A Classic Restoration Saga
A good friend I work with took pity on me and donated this project to "the cause". David McGlinn runs a busy and successful consumer electronics repair business in Ringwood, Melbourne. It was given to him by his uncle, some thirty years ago, and is a particular and valued treasure for him. For many years David used it "as is" as a workshop radio, in raw form, mounted simply on a slab of wood. I was extremely surprised inheriting it, and despite his oft repeated protestations, I've never really thought of it as "mine". At best it's our joint project, (he continues to take a keen interest in progress), and indeed I'd love to see it returned to him, restored, repaired and glorious!
David dropped it off to me one Saturday afternoon, in an old cardboard fruit-box. I think it had languished in his garage for quite some years, although not at all lost and forgotten. It was just a chassis, together with the speaker transformer loose in the carton. The chassis was certainly old, but aside from some apparently light surface rust, didn't look too bad.
There was a complete and apparently working set of valves - the output push-pull pair, 6V6's, were dissimilar types and both had cracked bases but otherwise later proved to be quite OK. Valves were, (RF onwards); 6SK7, 6J8, (which I learnt was a "triode-heptode"), 6AR7, (audio baseband about now), two 6SJ7's sequentially and then two 6V6's, in push pull. The dial scale was gone, but must've been rectangular, about (say) 5" by 10" – rear illuminated, with four lamps closely set behind. The dial scale would've got hot, so I suspected was probably originally of glass. David had apparently packed it away with some household goods placed in storage, as he was "between houses" at the time. Perhaps there was a chance that this key item might turn up during a later move, planned for about six months hence.
Meanwhile there was obviously plenty to keep me busy. Where to start? It seemed that so much needed doing!
The underside of
Research on the Internet
First I got on the net but got nowhere at all trying to find specific references to "Classic" brand Australian valve radio's. I knew from David that was the brand name, and that Classic had been an Australian manufacturer, but had no model number or manufacturing date, and certainly had no schematic nor ready access to one. While I was quickly able to map out the audio section of the radio, the RF section contained more adjustable coils and trimmer capacitors than I could imagine one AM and SW radio could possibly contain! Ugleeee! New to this game I certainly had no network of contacts amongst radio restorers or associated groups.
But after a bit of digging I found an extensive website by Tasmanian Paul Ledger. From his website came pictures of the Classic brand models Chieftain 10, Chieftain 2, Chieftain 3 and Chieftain 4. Apparently these were a "a series of advertising pictures featured in The Australian Radio Electrical Weekly. Classic were noted for their "Cocktail Cabinet" radiograms. Their radios were very basic designs, but sold well due to the attractive and innovative cabinetwork."
Paul Ledger built up a most remarkable Australian vintage radio website some years ago, with a substantial gallery of indexed images, perhaps now in need of another enthusiast’s interest. I gather that some time ago Paul lost most of the off-line website files in a computer crash, and the enthusiasm to build it all back up again has waned, overshadowed by more recent interests. He could well be in the market to let someone else take over the site. “Maintaining it means thumbing through old mag’s for piccies of old radios and gear etc. etc, and then editing them to go on the site” he told me.
Paul is a friendly, intelligent fellow living in Somerset, on the north coast of Tasmania. I was fortunate enough to be able to meet him for an hour or so in mid-2000, and enjoyed his relaxed take on life, generous spirit, and above all his ready but kindly wit. These days Paul is an intense and highly experienced eBayer, trading in valves and collectables very extensively. I purchased several 6EA7/6EM7’s from him in early 2001 and can heartily recommend dealing with him!
Paul Ledger's Vintage Radio Website
A motley collection of excised greasy waxy capacitors
Capacitor and Resistor Makeover
On the practical side I launched straight into a cut-n-tuck on all the grotty paper capacitors, electrolytic capacitors, and half-hearted repairs tacked in some time in the past. That all went pretty smoothly - there wasn't a lot in the chassis, so I just had to be basically thorough and careful. The output transformer was fine, as was the power transformer, (though it had ex-waxed a little bit, so some degree of over-heating must've occurred). While David recalled that the unit was working already when last he'd powered it up, albeit two or three years ago, I wanted to do a complete number on it first before applying electrons. Everything was sailing along enthusiastically and merrily, (it could only go downward from there), until I hit the first of several brick walls; I couldn't read the resistors! And how was I ever going to get help? What? Admit I couldn't read resistors? Ya gotta be bleedin' jokin' mate!
