Solutions needed for potentially failing NAS hardware

Question

I have an 8 drive Rocketraid NAS that seems to be having an issue with one of the "enclosures". Even though all 8 drives are in the same device the software apparently categorizes the top 4 drives in one "enclosure" and the bottom 4 into another, which may be because each set of 4 drives is on its own esata cable (since the NAS connects via 2 esata cables to a pci card). I had 5 drives (in a Raid 5) of the same make and model and thought it was an interesting coincidence that the one drive in its own enclosure failed and have now replaced it, only to see this new drive "fail" too.

The weird part is the smart data seems to be alright

ID  Name    Threshold   Worst   Value   Status
1   Read Error Rate 51  253 100 OK
3   Spin-up Time    21  176 176 OK
4   Start/Stop Count    0   100 100 OK
5   Re-allocated Sector Count   140 200 200 OK
7   Seek Error Rate 0   200 200 OK
9   Power-on Hours Count    0   100 100 OK
a   Spin-up Retry Count 0   253 100 OK
b   Drive Calibration Retry Count   0   253 100 OK
c   Drive Power Cycle Count 0   100 100 OK
c0  Power-Off Retract Count 0   200 200 OK
c1  Emergency Retract Cycle Ct  0   200 200 OK
c4  Relocation Event Count  0   200 200 OK
c5  Current Pending Sector Count    0   200 200 OK
c6  Off-line Scan Uncorrectable Sector Count    0   253 100 OK
c7  Ultra ATA CRC Error Rate    0   199 200 OK
c8  Multi-zone Error Rate   0   253 100 OK

I don't know much about how to read this, but they all say "OK". The device is picked up and used just fine, which again is weird, but about 5-15% of the way into the rebuild the rebuild fails and gives an event log of drive failed.

So here are my questions:

• Am I right to say that because the smart data reads all "OK" that it must be the enclosure and not the drive?
• What fixes can I try without buying new hardware?
• If I buy new hardware should I replace just the pci raid card? or just the device that holds the drives? How can I isolate where the issue is coming from?
• If I replace either or both with a different brand will all my data in the raid 5 be lost? Or will I be able to plug the drives into a new device and have it keep the current raid5 and data intact?

EDIT: To answer Dave's questions, my original setup was 5 samsung green drives, which I later realized is not recommended. It has caused some stuttering issues but the NAS worked fine for 3 years until one of the drives "failed" but still was detected fine on bootup, had good smart info, and would start rebuilds just fine, but fail before finishing the rebuild. The latest drive I got was a WD Red drive, since that is a NAS drive. I did try to research actually recommended drives this time, but the manufacture of the raid card had no currently available drives in their recommendation list, so I had to wing it.

I didn't know manufactures released their own diagnostic software, and while WD does apparently have some, there doesn't appear to be any for their WD Red drives. For my original samsung drives I've found some seagate diagnostic software (didn't realize samsung and seagate had merged) which I'm going to try.

I assume the firmware on all the drives matched before the failure as I order all 5 drives at the same time and were the same model number. Now I'm mixing and matching a bit since I want to migrate to NAS drives but don't want to pony up the cash to replace them all.