Valve electronics never reached the peak

I believe recent discussion about valve amplifiers has been missing several points. Valve technology has never been fully developed and your recent design of a valve preamplifier is a good example of how modern circuit design can be used with valves.

Just imagine how far valve design could have been developed if the transistor was not discovered for several more decades. I don’t believe valves were at the peak of their design life when the transistor came in.

Electronics was just starting and its growth in complexity still could have happened with valves as its active component. The engineers of that time would have loved to have the cheapness of components, the technology to manufacture small and complex devices that we have now and the test equipment and the Spice programs we have to work with.

Valves and transistors both have their good and bad points and should be used to their best advantage.

Grahame Macpherson, Redcliffe, Qld.

NOAA satellite receiver is excellent

Jim Rowe’s articles on receiving NOAA weather satellites in the December 2003 and January 2004 issues are superb and have come at a good time for me. I had been considering assembling economical NOAA weather satellite receiving stations for use in schools but the big sticking point has been an economical suitable wide-bandwidth FM receiver. Jim’s economical kitset receiver has filled the gap.

On the subject of suitable software to run a NOAA APT weather satellite receiving station, may I also mention a superb computer program which was absent from Jim’s comprehensive list in his first article. That program is "WXtoImg" which I currently use and which I believe is so exceptionally good that it rates a special mention.

WXtoImg has many clever features, including automatic Web page creation and upload, optional precision computer clock timekeeping via GPS receiver, and automatic Keplerian element updating. Also WXtoImg displays APT weather images as they are being received, which gives an illusion of actually flying with the NOAA spacecraft and looking down on Earth; quite exciting stuff for a school kid (and big kids!).

Any basic Pentium computer, combined with Jim’s receiver and antenna and running WXtoImg would make a very slick little educational or home receiving system. A well-featured version of WXtoImg may be downloaded as freeware from www.wxtoimg.com

Andre Phillips, VK2AAP/ZL3AW, Coonabarabran, NSW.

Dual supplies not a big factor in amplifier performance

I rather liked the concept of the dual power supply in the Ultra-LD amplifier design (SILICON CHIP, March & May 2000, November & December 2001, January 2002). That is, having a separate regulated supply for the front-end voltage gain stages and a conventional high power/current supply for the driver and output stages. I imagine that this configuration results in near perfect CMRR figures which remain constant even at loud levels.

I would like to know how much this contributes to the high performance specs of this amplifier – and if this could be adapted to other designs on the basis that this would improve the specs of these also; including the SC480 and now the Studio 350? Do you think that there would be any advantage in doing this?

Grant Saxton, Cambridge, New Zealand.

Comment: while the separate power supplies do offer a useful improvement in performance, it is not a major factor. Where they do help is in providing a higher signal voltage from the driver stages before the amplifier runs into clipping. This is why the Ultra-LD has such an abrupt transition from low distortion into clipping, as compared with the more gentle transition with the SC480 and Studio 350 designs.

By far the most important factor in the low distortion of the latter two amplifiers has been the PC board layout. We now think that if we applied the same PC board strategy to the Ultra-LD, the performance would be even better. Would the separate power supply rails produce a big improvement in the SC480 and Studio 350? Probably not.

Human-power LED torch variant needs no rectifier

Here is an idea for another variant of the human-powered LED torch presented in the February 2004 issue. During my initial testing of this great idea, I found that unless the stepper motor was turning fast enough, nothing happened. This is due, I believe, to having to overcome the forward bias voltage drop of two diodes (1.2V). It occurred to me that this is a waste of output, so two pairs of LEDs were connected directly to the motor as inverse-paralleled pairs.

The result of this is quite dramatic. The slightest turn of the shaft produces very bright, although flickering light. Of course, the flickering smooths out if you turn fast enough. The motors we initially used are quite small; with large ones, the output is awesome!

This application would be ideal for installation on bicycles, etc, since the objective is maximum light for minimum weight. The fact that there is always one LED conducting prevents any possibility of excessive reverse bias, although I have not measured the no-load output. Also, there appears to be a different output sequence depending on the rotational direction.

