This is an addendum to the "Storage" section, so please read that first, since it contains info you will need in order to understand this section.  Here we will cover the installation and replacement of hard drives.  CD-Rom drives and floppies are installed in a similar fashion, except they are not formatted, nor are files copied over.  

Preliminary Details

Lettering of Drives

DOS assigns letters to your drives, starting with A, and Windows follows suit.  The first two drives, A and B, are reserved for floppies.  About 10 years ago, the world was changing from the old 5 1/4" floppies to the newer 3 1/2" floppy.  Newer PC's needed to be able to read both types, and usually came with a 3 1/2" A drive, and a 5 1/4" B drive.  Today, since the death of the 5 1/4" drive, almost no one has two floppy drives anymore, and B is basically nonexistent.

When your system boots, the BIOS checks for hardware - let's assume you have two hard drives, each is partitioned into two logical drives.  Upon turning on your PC, the following occurs before Windows ever starts up :

• the system boot process begins, and finds the 3 1/2" diskette drive and assigns it the letter A
• looks for a second floppy and not finding one - retires the letter B
• looks through the primary IDE channels for hard drives and takes into account any partitions they may have by reading the MBR sector (Master Boot Record)
• the BIOS assigns the partitions letters starting at "C", "D", etc. (CD-ROM drives found are not assigned a letter).  In this example, your partitions are assigned the driver letters C, D, E, F
• your system startup files, config.sys and autoexec.bat, load CD-ROM drivers which assign it the next drive letter available - in this case, the letter "G"

How many Hard Drives ??  Two !!

Most PC's that I have seen use either one or two hard drives.  Unless you are a very casual user, you will definitely want two hard drives.  As mentioned in the storage section, you will want to partition your drives into two partitions.  The outer partition gives you blazing speed for applications and games, and the inner 2nd partition is a slower area, for storage of files such as videos, music (MP3's), etc.

Windows resides on the first partition of your first drive, and all of your Microsoft applications (such as Word, Excel, Outlook) deal very heavily with Windows operating system.  Therefore it is best to keep the applications on that partition  .  .  . your "C" drive.  You will also want to keep graphics applications, such as Adobe Photoshop, on the C drive, because they require speed due to the huge file sizes that are often encountered.

A second drive is needed, because you will have other files and programs that require fast speed - in addition to your applications.  This would apply to space hogs that require speed, such as AutoCad, and graphical games, such as Quake or Motocross Madness.

There is another advantage to two drives - you can reserve the 2nd partition of the 2nd drive as a backup area.  This is ever so much faster than using tape or CDR's to backup data.  In addition, the retrieval of lost data is fast.  Drives never fail in pairs . . . therefore, if your primary drive fails, you have all your important data backed up.  You can write simple batch program to perform the backup, or do it manually.  You will want to backup important documents, tax data, Quicken, images, resumes, receipts, and important projects. If your primary drive crashes, Windows and all the Applications can be reinstalled - so do not back those up.

NOTE: you can have up to 4 or more drives - but there is really no need.  If you feel you need more space - today's drives have capacity's of up to 72 GB.

Drive Cables (Ribbon Cables) :

There are two flavors of cables used for EIDE vs Ultra DMA.  The ribbon cable has a female connector on both ends - one plugs into the Motherboard interface and the other into the drive.  There is a third connector in the middle, which is used for the 2nd drive if one is installed, or a CD-ROM drive.  The middle connector is usually used for the slave device, but they can be switched.  It is the jumper (mentioned before) that dictates which device is Master and which is Slave. The Ultra DMA/66 is still a bit new, so the vast majority of you will have the 40-pin connector at the end of your hard drive ribbon cable, not 80 pins.

Here is a diagram of the two Hard drive cables :

The 80-pin Cable Myth -  many people think that the 80-pin cable has twice the number of data lines, and therefore can transfer 32 bits at a time, instead of the 16-bits that a 40-pin cable transfers..  It does not !!  Actually, the 40-pin cable has trouble with cross-talk between the closely spaced wires at high speeds, and therefore has errors at 66 MBps.  The 80-pin cable uses a grounded connection for every other pin, to eliminate cross-talk.  So functionally - it works 
identically to a 40-pin cable and can only transfer 16 bits at a time !!!

Note that there is a small tab in the middle of the plug - this prevents the plug from being inserted backwards.  Quite often, then end that plugs into the motherboard has no tab, and you have to figure out the correct orientation by trial and error.  DO NOT WORRY - of you insert the cable in the wrong way, no damage will occur (been there, done that - many times).  The following image shows the connectors on the motherboard which the cables plug into.  The two on the right are the Primary and Secondary IDE channels for the hard drives - the shorter one on the left is the floppy IDE connector :

Quite often, a ribbon cable will be too short to accommodate a new 2nd drive you have just purchased.  The original manufacturer cables are pretty skimpy when it comes to length - so you may need to buy a longer one.

Hard drive vs Floppy Drive cables and connectors :

Here is a close-up of the actual cables and connectors (plugs) of both types of ribbon cables :

Ribbon Cable Orientation:  each ribbon cable has a "hot lead" on one end, which is colored red.  The ribbon has a faint red strip running all the way down it's entire length.  In all cases, the cable plugs into the drive with the red side facing the 5-volt power plug. 

For this section, we will stick to a couple of basics and leave it at that. Most common is the EIDE (Enhanced IDE), and SCSI (Small Computers System Interface).

