Life span of a magic eye

Jacob Roschy Ernst Erb Martin Renz Bernhard Nagel 
Cliquez sur le bouton d'information bleu pour en apprendre plus sur cette page.
Forum » Tubes (Lampes) - Semi-Conducteurs » Autres lampes » Life span of a magic eye
Marco Gilardetti
Marco Gilardetti
I  Articles: 149
Schémas: 209
Images: 119
05.Nov.07 15:22

Remerciements: 10
Répondre  |  Vous n'êtes pas connecté, mais en mode visiteur   1

Gentle fellows, few years ago I purchased some russian magic eyes (I don't recall the code right now, but it's not important) as generic substitutes for 6E5 type tubes. As those tubes looked brand new, I then stocked a few.

The curious thing (at least to me) is that these are the first magic eyes of which I could observe the entire life span, from the box of the factory to the extinction of glowness.

I must say that I am pretty disappointed with the life of these tubes. The one which has been more in activity has now 2 years of operation, most likely 1 hour per day of usage - say 1000 hours. Its fluorescence is already so weak that it almost needs to be replaced. I figured that magic eyes were supposed to last much more that this, most probably because my family owned devices equipped with smaller units (DM70, tipically) which are still the ones which came from the factory, and they seem to work as they were new.

I would then like to hear from other technicians (more experienced than me, an easy-to-comply requirement...)  if 1000 hours is a standard life for a 6E5-type tube, or perhaps these russian substitutes are known to be poor and thus the original 6E5 tube would be a better purchase.

Are there magic eyes which are known to last very long, and other known to exhaust very rapidly?

Is there any electronic strategy to prolong the life of a brand-new magic eye? (aside from leaving it unconnected, I mean... ;-)

Emilio Ciardiello
Emilio Ciardiello
I  Articles: 530
Schémas: 165
Images: 790
09.Nov.07 19:06

Remerciements: 8
Répondre  |  Vous n'êtes pas connecté, mais en mode visiteur   2 Dear Marco,

I also had noted the short life of magic eyes, with some exceptions. In old radio sets round shaped magic eyes are often worn-out, giving no luminescence at all. On the contrary, light bar tuning indicators, such as EM84 or EM87, even after years and years of daily use, still retain some luminescence. I have no experience at all in the physical and chemical processes used inside magic eyes; nevertheless I devised an idea to explain the reasons for the different behavior of different tuning indicators.

In many and many indicators, as 6E5, EM4, EM34, EM71 or even EM80, the electro-luminescent layer is deposited directly on the anode surface. This implies that all the electrons emitted strike into the phosphor surface with considerably high speed. As result, the phosphor layer is slowly removed. Looking at an electron-ray tube out of its socket, I noted that in new tubes the target appears to have greenish tones, while it is metal gray in aged tubes, the luminescent layer appearing to have been removed.

The light bar tuning indicators, EM84 or EM87, work in a quite different way, much like a CRT. The anode, still called target, has a rectangular shape, all around the viewing area, while the luminescent layer, the so-called phosphor, is deposited on the internal surface of the glass envelope. The target current is in the order of one milliampere, as in the circular types, but these electrons do not reach the phosphor surface. Just a fraction of the electrons emitted by the cathode strike into the phosphor at considerably lower speed than in the circular types, because the acceleration ceases or changes direction once out of the anode window. This could explain the longer life typical of bar-type magic eyes.

I hope that some reader with experience can give his opinion on this idea.

Just few words to remember that the short life of magic eyes was well known, since some manufacturers, as Graetz, had introduced a switch, Mag-Auge, to shut-down the tube once the desired station was tuned in.
Michael Watterson
IRL  Articles: 969
Schémas: 650
Images: 2497
29.Nov.11 03:00

Remerciements: 12
Répondre  |  Vous n'êtes pas connecté, mais en mode visiteur   3

I think Emilio Ciardiello is entirely correct. I had read somewhere years ago that Magic Eye life was less than 2000 hrs.   I also believe that the side view types with coating on glass last longer, though I'm not so convinced of Emilio's explanation. But he could be right.

