Designing a laser cannon to work in orbit was no small engineering challenge. A hand-held laser pointer is a relatively simple, static device, but a big gas-powered laser is like a roaring locomotive. Powerful turbo-generators "pump" the carbon dioxide until its atoms become excited and emit light. The turbo-generators have large moving parts, and the gas used in the formation of the laser beam gets very hot, so it has to be vented. Moving parts and exhaust gases induce motion, which poses problems for spacecraft—particularly one that has to be pointed very precisely. The Polyus engineers developed a system to minimize the force of the expelled gas by sending it through deflectors. But the vehicle still required a complex control system to dampen motions caused by the exhaust gases, the turbo-generator, and the moving laser turret. (When firing, the entire spacecraft would be pointed at the target, with the turret making fine adjustments.)
The system was complicated enough that by 1985, the designers knew that testing its components would require more than one launch. The basic Skif-D1 spacecraft structure was proved out in 1987, while the laser wouldn't fly until Skif-D2, in 1988. Around the same time, another, related spacecraft went into development. Designated Skif-Stilet (Scythian-Stiletto), it was to be equipped with a weaker infrared laser based on an operational ground-based system. Skif-Stilet could only blind enemy satellites by targeting their optics. Polyus would have enough energy to destroy a spacecraft in low Earth orbit.
Work on these projects was proceeding at a furious pace throughout 1985 when an unexpected opportunity arose. The Buran shuttle had fallen behind schedule, and wouldn't be ready in time for the planned first launch of the Energia rocket in 1986. The rocket's designers were considering launching a dummy payload instead, and Skif's designers saw an opening: Why not test some of the components of their spacecraft earlier than scheduled?
They quickly drew up plans for a vehicle that would test the functional block's control system and additional components, like the gas ejection vents and a targeting system, consisting of a radar and a low-power fine pointing laser, that would be used in conjunction with the big chemical laser. They labeled the spacecraft Skif-DM, for "demonstration model." Launch was scheduled for fall 1986, which would not affect the launch of Skif-D1, planned for the summer of 1987.
Meeting such a tight deadline had a human cost. At one point, more than 70 firms within the Soviet aerospace industry were working on Polyus-Skif. In his history of the project, Lantratov quotes from an article by Yuri Kornilov, the lead Skif-DM designer at the Khrunichev Machine Building Factory: "As a rule, no excuses were accepted—not even the fact that it was almost the same group of people who, at that time, were performing the grandiose work associated with the creation of Buran. Everything took a back seat to meeting the deadlines assigned from the top."
The designers realized that once they launched the huge craft into space and it expelled large amounts of carbon dioxide, American intelligence analysts would observe the gas and quickly figure out that it was intended for a laser. So the Soviets switched to a combination of xenon and krypton for the Skif-DM venting test. These gases would interact with ionospheric plasma around Earth, and the spacecraft would appear to be part of a civilian geophysics experiment. Skif-DM would also be equipped with small inflatable balloon targets, mimicking enemy satellites, that would be jettisoned in flight and tracked with the radar and the pointing laser.
The launch of the demonstration satellite slipped to 1987, partly because the launch pad had to be modified to accommodate a rocket as heavy as Energia. The technical problems were relatively minor, but the delay had a critical impact on the project's political fortunes.
In 1986, Gorbachev, who had been general secretary of the Communist Party for only a year, was already advocating the sweeping economic and bureaucratic reforms that would come to be known as perestroika, or restructuring. He and his government allies were intent on reining in what they saw as ruinous levels of military spending, and had become increasingly opposed to the Soviet version of Star Wars. Gorbachev acknowledged that the American plan was dangerous, says Westwick, but warned that his country was becoming obsessed with it, and began challenging his advisors: "Maybe we shouldn't be so afraid of SDI."
In January 1987, with Skif-DM's launch just weeks away, Gorbachev's allies in the Politburo pushed through an order limiting what could be done during the demonstration flight. The spacecraft could be launched into orbit, but could not test the gas venting system or deploy any of the tracking targets. Even while the vehicle was on the pad, an order came down requiring several of the targets to be removed, but spacecraft engineers pointed out the dangers of interacting with a fueled rocket, and the order was canceled. Still, the number of experiments was reduced.
That spring, as the booster lay horizontally inside a vast assembly building at the Baikonur Cosmodrome in Kazakhstan, the Skif-DM was mated to its Energia rocket. Technicians then painted two names on the spacecraft. One was "Polyus." The other was "Mir-2," for the proposed civilian space station that Energia's leadership hoped to build. According to Polyus historian Lantratov, that may have been less an attempt to fool foreign spies about the mission's purpose than an advertisement for the Energia company's new project.