Standard EIDE Hard Drive / CD-Rom Configurations

As mentioned, there are two EIDE ribbon cables which plug into the motherboard.  Actually, there is a third ribbon cable, for your floppy drive - which is easily differentiated from the hard drive connectors on the motherboard - it is not as wide.  The two connectors on the motherboard are usually marked "Primary" and Seconday".  If you only need to use one - ALWAYS USE THE PRIMARY.  The following image shows two common configurations for the hard drive/s and the CD-ROM drive.  The first configuration can also be connected up using only the primary IDE cable, setting the hard drive as the master and the CD-ROM as the slave (which is also very common).

The following diagram looks as if each configuration has one long ribbon cable - actually there are two cables.  It looks like one because the connectors on the motherboard are right next to one another.

A lot of people (including me) place both hard drives as the Masters, using both primary and secondary - and the CD-ROM as a slave on the Primary channel.  The picture shown above is from PC World - who I tend to believe over myself, and from now on, I will place both drives on the primary channel.

Cages and Rails - most PC's come with a cage that allows a second drive to be installed without any special equipment.  Simply slide the drive in, and you will see some slots on either side, that screws go into.  The screws come with the hard drive.  If you already have two drives and are installing a third - it is possible that you may have to use one of the large, 5 1/4 " bays that were made for CD-ROM drives.  If so, you will need to purchase a special set of "rails" that extend the outer edges of the hard drive outward, so that it can be mounted.  The hard drive is only 3 1/2 " wide.

DRIVE INSTALLATION

1. decide what drive you need

2. physically install the drive (as the D drive in this example)

3. create a boot diskette

4. partition the drive

5. format the drive (if just adding a drive, that's it)

    If Replacing your C drive with a New Drive :

5. copy all files from old drive 

6. swap drives (C becomes D, D become C)

During my years as a network Integrator/LAN Administrator, and then later as a PC repairman - virtually everyone I ran across, had a PC with a hard drive that was almost full.

Decide What Drive You Need

There are several situations that can arise where you will need to buy a hard drive.  Of course, you are low on space.  If you have a very small or old drive I would recommend throwing it away once you install the new drive.  Older drives are notorious for crashing, which will completely devastate you - it has happened twice to me, and it is no picnic.

OK, now it's time to think big.  I don't care if you are a penny pincher - think BIG.  You really don't want to go through this experience next year, do you?  I thought not.  Get at least 20 GB, which with today's prices should run you around $200 or less.  Look for a rebate.

I mentioned I have had two drives crash - one was a new Fujitsu (bad mistake) and the other was a Conner drive (they have since been bought out by Seagate).  I had another drive go bad on me but I was able to salvage the data since it still operated marginally - ot was a Western Digital.

I have 3 PC's with a combination of IBM Deskstar's and Maxtor.  Both are excellent and I would highly recommend them.  In my opinion, the best drive is an IBM Deskstar.  They are fast, quiet, and they last forever.  A bit pricey though.  The Maxtor drives used to be of questionable quality, but they have improved them dramatically, into serious contenders.  Even Seagate and Western digital are now making drives that will last many years - but go with a Deskstar, what a drive !!

Caution - stay away from off brands and new entries into hard drive manufacturing, such as Fujitsu and Hitachi.  

Here are a few situations, (assuming you are out of space or close to it) and what I would recommend.  Remember, old drives are a liability and should be discarded.  Each place in the table where I recommend keeping a drive, I am assuming it is 3 4 years old or less :

NOTE:  in all cases where I recommend discarding a drive - you will of course, need to save the data and then wipe the drive clean.

Is my old drive OK ??  Tough question, but there are a few factors to consider :  

1. Age - If it is 5 years old older, it is generally a risk for failure.  

2. Bad Sectors - especially with older drives, run a surface scan using Norton Utilities, or Microsoft's "Scandisk".  If you find bad clusters - one or two is usually acceptable.  At that point, if the drive stays at that level with one or two bad clusters, over the years - it is OK.  If it begins to experience additional bad clusters, and over the years goes from 2, to 4 or 5, and then to 7 or 8  .  .  .  the drive surface or the read/write head or mechanism is decaying, and definitely - replace it.

3. Speed - if you drive has slowed down considerably - it is bad.  The drive should never, ever, slow at all. Be careful though - your system may slow due to fragmentation or lack of space on the drive.  Therefore, make sure to defragment and delete some stuff first before determining that your drive has slowed.

4. Sound - an experienced technician can detect a bad drive by the sound it makes as it grinds and chatters it's way through searching, reading, and writing data.  There are two types of "good sounds" and they are difficult to describe.  The best sound is a quiet yet solid clicking.  Another sound which I have found on even new drives - sounds bad, but it's fine  .  .  .  it is a louder, but very solid "crunching" sound.  These drives must have a larger, more powerful stepper motor which move the head to and fro.  The bad drive will have a tinkling sound, or an exaggerated and slow loud sound.  The tinkling almost sounds like tiny wind chimes, or Christmas tree bulbs touching one another - that is the only way to describe it, and I wish I had a bad drive to record the sound for you.

Physically Install the Drive

• Figure out which is C and D
• Figure out where to mount the new drive
• make sure cable reaches drive
• set Master/Slave jumper
• mount the drive
• connect the cables

If Replacing a Drive - figure out which is C and which is D

If you have two drives and are replacing one of them, it will be difficult to tell which is C and which is D.  The cable connection does not reveal this - either drive can connect to either the end plug or the middle plug.  The only way to tell upon inspection is to see which one has it's jumper as the Master, and that will be the C drive - but it is dark and you will be hard-pressed to see the lettering on the drive that tells you which pins are which.

