Tom's Guide | Tom's Hardware | Tom's Games
 |
 |
 |
My case is huge but it is very heaty and filled with drives and fans and many lights and such. Is there a way I could transfer my internal hard drives into another case? It would have adequate cooling ofcourse and perhaps even its own powersuppy just for those drives. My problem lies within the IDE cables. The longest ones I know are about 36" but I'm afraid I might need larger. Is there a way to buy a connector or adapter of some sort that can link 2 of these together midway? Also, will power and bandwidth transmission over the cable degrade because of the longer distance?
The problem with any parrallel cable is not loss of strength, but loss of sync. If the other end of the cable is not sure whether a bit is from this byte or the last, terrible crashes will occur. No two pieces of wire are identical.
What you might consider is setting up a modest computer (perhaps one made from parts lying around) as a file server and use network connection between them. Of course the network speed will create a bottleneck but shouldn't be a problem with archive data.
0 
I tried that in a home brew setup and the RF interference problem makes it self defeating. I think a full height server tower would be a better solution.
There are hard drive housings being sold but the cost of securing one is outragous to say the least. They're ment to be used in a totally RF dead enviroment.I may be crazy, but I'm not stupid...
0
Rimfire & setishock,
Very good points.
I thought about loss and RFI, but not sync.
M2
If at first you don't succeed, you're about average.
0
Hey M2, Can't think of anything funny right now!
I'm surprised that you somehow missed the discussions/arguements of parrallel versus serial. The limitation of parrallel is the differences between the individual wires. You may have heard the adage that the schematic diagram of a microscopic piece of wire is like a random collection of resistors, diodes, coilsandeven an occasional transistors. Monkeys with a limited keyboard have a good chance of replicating a real schematic!
This means that frequently, the path of least resistance may be a zig zag pattern, whilst in another wire, it might be a straight line. Even in a short piece of wire, this might cause problems. Increase the length of wire, you multiply the chance of error. Increase the frequency, you increase the possibility that latency might introduce a fault.
This is why the current technology is leaning toward serial. A single wire has the same charicturistics every time.
That's the short version. Did it make sense?
0
Yup,
It's a funny world we live in where serial is faster than parallel, CPUs are too fast, HDs are too big and Main memory is too big.
Didn't have these probs 15 years ago.
M2
If at first you don't succeed, you're about average.
0
Ah yes, I do miss my Commodore 64 with it's 1Mhz processor and 64K of RAM. (32 after the OS loaded). If I hadn't got pissed off with repairing the floppy drive, perhaps I'd still have it!
0
Hey guys thanks for all your responses. About the actual length, the maximum I have seen is 36" with the normal 2 devices, but here aparently is 46": http://www.trianglecables.com/48inide40pin.html
The RF interference among other signals I could possibly solve by wrapping it with some sort of sleeve or RF shield (I could swear I remember seeing them for sale). As for the data sync I really don't know. http://www.stonewallcable.com/dept.asp?dept%5Fid=122 has a list of different IDE styles of cables that you can customize to any length but obviously this isnt too logical (I can't expect having a 100 ft IDE cable to work properly). I really wanted to try this idea so I might buy 1 cable extra long and test a harddrive on it (with RF shield of course). So besides that, are there any other major probs? I was afraid I might need a signal booster of some sort midway to keep the line active. Also, this length isnt huge but it is indeed bigger than 46" (probaby)...the 2nd case would be right next to the first case so I would jus route it in a "U" around the backs...not too long I suppose.P.S. I use 36"s and I have almost have a terabyte of harddrives - no noticable problems.
0
Crosstalk is an issue but the real issue is at the speeds everything runs these days the time it takes the signal to travel 24" up and 24" back is already throwing things out of phase or sync. I have found that if the cables are shielded you can push the envelope a little further. I use a 24" tower and have 35" round shielded cables that seem to work fine. These cables are rated @133. Find shielded cables here: http://bestbyteinc.com/products.asp?dept=58&pagenumber=2
0
OtheHill,
While we're on the subject of trons...
I've been looking for an RJ11 opto-iso and having no luck.
Wrong keyword?
TIA
M2
If at first you don't succeed, you're about average.
0
M2
I am embarrassed to say the only part of your question I understood was RJ11. What the heck is a trons?
0
Sorry O,
I went off into the jargon zone; old war wound.
By trons, I mean electronics.
We were discussing wire RF interference etc.
A typical isolator has a sensing circuit, marbe a thyristor etc. And the good ones are pretty good.
But if you get a heavy hit, like a lightning strike, all bets are off.
An opto-iso is optically coupled so there's no electrical path from input to outpit.
Hope this msg has a lower "fog factor" than the last one.
M2
If at first you don't succeed, you're about average.
0
It does. I am a builder by trade and have installed Phone PBX systems in homes before. I have used the type of isolator you first describe to protect these systems. I have been apprised that there is a threshold of current that these devices can handle. Also is they are subjected to that much current they may become damaged. So, now you know what I know. As far as the optical sensor goes I couldn't begin to help you there.
0
M2
By the way, I am flattered that you thought I might have the knowledge required to answer your question. Unfortunately we form certain opinions and visions in our heads and can get fooled. A while back some folks posted pictures of themselves. I guess I unconciously had a vision of Dan Penny in my mind. When I saw his picture it was a shock to me because he didn't look "like a Dan Penny" to me. As for me, I am the proverbial jack of all trades.
0
O,
I was shocked at the picture of Dan.
He nad Neal Young are getting a little shaggy around the edges.
I'm a technician by trade; a shade tree mechanic at heart. I used to fix ships and airplanes; now I fix PCs. Doesn't look good on a resume.
Yeh, an isolator which has been stress tested, like an old fashioned circuit breaker, needs to go to the great parts bin in the sky.
M2
If at first you don't succeed, you're about average.
0
Don't get me wrong, I didn't find Dan's picture to be something bad. I thought about it and compared it to how you get an image of someone from speaking on the phone and when you meet them they aren't what you expected.
0
So, when are the first optical connectors coming out then?
Seems to me, that would solve the problem, right!
0
Why not a simple converter that would take the 40 signal paths and convert it to a 2 or 4 wire serial signal. Like the printer parallel to serial adaptor. On the other end it's converted back and sent to the drives. Of course it would have to be bi-directional. And mind boggeling fast.
I may be crazy, but I'm not stupid...
0
USB is theorically 480Megabits/sec. Fastest UATA is 133megabytes/sec. quite a bit faster. Besides that serial is staged to be ramped up substancially in speed. The real trick with all these standards is to get them to actually perform close to rated speeds.
0  |
 |
 |
This post is quite old and has been locked from receiving new replies. Please create a new posting instead.
| Ads by Google |
