Excessive brightness - help & suggestions needed

My tonearm does have a VTA adjustment but I prefer not messing around with such sensitive adjustments. As far as I know it is set to be level. 

Edited by Analog Kid - 26 Apr 2009 at 11:25pm

In my experience excessive brightness is caused by high frequency instability.

Something shared by the now editor of Hi-Fi News, Paul Miller.

I think Paul was the first and last to do comparitive tests during the 80's in Hi-Fi Choice where an independent listening panel would do the subjective review and then Paul would correlate their comments with the high frequency stability (or not) of the stage/amp/power amp/etc.

The "items" with the most active high frequency plots (we're not talking about treble here guys, but up into the RF spectrum) indicating either instability or the onset of it, were usually the ones the panel of listeners could not get on with.

I remember comments on those "items" being "overbright" and "very tight bass" to name but two.

So what in a system could be unstable?

I know we are the immediate suspect on the rare occasion (1 in 100 to 1 in 1,000) customers find things have gone excessively bright after buying one of our phono preamp stages. It's an obvious conclusion for people to make, but it's the wrong conclusion.

The brightness one can get with one of our phono preamp stages is a short lived thing which goes with "burn-in" (constant use/power-up over the first 3 weeks).

A continuing excessive brightness is because of the inability of the rest of the amplification to be stable once a high frequency stimulus has been introduced. Evidence?

A shocking number of amps these days are made by meddlers who are also entrepreneurs but not electronics engineers although they have to claim they are to sell, and usually do that through numerous forums (not this one though).

Now because they have elevated themselves to guru status butter cannot melt in their mouths...

Need I say more guys?

But the evidence is provided in exactly the same way as Paul Miller did in the 80's Hi-Fi Choice reviews - we test our products for stability which is to say the rate of closure to unity gain. Beyond where the average techy (trekkie/trainspotter/etc) stops thinking which is usually the high frequency -3dB turnover point, the output keeps falling.

Now, I know there are filter circuits which fall at a hefty rate but these are usually designed such that there is no oscillation - and with filter section stripped out of the equation, the output should fall at a rate of less than 12dB per octave (40dB per decade) otherwise it will be oscillating.

So, do we say 11.5dB per octave would be stable? No, the output will ring - it will ring on a long time before decaying - that long that whilever there is signal (a musical source in this case) it will ring. While it is ringing all the power resources are being used to sustain it, and the ringing frequency causes high frequency ripple on the power supply capacitors. The ripple rating is often exceeded and the capacitors don't work right and don't do the job of being an energy store for the low frequencies during this. Therefore the amp in question cannot supply bass power to the load.

What set off the ringing? Obviously it is a stimulus at that high frequency! And yes, our phono stage in this 1 in a 100 to 1 in a 1,000 case that supplied the stimulus!

You see, most "designed by brainless ear" amps were designed using a CDP or other source that doesn't have the high frequency extension of one of our phono preamp stages. Therefore there was no such stimulus during the "design".

Now, if an amp is designed using mathematics (seeing it IS a mathematical device) and then tested using test gear, and even better, if it is simulated provided ALL the parasitics are modelled in, BEFORE being ear-tweaked (AS ALL OUR PRODUCTS ARE), then the designer can design so that the rate of high frequency closure is going to be stable - and that is going to be much less steep than 11.5dB per octave.

Some dabblers talk to me as if I'm still "wet behind the ears". I have worked in R&D with 3 amp or line level product manufacturers - I know what goes on!

You see, as humans, we can imagine all sorts of things and convince ourselves that our imagination is right - but only by using real measurements designed to remove doubts that exist only in the mind, can we establish the truth.

Analog Kid wrote: In terms of being driven into high frequency instability, by for example the Reflex, are valve amplifiers worse in that regard compared to transistor amps? Or better? And when choosing an amplifier, how am I supposed to know whether it will have a tendency to go into high frequency instability? Maybe there is a particular specification I have to pay attention to?

Correction: the Reflex simply supplies the high frequency stimulus (the fast rising edges of music that do exist, are also at low level and cannot be heard anyway), the subsequent amp does the rest.

Best thing to do is visit a shop that is into vinyl in a serious way - they won't stock a "squealer" that was designed with a CDP or or other bandwidth limited source...