Plugging the LEDs into the connector on the motor cable, in a square arrangement of inverse-parallel pairs, the sequence is diagonally opposite pairs flash together while in the opposite direction, adjacent pairs flash. I have observed this with two different motors. There are probably applications for this such as directional sensing.

Geoff Hahn, via email.

Of valves and brakes

Thank you for the article on anti-lock braking in the February 2004 issue. I recently did an advanced driver’s course and most of the cars had ABS. Mine didn’t. At the start of the day I was able to out-brake and out-corner most of those using ABS. Sure it doesn’t take much to learn the basics, and once they were shown the importance of seating position and brake pedal use, they outmanoeuvred me with little problem. But it does raise the question of how many people receive the correct instruction in ABS use at the time of purchase.

I love the arguments about valves. I do not own a valve amplifier. I tried a few under the $5000 mark but they did not set me alight, however my CD player does have a valve output stage.

Specifications can be misleading and lots of preceding zeros in the distortion figures are not necessarily an indication of sound dynamics. I sold a well-known transistor based amplifier with distortion figures of around .002% at 170W and replaced it with a 70W amplifier with distortion of 0.12%. Yet there is a vast difference in what my ears hear. AND shock horror, it has zero negative feedback.

Andy Lee, Motueka, New Zealand.

Comment: it is safe to say that, in Australia at least, no-one is instructed about a car’s features at purchase and certainly not the best way to use the ABS.

Exciting time for DIY loudspeakers

Now is an exciting time for do-it-yourself loudspeaker construction as many important developments and notable names exist now. For decades the weakest part of the hifi system has been the loudspeaker. By the early 1980s sealed and especially ported manufactured loudspeakers have been accurately and scientifically designed using parameters from the work of Thiele & Small.

In addition to properly designed boxes, current drivers can be really quite good and need not easily be the weakest link. For example, Seas Excel range of mid/woofers offer a claimed THD of 0.1-0.3%, a remarkable and also figure (see www.seas.no)

It is an exciting era as information can be directly and personally retrieved from the internet. One outstanding source from the web, for example, the DIY Loudspeaker Designer’s Selection Guide (LDSG) lists all recommended drivers (see ldsg.snippets.org), as well as web retailers such as (www.madisound.com) and (www.e-speakers.com).

For excellent value for Peerless or Vifa drivers and cheaper varieties one can’t go past the Australian (www.wescomponents.com). E-speakers and LDSG are good to refer to ribbon tweeters, another exciting development due to cheaper production of rare earth magnets and fidelity demands of DVD-audio and SA-CD. Philips now market a consumer DVD surround system featuring their own leaf ribbon tweeters.

The web also allows downloading of great freeware to design boxes to one’s desire using Thiele-Small parameters (eg, see winISDbeta on www.linearteam.dk).

For ease in building speaker cabinets there are hardware items like the very useful Jasper Jig for routing perfectly round holes and recesses (see soundlabsgroup.com.au). I recommend buying a router with exchangeable 1/4 and 1/2- inch chucks, and perusing a router bit catalog (see www.carbitool.com.au).

The hobbyist’s effort in either building a budget or a world-class speaker system is practically the same, as is the cost of tools used and the timber which will more often than not be MDF. By spending hundreds, the hobbyist can build something of only better value than something in the shop. By spending a few thousand, however, the hobbyist can build a world-class system, giving many years of fine listening, and giving a system that would be many times more expensive if sourced commercially.

Sadly, this second choice has not been an option from the offerings of electronic magazines. This is a pity. One aspect of speaker building which has not been covered, is the crossover. These are quite tricky for the novice to design and the quality of the passive components in the crossover can be detrimental to the sound, especially for modern high definition drivers and amplifiers, etc.

This problem can be completely sidestepped by using active cross­overs and multiple amplifiers custom-designed for each speaker box. This will actually give the greatest sound quality and is perfectly matched for the realm of electronic enthusiasts. However, active systems dictate an expensive project not suited for budget drivers. Thus a complete feature for an active system, using the classiest drivers has not been featured in magazines, which have robbed enthusiasts of a class project, despite the classiest amplifier modules and active cross­overs being published.

