Grundig 2035 "Hastings"


 This rather fine looking radio arrived in the workshop in June 2015 for refurbishment. It came with a proper dust cover, manual and original invoice.

See the comprehensive paperwork.

As with many Grundig models the detailing of the controls and trim gives an impression of luxury. Those gold coloured bits are described in the brochure as gilt which really means GOLD plated (although in 1956 "gilt" may not have been entirely accurate as it would need to be in this enlightened age). No wonder the set cost 55 guineas in 1956...

It's claim to fame is it's 3-D sound. Unfortunately for the purchaser the set pre-dates FM stereophonic sound by 2 years, at which point the set was no longer up-to-date. Shortly afterwards VHF range was increased to end at 108Mc/s making the set much less desirable.
 I had a little time so I removed the back from this set and peered inside. I noticed a length of litz wire hooked onto a plastic bracket connected to a ferrite rod behind one of the valves. After puzzling a moment I realised the ferrite rod should have been clipped into the bracket and the litz wire was one of two. The second wire was broken... I resoldered it with the intention of switching on the radio for a quick check. Noting that the "Off" button hadn't been repaired in 1964 (or the fault had re-occurred later) I plugged it into the mains. The valves lit up and crackles and strange noises emerged from the speakers. Pressing the VHF button I briefly heard very distorted voices and noted the tuning mechanism was not very smooth and not a lot happened when the knob was turned. Clearly the set needs to be extracted from the chassis before any further tests are possible. The good news is the mains transformer works and the valves are at least partly functional. Only a single tiny dial lamp was lit (the main one was open circuit) so I imagine valve emissions will not be great.

 Above is a picture from the Net. My guess is the tuning fix will not be easy... Looking in the back of the set reveals more wires and strings than on the Cutty Sark... Click to see the circuit diagram of the similar Model 2033 (except the 2033 has no Short-Wave band and only a single VHF setting). The worrying thing is the proliferation of tuning parts. I counted more than 40 items so if the tuning isn't working (and it isn't) it will take an age to fix. Hopefully hardened grease is the problem (as always).
 Below a view into the back of the set. Not the usual rusty chassis, but typical German receiver design. Lots of thin wires.. in fact so many it looks like the speaker wiring is going to be a pain to disconnect.


The multitude of wires at the audio output transformer is to do with the range of frequencies passed to the loudspeakers.


The other end of the bunch of wires from the audio output transformer terminates at the top of the cabinet where connections are made to the twin side-mounted speakers and tweeter.


 An unusual tuning condenser. There are 21 vanes on the near section (oscillator) and 17 on the far one (RF). The spacing is different so the near section will have less capacitance per vane.

Superhet radios employ an oscillator which tracks the station tuning range by a fixed amount which determines the Intermediate Frequency or "IF" at which most of the gain is established. Many receivers use identical tuning condensers for RF and oscillator but employ a padding condenser to reduce the value of the oscillator tuning condenser hence making this higher in frequency than the received signals. Some sets have tuning condensers accurately designed to dispense with padding condensers. I suppose this minimises the amount of aluminium and saves a few parts?

Here, the spacing difference reflects the voltages used in the circuit. The circuit diagram indicates the oscillator tuner connects to the anode of the frequency changer triode and has a rather strange value of 15-433.9pF. The RF tuner has earthy connections and is marked 12-519.2pF. I can see a padding condenser in the circuit so no savings there.
 Time to remove the chassis. I found it best to cut the speaker leads (main + 2 side + tweeter) and unscrew the termination strip which connects to the output transformer, unclip the magic eye and remove the valve. Remove the four fixing screws and detach the chassis. This includes the dial and you don't have to remove the knobs. Because of lots of bits sticking out in all directions I used a wooden jig plus some blocks under parts of the chassis as you can see below.


 The first job I tackled was the tuning mechanism because, if this had been seriously damaged, the set wouldn't have been repairable. The tuning knob was very stiff in operation and the Long/Medium Wave tuning condenser inoperative. The FM tuning condenser worked but was very rough and slipped.

The set has two tuning mechanisms so that a station can be left tuned in say to Radio 4 on Long Waves and Radio 2 on FM. The selection of the appropriate tuning mechanism is selected by the push buttons. Pressing the Long, Medium or Short button disengages the FM tuning clutch and pressing the FM button diusengages the LMS clutch. There are two tuning pointers and loads of little pulleys. I oiled all the pulleys (taking care not to get oil on the cords) then oiled the gearing on the LMS tuning condenser plus its bearings, also the gearing and bearings on the FM tuning condenser. Tuning was OK but still a bit rough. This turned out to be the postion of the tuning knobs, pressing on the glass dial. There are two secured to the same shaft but both inner and outer serving exactly the same function. Maybe a different model has two independent functions for these knobs?