Damn Resistor Colour Codes!
Pretty soon I had a collection of resistors with markings I could make no sense of. There were even examples where I had three or four resistors in different parts of the circuit with the same colour coding, but all measuring radically differently both to each other and to any combination of numbers I could attach to the bands and body colours displayed! Quite extraordinary!
The first of the four colours in the sequences listed below was an "undercoat" appearing again as the third colour listed; bands two and four were over-painted atop this undercoat. The measured values of all the resistors in the chassis that supposedly were the "same" are then shown. How to read this colour code; you try!
Weird resistor colour codes!
An American, Brad Thompson, offered help in response to a query on this I posted on newsgroup rec.audio.tubes;
"Some older resistors used a "BED" code, an acronym for;
Body (first digit)
End (second digit)
Dot (multiplier as powers of ten-- black = 100, brown=101, etc).
While originally applied to radial-leaded "dog bone" resistors this code also carried over to early axial-leaded resistors. Thus, a brown body with a black end and a green dot defined a 1MOhm resistor."
Applying this didn't really work to deciphering the resistor coding system, but I sure appreciated Brad's thought and response. "Larry S. B." added "One odd thing I've noted with old carbon composition resistors is that big values sometimes drift down in value and small values always drift up." And while I was getting nowhere fast, I appreciated this respondent's support immensely too.
I sat and I stewed and I stewed and I sat. One day it just hit me like a ton of bricks. The classic brainwave! If I ignored the body colour, (the first of the four colours in the sequences I've listed), and then reversed the usual read-order of digit-digit-multiplier to multiplier first, then the digits, I guessed the following;
Green Yellow Green Black . . . measured 516kOhm, 617kOhm, 535kOhm, treat as 0,000 (yellow), 5 (green), 0 (black), i.e. 500kOhm. Errors against measured values would be 3%, 23% and 7%.
Red Green Red Black . . . measured 2.92MOhm, 2.41MOhm, treat as 00,000 (green), 2 (red), 0 (black), i.e. 2MOhm. Errors against measured values would be 46% and 21%.
Green Purple Yellow Purple . . . .measured o/c, 800kOhm, 740kOhm, 820kOhm. I now assumed I had been reading this one backwards, and it should be Purple, Yellow, Purple, Green. If I treated this then as 0,000 (yellow), 7 (purple), 5 (green), i.e. 750kOhm, errors against measured values would be n/a, 7%, -1.3% and 9%.
Green Orange Green Black . . . measured 55.7kOhm, treat as ,000 (orange), 5 (green), 0 (black), i.e. 50kOhm, and error against measured value would be 11%.
I then needed to suppose that the last one, Yellow Black Orange Black, was exceptional - on the basis that in this case the first and third colours were not repeats, it seemed reasonable to imagine that the coding system was different. The examples measured 33kOhm , 34kOhm, 77.7kOhm. Treated in "modern" fashion, as 4 (yellow), 0 (black), 000 (orange, i.e. 40kOhm, then errors against measured values would be -18%, -15% and 95%.
The scheme seemed to work; resistor tolerance may have been as wide as 20% (ouch!) at the time this thing was made. I guess also that the additional thing that was throwing me, (could there me more?), included the nominal resistor values I wasn't used to seeing. For me 470kOhm, 2.2MOhm, 680kOhm or 820kOhm, 39kOhm and 47kOhm rather than 500kOhm, 2MOhm, 750kOhm, 40kOhm and 50kOhm were more familiar. For the above scheme to be correct some of the resistors had to be quite "low" in value, (although technically not "faulty" if I allow the 20% tolerance). In my experience resistors almost invariable fail "high", not "low" in value. "Larry S. B."s words, quoted earlier. I replaced all the above resistors with values as per this scheme, and have little remaining doubt now, (using my keen 20-20 hindsight), that the scheme is indeed quite correct.
