Removable drive bays-->Corruption

R

Rus

If this has been covered before, I'm sorry. I searched all over the place
(including here) when I first encountered the problem several months ago.
Now that I've finally found a fix, I feel I have to share it. Might be
someone else out there in the same predicament.

For years I used removable drive bays. Then, I upgraded MB, CPU, etc. and
suddenly, my hard drives would freeze or corrupt data. I connected directly
to the hard drives and everything was fine. Obviously it was the removable
bays that caused me problems.

The bays I used had "Centronics" connectors. That's all there was at the
time I bought those bays. I supposed that it was those 25 relatively large
and unshielded copper ribbons radiating electromagnetic transmissions that
subsequently caused "cross-talk" and corruption.

I searched and searched. Finally, I found some IDE drive bays that use
"DIN" connectors.

I'm back in action again. The "DIN" connections work perfectly. I've tested
over the last couple of days doing all sorts of big and small data transfers
in several OS's. Absolutely works perfectly. If I was going to encounter
corruption again, it would have happened by now.

So, if you've had the same problem, look for removable IDE hard drive bays
that have either "32" or "101" in the model number. They will have the
"DIN" instead of the "Centronics." There's a bunch of brand names on these
bays. They all seem to use the same model designations. Any brand is
probably as good as the next. The "32's" are cheaper and mostly plastic.
The "101's" are a bit more expensive but they have aluminum trays and
aluminum rails.

I could only afford the "32's" but they work fine.

//rus//
 
R

Rod Speed

Rus said:
If this has been covered before, I'm sorry.

You've added some worthwhile observations anyway.
I searched all over the place (including here) when I first
encountered the problem several months ago. Now that
I've finally found a fix, I feel I have to share it. Might be
someone else out there in the same predicament.
For years I used removable drive bays. Then, I upgraded MB,
CPU, etc. and suddenly, my hard drives would freeze or corrupt
data. I connected directly to the hard drives and everything was
fine. Obviously it was the removable bays that caused me problems.
The bays I used had "Centronics" connectors. That's all
there was at the time I bought those bays. I supposed
that it was those 25 relatively large and unshielded
copper ribbons radiating electromagnetic transmissions
that subsequently caused "cross-talk" and corruption.

Its more likely to be the electrical discontinuity causing
reflections off that bit of the total connection between
drive and motherboard controller.
I searched and searched. Finally, I found some
IDE drive bays that use "DIN" connectors.
I'm back in action again. The "DIN" connections work
perfectly. I've tested over the last couple of days doing
all sorts of big and small data transfers in several OS's.
Absolutely works perfectly. If I was going to encounter
corruption again, it would have happened by now.
So, if you've had the same problem, look for removable IDE hard
drive bays that have either "32" or "101" in the model number.
They will have the "DIN" instead of the "Centronics." There's a
bunch of brand names on these bays. They all seem to use the
same model designations. Any brand is probably as good as the
next. The "32's" are cheaper and mostly plastic. The "101's" are a
bit more expensive but they have aluminum trays and aluminum rails.
I could only afford the "32's" but they work fine.

I think it makes more sense to use SATA drives in SATA format
removeable drive bays now, those dont flout the ATA standard.
That format supports hot swap too.
 
J

Jonny

Rus said:
If this has been covered before, I'm sorry. I searched all over the place
(including here) when I first encountered the problem several months ago.
Now that I've finally found a fix, I feel I have to share it. Might be
someone else out there in the same predicament.

For years I used removable drive bays. Then, I upgraded MB, CPU, etc. and
suddenly, my hard drives would freeze or corrupt data. I connected directly
to the hard drives and everything was fine. Obviously it was the removable
bays that caused me problems.

The bays I used had "Centronics" connectors. That's all there was at the
time I bought those bays. I supposed that it was those 25 relatively large
and unshielded copper ribbons radiating electromagnetic transmissions that
subsequently caused "cross-talk" and corruption.

I searched and searched. Finally, I found some IDE drive bays that use
"DIN" connectors.

I'm back in action again. The "DIN" connections work perfectly. I've tested
over the last couple of days doing all sorts of big and small data transfers
in several OS's. Absolutely works perfectly. If I was going to encounter
corruption again, it would have happened by now.

