I have been in email discussions with a few people recently about the ethics of carrying out modifications to sets. Clearly it is an area where there will never be complete agreement. On this page I will try to discuss the options and opinions. Comments and opinions are welcome.

Why modify a set?

There seems to be four main reasons why one would consider modifying a set from the original design:


Many valve radios could not comply with current electrical safety standards. Although this does not make them unsafe, it does cause problems when selling them as working electrical appliances.

Some sets are unsafe. They may have exposed voltage selectors, mains connectors which allow the pins to be touched when partly mated, high voltages on external connectors, etc.

My view here is that anything that is clearly dangerous must be sorted out, if you are actually going to use the set rather than have it sitting silent on the shelf.

Generally the modifications required would be fairly minor in nature. An external voltage selector can be disconnected and the connections permanently set to an appropriate setting (230V in the UK). A dangerous mains connector can be removed, a piece of plate fitted to cover the hole, and the mains cable connected permanently.

Further information about the legal position is given further down this page.


A valve radio will never be as reliable as a transistor set, and short of ripping out the chassis and replacing it with a transistor circuit, we aren't going to make it that reliable. However, some designs of valve set are more unreliable than others, and the main factor seems to be heat.

AC/DC sets having dropper resistors and small cabinets are, generally, less reliable than larger AC only sets - simply due to the localised heat build-up. Some manufacturers spread the heat-producing components around the chassis and add heat deflectors etc. Others group the hot components together and then place heat-sensitive components (like the smoothing capacitor) right next to them! It is not uncommon to find cabinets that have been damaged by the localised heat.

Once such sets have been repaired, and the cabinet damage sorted out, there is clearly a desire to prevent a repeat performance. You may consider replacing the dropper resistor with a capacitive dropper, which dissipates virtually no heat. You may prefer to add a diode into the heater chain, and decrease the dropper resistor to suit. If the set has a large dropper to allow it to operate from a higher supply voltage than originally intended (for example, to operate a 115V set from 230V) you may well decide to convert it back to the intended voltage and operate it using a transformer. You may even consider replacing the valve rectifier with a silicon type. The technical aspects of these ideas are discussed on this page.

Is such a modification a good idea? It all depends on the type of set, its condition and its value.

If the set has previously suffered cabinet damage due to the heat, then even with this repaired it is likely to be worth very little, so I can see no great objections to the types of modification outlined - as long as they are carried out in a technically competent manner. If the cabinet is Bakelite or plastic and has been repaired with adhesive or filler, this repair is less likely to be able to withstand the heat than the original cabinet, so the desire to minimise the heat would be even stronger.

However if the set is in good condition, valuable or collectable, you should think very carefully before carrying out any sort of modification that could affect its value. If you do decide to modify, make sure it can be reversed if required at a later date (to this end, it would be a good idea to keep any set-specific parts that are removed).

Component Availability

At one level it could be argued that virtually anything we do to repair the set is a modification due to component availability. We fit modern capacitors because wax-paper types are no longer available. We fit modern axial electrolytics underneath the chassis because the original cans are no longer available. We may choose to hide them inside the hollowed-out original cases, but they are still there.

Is there anything wrong with this? I don't think so. If the set was sent for repair ten years after it was made, the repairer would have ordered current replacement parts from Radio Spares - he would not have searched around for original ten-year-old parts.

We are doing the same thing, thirty to eighty years after the set was made. The only difference is that we are thinking of the set as a collectable object, whereas in the previous example it was being thought of as an appliance for daily use. We are applying more care and attention than the original repairer would, because we are doing it for the love of it, whereas he was doing it for the money. We have as much time as it takes to restore our sets, the original repairer may have had to repair ten sets each day.

What about replacing unavailable or expensive valves with more readily available types? A couple of people have asked me recently what I think about replacing a UL41 with a UL84. The UL41 is fairly rare - new devices start at £15 for modern imports and considerably more for new-old-stock Mullards. There aren't that many good used ones around, because they were not that reliable and were soon superseded by the UL84. The UL84 is a much more common device. New devices start at £5, and there are plenty of good used examples to be had from scrap sets etc.

Electrically and physically it would not be a difficult job. The valve socket would need to be changed, but it is the same size as the original so it would fit the hole in the chassis OK. The valve is around the same size too, so it wouldn't look out of place. The heater rating is the same. The grid bias voltage is a bit different, but changing the cathode bias resistor would correct this. The UL84 will drive a lower impedance anode load and deliver more power, but I would not expect any problems using it with the original output transformer.

So it could be done. But should it be done?

