IMPROVEMENTS FOR THE ELMO GS1200 XENON PROJECTOR DESIGNED BY UGO GRASSI.

Elmo GS1200 Xenon projectors are considered the best Super-8 projectors commercially available, as the far superior Fumeo and Beaulieu models (especially the Fumeo) were always sold in specialist or professional markets. 

Elmo GS1200 Xenon

I have a love/hate relationship with the GS1200 projectors, which leads me to own three units: a halogen model from the latest series, a rare PCom, and the extraordinary Xenon version. 

Elmo GS1200 PCom 

One might assume that the world's leading expert on Elmo GS1200 Xenon projectors would be Japanese or German, but no, it's Italian, a former Italian senator, Professor Ugo Grassi, who has just published an article about some modifications to improve the switch on of the Xenon lamps on the GS1200,  which he gave me permission to reproduce here:

For many years the Elmo GS‑1200 has been my fetish object, my Zen garden where every element must reach aesthetic and functional perfection.

It is precisely the machine’s intrinsic imperfection, combined with its enormous potential, that creates a challenge whose very pursuit brings me serenity and calm.

I have disassembled, serviced, and rebuilt dozens and dozens of Elmo GS‑1200 units, including many Xenon versions.

Over time I have accumulated various spare parts and lamps, and I even had the opportunity to buy some brand‑new original Toshiba lamps (the so‑called NOS). These are short‑arc xenon lamps, similar for example to the Osram XBO series, although in a reduced, “proprietary” 250‑watt format. These lamps are ignited by a high‑voltage discharge that ionizes the gas inside the bulb.

In the world of professional film projection it is well known that some lamps strike immediately, while others take a few seconds. Even though these lamps are extremely precise components, their ignition behaviour can vary noticeably even among “fresh out of the box” units.

This is why some fire up at the very first pulse, while others seem to “struggle” for a few seconds under the igniter’s blows.

1. Electrode Geometry

In short‑arc lamps, the distance between anode and cathode is critical.

Micro‑variations: Even with very tight manufacturing tolerances, a fraction of a millimetre in spacing or in the shape of the cathode tip can change the breakdown voltage required to ionize the xenon.

Emission points: The discharge always seeks the path of least resistance. If the cathode tip does not present an immediate “preferred point” (a trigger point), the train of high‑voltage pulses must persist until the gas reaches a plasma state.

2. Xenon Pressure and Purity

The xenon inside an XBO lamp is under extremely high pressure (about 8–10 bar cold, rising to 30–60 bar during operation).

Gas distribution: At ignition, the gas density between the electrodes may vary slightly due to ambient temperature or the lamp’s orientation.

Jitter effect: If the gas mixture does not ionize instantly in the central channel, the igniter continues sending high‑frequency pulses until the main arc stabilizes under the DC current from the rectifier.

3. Coupling with the Igniter

It is not always the lamp’s fault. The projector uses an igniter that generates pulses from 25 kV to 45 kV.

Pulse frequency: If the igniter is not perfectly matched to that specific batch of lamps, you may hear the classic prolonged “ticking” before the lamp finally catches and transitions to steady operation.

I never studied electronics formally, so everything I’ve learned about the projector’s circuits comes from raw, hands‑on experience.

The first modification concerns a small resistor in the ignition circuit, R704—an apparently insignificant component that is actually decisive in determining the cadence with which the igniter attempts to start the lamp.

Working on different units, I noticed that changing the actual value of this resistor very clearly alters the behaviour of the igniter. I first realised this when I found a 100‑ohm resistor installed instead of the standard 220‑ohm one, and everything suggested it had been set that way at the factory.

With lower values, the train of high‑voltage pulses becomes denser; with higher values, the pulses spread out and the ticking becomes slower and more deliberate. By ear, it also seems that in the first case each individual pulse is less “intense,” while in the second each discharge is stronger but less frequent.

I won’t attempt an academic theoretical analysis—that’s not my field—but the empirical fact is crystal clear: R704 controls the rhythm with which the circuit attempts ignition.

On some hard‑to‑start units, adjusting R704 toward lower or higher values (100 - 320 ohm) by a trimmer finally allowed easy ignition. But not all stubborn lamps responded to this alone: sometimes the lamp would start immediately, other times it needed two or three discharge sequences, each lasting a few seconds.

The second modification concerns a much cruder yet fascinating component: the air gap—the igniter’s spark gap.

Anyone expecting something sophisticated will be disappointed. It is essentially a small cylinder containing two metal points separated by an air gap. Nothing more.

The points operate in ambient air, and for this reason not only their spacing matters, but also small construction differences, surface condition, oxidation, even the mechanical “quality” of the piece.

And here a surprising but consistent fact emerges: the distance between the two points is not the same in all units.

Some spark gaps have closer points, others farther apart. Again, the consequence is immediately noticeable.

Closer points produce faster, more regular discharges; wider spacing slows the rhythm and, beyond a certain threshold, introduces clear irregularities.

The intuitive reason is simple: with a smaller gap, the air breaks down more easily; with a larger gap, the circuit must “charge up” more before overcoming the dielectric.

Usually these small differences are irrelevant for lamp ignition. Usually…

---

This modification is far more delicate.

You must open the igniter—an easy operation, since it is closed by a nylon cover held by three plastic screws.

Once opened, you discover that the components are not accessible because they are embedded in insulating resin (likely silicone).

The only visible component not submerged in resin is the air gap.

If you want to “play” with the spacing of the points, the original must be removed.

To do this, you sometimes need to carefully remove part of the resin to free the two contacts (two small metal tabs): one connected to a rigid braided wire, the other to a flexible lead emerging from the resin.

Once the two contacts are desoldered, the air gap must be extracted from its mounting base, keeping in mind that the two tabs pass through slots in the base.

At that point, you must insert the handmade air gap, ensuring that one of the points is adjustable so you can test different distances.

---

At this stage it is important to emphasize that I do not know why these modifications work.

And yet they work—and they work well—without endangering the circuits.