Therefore, to find out, simply unplug the ribbon cable from one of them (while the PC is OFF ! ), and try to boot.  If you are able to boot, you have just unplugged the D drive.  If you are unable to boot, you have just unplugged the C drive.  Regardless of which drive you want to replace, at this point :

• remove the old D drive
• set the small jumper on your new drive to configure it as "slave" - refer to the drive manual if the 
  Master/Slave" designation lettering on the drive is not clear (this happens a lot)

Figure out Where you will Mount the New Drive - if you are replacing a drive - then the new one simply takes it's place.  But, if you only have one drive currently and this will be an additional drive, you will need to locate the extra drive bay which comes with just about every PC.  It will be a metal cage that will exactly accommodate a 3 1/2" drive.  Sometimes there is an additional bay near the power supply.  Sometimes there is on near the front at the bottom next to the cooling fan.  Oftentimes, the only place to place an additional drive, is right next to the existing drive - which is usually directly above or below the floppy drive.

It is convenient for manufacturers to place the hard drive there, since the floppy drive is the same size (3 1/2 ").  They simply place a cage there that has 3 slots  .  .  .  1 for the floppy, and two for hard drives.  Right away you can see the difficulty in inserting screws on the one side facing the frame.  Sometimes you need to remove the cage from the frame, mount the drive, and then re-attach the cage  :  

Be careful here - drives can run rather hot, and you want to have some space between the two drives for air flow.  The bay shown, will butt the drives right against one another.  In this case, you would not want to fill the bay with the floppy and 2 drives.  Instead, get a rail kit, mount the floppy at the top of the bay and one drive at the bottom, and then mount the second drive above, in one of the CD-ROM bays.

This photo shows how a hard drive can be mounted in the large CD-ROM bays if necessary, but purchasing a "rails kit".  The rails extend the sides out to the edge of the large bay, where there are holes to accommodate screws :

DON'T MOUNT THE DRIVE YET (check cables first - see below) !!

IDE Ribbon Cables

ATA-33 Cables

ATA-33 hard drives and CD drives use a 40-pin cable while Floppy drives use 34-pins.

Hard Drive and CD drive ribbon cables normally come in either 18" or 24" lengths and have 40-wires, and three 40-pin plugs on them.  Some new computers come with one hard drive and a cheap, 2-plug cable.

ATA-66 ATA-100 Cables

ATA66/100 flat ribbons have 40 pin plugs, and 80 conductor wires. The higher speed means the length should not exceed 18".  The extra 40 wires are ground wires in between the original 40 to cut down on cross-talk noise.  The wires are signal/ground/signal/ground/etc.

Round Cables

Round cables are the new hot ticket item.  Some say they do not insulate against crosstalk as well as ribbon cables.  Some people actually slicing their ribbon cables apart, wire by wire, and twist them into a round cable.  The thinking is that the ribbon cables impede air flow, but it's such a slight difference that it is not worth it.

Make Sure Your Cable Reaches the Drives - Just realize that virtually all computers come with two EIDE drive connectors with ribbon cable, each cable will support two drives . . . a primary, and a secondary.  If you want SCSI you generally have to buy a separate SCSI cards which mounts in one of your computer's PCI slots.  Therefore, 99% of you will have  the EIDE drives in your system.  The controllers are in the drives themselves, and the ribbon cable connects the drive to your motherboard.

NOTE:  today, round cables are becoming popular

The image shows the hard drive cable has 3 connectors.  One of them, the one that is farthest away from the other two - plugs into the motherboard, and the other end, as well as the middle - plug into the two drives.  Before you even mount the drive, plug the cable into the motherboard (with system turned OFF) and move the plugs up where they must go - to make sure they will reach.  If they do not reach, you will have to either find a new spot to mount the drive, or go to the computer store to buy a longer Ribbon cable.  The stock cables are notorious for being too short (in order to save the manufacturer 1/2 of a cent !!).  

Orientation - the Hard drive cable plug has red strip painted on the hot lead (for pin 1) and a small tab in the middle, which prevents the cable from being inserted the wrong way into most hard drives.  Some motherboards do not have a plastic shell around the pins, which like hard drives - prevents wrong-way insertion.  For these motherboards, look for a pin number 1 on the board, and that will be the red lead on the ribbon cable. If there is no way to tell, simply try both ways - you will not hurt the drive or motherboard, as the manufacturer's already though of this possibility and built in protection for that.

 Most systems will only allow the hard drive cable plugs to be inserted one way.  BUT - they never did that with the floppy cable.  You can tell which way is correct on the floppy drive connection itself - the positive (hot lead) of the ribbon cable on one side is colored red - and that faces the power plug of the floppy drive.  But, on the motherboard it is more difficult.  If you are lucky, the motherboard will have a small, "1", printed on it to denote the hot lead, which is pin 1.

Floppy cable - unlike the hard drive cable, the floppy cable has a half-twist in the middle - that is how the system differentiates from A drive and B drive (floppy drives have no Master/Slave jumpers to differentiate them).  But these days, the middle connector is almost never used for another floppy, since B drives have gone the way of the dinosaur.  Instead, the middle connector can be used for a Tape backup drive.

The ribbon cables can also connect other IDE devices such as CD-ROM drives or Tape Drives for backup (most people simply refer to EIDE, as "IDE").  

Floppy Drive Cables

Actually, "Master" and "Slave" are terms which only apply to IDE hard drives, not floppy drives. The settings on a floppy are Device 0 or Device 1.  