I feel the pros and cons are very much the same as those in the Reliability section above. If it's a valuable or collectable set, think carefully. But if the set is basically a junker anyway, then the UL41 is probably worth more than the set, so it makes sense to modify it to use something more readily available and cheaper. My Ekco U245 is currently missing the UL41 because I needed it for another set. It will be a good candidate to try out this modification sometime, because the rest of the set is in fairly poor condition and is worth very little.

With older sets having octal-based valves, valve substitutions can be carried out more easily. If you have the old valve, remove the glass top from the base section, and use the base section as the basis of an adapter, with a valve holder to suit the replacement valve. You may be able to mount any components required for the modification in the "adapter", so that the whole lot could be unplugged and the original type of valve fitted easily at a later date. This sort of modification is easily reversible.


This is the area where things become more grey. Many sets could be made to sound better with some fairly simple modifications. But why should this be necessary?

1st Reason: Radio transmissions have changed somewhat since the set was made. MW stations use a lot more compression to make them sound louder and to help them cut through the background noise and interference. There are more stations packed into the band, so the risk of interference from strong neighbouring channels is greater. The audio frequency range is wider, and although AM cannot carry anything above about 5 kHz, broadcasters get as close to the limits as possible. On VHF too, compression is used much more, especially on the pop music stations like Radio 1 and the locals. Again the audio frequency response transmitted is wider than in the past, and we have the extra information relating to stereo and RDS riding on the top.

2nd Reason: The content of radio programmes has changed. Music in particular has changed considerably over the past 40 years (whether for the better or worse is a matter of opinion). A valve radio playing music from the 30s, 40s and 50s (such as some of the items played by Desmond Carrington on Radio 2 on Sunday afternoons or Saga Radio most of the weekend) sounds great. But the same radio on Radio 1 doesn't sound so good.

3rd Reason: Peoples expectations of good quality reproduction has changed. Many cheaper valve radios had no negative feedback to reduce distortion, and most listeners weren't that bothered about the shrillness bought about by third-harmonic distortion. The tone control (if fitted) reduced the effect, and there wasn't so much high frequency content in the music anyway. Wireless sets had a "mellow" tone, which by today's standards would sound muddy or muffled.

4th Reason: Sets may not have been designed to give the best possible quality, either intentionally or accidentally. The first VHF sets were designed and sold before VHF transmissions were available to listen to, so it would be understandable if these sets were not as good as later models. Many manufacturers used roughly the same circuit in a range of models. They had to do something to make the better sets in the range sound better than the cheaper ones, such as better quality speakers and refinements to the audio stages. The cheaper sets had smaller speakers and the manufacturers left out the niceties and produced a basic set. We may wonder whether it really saved them much by omitting a few resistors and capacitors, and in practice it probably didn't, but there still had to be the distinction.

So now we have some idea of why we might want to modify a set to make it sound better, but what can we do?

A simple modification to make a set sound better is to remove the cathode bypass capacitor on the output valve. It adds some negative feedback to the stage, which reduces distortion and improves the frequency response. It also reduces the gain, but with a good set of valves that's unlikely to be a significant problem. It's simple, reversible, and generally effective. It's a modification that I do fairly regularly, and don't have any problems with.

Then we can do further modifications to the audio stage, by adding or altering the negative feedback or tome correction components. A bit more subjective, but again generally easily reversed, so I would say they are not too objectionable. This is what I have done to my Philips B2G81U, and it sounds much better for it.

We can add damping to the IF stages, to flatten the response. Generally resistors across the primary and/or secondary of the IF transformers will have the desired effect. This reduces distortion on more compressed broadcasts, but maybe at the expense of a bit more interference from adjacent channels. Again it's easily reversed and if you listen to a station that is helped with this modification, then it could be worth a try.

The point is that a radio is intended to be listened to. And the better it sounds with the type of radio stations we like, the more likely we are to listen to it rather than leaving it sitting silently on the shelf. Radios need to be allowed to play periodically to keep them in good order, and it should be a pleasure not a chore to listen to them.

But how far do we go with the modifications? We will never make a cheaper set sounds like a more expensive model, and we shouldn't be trying to. But is there any harm in making it give the best it can for what it is?

To modify or not to modify...

Ranulph Poole (who has contributed two Recent Repairs to this website so far) makes the following comments on the subject:

Items of technology have to work. It is almost impossible to restore something to its 'original' state, since many original components (capacitors) may be faulty, and there is little choice but to use modern replacements.

Sometimes the manufacturers, in their rush to get the equipment on the market, 'got it wrong'. Some of the mistakes might be interesting, but are inconvenient if you wish to use the equipment. A case in point is the missing de-emphasis in the Baird (and the set really did sound horrible).