I think also the DM70/DM71 is different in operation from most magic eyes and it's a very late design as are EM84, EM87. I took a risk and bought five ex-equipment DM70 and they are suprisingly bright running off 6 x PP3 batteries to test. 

I was thinking a sensor connected to Tuning knob with a deactivation delay when the knob is not in use would have dramatically extended life. You do only need it while tuning.

I have also read though of people "stepping up" the voltage on the end display types that seem to fade more rapidly to 300, 400V, 600V or even 1000V. (Doublers and Triplers etc off the AC feed to main rectifier.

Certainly the VFD which started with DM160 (not a magic eye) is still prone to fading with time though seem longer lasting than Magic Eyes. Uneven wear is an issue with VFD as they are discrete on/off patches so the bottoms of bargraphs become dimmer than the tops. But you see the same on some Magic Eyes. If at fully closed it's not brighter than the part always on then emmission could be dropping rather than coating wearing out.


Rüdiger Walz
Rüdiger Walz
D  Articles: 664
Schémas: 142
Images: 197
29.Nov.11 23:19

Remerciements: 12
Répondre  |  Vous n'êtes pas connecté, mais en mode visiteur   4

The difference in lifespan is caused by a different composition of the luminicent layer. The relative slow electrons in a magic eye tube (200 V) destroy the luminiscent in the elder tubes quite fast. The pre war used  Zinc based luminiscent recrystalized under the influence of the slow electrons at approx. 200 V. In a CRT with sevaral thousand volt anode voltage the effect is much less.

The elder luminiscent can be recognized of its warm yellow green colour. (6E5, EM4, EM34, AM1, AM2, early EM80). 

After the war a new composition of the Zinc luminiscent was used. The colour changed to a more bright blue-green and looks cold due to the lack of a yellow component. This luminiscent was much more stable than the old one (DM 70/71, EM 84, EM87).

I do not know which type the new 6E5 from Russia use but it can be determined from the colour of the luminiscent.

Due to the recrystalisation you see markings on the screen of the elder valves. Also the decribed change in construction increased the lafespan because not all slow electrons hit the screen as described by Emilio.

Michael Watterson
IRL  Articles: 969
Schémas: 650
Images: 2497
30.Nov.11 02:00

Remerciements: 13
Répondre  |  Vous n'êtes pas connecté, mais en mode visiteur   5

The DM70 structure seems more related to a VFD, which is operate only on/off as low as 9V static and 22V multiplexed. The modern VFD default colour is the blue/green of DM70, EM84 and EM87.

The variable length bar on the DM70 appears to be due to the grid aperture shape, There are only the 3 electrodes, Anode, grid (which is a keyhole plate) and the point source direct filament, which is at the "front" and looks just like the VFD filaments.

Indeed the DM70/DM71 seems long lived and at 18V HT behaves more like a VFD, with 1/3rd part at 1.5V, or nearly off at 0V, and 2/3rds @ 9V "grid". At 18V the brightness of old ex-equipment DM70 is still usable! But 9V is very dim. At 18V grid the beam defocuses but fills entirely. VFD grid is normally equal to anode for on and slightly negative to filament for off. The VFD "defocusing" isn't apparent as the "grid" is a perforated foil / mesh just above the anode. Test circuit uses a series 4.7M grid resistor to wiper of 470k pot which has one end as voltage source and the other to pin 5 f+. The HT connected to pin 4 f- and to anode via 390k. The 1.5V LT battery was reversed for shortest bar.

VFDs do "wear out" but very much more gradually than old style Magic Eyes. On VFDs and DM70 or DM71 the luminicent layer is on the anode surface.

The EM84 / EM87 appear to operate more like miniature Electrostatic CRT with the luminicent layer direct on the glass.

The Ukrainian sellers of the 6e3s (6E3C in Cyrillic) seem to offer two types, a 1960s NOS and a 1980s NOS. Strangely the 1960s type is more expensive (from same seller). It would be ironic if the cheaper 1980s model is the improved blue/green luminicent layer and the 1960s model the original type with more emerald green.

Cet article a été édité 30.Nov.11 02:07 par Michael Watterson .