Exray's Radio LL Super Baby Restoration

First order of business is to explain what is a Radio LL and what is its significance.

Lucien Levy (LL) was a Frenchman who has been at times credited with the invention of the Superheterodyne circuit along with the American inventor Edwin H. Armstrong.  I'll not enter the fray claiming of who was first.  There's a good article by Alan Douglas that gives the details.

Radio LL remained a major French manufacturer - and broadcaster - all through the history of radio.  Some 255 models are listed in the Grand Livre from the earliest days of radio right into the transistor era of the 1960s and thats only scraping the surface.

My set is a Super Baby, Model B, Type 3351, Series 14, s/n 29.  I understand it to be from 1926 or 1927.

Many European countries charged a tax based on the number of tubes.  Here is the tax stamp found on this set.

It is a 6-tube battery set.  I'm not sure of the exact tube lineup but it would appear to be an A441 tetrode for the oscillator-mixer - Philips calls this a Bigrille or Dubbelroosterlamp - four A410 for the IF amps and detector and a B406 for audio output.  The IF is 60kc and it uses 4 plug-in transformers for the IF.  On the front panel are two plug in coils for band selection, two very nice vernier tuning controls HETER and ACCORD - basically osc tuning and antenna tuning - ALLUMAGE which is the filament rheostat (all tubes are controlled) and RENFORCEMENT which serves as a bias potentiometer for the three IF amplifier tubes.

Antenna connections are on the lower front and provisions are made for it to be used with a loop or random wire.

The front panel is heavy aluminum with a swirled finish in the aircraft style.  Chassis is wooden.  The case is wooden with a faux covering

*   *   *   *
Replacement tubes

The first problem encountered was bad tubes.  All 6 of them.  Two had open filaments and the others had virtually zero emission.  A bad way to start the project.  There's a number of  equivalents for these tubes but they are equally difficult and expensive to locate.  In those days different manufacturers used their own numbers for the same tubes.  I made a few half-hearted bids on ebay.de and decided I best embark on a new plan.

That plan was to use sub-mini replacements (5676 and 5672) as are sometimes used to replace WD-11/12 and 99s in American sets.  I did not want to destroy the original tubes to salvage the bases so I fabricated my own.  The original tubes use a thin banana-pin pinout and the socket itself simply has hollow tubes with no gripping mechanism for the contacts.  I made my base out of 1" ID phenolic tubing and fashioned a disc for the bottom out of plexiglas.  The actual pins are a miniature banana pin found thru Jameco.  Expensive little devils!  The glass is made from 25mm test tubes (shooter tubes!) cut to size with a Dremel tool using a diamond cutting disc.  I tinted the glass with a solution from a craft store.  The glass fits the tubing perfectly and is held in place with silicon sealant so they are easily re-enterable.

(this base is from another project)

Will they work?  I sure hope so.  I'm not there yet.

A note on the tube sockets.  The same sockets are used for the tubes as well as the IF transformers.  These use a piece of brass tubing peened out on the bottom to make contact with the lugs.  I had three that had lost electrical contact due to oxidation and one other that read a high resistance.  I had to remove them and clean and solder the 'riveted' connection.  I'm not sure what material these are made from - its not Bakelite...and the IF coils use the same material.  They will crack with little or no provocation.

*    *    *    *
The Transformers and coils.

It was clear I had some problems to deal with regarding the IF xfmrs.  Two of them rattled inside and there was no continuity on one of the windings.  The screws apparently tend to work loose inside of these things for lack of a lockwasher having been utilized.
Fortunately, the 1st IF was intact although it too rattled.  Good thing because its a fairly complicated one.