Old cables had three plugs with a twist just before the last, end plug (see image above) that were used for both the 3.5" and 5.25" floppy drives.  ALL 3.5" floppy drives available today have only one drive plug and come pre-set to Device 1, which is the 3.5" drive.  They may still have the twist in the cable, since the motherboard pins support both Device 0 and 1. 

Hard Drives Master/Slave/Cable-Select Jumper

The old MFM/RLL drive controllers could support two drives, just like the new ATAPI/IDE controllers.  However, the drives had no jumpers.  Instead. the cable and controller used the same technology as floppy drive cables - they had a twist in between the middle and end drives.  The second drive had to go at the end connector with the twist.

Today, each Hard Drive and CD-ROM has some very small plugs on the back, called "jumpers", which you can remove and change the position of - to make the drive a "Master" or a "Slave"  .  .  .  or "Cable Select".  A jumper is a small plastic rectangle with two metal tubes that are shorted together inside.  When you place a jumper over a set of 2 pins, the pins slide up inside the tiny tubs and are shorted together (see pic below).  When the jumper shorts the two pins, it is considered in the "ON" position.  When the jumper is not connecting the two pins it is considered "OFF".  To place a jumper in the OFF position you can remove it, but instead, simply place it over one of the pins, in case you need it later.  Jumpers are tiny and once removed - they are as good as lost.  If you do not have long-nosed pliers, you can use a small flat-blade screwdriver to pry the jumper out from the bottom edge.

Master/Slave Settings

If you use the Master or Slave jumper setting, you will need the normal IDE, 3-plug, 40-pin ribbon cable, which all plugs are the same color (usually grey, but some are black or light blue).  It does not matter A CD-ROM drive is shown below with the jumper to select it as Master or Slave.  On hard drives the lettering is less obvious - you will only see CS, SL, and MA listed, and sometimes just C, S, and M.  

Note: the audio Out on the CD-ROM (hard drives have no audio out connector) -  this is often neglected, but requires a special cable that comes with new CD-ROM drives, and connects to your sound card (to the last audio amplification stage).  If this connection is forgotten, you will have system sound which feeds into the earlier amplifier stages, but will not be able to hear any sound with audio CD's.

I mentioned that each ribbon cable can support two drives - it has two plugs on it . . . one at the end, and one halfway down the length of the cable.  If you have only one drive, then you will set the jumpers so that the drive is a Master.  If you have two drives on the cable, then you set the jumpers so that one drive is a master, and one is a slave.  The drives will not operate if you fail to do this.

Cable Select

The "Cable Select, or "CS" option is rarely used - STAY AWAY FROM IT UNLESS YOUR SYSTEM IS ALREADY PRESET WITH IT.  It means that the position of the drive on the cable, "selects" whether the drive will be Master or Slave.  The end Black plug is Master, and the middle Grey plug is Slave.  The motherboard plug is normally Blue.  99.9% of all computer system are pre-configured using the Master/Slave jumpers.  You can still use CS on any ATAPI/IDE compliant system, but why?

Placing the jumper over the two CS pins, allows you to flip the master and slave drive by simply switching which plug goes into which drive.  Only special cables with a Master plug and a slave plug will work with a CS setting.  The cable must be an ATA If your Master drive is jumpered as CS, then your Slave drive should be jumpered as CS. Cable Select means that the relative position on the drive cable determines whether the drive is Master or Slave. In this case only, the ribbon cable that connects to the Master drive should have a black plug on it and the end of the cable that connects to the Motherboard should have a blue plug on it. The grey plug is connected to the Slave drive.

Cable Select uses the presence/absence of a signal on line 28 of the cable to identify the Master or Slave drive. To use this, the hard drives must be jumpered for "Cable Select", and the Slave drive is plugged into the cable connector where line 28 is absent (the middle plug).

ALL equipment that calls itself "ATAPI/IDE compliant" has to support Cable Select, since it's in the Standard. Ditto for Jumper Select, which is part of the Standard too. 

You can always use a Cable Select cable in a Jumper Select application, but not the other way around. In a Jumper Select application, line 28 is ignored. The other line that's always ignored is line 40. This connector position is used for keying, and may or may not be plugged to prevent incorrect insertion of the connector.

On Compaq equipment, you can convert from Cable Select to Jumper Select. Simply set the jumpers on your IDE drives as required. You can even use the same IDE signal cables. This should be possible on other manufacturers' equipment, but "YMV".

In order to be truly compliant with the ATAPI/IDE specifications, you should always plug a single drive into the end of the cable. This is intended to prevent signal reflections in the IDE cable by avoiding unterminated cable stubs. If you do this, you can use IDE cables up to 40" long instead of the very short 28" cable mentioned above. If you use a Master and a Slave drive on the same cable, and you use Jumper Select, the position on the cable of the Master and the Slave drives don't matter.

With the "improperly made" Cable Select cables that I've seen in some equipment, the "Master drive at end" rule obviously can't be implemented.  In these cables, line 28 is simply cut between the middle and end connector. Not good -- missing line 28 identifies the "slave" connector.

A "properly made" IDE cable that supports Cable Select looks like a normal IDE cable, except that the connector in the *middle* of the cable has pin 28 removed before the connector is clamped to the cable. These cables should also be clearly identified as to which connector is the controller end, which is the Slave, and which is the Master. Furthermore, pin 40 of
the Master and the Slave connector should be plugged so that only the "controller" end can be plugged into the motherboard (assuming that the male connectors on the motherboard do NOT have pin 40 missing).

