You certainly can copy the to a virtual machine. ddrescue will make a bit-for-bit image of the drive, and for that you will need 500GB of space regardless of how much space on the source was actually in use.

Yes, you can just copy the rescued image back to that drive. For that, reformatting would be superfluous. Or, you can reformat and copy the files back to the new filesystem. But from the outside, a dying drive looks exactly the same as a factory-fresh new one. If your data is unimportant enough to trust to a drive in unknown condition, go right ahead.

Could i use CrystalDiscInfo to determine whether the drive is failing?

I have no way to know whether that program might have a way to work around the inability of the bridge chip in your enclosure to pass SMART commands to the drive. All of my various USB enclosures, docking stations, and adapters do pass the SMART commands properly, so I can't meaningfully test that program to see if it might work for you. Frankly, I doubt it.

Your best bet would be to go to the Web site of the manufacturer of the enclosure and see if they have any sort of diagnostic program available.

Also i found this, http://www.easeus.com/datarecoverywizard/download.htm

If i can make an image of the drive in windows, then apparently this program can handle all the data recovery. Seems simple, i just need to make an image of the failed drive. How would i do this in windows?

I'm not an expert on doing things in Windows, but I'm sure Google can point you to a solution. Note that the free version of Data Recovery Wizard is limited to recovering 2GB maximum.

Well I've tried using DriveImage XML which is apparently the best program for creating an image of a hdd but when i select the portable drive, it does not display any information such as the filesystem, total sectors, free/used bytes and so-on and is unable to create an image of it.

So the portable hdd is so buggered up that these programs can't even determine its filesystem... Is there any sort of program that just reads the raw data to an image regardless?

In Linux, dd and ddrescue would be the choice. In Windows, as I said, I'm no expert there.

R-Studio.

http://www.r-studio.com/

It recovers whatever is recoverable. Everything from holiday pictures on a flash memory that lost its FAT, to a HDD with lots of badblocks.

Oh wow, this program is fantastic.

Unfortunately i think this drive is toast. I ran this program on it all night, and all it was able to recover was folders. For everything else it gave me CRC errors and read failures. I haven't been able to recover anything that isn't a folder.

Whatever is wrong with this drive, it is something serious. I might as well just throw it in the trash.

Yes, R-studio really is an impressive HDD forensics software.

About USB HDD enclosures and reading disks with bad blocks - Many of these USB enclosures give up after a certain amount of bad reads, and thenafter they stop reading, and instead start pewing the syslog with read errors even for good blocks. Detaching and reattaching the USB makes it behave for a while again. So if you have been trying to read the HDD through USB, chances are that if you instead connect it to a SATA port, it will recover much more data from the HDD.

Also, recovering data from a bad HDD makes it run hotter than usual. Ensure proper cooling for the HDD. It will let you recover more of the data, and the HDD will last a tad bit longer before it "walks into the light".

I managed to recover a handful of files with a little diligence, but the files i was trying to save (mostly music) are just too damaged to be worth saving. Most of the albums i managed to recover only have about half the songs without bad blocks. So i'll have to re-acquire the music regardless.

The sounds it is making leads me to believe that there is some sort of hardware issue. It will sit silent for 3 seconds, then vibrate quickly (making a purring noise) for half a second, continuously, while it is being read from.

Even if i did reformat it, i don't think it would be usable. Thanks for the help, all.

One thing to be aware of about recovering data from a HDD with NTFS filesystems and alike is that everything is handled by "clusters".
A cluster is an allocation unit that most often comprise eight consecutive sectors. When any of the sectors, (normally 512 bytes in size), is bad, then the cluster is deemed as bad.

If one uses DIRECT_IO to read the sectors one by one, (can be done with the shell command 'dd'), the chances are that 7/8 of the cluster is good sectors. And surprisingly often the bad sector is after the end of the file stored in the cluster.

Also, if the NTFS filesystem has been defragmented, there usually are old copies of a file stored in one or more of the unallocated clusters. So, if one has a file of known type that contains a bad cluster, then it is fully possible to look for similar clusters to the damaged one in the unallocated clusters, and check if the cluster copy will make the known file complete and working. Say it is a zip-file we are looking for. Then we can test with similar clusters in the unallocated space to see if the file integrity check is OK on the zip-file if we patch in the proposed cluster copy.
Too look for possible old cluster copies we can simply match the seven good sectors of the bad cluster to the unallocated clusters. Also, if the damaged file is a system file, like a DLL, we can often see what DLL name/version we are looking for, and simply replace the damaged DLL file with a fresh one from an install disk or other source.

It is very often possible to restore more than 95% of a damaged NTFS filesystem, and even 100% restore. It depends upon how important the data is, how much effort we put in.
Also know that if we have a damaged raid with parity, then we will have an overhead of several alternative content for the damaged data. Then we only need a really fast computer and some clever coding to recover the minced data. And a bad sector on a HDD will often produce the correct sector contents if we only do enough retries. The HDD has its internal error correcting algorithm, and sometimes the data is correct despite the read command returns a BAD status just because the HDD doesn't know if the data is correct.