Cheap rockets with fast deployment would be game changer.
Satellites are built to narrow margin to reduce weight because of the weight restrictions on satellites. Right now we have reusable rockets but the rockets aren't any bigger so while the cost of launch has come down the volume/mass that can be launched in one go has not gone up. If laser interferometry gets better then large telescopes can be launched as constellations. Larger reusable rockets are also being worked on. When that happens they can care less about mass margin and make cheaper lower margin telescopes.
Telescopes are also generally built as a one-off which also increases cost. I hope the technology for telescope constellations will improve.
Size: There's still a finite fairing size. Even the biggest upcoming rockets can't fit a 30m mirror in one piece.
Weight: say you want to put something like ALMA in space, which weighs over ten million pounds. Even with platforms like starship,you're not going to be able to do that in one go. Maybe one launch for each of the 66 antennas,but even that is iffy.
Power: Hubble's power consumption is around 2Kw. Using ALMA as an example again, each individual antenna uses about 10x more electricity than that on average. More electricity means you need RTGs or nuclear reactors (not politically popular, complicates thermal regulation, potential for spreading radioactive waste across orbit and earth a la RORSAT), or massive arrays of solar panels. Massive solar panels means even more weight and space required
For interferometry, knowing your location to the nanometer is just the first of many challenges. You have to station keep across the constellation (tricky to do continuously while you're in orbit!), it becomes even more important to minimize vibrations, etc
> For interferometry, knowing your location to the nanometer is just the first of many challenges. You have to station keep across the constellation (tricky to do continuously while you're in orbit!), it becomes even more important to minimize vibrations, etc
I didn't say it wasn't an engineering problem, but the major reason it hasn't been used yet is there was no real benefit to doing that over launching a single large telescope in terms of launch cost. I hope to see these problems solved soon in the future.
>no real benefit to doing that over launching a single large telescope in terms of launch cost
It would be hugely beneficial to do it in space. When you do interferometry on the ground, your maximum aperture size is the diameter of the Earth (if you had two observatories at opposite points on the Earth). When you do it in space, the size of your synthetic aperture is practically limitless.
I'm not convinced that's true. If you look at past space telescopes, launch costs are not a large part of the budget. Even if SpaceX gave everyone free launches as a goodwill measure, you'd only make space telescopes maybe 2-5% cheaper. There's very real challenges that space based telescopes have that Earth based ones don't (how do you make a cryocooler that doesn't vibrate your optics? How do you make a machine that can run for years without maintenance? How do you get precise enough positioning to interferometry?). My understanding is that due to adaptive optics, being ground based isn't much of a problem for visual telescopes. And for large radio telescope arrays, I don't see how it's possible at all to move to space