Grundig 2035/W/3D Hastings

This "Recent Repair" was kindly contributed by Tim Pullin.


Luxury table set. MW,LW,SW & VHF C.1956. Valve line up: ECC85, ECH81, EF89, EABC80, EL84, EM80, Metal rectifier. Features 4 speaker "3D" system, Bass and Treble controls, Optional remote controls. A top of the range set, manufactured in West Germany for the British market.

The owner left me with two of these for repair, the one pictured here and a scruffier example for spares. Overall condition isn't too bad, with all knobs, back and trim etc complete. Given the complexity of this chassis (see "Other stuff" below) I was glad to see a copy the Grundig service manual accompanying them.

Although not clearly visible in the picture, the top left hand corner of the glass tuning scale is broken (courtesy of Royal Mail) but fortunately the donor set has a perfect scale which can be used.

It's always worth taking the back off and having a quick look inside, just in case anything is immediately obvious. The damage was not confined to the tuning scale; the posties efforts had also broken the internal ferrite rod aerial. Hopefully this too can be replaced from the donor set, so all tests were carried out via the conventional aerial socket. Repairs had been carried out in the last few years and all looked to a reasonable standard. The existing two core mains flex needs to be replaced too, as it is the flat twin type, is damaged in places and has no additional sheathing. This is an AC set so the chassis should be earthed.

The mains comes straight onto the voltage selector panel at the rear of the chassis where a 20mm fuse is used as a link to select the correct mains voltage. (A very neat solution in my opinion. Clever those Germans.) Although normally safely covered by the fibreboard back, these surface mount fuse holders are bare and hence DANGEROUS anyone working on the unprotected chassis. I thought it prudent to insulate the panel while working to avoid mains coming into contact with anything (especially me!) on the bench. Thick cardboard was wrapped around the panel and secured with insulating tape. In similar past situations I have formed a shield from the side of a plastic milk container, which is a good insulator, easy to cut and very flexible.


The usual checks revealed no mains problems (other than the all too commonly found 13 amp fuse in the mains plug - I don't think I have had to buy any 13 amp fuses for the last 10 years!) so the set was run up with the Variac and the voltage gradually increased over about half an hour, during which time the DC output from the metal rectifier was monitored. The smoothing capacitors reformed with no problems. Eventually with all heaters and dial lamp lit the set slowly came up, with no appreciable hum but lots of crackling and no signal on any band.

The crackling was present at every switch position including P.U, which led me to the conclusion it was being generated in the audio stages. The volume control seemed to have a part in this crackle, so was given a good dose of switch cleaner, which improved matters slightly. Another common cause of crackling volume controls is DC on the track (often caused by a leaky capacitor - this is quite possible, as several had already been replaced). Access to this part is not possible with the chassis installed so would have to wait till later. A few tens of volts could be measured with a digital meter at the terminals of the volume control. Hmmmmmmm.

Fortunately this rather large set has an access panel in the base, through which it is possible to access nearly all the components without the need to remove the chassis, which because of the four speaker "3D" system, would be rather involved. S ome valve electrode voltages were incorrect, notably the anode of the EABC80 audio triode, which was about 30V down at around 50V. One of the 200K feed resistors had gone high in value so was renewed.

The 30K anode feed resistor to the triode anode of the MW oscillator (ECH81) proved to be open circuit, giving no voltage at the valve socket. The nearest I had for a replacement was a 33K which seemed OK as the voltage was very near the figure stated in the service data when the set was switched back on.

This time SW worked, picking up stations all across the scale, but wasn't terribly loud. MW and LW however gave just a hiss but no stations, with the ever present crackling.

A check at the grid of the oscillator triode while the set was switched to MW & LW showed a couple of volts negative, suggesting the oscillator was running. I selected MW and tuned to the HF end of the band, where there are usually some stations, and connected the workshop aerial to the sets external aerial socket. Still nothing was received. As the ferrite rod had been damaged I suspected this might have made a difference to the aerial circuits.

A set with faults in the aerial stages will often receive stations when a multimeter lead or similar length of wire is connected to the mixer grid. This old trick worked here too, the set bursting into life whenever the test probe was applied at the mixer grid. Just to eliminate a faulty valve holder contact, I applied the test probe at the other end of the wire connecting the grid to a series capacitor. The MW band worked with the wire connected here too, but when the wire was connected at the other side of the 150pF capacitor and contacts of the wave-change switch nothing happened, suggesting a faulty component.