Perhaps the latest & greatest exciting development is a resonant free speaker frame which is boxless and (necessarily) active to give the highest available fidelity without outrageous costs – a perfect project for the electronic audiophile enthusiast (see www.linkwitzlab.com). The credentials for such a design comes ironically from one of the developers of commonly used passive crossover design theory. I predict that this active open-baffle system could be, or should be, the future for home audiophile enthusiasts with electronic skills.

Would it be suitable for SILICON CHIP to bring together past articles with fine drivers (eg, Seas) for a fully-featured active speaker system and also perhaps flirt with a Linkwitz-style active open-baffle non-resonant system?

Paul R. Rohde, via email.

Comment: the major reason why do-it-yourself speaker systems with such elevated prices have not been described is that they are far beyond the budgets of most readers.

Making PC boards with transparency film

I use a different technique to that described in your April 2004 issue on circuit board preparation and this yields much better precision then the "iron on" method.

I use a 600 dpi laser printer together with Hewlett Packard LaserJet monochrome transparency film (available from the larger stationery suppliers like OfficeWorks). When I set up the printer out of Autotrax, I use the "mirror" setting which puts the image on the film in reverse – this will be explained later. I also use the pre-coated positive-acting laminate from Kalex together with their liquid developer. This system, although photographic, can be used in subdued light, so there is no need for a darkroom.

When the transparency is ready, the protective membrane is peeled off the circuit board and the transparency is then laid face down (image towards the board) on the pre-sensitised surface of the board. With the image printed in reverse, the image will now appear "right way around" on the board.

The board is then exposed through the film in a light frame (with four 20W UV fluorescent tubes) for about 90 seconds. This time will need to be determined experimentally using scraps of circuit board and depends on the light frame configuration.

Once exposed, the transparency is separated from the board and the board then developed in the solution supplied by Kalex. When the image is fully developed, the board is washed off in hot water and then etched as in your article. I prefer to use ammonium persulphate (even though it must be heated) because I find the ammonium much cleaner and more pleasant to use – besides I get into far less trouble from "she who must be obeyed" about stains in the laundry and on my clothes!

The image obtained by this method is very sharp and very well defined. I can use tracks down to 10 mil with confidence and making a double-sided board is basically just as easy. You just produce the two sets of artwork (both sides), carefully align them, lay them up together and slip the circuit board in between using adhesive tape to secure the board to the film so that it won’t move when you turn it over to expose the second side.

The secret to this method is that you cannot use "just any old transparency film"; it has to be "good" (read, a little more expensive) and be capable of producing a good solid black image, such as the HP film. Once set up, I can produce a double-sided circuit board in about 10 minutes from printer to etched board and be confident about the finished product every time.

Jeff Thomas, via email.

Comment: we described this method in detail in March 2001, although we suggested that "... the density (of transparency film) isn’t quite good enough for PC board making". Guess it is, if you have the right film & printer!

EGO sensor as kiln monitor

Pages 64-65 of the April 2004 issue have a truly excellent description of the operation of an EGO oxygen sensor. I am a potter with a science degree background and a long-time reader of SILICON CHIP.

About 18 months ago, I spent time devising an instrument for measuring the degree of reduction or oxidation in the atmosphere of a gas-fired kiln, making use of an EGO sensor mounted at the kiln flue. This was written up as an article featured in the American pottery magazine "Ceramics Monthly", in their March 2003 issue, under the name "Kiln Exhaust Sniffer". This was later included in the "must read" pages of their website. The URL is www.ceramicsmonthly.org

Disappointingly, in their editorial wisdom they didn’t include the graphs of EGO readings at various stages in a kiln firing, showing how the millivolt output from the salvaged EGO sensor ($5) compared with that from a commercial oxygen probe ($800) but the whole idea came out plainly enough.

Roger Graham, Pottery at Old Toolijooa School, Toolijooa, NSW.

Comment: The EGO description was part of the article entitled "A Smart Mixture Display For Your Car". Readers can access the article on-line at www.silicon.chip.com.au