Mount the Drive - slide the drive into the Bay, and insert the screws that come with the drive.  The discount drives often, and infuriatingly, come bare in a box with no screws - so you will need to make yet another trip to the computer store.  Do not try to insert the plugs before you mount the drive - it can make it impossible to maneuver. 

The screws, like the cable, can be very difficult to get in.  If it is extremely tough - just insert two of them - but you must have one screw on either side.  The right side is the hardest, because it is often blocked by the frame of the PC, with just a small cutout to access through.  Well, OK, if the drive is simply too hard to access on one side, you can mount 2 screws on the other side and leave it - but just make sure the screws are in very tight - ONLY DO THIS IF THERE IS NO OTHER WAY.  Personally, I will dismantle the PC to get both sides secured.  You really do not want the drive to be able to vibrate - that shortens it's life.

The following image is from a Dell system with one floppy (top) and two drives (middle and bottom right) - connected as Master and Slave on one ribbon cable.  Of course, it is the Primary IDE channel.  You can have additional drives on the secondary channel, but you never use it for your boot drive.  The drive bay on the bottom right is unusual, but there are many unusual drive bays - depending on the case.

Note that there is some space between the floppy drive (top) and the hard drive, which allows air to flow between the two drives for cooling.  Also note that the ribbon cable is stretched just about as far as it will go.  This is why you always want to check the cables first - to make sure they will reach.

Connect the Cables - When you install a new drive, there are two connections to make . . . the ribbon cable, which the data flows through, and also a 5-volt power cable from the Power Supply (see the Case section).  Both cables will only go in one way :

If there are no 5-volt plugs available, you can purchase a "Y-plug" that takes one plug and turns it into two.  You should have plenty, unless you are dealing with an older, or a cheap system.

NOTE: the image above is a bit of a fairy tale.  You never have open space.  Inserting the ribbon cable can be very trying indeed - you may have very little room to work with and it is hard to see. Also you are pressing in sideways, and it is very awkward. Shine a flashlight in to make sure you are inserting the cable in the correct orientation (look for the tab in the middle).  Just be patient.  Do not "force" the cable in. !!! That can damage and bend the tiny pins.  Been there, done that.  The cable does take some force to go in, but it is just a moderate amount of force. 

The floppy drive power plug - this is a smaller, 5-volt plug, which thankfully (unlike the floppy ribbon cable) - only allows you to insert it one way.  Here is a picture of the floppy power plug, which you will find bundled along with the hard drive plugs :

Create a Boot Diskette

The standard boot diskette that Windows supplies you with may or may not have all the files you need.  I have found that most people have lost that diskette, and even if they have it, they need to add CD-ROM drivers and a few other files to it.  There are a few CD-ROM drivers that work with the vast majority of CD-ROM drives.

Instead of a long explanation - I am providing you with some files called "images" which, when run, create a bootable diskette with some additional files.  You will want to download the image file that matches your operating system (sorry, no Mac, NT, or WIN2000), and the Boot Files zip as well.  Place a blank diskette in the floppy drive and run the file - then add the three additional files that I am also proving (writing over the existing files on the floppy) - an, viola !!

Boot Diskette Creation Image Files :

Win95 A (Standard)    Win95 B (OSR2)    Win98    Win98SE    WinME

The Three Additional Files to Place on the Boot Diskette

These three files need to be copied onto the disk . . . config.sys, autoexec.bat, and teac_CDI.sys (writing over the existing ones).  The TEAC file is a basic, CD-ROM driver that seems to work with 95% of all CD-ROMS.  You will need to access your CD-ROM upon  startup, in case you need to reinstall Windows :

Download the 3 basic Boor Diskette Files

NOTE: here is a list of files that you need to have on any boot disk - just in case the image files are missing anything (you can just copy the missing files from your C drive to the boot diskette):

AUTOEXEC.BAT
COMMAND.COM
CONFIG.SYS
DELTREE.EXE
ATTRIB.EXE
EDIT.COM
FDISK.EXE
FORMAT.COM
MSCDEX.EXE
SCANDISK.EXE
TEAC_CDI.SYS
XCOPY32.EXE
HIMEM.SYS
SYS.COM

How to make your own Boot Disk from Windows :

Control Panel/Add-Remove Programs . . . click on "Startup Disk" tab  . . . click on "Create Disk" button

Partition the Drive - use Fdisk

*** use extreme caution ***

Description of Partitions

The new drive is an empty platter - just magnetic metal with no areas defined.  No cylinders or tracks - no magnetized South and North bits - no sectors.  Partitioning changes that and sets up boundaries on the drive.

You must partition the drive, into one or more partitions.  Use two partitions on each drive - absolutely, use two.  I prefer to use one partition for the C drive on system that have two drives - and will explain that shortly.  When you partition a drive, it obliterates all the data on the drive.  This is no problem for us here, though, since this is an empty drive.  

Partitioning sets up the drive to use the file system that your operating system can use to write data.  Partitioning also allows you to split a drive into multiple logical drives if you wish - multiple partitions. Its creates the Master Boot Record (MBR) on the new drive. The MBR stores the partition table as well as a special program or programs your operating system needs in order to boot your PC. In other words, it contains the important data that tells your OS how to store data on the drive and retrieve data from it.   The partition table itself is simply a record of the beginning and ending sector of each logical disk that you have created when you have partitioned the drive. It does not define all the millions of sectors in between, so the actual process of partitioning a drive is very quick.