This capacitor was a polystyrene type (which I have found are usually pretty reliable) and would not have been high on my list of suspects. This one must have gone open or more likely, since SW worked, low in value. For information these are made by Siemens and are red in colour, so look out for them.

Replacement of this capacitor restored MW and LW at good quality and volume, but the performance was not as good as one would have expected for such a posh set, or indeed as good as that with the wire applied straight to the mixer valve, when distant MW stations could be received.

Here in Wiltshire, Virgin AM is quite feint and can be used as a "distant" station. I know if a radio is working really well as it will receive BBC Radio Scotland ( 810kHz, 370m), and I often use this for a distant test. From a quick glance at the circuit diagram it was apparent the ferrite rod aerial played an important part in reception whether or not the external aerial was used. The complete ferrite aerial from the other set was fitted in place of the original, taking care to centre it in it's mounting, to ensure it was able to rotate freely when the "Aerial rotator" knob on the front panel was turned. A few turns of tape were wrapped round the rod to ensure it was a tight fit in it's mounting. The rod was then secured to the mounting using waxed lacing twine.

The broken ferrite obviously affected performance, with reception of MW and LW much improved using both the ferrite alone and an external aerial. All bands were jammed full of stations far and near including some interesting short wave ones.

Right then, now for the VHF. The band worked after the set had been on about 15 minutes but also wasn't very loud. Touching the ECC85 VHF valve brought the band to life instantly, eventually settling down to reasonable sounding VHF (but for the audio stage crackle). The VHF fault may well be down to dirty contacts, dry joints, or possibly a faulty holder. This chassis uses those horrid paxolin valve holders, which is a shame as the set was clearly very expensive. The connections on the chassis side of the holders are liable to touch the chassis if moved very much so one should take care when soldering or taking voltage readings. Due to an awkwardly placed screw, removal of the cover to the VHF tuner head is well nigh impossible with the chassis in the cabinet, but the chassis will have to be removed in any case to swap the tuning scale glass.

I carefully noted the connections to the speaker crossover network at the output transformer. The colour of the connecting wire was marked on the transformers connection panel with a fine permanent marker before removal. (Pencil is OK too but don't forget it's conductive!!)

DON'T FORGET THE WIRES TO THE MAIN SPEAKER. They are only about three inches long and easily overlooked. The small paxolin panel that supports the connection tags of the speaker was broken on my set, probably after someone trying to remove the chassis while the speaker wires were still connected. Fortunately I have a spare! The "magic eye" is held to the speaker panel by a simple spring and easily removed. It was carefully unclipped from its holder and put somewhere safe; they are rare and expensive enough already!

There are so many gadgets and pulleys on this chassis I found the best way to lay it on the bench was on it's back, taking care not to damage the Ferrite rod mains and aerial panels. An aerial can be connected to the back of the socket with a crocodile clip. While on the bench the good, spare main speaker was linked to the set via about 18" of the old mains flex.

Replacement of the original mains cable was not difficult as it only runs to the mains panel at the rear of the chassis. However the single pole mains switch is on the opposite leg of the incoming mains supply to the fuse. I suppose this was done because the mains flex is simple non-colour coded twin, and the mains could be connected either way round. Fuses should always be in the live side , so this means the switch would end up in the neutral. It is not possible to change the switch to double pole due to lack of space (see picture of underside), so a slight modification to the wiring was carried out. The live mains now comes straight to the switch, then the 500mA fuse/selector link and then on to one side of the mains transformer primary. The other side of the primary winding is connected directly to neutral. The earth was soldered directly to a nearby chassis tag.

Next a few likely capacitors in the audio stage were tested and found to be leaky (including the dreaded DC blocker at the grid of the output valve actually measured about 10 M at 250V DC.) so were replaced, along with some that were the same type as others that were leaky, as a precaution. Due to the "3D" system in this model, one would not relish trying to find another output transformer, let alone rewind it lest it burned out as a result of additional anode current flowing caused by the output valves faulty grid capacitor. This photo shows the types of capacitor that were found to be leaky.

Upon retest the DC on the volume control and crackle was gone! Just to make sure, the set was switched to gram, the P.U input sockets shorted and the volume turned up. No crackling was heard for several hours. The set sounded fine on the AM bands too, so now it was time to look at the VHF side.

The cover to the VHF tuner is easily removed, once the chassis is out. As far as I could tell there were no loose wires or dry joints at the ECC85 socket. The anode resistor for the VHF oscillator section was a little high so was replaced. As is common in tuner heads like this, the socket is well earthed with a few of its pins soldered directly to chassis. Seeing as it would be a long job to replace it, I thought it might be worth trying another valve. The replacement ECC85 worked perfectly, with no crackling when it was touched or moved. VHF stations were received with excellent quality at about the right points on the scale.

