In my collection comprising thousands of germanium transistors, I have a few branded "Impex". Searching the Web reveals that I am not the only person having difficulty finding out about this name. It turns out to be rather mysterious. A UK company has existed since 1939 currently called "Philips Impex Limited", with its head office in "Philips Centre, Guildford". However, that Companies House entry shows that this name is relatively recent, and the company was called something quite different in the era of germanium transistors.

On the UK Vintage Radio forum someone has suggested, quite plausibly, that "Impex" is a portmanteau for "import-export". A web search reveals many modern companies using the term.

The Radiomuseum site has an entry for a UK company called "Dario-Impex Electrical", but this appears to be unrelated. There is potential for confusion here, as Dario-Impex manufactured radios branded "Impex" but not containing Impex-branded transistors.

Impex-Mullard numbers

The Vintage Radio thread shows a very interesting label found inside a Perdio "Spinney" radio. It shows that some Impex part numbers can be related to directly equivalent Philips/Mullard parts. Rather interestingly, replacement by Mullard parts is recommended! In the same thread, a correspondent states that his Impex transistors have "Mullard markings that have been mostly rubbed off".

Searching for valves/tubes branded Impex only adds to the puzzle. The so-called National Valve Museum, which is a UK virtual museum site run by Allan Wyatt, shows two valves marked "Impex Radio". However both are apparently of American manufacture!

If you can explain any of this, please


S45T transistor

Here is one of the aforementioned Impex S45T transistors in a JEDEC standard TO-1 metal can with an orange paint spot on the top. According to the above label, that is equivalent to a Philips/Mullard OC45. However, it does not use the standard Philips/Mullard SO-2 glass tube, so strictly is the same as an OC45M.

The label in the radio shows that a type S44T also exists and the Vintage Radio thread shows it: again it's a TO-1 metal can. I would like to obtain some. I would also be very interested to hear about any other Impex part numbers.


S11T transistor

Here is an Impex transistor in a Philips/Mullard standard SO-2 metal over glass encapsulation, poorly printed S11T. This cannot be a re-badged OC11 as that was a rare early type in a plastic cuboid outline. I have no information about S11T at all, so if you can provide data about it, please


S18T transistor

Here are two Impex devices, marked S18T (composite image shown) and S19T and using the early GEC encapsulation of a copper can with a plastic sleeve. This shape was continued briefly by Mullard/Philips after it absorbed GEC's semiconductor business. I hazard that the equivalent GEC part numbers are GET118 and GET119 as there were no types GET18 or GET19. Intriguingly, GET118 and GET119 are quite obscure in their own right, and I have no solid data for them although I do have examples, which are branded Mullard. Suspiciously, my GET118 is the only GET-series type I have that uses a white sleeve, just like my S18T. Of course, this supposition could be completely wrong.

S19T transistor
S15T transistor

Well, here's a type S15T which does not fit that numbering theory - GET115 is a medium-power transistor in a black metal heatsink.

If you can provide data about Impex transistors or know where I can obtain more types, please


S30T transistor

Impex transistors can be found marked S30T and S31T in the standard Philips/Mullard SO-2 metal over glass tube. My image shows the S30T. However, these are certainly not equivalent to OC30 and OC31, as the first of these is a power type and the second does not exist!

I am keen to find examples of S31T. If you can tell me where I can obtain some, please


S3R diodes

I am indebted to Chris from the excellent Silicon Ark site for these two Impex diodes. They are printed, rather poorly, S3R, but one is also faintly printed OA79, the other OA81, and indeed they have the same encapsulation as those Philips/Mullard types. Chris has tested a number of these diodes, and the results are puzzling to say the least. The forward voltage at only 5mA varies between about 0.6V and 1.0V, which is far too high for germanium. Also, some of them exhibit Zener breakdown at only 10V reverse voltage. These are bizarre characteristics, totally wrong for OA79/OA81.


All of this uncertainty leads me to suspect that the name Impex has no direct connection with Philips/Mullard. It is possibly a brand that someone applied to rejected devices in order to resell them. However, that raises the question of why Perdio used them in their radios. I have no definite information. If you can help resolve this, please