Partitioning also creates one sector at the beginning of each volume (a volume is another name for partition, taken from IBM Mainframe lingo) - called a "Volume Boot Sector".  This sector has information about the disk and the volume parameters. 

Primary Partition, Extended Partition, and Logical Drives  - there are just two types of partitions.  The Primary partition, and the extended partition.  The Primary partition is assigned to the C drive, and you can only boot from a primary partition. For your first disk, usually called "Disk 0" (but Partition Magic calls it "Disk 1"), if you have multiple drives on it you will normally do the following:

Use Fdisk to create 2 physical partitions - one Primary Partition which is auto-assigned as your C drive, and one Extended partition which initially has no drive letters assigned - but Fdisk will automatically prompt you to create one or more Logical drives in the Extended partition.  The extended partition takes up ALL of the space leftover after you configure the Primary partition.  By default, the primary partition is it's own, separate physical partition and takes up to outer portion of the drive platters (there are multiple platters).  The extended partition is also it's own, separate physical partition and it takes up the inner portion of the drive platters, which is slower.  The farther you move inward on a drive, the slower it is because less surface area is moving under the heads.  Now, the primary partition is always just one drive - your C drive.  The extended partition is one or more drives.  If you have 2 drives total then each physical partition has a single drive.  If you have more than 2 drives to go onto this disk, then the extended partition will contain multiple drives.  These are "logical drives", and they are formed in some complex manner throughout the platters of the extended drive.  They are NOT separated out physically, as in "drive D is the outer portion of the extended partition platter space, drive E in the inner portion of the extended partition platter space, etc.  I do not know how they are distributed, but they are not done as physical partitions are done . . . from outer to inner on the drive.

Fdisk and Large Drives - Fdisk, even the Fdisk that comes with Win98, will work on HUGE drives !!!  BUT it will list odd amounts of space that are completely inaccurate - basically as far as presenting the amounts of capacity and space on the drive, the partitions, and the Logical drives - it chokes and lists very odd numbers.  HOWEVER - IN REALITY IT WORKS FINE SO LONG AS YOU TYPE IN PERCENTAGES FOR THE AMOUNTS OF SPACE YOU WANT TO ASSIGN !!!  I would not trust it to understand large numbers inputted by you (i.e. you can input the number of bytes instead of % space - but don't do it).

It is just the display that is off.  For example, I took a 400 GB drive and partitioned it into one primary, one extneded, and 4 logical drives in the extended.  It listed the drive capacity of the extended which was actually 350 GB, as 8085 MB, which is only 8 GB !!!  Then after the logical drives were created, it listed the capacities as follows:

Fdisk Displayed Logical Drives in a 350 GB Extended Partition
MB is all wrong, while each Usage % is correct

NOTE:  Win98 has a limit of 127,5 GB per partition.  However, WinXP's limit is approx 2 Terabytes (you will never need that much space !!). 

The Drive Letter Shell Game - a common mistake, is to partition the D drive into two, and making the first partition a primary partition.  If your C drive had two partitions (C and D), next time you boot, the system will give precedence to the two primary partitions, and will give the first drive's primary partition the Letter C and the 2nd drive's primary partition the Letter D.  Now your old D drive has been renamed to E.  (see "After 1" in the diagram below).  If your first drive has only one primary partition, this problem will not occur - but it sure will if you ever repartition it into two. The fix:  the second hard drive should be partitioned as ONE EXTENDED PARTITION - and then two logical drives, E and F will be created on the one partition.  Now your first drive has one primary partition, labels it C, and keeps the it's extended partition labeled as D.  The second drive has no primary partition, and just one large extended partition - which can then be broken into two logical drives.  Since the second drive has no primary partition, it does not compete with the first drive, and politely takes the next two available letters  -  E and F - for it's two logical drives (see "After 2" in the diagram).

The Age-Old "Drive Letters Shell Game" Problem
with the easy Fix (shown in the "After 2" configuration)

Perform the Partitioning

There are two methods :

1)  Fdisk - you can use the boot disk we created, and run "Fdisk" - this is my own perferred method because Fdisk is a "pure" and low-level partitioning program with no bells and whistles.  However, it is just a tad tricky to use.  If you are a novice, either have a friend that knows how, run fdisk for you, or use the disk that came with your drive using the manufacturers "overlay software". 

If you must use FDISK (and that is what I use), you can try this link for details :   FDISK INSTRUCTIONS  you must realize that these instructions are for a single drive with Win95 Fdisk, so you will have to figure out how to modify them. Here is an even better set of FDISK INSTRUCTIONS.    To run fdisk, simply boot the PC with the boot disk, and type in "Fdisk".  For our example, where we are using the 2nd drive, and plan to partition the entire drive an one extended partition with two logical drive letters - E and F.  I think you will do fine on your own - here is the basic method 

First of - there are two possible situations you may be trying to acheive :

Situation 1)  Old Drive will Stay as your Boot Drive, New Drive is add-on.  In this case, the new drive will not be used to boot from - and to prevent the old "Drive Letters Shell Game" problem described above, you will want to partition it as one extended partition with two logical drive.

Situation 2)  New Drive will become your Main, C Drive.  In the case, the new drive will still have two partitions - but the first one must be a primary partition (to ba able to boot from it) and the second drive will be an extended partition.  The drive letters will be automaticall be added to the drive.  If your old drive has two partitions, we will initially have the shell game problem but will fix it later.

