Philips 834a 'Super Inductance' LW MW 1933.

Bakelite fronted, steel framed with a sort of painted card cabinet that gives it a unique style, kinda like either a double bass guitar or a cathedral. The card reduces the weight of the set although it is still quite large at about 17 inches tall. Coming up for sale at silly prices, similar to the round Ekco's I found this one badly rusted and missing knobs, speaker and wooden base and sold as two different items; the cabinet and the chassis. The cabinet was a buy now price and the chassis was bidding only, strange as cabinet is the thing of beauty, although cleaned up the chassis shine too as we'll see. The Super Inductance tag basically means extra care in amplification of the stages has been done instead of using an IF frequency or TRF method, the use of screening with brass a copper is all over. The volume is controlled by varying the the gain of the first stage, this also removes the need for AGC for low signals, you just turn them up.

First off I'll show you the sellers photos and you can judge for yourselves, although nobody bid against me, I think most people were put off.

Two pictures before I started

The first thing I must say the seller did a wonderful job (after I messaged him) of packing the radio, putting plastic milk cartons around for extra protection from the outside kicking in, also he wrapped all the valves in cotton wool, packed the chassis inside the cabinet on a child's pillow to stop it pushing against the Bakelite, I was well impressed.
The cabinet had dozens of spots on it that I feared was woodworm, but it just turned out to be paint spots that I removed easily with a credit card edge.
The Repair
Where does one start?  First I checked all the valves for continuous heater windings, all were good. One valve had it's top cap broken off and another had been pulled from it's base. I used a glass cutting stone drill to remove some of the glass and reveal the connecting wire again. Once done I carefully soldered 1 braid of copper wire to it and to the screw cap and glued it back to the top of the E162 RF amp valve, a mega rare RF pentode saved. The other RF valve (E455) equally are rare to find, I also repaired by solder sucking all the pins of it base and carefully re-inserting the 6 wire back in and re-soldering them, another valve saved.

Left: shows where I etched the glass back and added a hair thin wire Above show where it was attached to top cap again, after I glued the cap back

Next I remove the rotted main smother capacitors and fitted modern 22uF in place. I had to saw off the rusted bracket and make some new fixings for the choke.

Photo left shows 2 new capacitors where old smoothers were fitted. Some restorers like to fit them inside the old ones, but here they were too badly rotted.  Next I checked the continuity of the mains transformer, all winding were good, so next I powered it up to see if I had any HT and heaters. All valves lit up, but no HT. The double diode rectifier valve B5 a 1823 was not conducting. This valve is impossible to find, one for sale at £250, so I decided to fit 2 x silicon diodes (BY223) I've used these before and they work well, but BY127 or similar can be used, but with little forward resistance now the HT will be higher, so a dropper resistor needs to be added in series with the load. I used a 5 watt 4k ohm. This brings HT to around 150v, lower that the original HT, but will be less stressful on the ageing components in the set. I disconnected the heater 4 volt voltage from the valve and HT wires from cathode and anode, but left the valve in place (looks pretty being a balloon type valve).



Circuit I've drawn here shows the new wiring of the silicon diodes and resistor that replaced the 1823 double rectifier valve. Take care to make wiring secure and safe for future repairers, that a must, might be me. I've shown the wiring of the diodes as this can be confusing and easy to get the wiring back-to-front. The negative from the mains transformer goes through a series of resistors before reaching earth or chassis, this creates a negative bias voltage for valve control grids. In this set the volume control is tapped at about 150 ohms, this makes it hard to find a modern equivalent, so extra care is needed to restore this component.

Next I cleaned the pins and sockets on all the valves and holders. Power was applied and noises from the speaker that I added along with a LS transformer was heard. With aerial applied no signal was happening. Volume control and on/off switch were seized-up, so I removed the control and spend nearly a day trying to restore the unit. Spindle was complacently jammed and had to be hammered free with heat from iron and oil. Most the time was spent trying to refit the spring to the on/off switch, but try as I did. I had to give up with that and decided to fit a micro switch on the outside of the chassis. Here is a few images to help if you have to dismember a similar one.

on-off volume and power switch in situ	stripped down showing internal rust
complicated mechanism for the switch- bar slides	I cleaned the contact and track OK
new switch fitted outside using a bolt to activate	chassis was de-rusted and treated with silver paint

After on on/off rebuild I rust treated the chassis giving it a nice shiny look. Inside I checked most resistors and capacitors for values. No layout is provided on line, a few other repairers had uploaded some component layouts, but I found they didn't always match mine, but were partly useful. Now the volume control was restored I checked signals with my oscilloscope to see where they had gone missing. To my great relief all valves were working and I had signal right up to the insterstage transformer between valves B2+B3 (S13 and 14) on diagram above. I checked the windings and one was open circuit.

