Epic of the Cuban Doppler-4. The Need, the Obstacles, and the Technological Challenge

"Cuban Doppler Epic-4" reveals the strategic reasons, the internal obstacles, and the colossal technological challenge the Cuban team faced upon taking on the mission of building a Doppler radar. This chapter explains why Rubiera asked for the Doppler, the controversial decision to build it in Cuba instead of buying it, and the clash with skeptical voices within the Instituto de Meteorología (Cuba's national weather service) itself. Learn about the two major technical challenges —the high-stability local oscillator (STALO) and the real-time signal processor— that turned this project into a bet on technological independence, placing LADETEC among the few institutions in the world capable of designing a complete Doppler weather radar with homegrown technology.

"The need, the obstacles, and the technological challenge" covers the rationale, the resistance, and the technical complexities surrounding the decision to build a Doppler radar in Cuba. The chapter opens by explaining why Dr. José Rubiera asked for the Doppler: his experience at the U.S. National Hurricane Center and his familiarity with NEXRAD radars had shown him that Doppler and polarimetric radar is essential for nowcasting, beyond what satellites can offer. However, no sooner had the assignment become known than opposing voices arose within the very headquarters of the Instituto de Meteorología: from skepticism about the Camagüey team's capabilities to a preference for buying new technology abroad. Fortunately, INSMET's leadership, the Agencia de Medio Ambiente (Environmental Agency), the CITMA minister, and the Comandante en Jefe's Support Group placed full trust in LADETEC, backed by the track record of modernizing Cuba's radars (Camagüey, Pico San Juan, Casa Blanca, and La Bajada). The technological challenge was enormous: two critical elements defined the leap to Doppler —the high-stability local oscillator (STALO), under military control and hard to obtain because of the embargo, and the real-time signal processor, which demanded theoretical command of algorithms along with hands-on skill programming programmable devices. On top of that came the open question of whether conventional magnetrons would be stable enough compared with costly coaxial magnetrons or klystrons. The text stresses that "assembling" a radar by buying OEM components is trivial, but designing your own transmitter, receiver, antenna system, signal processor, and data processor is a feat only about seven institutions in the world have pulled off —and LADETEC is in that fight, proving that technological independence carries a price, but also an invaluable reward.

Why did Rubiera ask for a Doppler radar?

Rubiera, obviously, knew perfectly well why he wanted a Doppler radar. We knew it too. However, writing the reports making the case fell to me. I had to resign myself to my fate as a "writer" (OK, hack writer, that's what there is) and write. First I had to write, why did we need a Doppler radar? Then I realized I had to go back a bit further and argue, why do we need weather radars?

Later on, faced with the confusion, I decided to clarify, what we can expect from satellites and what we can expect from radars ? Although some are dazzled by GOES-16, Doppler weather radars and Polarimetric radars are essential for nowcasting.

The curious thing is that in the United States that controversy had already been put to rest — we settled ours by talking and arguing, but they had to develop an expensive project just to prove that Doppler was indispensable, and Rubiera knew this very well, because he has visited the U.S. National Hurricane Center many times and had seen firsthand the work of the NEXRAD radars, not to mention everything he must have read and studied on the subject.

Why do it in Cuba?

No sooner had the Doppler question and the visit from the Support Group people come up than the "opposition" appeared. Right at the main headquarters of the Instituto de Meteorologia in Casa Blanca (the Havana neighborhood where it's based), voices could be heard: what the hell do they know about what a Doppler radar even is? Why, if the Comandante en Jefe said to buy it, are they going to go and screw it up? New is the best brand. What they're used to is patching things together with old Russian parts. A Doppler radar is high-tech business, all they're going to do is waste time and money, and in the end we'll have nothing at all. And much more was said, including some ugly things. The funny part is that it was serious people saying it, researchers with real standing, but… who had never once come near us. Fortunately, they were a minority, and even more fortunately for us (and for Cuba), they weren't the ones making the decisions.

Meteorology's leadership (center chiefs, with Rubiera at the head, and directors, of course, including the director general), the leadership of the Environment Agency, and the CITMA Minister backed us fully, completely — and the Comandante en Jefe, through his Support Group, put his trust in us as well.