The valve, holder and wiring were gently tapped with a piece of rubber sleeving stuck over the end of a pencil.

No fault could be reproduced so the tuner cover was replaced and the chassis tested for several hours, during which wavebands were switched often. No further faults showed.

Cabinet and Fittings

The grub screws securing the control knobs came undone with no problems (do be careful not to over tighten when refitting!). Metal clips at each end of the chassis then secure the tuning scale. Once the old glass was removed I took the opportunity to clean the backing plate with a small brush and some hand cleaning wipes, taking care not to dislodge the tuning cursors from the drive cords. The back of the replacement scale was cleaned gently with Baby Wipes and re-fitted along with the knobs. The chassis was then tested again to ensure stations were accurate on the scale, and the chassis refitted into the cabinet.

Note It is possible to fit the cover to the VHF tuner upside down so that the holes do not align with the adjusters beneath. Check this BEFORE the chassis is refitted, as the cover is difficult to remove once the chassis is replaced! You have been warned! The wooden cabinet was in pretty good condition anyway, so was just given a good clean and wax polish.

Other Stuff

The magic eye wasn't very bright when the set was first tested and didn't open fully when tuning. The anode resistor was checked and found OK, operation of the indicator much improved once the AGC decoupling capacitor had been replaced. Somehow you can just tell it's a Grundig´┐Ż..

The original was hardly visible in normal light so was replaced with a used good EM81. Although not brilliant, it was visible in normal daylight and certainly very usable.

This set has a few "luxury" features. In addition to Bass and Treble controls, the ferrite aerial is rotatable via a length of cord connected to one of the front panel knobs.

There are two tuning cursors, the largest one for AM wavebands and the other for VHF. Separate cords drive these from two pulleys around the tuning knob shaft. When the VHF waveband switch is selected the mechanism slides up and down the tuning knob shaft and it's clutch engages with the flywheel (while the AM drive moved away from the flywheel and disengaged) driving the cord for the VHF tuning cursor and gang. MW is the reverse of this process. And all with a single tuning knob! This of course means you can leave the set tuned to your favourite VHF station and still listen to the MW golden oldies. Other sets employ separate tuning cursors for the VHF waveband such as the Philips B3G63A, but that has a (slightly!) less complex mechanism and uses separate concentric controls for each band.

Three of the four speakers are clearly visible in the picture, but the fourth an electrostatic tweeter, is partially hidden behind the ferrite rod swivel mechanism and the aerial panel to the left of the main speaker. Some of the crossover network for these speakers is also visible at the "ten o'clock" position from the main speaker.

Also just visible is the 3-pin din socket used for the Gram or pick up input. There is also an output available from the audio triode via this socket, for your (they hope, Grundig) tape recorder.

The owner wanted to listen to his HI-FI through the set so a lead was made up, utilising the Din input. I used two 10K 0.25W resistors to common the stereo channels together creating a mono input signal.

There is also a remote control available for this set, the Grundig "Majestic Baton". This plugs into an umpteen way socket at the back, it's five "piano" keys (with illuminating icons of course) enable the user to change the tone and 3D sound switching.

Information thanks to Martin Francis Vintage wireless database. This picture shows a set and remote recently offered for sale on eBay.

The set certainly sounds very good. I don't know whether this is due to the extra speakers or a quality design, but would have certainly been the talking point of any room back in 1956.

The "spare" set worked too, and also needed some capacitors changing. It actually had fewer faults than the one I had been asked to repair! Although the dreaded grid capacitor for the EL84 had already been replaced with an old style grey "Radiospares" type (with the little man), that also proved to be leaky so was replaced.

The "spare" chassis sounded good on the bench, and will be assembled until I can find a new tuning scale.

The paxolin panel on the speaker was replaced by a piece of tag strip and the voice coil leads carefully soldered back on. It seems a shame to use it for spares, as it came in handy for testing the CD listening lead, which was made after I had handed back the repaired set.

Text and Photos Copyright © 2003 Tim Pullin.

This website, including all text and images not otherwise credited, is copyright © 1997 - 2006 Paul Stenning.
No part of this website may be reproduced in any form without prior written permission from Paul Stenning.
All details are believed to be accurate, but no liability can be accepted for any errors.
The types of equipment discussed on this website may contain high voltages and/or operate at high temperatures.
Appropriate precautions must always be taken to minimise the risk of accidents.

Last updated 14th April 2006.