Keith Howard meets the man behind one of the UK's few output transformerless valve amp brands

Glenn Croft might justifiably be called the recluse of British valve amplifier manufacturers.

A man who works until Sam and then sleeps until the early afternoon in order to hand?build his amps with minimal interruption from the outside world. A man so notorious for not answering the telephone that, until Eminent Audio took over distribution of Croft's products, dealers had to communicate with him via the post. Taking a mug shot of him for this piece, he says: `is probably not a good idea'.

In other respects, he could not be described as being backward in coming forwards. He designs and builds the only output transformerless valve amplifiers made in the UK and has strong, not to say controversial, opinions about most hi-fi components.

He recommends electrostatic and horn loudspeakers - uneasy bedfellows for most of us - and decries the popularity of moving coil cartridges, which he says are over-priced as well as overrated.

I didn't meet Glenn Croft face to face, but I did succeed in talking to him by telephone. He insisted that I send him written questions beforehand, to which he would supply written answers. A first in my interviewing experience, but a deal we both kept.

What follows is, as a result, an amalgam of his written responses and verbal clarifications, where I asked for them. I first asked him to explain what effects the well-documented technical limitations of output transformers have on the sound of transformer-coupled valve amplifiers.

`The general criticisms of transformer-coupled designs according to our customers are the grainy sound, muddied bass, lack of definition and lack of transparency.
`It is difficult to pin-point which specific transformer characteristics are responsible for these sonic deficiencies. It seems, howl ever, probable to me that if the bass is lacking somewhat in authority and depth, the primary inductance is insufficient. If the treble is rolled-off or coarse, then I would expect that the leakage inductance is too high or the interwinding capacitances are too high or imbalanced.

`Generally, though, the warm and rosy bass response associated with modern valve designs can be attributed to appallingly inadequate and unregulated power supplies. High frequency problems are further exacerbated by inadequate circuit design in not controlling the high frequency amplification to feedback relationship.

`You can run into transient stability problems, you can have a peak in the top end response, depending on the capacitive load presented by the speaker, which can vary according to output level as well. This is without considering whether an amplifier has a regulated high tension supply, a good driver stage and phase splitter, correct operating voltages and decent component quality.

`It may be true that careful attention to detail in transformer design may lessen some of the myriad problems, but the factors which I influence the ultimate sound quality are not definite. The obvious solution is to employ the perfect transformer: no transformer at all.'

Croft now offers two output transformer-less power amplifiers, the reworked, 1600 Series IIIR stereo amplifier and the brand new, 5500 Series IIR monoblocks. One difference between them is that whereas the Series IIIR is an all-triode design, the more costly IIR uses pentodes in both the voltage amplifier and phase - splitter/amplifier stages. A move which some would regard as retrograde.
 
"HE HAS STRONG, NOT TO SAY CONTROVERSIAL OPINIONS ABOUT MOST HI-Fi"
 
`There is a school of thought that attributes a purer sound to triodes due to their more benign distortion characteristics. But pentodes~can give much higher gain and, due to the much lower inter-electrode capacitances, more frequency response.'

This allows the application of greater loop negative feedback, which increases from a low 14dB in the Series IIIR to a modest 34dB in the IIR. High values of loop feedback, Croft says, `tend to give a dead, closed?in sound quality'.

Triodes are used in the IIR's output stage, however, in preference to the pentodes that are used in the classic Futterman transformerless designs. `The triode output valve we use has an inherently lower impedance than the pentode used by Futterman, enabling us to extract more power into lower impedance loads. The Futterman output stage used in a stereo amplifier also requires three more regulated 250V power supplies to drive the screen grids, which greatly complicates matters.'

With feedback connected, the output resistance of the IIR is 0.2ohms. A value no greater than that achieved by most transformer?coupled valve amps, but higher than most transistor designs. Devotees of high damping factors will tut?tut, but Glenn Croft is unimpressed.

`There is more scope for a high damping factor with a powerful out transformer-less design but we do not pay lip service to damping factor. In general it does not appear to correlate well with sound quality.'
 


Home page Reviews main page
More Reviews........
Previous review Next review
 

© Croft Acoustics 2008 (design by EnterpriseDesign) t: +44(0)1723 355111 e: glenn@croftacoustics.co.uk

 

Back to homepage Send email