What you see here TWO individual coils in parallel making up both the primary and secondary side.  Wire gauge is in the mid-40s and separation between the two pairs is 7.5mm.  On the primary I measured 9.0mH with 56 ohms DC resistance.  The secondary measured 9.7mH and 63 ohms DCR.  (or vice versa?) I made no attempt to separate the two coils on either side.  You can see the slight difference in side in the photo so it would seem the slight difference in measurements is intentional.
The caps utilized in this xfmr appear to be 540pf (marked 0,54)  Crunching the numbers suggests that this gives a slight stagger tuning of about 2.5 kc.  One of  my caps is defective.
Note that the wires lead up to the 'hot' side of the individual caps but they share a common 'cold' side by virtue of the mounting screw which is connected to the secondary cold side.  I'm not really sure why this was done.

Here's the #2, 3, 4 transformers

And here's a pic of the coils, primary is inside of the secondary.

Again, very fine mid-40s gauge wire.  The inner primary was open on the first one and testing revaled it was open in more than one place.  I was able to snag a wire at the very inner and outer extremes and tacked on new leads with a strand of #46 from a litz wire.  I might add that this repair took about 2-3 hours and doesn't qualify as 'radio fun'.

The specs on the xfmrs is as follows.  Primary ~8.7mH, untuned, 100 ohms DCR.  Secondary 42mh tuned with a 160pf cap in the top, 245 ohms.  This gives resonance at ~60 kc.  My three calculated to be at 60.6, 60.9 and 61.8 when measured out of circuit.  The caps (marked 0,16) ranged 66, 98 and 142pf.  Safe to say that these caps have not held up well over the years and I will fudge in the additional capacitance with some ceramic discs hidden underneath the original assembly.

In comparison, the front panel plug in coils are a piece of cake.  They are wound with a good size wire - 22 or 24 gauge - and relatively simple.  Here's the BCB RF coil (120-500 meters).  It too had come loose internally but no damage occured thanks to the large wire.  It measures 67uH which corresponds with the specified tuning range using the 70-1100pf tuning cap.  I also have another RF coil missing its cover,  it measures 3.2mH which suggests ~85 to 335 kc or maybe it was originally listed as 1000-3500 meters.  Note that although the variable capacitors both appear to be two-gang units, the gangs on both are connected in parallel.

The oscillator (HETRO) coil is wound with the same large wire.  It is marked as 180-520 meters.  65uH and 296 uH respectively on the windings.

Well, that was a fun exercise.  Now we know whats inside of them and the repairs have been made.  One less thing to troubleshoot at fire-up time.

*    *    *    *

*    *    *    *

It Works!!!

This evening I put it all together for the smoke testMuch to my surprise it fired right up!!!   I used 90 volts as the B+ voltage and 67.5 on the oscillator.  I found the oscillator wasn't real happy with that voltage and increased it to 90.  Actually the circuit does well if the B+ is at 67.5 but the oscillator definitely wants 90.  The C-minus for the output tube seemed best at -3 to -3.5 volts and when it was all trimmed out and playing a filament voltage of around 0.8 volts at the tubes seemed to be the ideal setting.  Audio quality is not remarkable.

I made a quickie tuning chart for the oscillator and it tunes from about 600kc to 1750kc so that gives full coverage from 540 to 1690 kc.

Operationally the set is very well behaved and not at all difficult to tune.  The verniers are some of the nicest I've ever used.  Of course there's the image issue with the 60 kc IF but the rf input tuning is quite sharp even with my random wire antenna and does a fair job of image rejection - noticeably better than my homebrew Ultradyne.

After about an hour of operation it died.  Something was pulling down the B+.  I disconnected it and re-connected again and it ran for a few minutes more.  I looked under the chassis and lo and behold the 2mf paper cap had gotten hot and swollen up.  Duh, I should have restuffed those caps to start with.  There's only two, the 2mf and a 0.5mf and that situation is now corrected.

In summary, I'll say that Mr. Levy did a good job with this set.  I don't know how it would compare using the original tubes but I would suspect that they weren't as hot gain wise as the modern sub-minis.  It receives well, comparable to any typical 5-tube TRF set with the obvious benefit of having selectivity at the top end of the band.

-Bill  Meacham, 10 May 2006