It's a small copped shielded coil looking more like a bell with a centre nut to give access, I removed it and with a magnifying glass I could see the one wire had rotted, but left about 2 mm pocking from a hole in the Bakelite former.  I only had one chance of retrieving this, so I carefully scraped the insulating paint from it and attached a thin piece of were to the original contact.

inside interstage transformer 4 wires (2 in/2 out)

tiny hole where wire comes out-joined back

The resistance of the transformer winding's are now identical, showing that this is just an isolating unit rather than a step-up transformer ie, 1:1 ratio. After res assembly the set finally started to work, the the volume control didn't do anything, max volume all time. I spent quite some time checking voltages as this design is a little unusual, it uses negative voltage applied to the control grids of the sound output and first RF amp valves. I found the only way to get it to work was to reduce the 2.2 meg feed resistor (R8) to 56k, quite a major change, but it worked fine then.

Next I cleaned the cabinet with soap followed by some 3-in-1 lubrication oil, this brings back the shine to the Bakelite and the painted wood sides. The next task was to make a new base for the set as the original one had rotted and gone missing with the last owner. I found some photos of what the stand looked like and decided that the identical base would be a little hard so I found a picture frame in a charity shop for a pound that was around the same size as the base and attached some wood with similar internal steps to hold the wire frame and chassis to the cabinet. The wire frame had some holes rusted in it, so I treated and filled the holes and sanded and re painted, frame only shows at rear of set, most of it is bolted to the insides.

I painted the card surround black to make it look tidy and cleaned inside and rust treated frame

I used a speaker I had to hand and found some cloth to cover front. Brackets were treated and a new TX added, old one open circuit.

Above I've added some more useful information, if you need PDF circuit in full email me, but most is shown here. Picture bellow shows 'A' being the Radio/PU switch via a plug and 'B' where the aerial should be inserted. I've noticed that an in build room aerial could be purchased for an extra cost and would have gone in the other sockets provided, and fitted inside the top arch of the set where holes are fitted in the back.

I forgot to mention the band switch was jammed. It's a rod, same spindle as tuning knob that runs the whole depth of the cabinet and is 'push-pull' operation. It even has a hole cut in the rear of the set to move back, if stiff you can push and oil from there. The band switch is quite simple just 3 sets of 3 contacts that are large and easy to clean. I read one guys account of somebody sawing this rod in half, so take care in restoring that. It can be seen below, just to the left of the manufacturers label. This is a very dangerous set to work on, most of the internal wires are 18 S.W.G gauge wire without insulation and the voltage selector is in open air space. To protect the public Philips mounted the mains plug as part of the rear cover, so when removed the power is also disconnected.

The back had developed a curvature over the years and needed to be flattened, i did that by making the back damp and heating it with a hair dryer, so the shape could be re bent.

Picture frame I cut in half that I found in a charity shop and used as the outer of the radio's base.

Model makers drill kit used to remove lose rust and then painted with anti-rust paint

The dial mechanism was oiled and cleaned up nicely, I found 2 knobs in my spares box and the 6.3 volt dial light was working? good job as it was rather a large un-common type. In the end I was quite surprised how quick this restoration went, under a week and quite enjoyed doing it, the clean inside will never be seen but I know it's presentable and that's all that matters.

There is a video of this working on my Facebook page :- https://www.facebook.com/100001396195677/videos/pcb.643129939948984/3352980251425148/

Update

A couple of years on from the above repair, she died on me, and for a while had been slow to warm up and and with reduced volume each time I turned on. So I decided to bring it back into the work bench and strip down. I extended the loudspeaker cable to detach cabinet from chassis. With the oscilloscope I found the pre-amp audio valve (E499) a 5 pin triode was not working, all the bias and coupling capacitors checked, I decided the valve must be duff. The valve dates from 1929 and as with all early triodes the hi-fi builders buy-up all the old stock and make it a rare item to find (£80-250) so as a last result I converted the old valve body to hold a ECC83 with its 9 pin holder also fitting inside. The only problem is the valve needs 6.3 volt heater, but because the gain is 3 x more that original valve, reducing the heater to the chassis 4 volts brings the gain in line with the need. This way at a later date the E499 valve can still be fitted.

Arrow showing ecc83 replacing tired 100 year old triode