Now, let's partition the drive :

Situation 1 - Old Drive will Stay as your Boot Drive, New Drive is add-on

• start up Fdisk, and say "Y" (Yes) when they ask if you want to allow support for large drives (that simply means that it will be partitioned as  FAT32 drive - not the old Win95 FAT16).  FAT16 only allows 2 GB per partition.
• change to the fixed disk you want to partition - in this case it will be the 2nd fixed disk - make sure you do not have the first disk selected !!!!  You will lose all disk data when you partition it
• select "add an extended partition", and then it will ask you if you want the maximum size allocated to that partition - say "Y"
• once the drive is partitioned, select "Create a Logical Drive"
• it will ask you how much space you want allocated to the Logical drive - you can type in a percent (easier than  figuring out bytes)  -  50% is a good figure, but it is up to you - the first one will be the outer portion of the disk, and will be Drive E
• again - select "Create a Logical Drive"
• it will ask you how much space you want allocated to the Logical drive - you can type in a percent (easier than  figuring out bytes)  -  add the remaining portion of the drive, so that both partitions take up the entore drive
• done - now you must reboot for the changes to take effect

Situation 2 - New Drive will become your Main, C Drive

• start up Fdisk, and say "Y" (Yes) when they ask if you want to allow support for large drives (that simply means that it will be partitioned as  FAT32 drive - not the old Win95 FAT16).  FAT16 only allows 2 GB per partition.
• change to the fixed disk you want to partition - in this case it will be the 2nd fixed disk - make sure you do not have the first disk selected !!!!  You will lose all disk data when you partition it
• select "add a primary partition", and then it will ask you if you want the maximum size allocated to that partition - say "N", and instead enter a percentage to be allocated (50% is usually fine)
• once the partition is created, select "add an extended partition" and then it will ask you if you want the maximum size allocated to that partition - say "Y", and it will receive the remaining allotment of space
• the Drive letters should be automatically created for you
• done - now you must reboot for the changes to take effect

NOTE:  if you are doing this to the C drive (the boot partition) you must create it as a "Primary Partition", and then at the end - select "make a partition active" and select the C partition.  Otherwise you will not be able to boot from C drive.

2)  Overlay Software (overlays your BIOS)  -  virtually all drives come with  diskette containing overlay BIOS software, usually called either "Disk Manager" or "EZ Drive".  There are a few others as well.  They load software that acts as an overlay on top of your BIOS.  It is BIOS translation software, and can slow the drive down - although usually just a very slight slowdown.

Windows98 comes with an updated version of a program called "FDISK" that performs partitioning, and can read the LBA drives (the old FDISK could not).  I usually use FDISK because I am familiar with it and trust it completely.  The overlay BIOS that comes with your PC sometimes act strangely, but is usually fine.  One type from Quantum, forces you to keep a special diskette, and does not allow you to boot with a regular boot diskette - which I absolutely hate.  Disk Manager has been around the longest.  The very large drives (27 GB and up) may actually require you to use the translation software that comes with the drive.  I am not sure of the drive size limitations of FDISK.

Format the Drive -  Copy Files - Swap Drives

*** use extreme caution with format ***

 Definition of Format

You now have a drive with two empty partitions.  Formatting will lay down the tracks and sectors.  The sectors will all be empty - but format lays down the start and finish points of each sector.  The image below displays what partitioning and formatting does to a drive (there are many more tracks than shown, as well as multiple platters vs the one shown).  I am using C and D drive, even though in the examples on this page, we are adding a second drive and using the letters E and F :

Format Command

As we mentioned earlier in the storage section - 32 kB clusters will waste a small amount of space, but your drive will run faster.  The following commands will format both drives.  Each drive will take a while to finish formatting - the screen shows you the percent completion as it works it's way across the drive :

The Generic command is as follows :

format x: /z:64
where x= your hard drive letter, this will format 32k clusters under fat 32.

for other sizes of clusters :
FORMAT x: /Z:n where n is the number of 512 byte sectors in a cluster;
    > n must be a number 2^x, so: 1, 2, 4, 8, 16, 32, 64, ...

Example:
format c: /z:32 (for 32 - 512 byte sectors per cluster = 16 bytes per cluster)
format c: /z:16 (for 16 - 512 byte sectors per cluster = 8k bytes per cluster)
etc.

NOTE: if you do not specify with the /z switch, it will default to 4 kB per cluster if possible (unless the drive is too big, then it may have to use larger cluster sizes). The minimum size of the drive partition for the default 4 kB clusters is 504 MB. This minimum size changes if you use larger cluster sizes, as does the maximum size.

Perform the Formatting

There are two possible situations - you may want to simply add another drive, keeping your original boot drive, or you want to move your old drive down the pecking order as E and F  -  and make your new drive the boot drive  C and D.

Caution - "Low-Level" formatting -you may see this option in some drive utilities and old BIOS's  -  it only applies to old drives - never low-level format a drive !!

Situation 1)  Old Drive will Stay as your Boot Drive, New Drive is add-on

In this example, we have an empty partitioned drive, with two logical drives on it - E and F (C and D are the two logical drives on the existing drive).   Therefore you will need to run the format command twice.  When you issue the command, there are a couple prompts that you need to answer.  One asks are you sure (Yes - but double-check to make sure it is not a drive with data on it - formatting, like partitioning - removes all data from a drive).  The other prompt is at the end of the format process, and it asks you if you want to assign a volume label.  Typically, just say no - there is really no need for labels with hard drives.