So, if you've had the same problem, look for removable IDE hard drive bays
that have either "32" or "101" in the model number. They will have the
"DIN" instead of the "Centronics." There's a bunch of brand names on these
bays. They all seem to use the same model designations. Any brand is
probably as good as the next. The "32's" are cheaper and mostly plastic.
The "101's" are a bit more expensive but they have aluminum trays and
aluminum rails.

I could only afford the "32's" but they work fine.

//rus//

Good deal. Didn't think about that before. Makes sense. Noticed you're
using ide.
 
R

Rus

You've added some worthwhile observations anyway.

Well. I said I'm sorry only to be polite not because I'm looking for your
approval. Truth of the matter is that people pop in and out of these
newsgroups and most don't follow the entire history of postings from day one.
Many come looking for help and they go once they've solved a problem. I'm
just trying to help those who are trying to solve a problem. I figure UseNet
has a purpose and every problem shouldn't be solved in a Google search.
UseNet allows humans to interact to a greater degree.

I scanned all the current posts and saw nothing related to this thread, so, I
posted for the benefit of those who would approach this NG looking for help
on the same topic. It's only valuable to those individuals until the thread
is dumped from the server unless they look in the Google archives and such.
I figure if I can help one person, I've done something worthwhile.

To those who moderate this newsgroup or participate regularly over a long
period of time, I say "I'm sorry" because I can understand that from their
perspective, repeated approaches to the same topics can become tiresome.

Actually, I'm not sorry. I post with pride because I know that somewhere,
someday, someone might benefit from the things I have learned the hard way.

Its more likely to be the electrical discontinuity causing
reflections off that bit of the total connection between
drive and motherboard controller.

Wow. Electronics has been a hobby of mine for over 20 years and I have
discussed theory with technicians and engineers alike. I've got to say,
however, I can't make any sense out that statement.
I think it makes more sense to use SATA drives in SATA format
removeable drive bays now, those dont flout the ATA standard.
That format supports hot swap too.

My IDE drives just won't quit. I've got SATA ports on the MB but there's no
reason for me to upgrade just to be able to say that I've got SATA drives in
my computer.

A lot of people (probably you included) jump to SATA drives and don't even
know why. They just do it so they can say "I've got SATA."

In a practical sense, there's not much to gain. Current hard drives can't
yet saturate an IDE channel with sustained reads or writes so there's no need
for me to upgrade.

The only "speed" benefit of a SATA is just a brief fraction of a second when
data is "burst" to and from the memory chip in the hard drive. Once that
chip is saturated, everything slows down to the speed at which data can be
electromechanically transferred through the heads to and from the spinning
platters. Until hard drives are packed densly enough and spin fast enough to
provide a sustained read/write speed in excess of 100MB/sec or 133 MB/sec,
there's not much to be gained with a SATA channel.

IDE is 100 (or 133). SATA is 150. For the brief fraction of a second in
time that I'd gain even for a burst write, it's not worth it to me to go to
SATA right now.

I still boot to DOS, W98SE, Win2K, various Linux distributions, as well as
WinXP. Only XP and some Linux kernels are fully satisfied with a combination
of SATA and IDE hard drives connected at the same time. I also boot from a
secondary hard drive sometimes. All these things make a SATA + IDE
combination difficult to manage, I think. I know that for what little gain
there is, it's not worth scrapping all 6 of my IDE drives for SATA just to
say I've got SATA. If I try to combine SATA HD's and IDE HD's on the same
MB, I'm going to add problems with some of the other OS's in which I play and
experiment. It's not worth those problems just to gain 150 - 100 = 50MB/sec
for a fraction of a second.

Now if I was building a killer gaming machine or if I was building a server
that would manage lots *.gif, *.jpg, *.js, *.txt files and other tiny files,
I think SATA drives would be beneficial to some small degree.

Since my main computer usage involves video editing where I transfer mostly
huge files over long-term sustained reads and writes? SATA offers very
little benefit for me.

//rus//
 
R

Rod Speed

Well. I said I'm sorry only to be polite not because I'm looking for
your approval. Truth of the matter is that people pop in and out of
these newsgroups and most don't follow the entire history of postings
from day one. Many come looking for help and they go once they've
solved a problem. I'm just trying to help those who are trying to
solve a problem. I figure UseNet has a purpose and every problem
shouldn't be solved in a Google search. UseNet allows humans to
interact to a greater degree.