A Wild Goose Chase, Identifying the Radio's Age
Meanwhile I was in the process of wandering off on a wild goose chase over the units age. I'd learnt from a friend that for many many years in Australia, manufacture of a radio receiver required a license, and circuits of all licensed radios were published in an annual almanac-type book. Apparently sets of these books were rare and obviously valuable, but they were around. If I could identify the year of manufacture, then maybe I could find someone with a set of such books, and then maybe I could get a circuit! Seems convoluted, and in hindsight perhaps there would have been alternate ways of going about it, but it made sense to me at the time!
I started searching for a vintage radio club or something like that; maybe I'd get lucky and there'd turn out to be one right here in Melbourne. There's an Audio Club, so why shouldn't there be a Vintage Radio Club or some-such? Of course there indeed was, the Historical Radio Society of Australia. And from one of their newsletters came a very interesting piece about the "ARTS & P" licensing system.
"The ARTS & P system was a licensing system that was used in Australia and New Zealand between 1934 and the 1960’s. The system was introduced to verify that radio manufacturers payed royalties for the items they were using that were covered by Patents. Some radio makers avoided this by selling parts, rather than complete radios, even though the parts could be easily assembled into a radio. Each licensed radio was fitted with a small sticker attached to the back of the chassis. The colour of the sticker is a useful way of determining the age of manufacture.
1934 . . . white, serial number prefixed by the letter A.
1935-1936 . . . pale blue, serial number prefixed by the letter B.
1936 . . . pale blue, serial number prefixed by the letter C.
1937 . . . pale blue, serial number prefixed by the letter D.
1938 . . . pale blue, serial number prefixed by the letter E.
1939-1940 . . . pale blue, serial number prefixed by the letter F.
1940-1941 . . . pale blue, serial number prefixed by the letter G.
1942-1946 . . . pale blue, serial number prefixed by the letter H.
1946-1952 . . . dark green with red letters, serial number prefixed by the letter T.
1952-1955 . . . orange with dark green letters, serial number prefixed by the letter T.
1955-1960’s . . . small pale blue, with dark blue letters, no prefix to the serial number.
Reference; "How Old is Your Radio", H.R.S.A. Newsletter, October 1987, p.20
The sticker on the Classic was pale blue, and the serial number was stamped into the metal at the rear of the chassis; CL9245. Using the preceding table, this dated the radio at 1936. Over 60 years old. The totally unfamiliar resistor colour coding scheme seemed to make more sense in the light of this. I then set about asking about on the net on this basis, and received a wonderful and illuminating reply from immensely experienced Queensland vintage radio enthusiast, Brian Smith. Brian operates a very successful vintage radio restoration parts supply operation; Brian Smiths Wireless Workshop . The range of stuff that Brian carries and can supply is simply astounding; items that I for one would never imagine could be made available. Well worth a cyber-visit. I later obtained a copy of Brian's catalogue - and then it was easy to see that his prices were very reasonable. A great find! ("I'll be baaack").
Brian Smith's Wireless Workshop
The name Classic rings a bell but it is not listed in the AORSM's, the books you refer to - first volume is 1937! I think that Classic was a brand name used by another manufacturer or perhaps it is the model name! I recollect that someone, possibly Eclipse, used this brand, but Eclipse sets do not feature in the books either! Earlier other info’ I have does not assist, also! If you can send more details, such as number of valves and types, number of wave bands and types, number of front panel controls and their use, etc. I can search through the books to see if I can find something similar in the way of a circuit. Circuits of the period had many similarities between brands, models etc, and a similar circuit could assist you. Radiograms were not common at that time, although many mantel and table-top sets and a few consoles had connections on the rear of the chassis for a pick up to be connected, allowing the user to purchase a turntable and pick-up in a stand-alone box at a later date. I suspect that this is the sort of thing you have. Tuning eyes were generally fitted to up-market models of the console or table-top type."