Resetting the knobs clear of the dial glass sorted the problem and the tuning then responded nicely to the built-in flywheel.

 The end push button is supposed to operate the on-off switch and I noted this had been reported faulty in 1964. There's a standard toggle switch with an odd shaped lever mounted under the chassis on the inner edge of the box containing the LMS tuning and oscillator coils. It's supposed to connect via a lever and at least one spring to the push button, but there's something not quite right. A shaped piece of metal is wedged in place and there was a loose spring but seemingly no way to connect the parts together. My guess is an abandoned attempt at repair when a vital part was removed.

The first electrical problem was weird audio. This was due to an open circuit burnt cathode resistor at the EL84 and current being supplied by a long suffering electrolytic. I fitted a new 180 ohm plus a new 47uF electrolytic.

The incessant crackling present at all settings ogf the push buttons could be reduced by carefully adjusting the two tone controls and I suspect there's some undesired DC present, perhaps from a leaky condenser. I spotted an intermittent positive voltage of about 7 volts on the EL84 grid after changing the 4700 pF coupling condenser and this went away after swapping the valve for a brand new old stock Russian one.

I found a burnt resistor measuring zero ohms in the RF box. This was fed by a 0.1uF condenser which I replaced. The DC voltage that had been present disappeared. This is something to do with tone adjustment? I also changed a 300pF condenser in the tone circuit but that I think was blameless.

Turning the tone control knob below the volume control seems to kill the audio for a few seconds. This can't be right, so I swapped a 1000pF condenser in the circuit but the fault remained.


 Opposite are the initial parts removed. The resistor is open circuit and is the cathode resistor for the EL84. I fitted a 180 ohm replacement.

The set came on with incessant crackling which I think is a combination of problems including dirty valve pins and leaky condensers. Oddly none of these showed up as leaky using a multimeter but I'm sure one or two leaked when high voltages were present.

The large condenser is marked 0.1uF 125 vw and measured 0.17uF.

I fitted a new 0.068uF. The two smaller condensers were marked 1000pF and 4700pF at 500vw. These measured 6.25nF and 28.5nF respectively, both being about six times their marked values.

The electrolytic marked 25uF measured 16uF and was 11 ohms on my ESR meter.

An annoying hum was present I temporarily added a new 22uF 400vw electrolytic across the smoothing condenser. The reservoir voltage was OK so I left this condenser as serviceable. The HT voltage improved from 170 to around 180 volts and the FM band came to life.

 The incessant crackling was driving me round the bend so I decided to track it down methodically. I unplugged the EABC80 and tried a new one. No good, still crackling. Unplugged the new one with a view to refitting the old one and noticed the crackling didn't change. I made up a lead with a 0.22uF capacitor to ground and touched it on the EL84 grid... silence, and then the series grid resistor (R36)... silence again. Then the anode pin of the absent EABC80 (PIN 9). The crackling changed in tone but was still present. I could see two 200kohm resistors (R34 & R35) between the anode of the EABC80 and HT. Their mid-point was decoupled by a condenser (C55) so I disconnected it.. still crackling. I fitted a new 200kohm in each position and still the crackling persisted. The only things I hadn't eliminated were the B9A base for the EABC80 and a tie-off point at the earthy side of R35/live side of C55. The valve socket proved OK so I disconnected the tie-off point and amazingly.... silence apart from a slight hum, so I re-routed things and used a mid-air tie-off point.

What had caused the problem? Well, underneath the tie off point was a piece of loose metal trapped in limbo. Sort of cooped up except it can probably bounce around if the set was thumped. This must have at some time resulted in tracking to ground? The metal is part of the "off" mechanism which according to the paperwork with the set stopped working in 1960.

Below, the bad tie-off point is the pin immediately below the centre of the insulated lead from the yellow condenser and to the left of the pin carrying the red wire.

 Turning on the set with its EABC80 back in place I pressed the button marked "AD". The dial light went to half brightness and my first thought was AD must mean dimmer, however a cloud of smoke caused me to switch off promptly. AD switches in the ferrite rod circuit and lights a small lamp above the ferrite rod rotate knob. The spring supplying 6.3 volts to this lamp was shorting to ground and the insulation covering a short length of wire had melted. I fixed this and continued to look for more problems...

Roughly at this point I decided to make a test jig to hold the chassis. I cut three pieces of aluminium angle and bent them and drilled them to support the set upside down. This was so I could access the various alignment trimmers and coil slugs.