That last is arguable. I prefer usenet over say forums, but that
is because groups.google provides a rather more homogenious
collection of data than does google over forums etc.
I scanned all the current posts and saw nothing related to
this thread, so, I posted for the benefit of those who would
approach this NG looking for help on the same topic. It's
only valuable to those individuals until the thread is dumped
from the server unless they look in the Google archives and such.

Anyone with any sense uses groups.google for the entire collection
of what has been posted on usenet over decades now.
I figure if I can help one person, I've done something worthwhile.
To those who moderate this newsgroup

No such animal, its an unmoderated newsgroup.
or participate regularly over a long period of time,
I say "I'm sorry" because I can understand that
from their perspective, repeated approaches to
the same topics can become tiresome.

I dont recall the din connector being discussed with
removable drive bays. There has been quite a bit
of discussion of removable drive bays in general tho.

If some arent interested in reading discussions on
removable drive bays, they are welcome to not read them.
Actually, I'm not sorry. I post with pride

Isnt pride supposed to be a sin ? |-)
because I know that somewhere, someday, someone
might benefit from the things I have learned the hard way.

They may well indeed. Thats one big downside of the
groups.google/usenet approach, no one will ever know
how many answer their question without bothering to ask.
Wow. Electronics has been a hobby of mine for over 20 years

Its been mine for much longer than that, more than twice that in fact.
and I have discussed theory with technicians and engineers alike.

Maybe some of those mentioned it in a way that you didnt understand.
Or they dont know much about electrical transmission lines.
I've got to say, however, I can't make any sense out that statement.

Electrical transmission lines produce reflections on signal
edges at discontinuitys in impedance of the transmission line.
Thats the reason electrical tranmission lines are terminated
with their characteristic impedance, to avoid reflections.
That is the reason for termination with SCSI.

While the ATA system doesnt use terminated lines,
its also the reason that the most recent variants have
the only drive on a ribbon cable on the end connector.
Otherwise you do get reflections off the stub which
is the end connector that doesnt have a drive connected.
My IDE drives just won't quit.

Yeah, mine dont either, but I prefer to use a system which
adheres to the formal standard so you dont end up with
fangs in the bum when you increase the speed of the system.
I've got SATA ports on the MB but there's no reason for me to upgrade
just to be able to say that I've got SATA drives in my computer.

No one was suggesting you do that for that reason.

The reason to use SATA with removable drive bays is because the
system overall is standard compliant. Those removable drive bays
with din or centronics connectors flout the ATA standard and thats the
reason they can and do bite you on the bum like happened with you.
A lot of people (probably you included) jump to SATA drives

I didnt, actually.
and don't even know why. They just
do it so they can say "I've got SATA."

They are completely irrelevant to whether SATA is the way
to go *if* you are using removable drive bays and want to
retain the advantages of ATA over USB or firewire etc.
In a practical sense, there's not much to gain.

There is with removable drive bays.
Current hard drives can't yet saturate an IDE channel with
sustained reads or writes so there's no need for me to upgrade.

It aint just about the speed.
The only "speed" benefit of a SATA is just a brief fraction of a
second when data is "burst" to and from the memory chip in the
hard drive. Once that chip is saturated, everything slows down
to the speed at which data can be electromechanically transferred
through the heads to and from the spinning platters. Until hard
drives are packed densly enough and spin fast enough to provide
a sustained read/write speed in excess of 100MB/sec or
133 MB/sec, there's not much to be gained with a SATA channel.

Wrong with removable drive bays. The big advantage is that the
SATA standard is adhered to. Its flouted with any ATA drive.
IDE is 100 (or 133). SATA is 150.

Not with SATA II.
For the brief fraction of a second in time that I'd gain even for
a burst write, it's not worth it to me to go to SATA right now.

You clearly dont understand the whole point of standards.
I still boot to DOS, W98SE, Win2K, various Linux distributions, as
well as WinXP. Only XP and some Linux kernels are fully satisfied
with a combination of SATA and IDE hard drives connected at the same
time. I also boot from a secondary hard drive sometimes. All these
things make a SATA + IDE combination difficult to manage, I think.
Nope.

I know that for what little gain there is, it's not worth scrapping
all 6 of my IDE drives for SATA just to say I've got SATA.