Nobody asked us how much we knew for that decision — the only visible résumé was having modernized and automated (full teardown and rewiring of all the cabling included) the radars at Camaguey, Pico San Juan, Casa Blanca, and the one at La Bajada, which was in progress at the time. The results of that modernization were there for the whole world to see. Anyone who cared to could have watched us in action at the Casa Blanca radar back in 2001 (Tomas saw the whole thing in detail — the entire process unfolded "right over his head," that is, one floor above management, day by day, in under a month). Nobody knew how well-prepared we were in theory for the task. They simply trusted us. Res non verba.

Even so, I argued this too, and I've kept arguing it for years: Why Cuba shouldn't buy weather radars from foreign manufacturers? Why Cuba should make its own weather radars? Manufacture, or buy what's already manufactured? As for that business about New is the best brand, life gave its own answer with the Holguin radar: the new one, and it has the worst historical performance, losing by a wide margin to the old Japanese and Soviet radars. The problem isn't whether a new radar built by us is better or worse than a new radar bought from a foreign manufacturer. The problem lies in sustainability, and in technological dependence on the manufacturer.

There were some people who believed the job should have been given to them, and they put up a certain amount of opposition. I won't name them. René Hidalgo, with tremendous tact, took care of managing that and letting us stay focused on the task. We were given the job and we did it. We built a prototype at the Camagüey radar, built a serial unit from scratch at Casa Blanca, and another on the salvageable base of the Holguín radar. That's the story it tells.

The technological challenge

I've explained this in detail before: we already had 90% of the Cuban Weather Radar; but it lacked the ability to capture the Doppler effect (the shift in the frequency of the received signal relative to the transmitted signal, caused by wind acting on weather targets). Getting that remaining 10% was no joke — it was, indeed, a high-technology problem. It wasn't a matter of "little patches" or "parts salvaged from old Soviet televisions and radio receivers."

The technological challenge centered on two fundamental issues we had studied thoroughly: it requires a heterodyne oscillator (local oscillator) of very high stability (in English, STALO, for Stable Local Oscillator). We couldn't manufacture that in Cuba, at least not at that time. We needed to acquire it. No, you don't buy that at a neighborhood hardware store. The firms that produce it are required to control the end user — it's highly controlled military technology. We already know that Cuba is barred as a destination the world over, "thanks" to the United States blockade. It might never have been solved; but it was.

The second element — less of a problem in terms of blockade control over it — is a signal processor capable of running, in real time, the complicated algorithms for calculating the moments of the received power spectrum (commonly, and incorrectly, called the Doppler spectrum). This requires theoretical command of the algorithms, which we'd already had since 1993, and practical skills in programming programmable devices, which we had partly been acquiring over the course of a decade. Read the story of our six generations of processors. This was a huge leap, a paradigm shift.

There was a third element to take into account: the stability of conventional magnetrons. Could we use the old, very cheap Russian magnetrons? Or would we have to acquire costly coaxial magnetrons or klystrons? During my internships and trips abroad I had inquired about this. Nobody knew for sure — they had simply used coaxial magnetrons or klystrons. To answer that, we ourselves would have to study the stability of magnetrons and work through ruling out the instabilities generated by the heater power supply, the main power supply, and the modulator.

There was another problem to solve, also a technological one, and that's the secondary processing software; but I'll devote a separate post to that.

---

No, however much some talked badly, we did know the problem we would face, in all its facets, and let's say we had it well "cornered" when we accepted the challenge of building a Doppler radar prototype … I should clarify "building it at the lowest possible cost," since buying all new parts would have made everything very easy. True, there were no guarantees, it COULD have gone wrong, and true, it took eight years — that is the price to pay for technological independence,

Assembling a Doppler radar by buying the transmitter, the receiver, the antenna, the antenna drive from OEM manufacturers, a Doppler signal processor for weather radar, and a data processor … anybody can do that. The control software (SCADA) part is trivial, anybody does that. But designing your OWN transmitter, receiver, antenna system, an OWN signal processor, and an OWN data processor/display … believe me, there aren't seven institutions in the world that pull that off. Honorably, LADETEC is in that fight, and, in spite of everything … today it is still standing, with new solutions.