Meetings Program
 and Reports
Contact Us
Headlines List

Page updated Friday
23rd 2002f August 2002
04:01:44 UTC
Now 16:43:11 UTC

  Halcro $50,000 Audio Amplifiers - Extraordinary Technology

Date:July 12, 2002
Time:7:00 pm.
Place:WEA, 223 Angas Street, ADELAIDE
Presenter:Mr Bruce Candy

THE HALCRO AUDIO AMPLIFIER Bruce Candy is the designer of the Halcro audio amplifier which many regard as the world's finest. He is also the designer of Minelab's land mine detectors used in 70% of the world's land mine detecting activities. We had an excellent presentation on his mine detectors a few years ago so this time Bruce provided a fascinating insight into the Halcro amplifier design for our July meeting at the WEA.

Bruce has been all over the world displaying the Halcro at exhibitions and the amplifier has been featured in many audio magazines, some with stunning photos of the Halcro on their covers. It has been spectacularly successful, particularly in Japan and the USA, which, with a price tag of about $US40,000, is no mean feat. Also, as Bruce pointed out, Australia is the last place on earth that overseas customers would expect such excellence in electronics' design. Our national image of the outback and kangaroos was just another hurdle that the Halcro has had to overcome in addition to our remoteness and lack of local component manufacture. However, there is one advantage Bruce has that is hard to find anywhere these days and that is good, analog design skills in a world that has gone completely digitally mad.

The Halcro amplifier is a fully solid state design. As Bruce said, in spite of all the hype about valve amplifiers, they still have measurably higher distortion than the best transistorised designs such as the Halcro, but that is often the way listeners likes it! In Bruce's and many others' opinion the best quality classical music needs to be heard with the very least possible distortion. Other kinds of music such as pop music may actually benefit from the harmonic distortion introduced by valves. Audio still remains largely subjective. Nevertheless, if you truly want low distortion, for whatever reason, then you will have to go for a design such as the Halcro to achieve it.

Bruce then got into some design detail starting with an area that is frequently neglected by amplifier designers which is vital to ensure the highest possible performance and that is the power supply. The usually wildly fluctuated current demands that occur in audio amplifiers makes conventional designs using transformers and diode bridges, as well as most switching power supplies, a source of distortion. Bruce has put a lot of effort into the power supply design in the Halcro and it is a major contributor to the overall high performance (as well as a major contributor to its overall bulk).

Another area that needs careful consideration is the front end. This needs to be very low noise and very well isolated from the power supply. Bruce's design features constant current and current mirroring to achieve this.

A design factor that would not normally be of any concern in lesser amplifier designs is where the high currents circulating in the output stages can induce unwanted feedback into the front end. Bruce's solution to this problem in the Halcro has been the effective use of thick aluminium eddy current shielding.

Next Bruce moved onto measurements. As it turns out measuring distortion and noise, which are the most important characteristics of an audio amplifier, proves to be fairly simple. Bruce uses the Baxendall method where the amplifier under test is placed in a unity gain resistor bridge such that the signal is completely cancelled out leaving only distortion and noise products. There are some phase compensation and fine resistor adjustments needed, but when they are achieved you can read incredibly low values of distortion and noise.

Unfortunately for us Bruce was unable to discuss any details of the output stage which is commercially secret and holds the key to much of the Halcro's success over its rivals. Oh well, we will all just have to rush out and buy one or two and have a look inside . . . maybe.

Bruce's next venture looks like being home theatre audio systems. Presently, the most favoured design is the digital-analog, class-D, or delta-class amplifier. This design has the advantages, especially where maybe 7 or more channels are required, of being small, having low power dissipation, and being more affordable. This will compromise performance a little, say compared to the Halcro, but should still be acceptable. The digital-analog amplifier uses a comparator to compare the analog input signal to a triangular wave-form to create a pulse-width modulated squarewave. Relatively small output transistors are used in a low power dissipation switching mode instead of needing to use much bigger transistors in a far higher dissipation linear mode. The pulse width modulated output is filtered before being fed to the loudspeakers.

Bruce believes this mid-performance home theatre market where there are some really big players such as Sony and Philips will be much harder to break into than the top end that the Halcro enjoys - although Halcro's good name will certainly help.

Bruce listed a lot of audio amplifier design references such as the one he most recommends, a book by Douglas Self. He also had praise for the many articles over the years in Electronics World (the long running British publication formerly known as Wireless World). The number of Japanese publications Bruce had on hand (unfortunately all written in Japanese) was amazing, but it was still possible to often see the word "Halcro" and photos of the Halcro. There was also mention made of what Bruce regards as a fraudulent book on audio design which will remain nameless here (sorry guys, you should have been there!).

There was a lot of general discussion on audio such as choice of loudspeakers (Wharfdale, Duntech, and others), oxygen-free and Litz-wire speaker cables, vinyl versus CDs, live versus recorded music, connectors (a major source of noise), and room acoustics among others. This was, in fact, in nearly 30 years of existence, the first time we have had a presentation where not one single chip or even a single byte of code was mentioned. It was purely analog. Many thanks Bruce for this introduction to some, and a welcome refresh for others, to good, solid-state analog design. Good on yer Bruce, for keeping Australia on the "audio map".

. . . Rick Matthews