You arent doing it just to say you've got SATA. You'd be doing it
with the removable drive bays so that you arent flouting the standard.
If I try to combine SATA HD's and IDE HD's on
the same MB, I'm going to add problems with some
of the other OS's in which I play and experiment.

Not necessarily, depends on whether you want
to use the removable drive bays with those.
It's not worth those problems just to gain
150 - 100 = 50MB/sec for a fraction of a second.

That isnt the reason for using SATA.
Now if I was building a killer gaming machine or if I was building a
server that would manage lots *.gif, *.jpg, *.js, *.txt files and other
tiny files, I think SATA drives would be beneficial to some small degree.
Nope.

Since my main computer usage involves video editing
where I transfer mostly huge files over long-term sustained
reads and writes? SATA offers very little benefit for me.

Wrong again, it eliminates any problem with removable drive bays.

You may well end up with fangs in your arse with those din connectors
in the future, because they STILL flout the ATA standard. They just
dont flout it as badly as centronics connectors do.

And you havent tested the din connector bays properly either.
 
T

Timothy Daniels

Rus said:
For years I used removable drive bays. Then, I upgraded MB, CPU, etc.
and suddenly, my hard drives would freeze or corrupt data. I connected
directly to the hard drives and everything was fine. Obviously it was the
removable bays that caused me problems.

The bays I used had "Centronics" connectors. That's all there was at the
time I bought those bays. I supposed that it was those 25 relatively large
and unshielded copper ribbons radiating electromagnetic transmissions
that subsequently caused "cross-talk" and corruption.

I searched and searched. Finally, I found some IDE drive bays that use
"DIN" connectors.

I'm back in action again. The "DIN" connections work perfectly. I've tested
over the last couple of days doing all sorts of big and small data transfers
in several OS's. Absolutely works perfectly. If I was going to encounter
corruption again, it would have happened by now.

So, if you've had the same problem, look for removable IDE hard drive
bays that have either "32" or "101" in the model number. They will have
the "DIN" instead of the "Centronics." There's a bunch of brand names
on these bays. They all seem to use the same model designations.
Any brand is probably as good as the next. The "32's" are cheaper and
mostly plastic. The "101's" are a bit more expensive but they have
aluminum trays and aluminum rails.

I could only afford the "32's" but they work fine.


I've been using Kingwin's aluminum removable drive trays for PATA
drives for almost 3 years, now, and I've had no problems. They range
between $15 and $25 for the assembly, and between $11 and $16 for
extra trays. The flexibility and convenience that they provide is amazing.

For PATA drives:
http://www.kingwin.com/pdut_Cat.asp?CateID=25

Also for SATA drives:
http://www.kingwin.com/pdut_Cat.asp?CateID=47

*TimDaniels*
 
J

Jonny

If you're not seeking approval, why the big long explanation. Just leave it
alone is usually best choice, especially when phishing is suspected.
 
R

Rod Speed

Jonny said:
If you're not seeking approval, why the big long explanation.

He got his nose out of joint when it was pointed out
that some of what he suggested in another post
couldnt produce the symptoms the OP is seeing.
Just leave it alone is usually best choice,

No it isnt. You wont learn a damned thing that way.

In this case what reflections on cables are all about.
especially when phishing is suspected.

Only by those who jump at boogeymen.

 
F

Folkert Rienstra

Rod Speed said:
You've added some worthwhile observations anyway.




Its more likely to be the electrical discontinuity causing
reflections off that bit of the total connection between
drive and motherboard controller.





I think it makes more sense to use SATA drives in SATA format
removeable drive bays now,
those dont flout the ATA standard.

The ones that don't, dont. The ones that do, do. (Yeah, duh.)

There are those, probably the majority, that just use another proprietary
convertor than the IDE ones and those obviously flout the SATA standard.

http://lib.store.yahoo.net/lib/cooldrives/IMG043155.jpg
http://lib.store.yahoo.net/lib/cooldrives/IMG04244.jpg

Only those without a proprietary convertor -where the drive plugs directly
into a SATA backplane in the case chassis part- are in accordance with the
standard.

http://lib.store.yahoo.net/lib/cooldrives/P512197566.jpg
http://www.cooldrives.com/morafor3sahd.html
That format supports hot swap too.

Only the latter ones.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top