I took the opportunity to ask about doing an alignment on such a radio; I don't have a strong practical background in "RF" and the thought of mucking up all those trimmer capacitors and coil slugs was weighing heavily on my mind.
"With regard to the alignment, I have found that as a general rule the original settings are usually almost bang on. It is really only necessary to touch things if you have to replace a coil/IFT etc, so I would leave the original settings alone until you have done all other necessary restoration work. Just a point about the IF transformers; sets of the period used frequencies in the 150kHz~250kHz range, not the 455kHz we are used to these days!"
Brian continued, with helpful, general advice to help introduce me to restoration work. What an amazing degree of forthright support. I hope that one day, in my turn, I get to help a novice the way Brian so openly gave of his time and knowledge to me!
"Most troubles with these old sets can be cured by replacing all the paper and electrolytic capacitors. Mica capacitors are generally very reliable. Output transformers are often open circuit and sometimes the power transformer has problems, having been overloaded by all the leaky capacitors. After checking continuity of all coils, transformers, speaker etc. and replacing paper and electrolytic capacitors, the set will usually fire up OK and then you can fix all the sundry other niggles! I doubt that a dial is available, but if it is a small round one the Radiolette 27 may be able to be adapted".
The fact that Brian couldn't immediately identify the unit was worrying me, and strongly suggested I'd fed him some wrong information. At this point I sent Brian details of the valve compliment, and the Chieftain pictures. I knew that my chassis agreed completely with these.
Brian came back all but immediately: "The extra details you supplied have been of use. You appear to have a "Classic", model CL, made by Classic Radio of Malvern, Victoria, in 1950, and marketed for some years after. The model was updated in 1952 to include some miniature valves. The circuit and parts lists for both models are available."
A solid, positive identification. Wow! And almost fifty years old rather than sixty, was still a fair innings. So how had I mislead Brian so woefully?
"Your decode of the ARTS & P must have been incorrect as the following points should be noted:
Valves of the 6S series were first released in metal form in the US in 1938. Glass types followed about a year later in the US, but by this time the war had started and most Australian manufacturers continued using older types as the newer designs were going into military gear! Domestic radio manufacturing virtually ceased in Australia for the period 1942~1946 due to the military's requirements. It was therefore not until post war that Australian manufacturers started using the 6S series valves.
Rectangular dials had just started to appear prior to the war, but were generally of smaller dimensions. In pos-war times the larger rectangular dials were the norm.
The 6AR7GT valve is a unique Australian type designed after the end of the war. Its characteristics are very similar to an EBF32/35 with the same pin-outs - the CL circuit actually specifies the EBF32 valve.
The IF of this set will most likely be around 455kHz. My previous comments about low IF were on the assumption that your set was pre-war."
Brian later clarified what speaker impedance I should and could use; I'd so far assumed that in days of yore relatively high speaker impedances were the norm, and I'd told him I expected to use a 16 Ohm unit or series two 8 Ohm types.
"The speaker would most likely have been 3.5 Ohm impedance; 15/16 Ohm speakers started to appear from the mid 1950's onwards. Using your 16 Ohm will not cause problems; a slight lowering of output power, (5% approx.), and a slight increase in distortion products from probably around 6% to 6.5% - you will not notice these changes! "
I was thus able to get from Brian copies of both the 1950 Classic Model CL schematic and 1952 Classic Model CL schematic, and a suitable set of RF alignment instructions. I remain immensely indebted to Brian.
Disaster at the Paintworks
No way was I out of the woods yet, for absolute disaster now occurred. The chassis was baked enamel on steel, with some steel exposed and lightly rusted. I had several goes at trying to clean the rust away with emery tape, but there was just too much of it, and in too many inaccessible places. Articles in Silicon Chip and Electronics Australia on vintage radio restoration indicated that the done thing, by the experts, was to completely strip all the components off the chassis, completely redo the bare metal, and then rebuild the entire radio again. It was an absolute joke to think I had anything like the experience at this to do that sort of thing! Did that therefore disqualify me from tackling a project of this nature? Some folk might well answer "yes!"