• install the drive as detailed above - partition the new drive as one extended partition with two logical drives on it - which will be E and F  (C and D are taken by your original drive).  If your original drive only has one partition, C  -  then your new drive logical partitions will automaticallt be given the next two available drive letters - D and E.  
• format the two logical drives :

format e:  /z:64 

format f:  /z:64

Situation 2)  New Drive will become your Main, C Drive

OK fine - but first, you will need to initially allow the drive to be your 2nd drive, copy the files over from C, and then swap drives and Master/Slave settings.  Most people think you need to buy software such as Partition Magic or Norton's "Ghost" program (tricky to use).  But you don't - here is a very simple method - and I guarantee you - YOU WILL NOT MISS A SINGLE FILE !!

For this situation, we have to look at the two possible cases :

Case 1 - old drive is one partition, C drive 

Case 2 - old drive is two partitions - primary and extended

For case 1, your existing drive will become E, and you do not have to repartition or change it.

For case 2, your existing drive has two partitions, it will be a bit tricky, due to the "Shell Game" problem mentioned above.  You cannot change the primary partition to an extended partition easily.  If you buy "Partition Magic" you can do it on the fly.   Otherwise, the best method is to simply copy all the data over to your new drive, repartition the old drive into one extended partition with two logical drives, and copy the data back over.

But first - let's cover the process of making your new drive . . . your C drive.

• install the drive as detailed above - but partition the outer partition as a primary partition, and the inner partition as an extended partition.  Do not make the primary partition active yet (actually the system won't let you).  You must realize that if you old drive has two partitions, this will change the drive letter of your old drive, due to the drive letter shell game (see above).  So now, if your old drive has two partitions . . . your old drive is now C and E and your new drive is D and F .   If your old drive has one partition, then your old drive is now C and your new drive is D and E
• format the drive, but when you format the outer partition (now the E drive) you need to format it so that it has the bootable system files.  When you change it to a C drive it will need these system files.  Issue these commands for the two drives.  Make sure to type in " dir e:<enter>  and  dir f:<enter>" first to list all the files on the drives  .  .  .  both drives should be unreadable with the dir command - since they are not yet formatted - this means it is safe to format them.  But, if you see a bunch of files, then it is your old drive and something is wrong - you must stop and troubleshoot to see what happened. 

Case 1  (new drive is D and E)

format d:  /z:64 /s        (the /s means "system")

format e:  /z:64

Case 2 (new drive is D and F)

format d:  /z:64 /s        (the /s means "system")

format f:  /z:64

 

• start up Windows normally
• close any programs that may have autostarted
• click START/Run and type in "command" to open up a DOS box
• make sure the prompt says :    C:\>   than means you are at the root directory.  If it says anything else, such as C:\windows>  then you can go back up to the root by typing in:    cd\ <enter>
• type in the following to copy all files (except the swap file) :   

Case 1 (old drive has just one partition)

xcopy c:\  d:\  /c/h/e/r/k/y/s  (will copy all except swap file, win386.swp)

 

Case 2 (old drive has two partitions)

xcopy c:\  d:\  /c/h/e/r/k/y/s  (will copy all except swap file, win386.swp)

xcopy e:\  f:\  /c/h/e/r/k/y/s  (will copy all files)

*** You may think you need to use xcopy32, but actually xcopy does the same thing - and it will retain your long file names, surprisingly

• NOTE:  all those letters after the slash marks are called "switches" or "options".  There is an easy way to remember these . . . they spell out the rather unusual word "cherkys"
• wait for all the files to copy over - this will take quite a while
• shut down the system
• move the jumpers on the drives so that the new drive is the Master, and the old drive is the slave
• insert your boot disk, and start the PC
• type "Fdisk" to start the fdisk program, and select "make a partition active"
• make the C drive active

For case 1 - old drive is one partition, C drive  - you are done - reboot the system, and Viola !!!  It will then boot into windows as it always has

For case 2 - old drive has two partitions - you still need to repartition the old drive - since each drive will now have a primary partition, the old drive (now your second drive) will unfortunately grab the letter D for it's primary partition.  You will want to change it, so that there is one extended partition, and two logical drives.  You can buy Partition Magic to do this - but no need - do the following :

• realize that you have copied both partitions on the old drive to the new drive, so you do not need to worry about losing data - other than to make sure that you repartition the correct drive.  The old C drive partition is now your D drive, and has redundant data on it - so you do not need that data.  The old drive's 2nd partition is now F, and also has redundant data on it. Remember - partitioning deletes all data - so be careful !!

• reboot using the boot disk, using Fdisk, and make sure to switch to the 2nd drive once you get into Fdisk !!!  It defaults to the first drive !!!  Repartition the old drive into one extended partition, and then create two logical drives - when you do this the 2nd partition on the new drive will be renamed from E to D, and your old drive will now have two logical drives on it which will be assigned the letters E and F - all the redundant data is now cleared 

• format the old drive (the two logical drives on it) - make sure to type "dir e:" and then "dir f:" first, to make sure the drive are indeed the newly partitioned drives they should be unreadable with the dir command since they are not yet formatted, and that will confirm they are safe for formatting

format e:  /z:64

format f:  /z:64

IMPORTANT NOTE:  adding a new drive and making it your C drive is a monumental opportunity to reinstall Windows from scratch, and to not copy the file over.  This will be a hassle, but will make your system much, much, much faster.  If I were you - that is what I would do - and actually I have always done that, unless the old drive has a fairly fresh install of Windows on it.

Done  .  .  .  Tada !!!  If you have any questions - please email me !!

Email