It seems acceptable to me to correct such mistakes, provided the modifications are reversible and documented. Complete refinishing of cabinets is highly undesirable - as I have already admitted. However, if the original finish has been destroyed through careless treatment, one has to do something. One does not want to live with an eyesore.

The final point is that the sets I have dealt with were 'rescued' and would have ended up in the dustbin otherwise. It is not a question of having desecrated valuable antiques. Better modified than dead...

The final point is a very good one. Many of the sets in our collections (or "accumulations" as Jon Evans rather aptly calls them) are not especially valuable and are not in first-class condition. They probably cost only a few pounds, or were rescued from the dustbin or skip. Certainly many of the sets in my collection are in considerably less-than-perfect condition, and were bought simply because I liked them and they were cheap. They are not rare, valuable or highly collectable items, so I am hardly depriving future generations of an original important piece of history.

John Sykes added:

If a set (in extremis) was originally marketed with a gross design flaw, to the extent that it was (in modern parlance) not fit for purpose, then a restoration which left the radio in a similar unusable state has only a bleak, dry and academic attraction. I well recall my father purchasing (c 1956) one of the first FM hi-fi tuners to appear on the market (I cannot remember the name of the manufacturer, but it was not Armstrong, Leak or Quad): it suffered from either a complete lack of or flawed AFC, to his immense frustration (as it constantly drifted off tune). Radios are to use, not to sit in a museum in their original but non-working condition.

All opinions welcomed! The aim here is not to reach a consensus (because I am sure this is impossible), but to allow differing opinions to be stated so that visitors can come to their own conclusions.

Legal Position

I am no legal expert. The following information is my understanding of the legal and regulatory position, but you must check it with an appropriate professional before relying in it.

Trading Standards stipulate that any second-hand item of electrical equipment must be fitted with a correctly wired approved mains plug, fitted with the correct fuse. The item must be inspected and tested for electrical safety with appropriate testing equipment, and fitted with an appropriate label showing that it has passed the test. It must therefore comply with current electrical safety standards.

To achieve this, it is likely that a valve radio set will need to be modified in some way. This is where it gets tricky.

All new electrical goods must be tested for compliance with the Low Voltage Directive and for EMC susceptibility and emission. If it passes all these tests it may be fitted with a CE mark and sold in the EU. The cost, time and paperwork involved in complying with these regulations, and proving compliance, is considerable.

Second-hand refurbished goods do not need to go through this testing, and simply need to be safe and to comply with whatever standards were in place when it was originally sold. However, items that have been modified or adapted do need to comply with current regulations, and require CE marking the same as new goods.

Clearly this is contradictory. To make a valve radio safe would probably require a modification. If we make a modification the item is then suddenly subject to a different set of regulations, which it is impossible to comply with for a single item. So effectively, it is impossible to sell the set as a working electrical appliance.

This is why many antique dealers simply chop off the mains cable (or coil it up inside the set) and sell the set as an antique for display purposes, and not as a working electrical appliance. At the twice-yearly NVCF in Birmingham, all items offered for sale are understood to be "collector's items only" and not "second-hand electrical goods or appliances", unless the seller states otherwise.

But what do you do if you want to sell sets as working items? To the strict letter of the law, you probably can't. However from speaking to an EMC expert at a recent seminar, I obtained the following general advice.

The most important thing is to make sure the set is safe, and has the correct mains plug, fuse and flex. Carry out the minimum amount of modifications to achieve this, but make sure it looks safe and really is safe. Test the set for safety using the appropriate equipment, or satisfy yourself some other way that the set is indeed safe for sale to the general public. Apply an appropriate label stating that it has been tested for electrical safety, and sell it. If Trading Standards are investigating, they are going to be looking for items that are unsafe or untested. They are extremely unlikely to be looking for signs of modifications and insisting on CE marks, so don't worry about this one. However, keep the modifications to a minimum, and make sure any modifications do not affect the set's ability to comply with whatever standards might have been in force at the time it was sold. Keep full records or what work has been done, and why, so that you are in the best position if someone does subsequently complain to Trading Standards.

Remember though, that this advice was from an expert in EMC and electrical safety testing, not a legal expert. If in doubt, seek qualified legal advice.


One final point. If we are selling or otherwise parting with a set that has been modified in any way, the new owner must be made aware of the work that has been done. It's only fair - you would be unhappy if you found a set you had bought had modifications that you were not made aware of. Treat others as you would like to be treated.

It would be a good idea to keep a written record of the work done and any modifications, so that if your relatives need to dispose of your collection when you are no longer around, the full information is available.

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.