 Left you can see some more parts that had failed.

The large condenser has one section of about 30uF and half an ohm ESR but the other half (used for smoothing was open circuit. I fitted a new 47uF reservoir condenser and a new 220uF smoothing condenser which raised the set's HT lines from 180 to 245 and 220 volts for the EL84 and RF circuits respectively (and removed the hum).

The middle component says it's 4uF but measured 7uF and some ohms ESR whilst the small condenser was decoupling the EL84 anode and had risen in value considerably from its marked 2200pF. If this had failed short-circuit it would have ruined the output transformer.

 Below: The audio circuitry with some new parts fitted. The old reservoir/smoothing condenser used to be fitted above the tag strip holding the blue condenser. The new smoothing condenser is secured to the side of the RF box.


 According to the circuit diagram the IFs are 468 Kc/s and 10.7Mc/s. The coil slugs were secured by wax which I removed using a small blade screwdriver heated by my soldering iron. There's a 468Kc/s reject coil in the RF section which I adjusted after checking the four slugs in the IF transformers, which where pretty close to optimum. When I tried aligning medium waves the set seemed pretty deaf to my signal generator so I'll need to figure out what's wrong. VHF FM signals are OK. I'm suspicious about the AD switch. Maybe it's gone high resistance? The circuitry isn't shown below, so I'll have to trace it.

Below you can see the circuit diagram, although the RF circuits are those for the Model 2033 which is the version without the shortwave band. The 2035 has an extra push button for Short Waves between the MW and UKW buttons. It also has the AD button on the right of UKW for switching the aerial input from the rear socket to the directional ferrite rod.


 Commenting on the first page of the circuit diagram above and the third page you'll see that Grundig have simplified (and no doubt reduced the potential performance of the receiver) by not fitting padding condensers to the long waveband and the short waveband. Instead of tracking the dial readings they merely peak the performance, for example on long waves, at 175KHz.

After testing the set using a long wire aerial and finding it pretty good, I tried it on the internal aerial and found on medium waves it was awful. This sort of lines up with what I found when I used the signal generator during alignment. I'd expect to hear a few microvolts but the best I could do was around 100 microvolts. The set lined up fine as far as its local oscillator is concerned, but RF-wise I just couldn't get it to track. The circuit is not straightforward because you need to refer to (and understand) the push-button arrangements. This shows on medium waves that the ferrite rod coil is in series with the RF coil and is tuned by a 3-30pF beehive trimmer, whilst on long waves the medium wave RF coil and ferrite rod coil tap into a tuned circuit peaked at 175KHz. This means that as far as medium waves is concerned the ferrite rod coil needs to have an accurate inductance in order for medium wave RF tuning to track properly. Unfortunately the ferrite rod coil was a bit of a mess so I'll need to carry out some experiments to get its number of turns number right. I could maybe do this by detaching the ferrite rod coil and substituting for this a variable inductance. I can then use this to get the tracking right, measure the final inductance, and then set the ferrite rod coil to this value.

 Detaching the ferrite rod and measuring the inductance of the coil revealed it was around 70uH. This inductance adds to that of the coil in the RF box and together (with a 3-30pF trimmer) they tune from about 1.5MHz to 550KHz. To get them to tune correctly I had to add a fixed condenser of 30pF across the trimmer whist fully inserting the tuning slug, so clearly the 70uH ferrite rod coil was too low in value. The old coil had broken away from its connections long ago and lots of its cotton covering over litz wire had been damaged so I removed it and substituted a length of 28SWG enamelled wire. Not ideal but much easier to work with. I wound about 60 inches and found it measured about 240uH. This failed to tune so I removed turns until the inductance had dropped to 169, 147 and finally 136uH at which point I was able to align the RF circuit across the medium waveband. I was now seeing a good response to test signals of 20uV.


 Above, the cabinet cleaned of 50 odd years of dust. There are 3 loudspeakers and a tweeter. The white cable is the built-in FM aerial. Once the set had been refitted into its case I finalised alignment and secured the lower and rear panels. The set uses a 2-conductor mains cable which was in good order. Long and medium reception on the internal ferriterod aerial was OK but better results, especially for shortwaves will be gained by using an external wire aerial. FM results are really good and Classic FM nicely positioned at the highest frequency on the dial. Alas the EM80 tuning indicator has long since lost its brightness so will have to remain so unless a replacement at a sensible price pops up...

The AD switch is sticky and is certainly not replaceable so the lamp which operates in the space over the RH knobs will need to be used to indicate whether the switch has operated properly. It should improve with use. No mains switch repair is possible so the set will need to be turned off at its 13-amp socket.

Return to entrance