance (the reverse of traditional bootstrapping).

Referring to:

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


which is close to the schematic of a commercial Phono Stage, the greatest improvement in sound quality of this circuit is attained by removing R6 and shorting C2, at least for the specific product (names shall be omitted) I had repeat chances to modify.

Ciao T

This just looks like synthesized cartridge loading, a well-understood concept. It can sometimes provide a noise advantage. If you have a net gain of -9 to the far end of a 470k feedback resistor, you will have synthesized a 47k resistor.

This technique is especially useful if you heavy-load an MM cartridge so that it sees a low load resistance on the order of, say, 5k, so that the 75us RIAA time constant is achieved by the input load resistance working against the cartridge inductance. The use of a direct loading of, say, 5k, would create a noise disadvantage with respect to the conventional 47k load. You can avoid the noise disadvantage by synthesizing the 5k load with a 47k resistor with feedback. Such heavy-load EQ approaches have some interesting advantages, but the synthesized load resistance must be adjusted in accordance with the cartridge inductance to get the 75 us RIAA time constant right.

One interesting advantage of the heavy-load approach is that it damps out the cartridge electrical resonance and largely takes out the resulting double-pole electrical rolloff, leaving primarily the cantilever resonance to deal with. If the cantilever resonance is at a high frequency like 30kHz, you get a fairly wideband response out of the MM cartridge with largely a second-order rolloff instead of the usual 4th order rolloff (cantilever resonance LPF + electrical resonance LPF).

However, for cartridges that depend on a balancing act between the cantilever resonance and the electrical resonance to achieve a relatively flat response, this technique is not as useful.

Cheers,
Bob
 
Hi,

This just looks like synthesized cartridge loading, a well-understood concept. It can sometimes provide a noise advantage. If you have a net gain of -9 to the far end of a 470k feedback resistor, you will have synthesized a 47k resistor.

That's what I said, except I I referred to it in a more generic terminology.

BTW, the "Actidamp" Circuit synthesises a 45.5K load...

The "heavy loading" applications rarely work with real MM Cartridges, though they can be used with other transducers and sensors to overcome similar problems with more success...

Ciao T
 
Hi,



That's what I said, except I I referred to it in a more generic terminology.

BTW, the "Actidamp" Circuit synthesises a 45.5K load...

The "heavy loading" applications rarely work with real MM Cartridges, though they can be used with other transducers and sensors to overcome similar problems with more success...

Ciao T

Hi ThorstenL,

Can you describe in more detail the limitations of this approach that you are aware of?

BTW, I think the Actidamp term may be misleading when used in conjunction with the heavy-load approach I described. By this I mean that the objective in the heavy-load approach is not to provide any mechanical damping of the cartridge by electromotive effects. The only damping that the heavy-load approach provides is damping to the electrical resonance to the point where it largely becomes a first-order rolloff. The active part of the heavy-loading approach is only there to get back the S/N lost at HF by heavily loading the cartridge.

I have experimented with the heavy-load approach for quite some time, and my main observation is that it is of questionable practicality because of the fact that the load resistance must be tailored to the inductance of the particular cartridge, so how that resistance is set becomes an issue. In principle, it could be done using a good frequency response test record and output level meters, but that is easier said than done. Obviously, it can also be done by ear to taste (as is usually done for conventional cartridge loading R and C selection), but the adjustment here has more mid-band implications. One can also easily build the preamp to be switchable between the heavy-load architecture and a conventional architecture, and then comparisons can be made between the two by ear with program material or a test record. In any case, the proper setting of the heavy load can be a PIA subject to error.

Cheers,
Bob
 
Bob,

Can you describe in more detail the limitations of this approach that you are aware of?

Many MM Pickups use mechanical as well as electrical resonance to get a flat response. Heavy loading will disable one of the two mechanisms but not both.

The reason is that MM pickups are (or should be) designed for a specific set of load conditions, changing the load conditions means the original flat response is no longer achievable.

Ciao T
 
This just looks like synthesized cartridge loading, a well-understood concept. It can sometimes provide a noise advantage. If you have a net gain of -9 to the far end of a 470k feedback resistor, you will have synthesized a 47k resistor.
:confused: IIRC, there was a previous discussion about using a feedback capacitor with an amp whose gain rolls off 6dB/oct to synthesize e.g. a 47K input impedance. The conclusion then was that there wouldn't be a noise advantage. (I think Scott had the explanation).
 
Member
Joined 2002
Paid Member
Hi,



It really depends. IF the lateral and vertical compliance are the same (not a given, try a Decca pickup) AND the arm has the same effective moving mass in both planes, yes, the resonances will be the same.

In practice I cannot remember a single arm/cartridge combo where they where the same, though usually quite close (using HFN Test Record).

Ciao T


With the SME 3OO9 Improved and the DL103 (both a bit “touched-up”) using the HFN Test Record, I had the peaks on 9Hz and 12 Hz, but I do not remember which was of the vertical and which of the horizontal resonance frequency.

Regards
George

> Edit: The cartridge was just burned in (~25 hours). The arm has some 30-35 years on it's back.
 

Attachments

  • Picture 057.jpg
    Picture 057.jpg
    162.1 KB · Views: 254
Last edited:
Status
Not open for further replies.