Rear view of the completed and painted chassis
Instead I purchased and applied some "rust restorer". I'd never used such stuff before, (why not I don't know), but understood it to react chemically with rust to form an inert but stable compound which would halt further rust attack, was black or grey, rather than the distinctive rust colour, and if needs be could be painted over. Where I'd acquired all these half-baked notions was anybody's guess. Anyway, rust restorer went on and seemed to be working well; I tried a small area, which seemed to go OK, and proceeded minutes later to liberally coat the whole of the upper surfaces.
What a shambles! The partially missing coating the chassis had on wasn't baked enamel at all. It was in fact paint. It'd had me completely fooled (OK, not a particularly difficult thing to achieve!). The paint reacted with the rust restorer, to form a sticky, streaked, resinous black gunge. I'd made one hell of a mess! In a blind panic to clean it off I destroyed the marking sticker on the back. I was dumb-struck, and sat there stunned. The whole project languished on the workbench, during which time I harboured the terrible secret of what I'd done . . .
Climbing back out of this hole was slow and painful. I spent hours cleaning every inch down and painting it, using a ton of patience and a very small brush, in a metal primer, (fortuitously very nearly the very same pale-greenish colour as the original). It came up pretty close to the original look in the end, sans the sticker, which I remain really annoyed about destroying. It was a humiliating and chastening experience.
Testing Time is Here!
Finally the day came to power it up. Away it went, no trouble at all. I was thrilled. The RF side worked fine, as did the audio, and I decided to leave well enough alone as far as alignment was concerned. A few voltage checks to confirm that all was well, and that the valve pin voltages made sense compared to spec. data I'd assembled for each valve, and there it was. Things could only get better from here! Even the magic eye tuning worked fine, and I'd added a different pair of 6V6 outputs I had to hand - they matched each other and had intact bases which the ones I found in there hadn't.
The dial scale itself - in near perfect condition.
Surprise, surprise, look what's just shown up! David found the dial-scale, and it is in near mint condition. What a piece of artwork! It's complete with the manufacturer's logo "crest"; just the ticket. It appears it would have fastened, (by means of four screw holes, one at each corner of the glass), to the wood of the cabinet, as there's no mounting evident on the radio chassis itself.
I discussed with him what to do next, by way of mounting in a cabinet. It never occurred to me to start chasing down an original cabinet; I couldn't see that as a realistic possibility. And, judging by the Chieftain pictures referenced earlier, they're ugly looking designs anyway, (just an opinion . . . )! But David holds out some hope; he's more determined and focused than me, so it perhaps is indeed the right course. I believe he and his wife are into cruising out-of-the-way antique shops, (must've far too much free time on their hands ), so who knows! Meanwhile I've a friend who does some amazing things with cabinetry - I think I'll just have a word with him about making something that "looks the part" - those Chieftain pictures will come in useful yet again!
I may yet run through the RF alignment, with a little help. Another very helpful and supportive friend, well known RMIT senior lecturer in electronics Rod Humphris, has sort-of offered to help - I think he wouldn't mind a "bit of a fiddle", recalling many years work with valve gear himself.
It never ends, does it! I'll keep you posted!
Update, September, 2002
Well, well, well! Thanks to restorer and enthusiast Colin Rutter, of Boronia in Melbourne, I know know what the cabinet that my CL model chassis may have gone into might have looked something like, for the first time. Colin purchased this unit locally, and learnt that it had originally been purchased at Myers in 1954. Amazingly, he recognised the chassis as being apparently the same as he'd seen a considerable time before, in this article here. It's very, very close but it's not quite the same. A later edition or variant perhaps? This model was in production for some few years, and may well have appeared with chassis structural variants . . . Anyway, here are some pix of Colin's unit. Note the connector variations at the rear of the chassis (lower right picture), and the dial-scale, which is substantially different in marking though not in overall shape.