Strictly speaking, the RTGs will continue generating power past that time -- it just won't be enough to power any of the instruments or the radio; the thermocouples also degrade, which further impacts the power produced. (tomato, tomahto) Attitude propellant is also an issue. Around 2020, they'll start to turn off instruments until they don't have enough power to run any of them anymore. Apparently, they haven't picked the shutdown order yet -- but they have 10 years to decide, so no rush. =)

NASA's timeline for the Voyager lifetimes: http://voyager.jpl.nasa.gov/spacecraft/spacecraftlife.html

It would be cool if they could shut it down for a long time then have it wake up and send some data back with the remaining power.

Neat idea, but I don't think there were any onboard batteries -- AFAIK power was just fed straight from the RTG thermocouples, more or less. Batteries would have just added weight, would have had a crappy recharge capacity after a while, and wouldn't have made much sense for a deep space probe which relied on an RTG instead of solar panels.

Also, the RTG might be powering instrument heaters, keeping them at warm enough temperatures to prevent failure. If that's the case, once those are shut off, there's no turning those instruments back on.

EDIT: It appears that Voyager 1's ultraviolet spectrometer instrument, at least, is pretty tough: as of January 2012, it has survived having its heater shut off for power conservation: http://www.nasa.gov/mission_pages/voyager/voyager20120117.ht...

That's one hardy spacecraft! =)

Should have went with the extended extended-battery.

Ha, true. From what I can find, each of Voyager 1's RTGs is about 5% of the total mass of Voyager 1, so about 15% of it's mass total. Seems surprisingly low to me, considering they are basically trashcans full of plutonium. I'm not sure how many more you would have to add to keep Voyager 1 operating longer, but the more you add the slower it would ultimately be going (well, the harder it is to get it going the speed the mission required). My guess is that additional/larger RTGs couldn't be justified or technically pulled off at the time.

I wonder what the expected lifetime of the crafts were when they were designed back in the 70s. Maybe they thought the RTGs would outlive the other components?

My thoughts exactly -- they probably didn't bother putting a whopping big 100-year RTG on there because they didn't figure the electronics would last long enough, or they figured they'd run out of attitude propellant long before then.

You could look up the planned mission duration, but all this would tell you would be the duration that they were absolutely designed to work for -- kind of like how the Mars rover missions were 90 sols, but in actuality they've gone over 2,000 (albeit with solar panel dust issues, busted wheel motors, etc).

Well I mean, Voyager 1, launched in 1977, has been on the "interstellar mission" since 1980 (32 years) and is planned to go into the 2020's (total of more than 40 years). That doesn't quite work out to the (2000-90)/90 ratio (though the distance traveled by Opportunity is an even more impressive ratio), but I think it is pretty damned impressive nevertheless. It was definitely engineered to last a lot longer than the Jupiter/Saturn observation mission required; I feel like both came from the same "school" of engineering and mission planning.

This sort of thing is a natural consequence of high cost and high risk missions. You have a multi-billion dollar mission, and it's doing something unprecedented and using a brand new design and construction. When you build something like that what tends to happen is that you try to keep the mission scope reasonable and then you build the vehicle to have a very high (say >95%) chance of satisfying that minimum mission requirement. But what happens when, say, you create a Mars rover with a 95% chance of working for 90 days you'll likely end up engineering a vehicle which might have, say, a 50% chance of lasting for 5 years. If you set out to do that from the start it would be a non-starter, because a 50% mission success rate is too low for a multi-billion dollar mission.

I'm guessing they figured there'd be other much more advanced and capable space probes overtaking the Voyagers well before the power got this low…

Not at all, actually. The whole interstellar mission aspect of the mission was entirely a beneficial side effect. In fact, the only part of the mission that was planned from the start was the Jupiter and Saturn encounters. The mission ended up becoming an outer Solar System "grand tour" due to a fortuitous planetary alignment, but only Voyager 2 was able to visit Uranus and Neptune. The interstellar mission was entirely unplanned and was funded mostly because it has been relatively inexpensive and for various reasons the Voyager craft have suitable instruments for studying the magnetic, particle, and dust environment in the outer Solar System and interstellar space.