||Tools, Jigs and other bits (page 2)
|This section is for the various bits of tools
and fixtures that I have made to help me build Britannia. There is nothing
particularly clever or innovative in any of the following items and all
will have been made before by someone else. If I have been directly influenced
by someone elses design I have acknowledged that person but, just because
something I have made looks like something that has been seen elsewhere,
if I haven't seen it then I haven't copied it and cannot offer a credit
for it. At the end of the day, there is nothing original to be found in
day-to-day engineering, it's all been done before one way or another.
|Modifying a pair of Drill
||Having a need to hold some long flat bar on the mill,
I purchased a pair of drill press vices from Toolstation for the princely
sum of £ 12.90 each. They are pretty rough and ready, definitely not square
or level but they are made of cast iron so I decided to re-machine them
as a matched pair for general purpose milling jobs. I first spent ten minutes
or so on each one flatting the bases on some wet and dry on my surface table
and then dismantled them to leave the main casting. The jaw facings that
have been provided are just strips of some unknown gash material as are
the sliding jaw catch plates. These are going straight in the bin, along
with the screws.
||I trued them up as best I could on the mill and machined
the slideways first, the difference in heights being seen by the size of
the chips on each one. Then I put the large angle plate up and milled for
a clean-up on the two sides followed by squaring up and milling the front
and back. This gives me the option of aligning the vices using a square
or butting up to my tee-slot packers if I don't need super accuracy. In
hindsight, it would have been smart to do this set of operations first.
While I had the angle plate set up, I also drilled and tapped an M5 hole
in each end of the fixed jaw section to allow work-stops to be bolted on
at either side.
||Next, I trued each vice body in turn and milled the fixed-jaw
face for a complete clean-up and also made the lower step where the jaws
sit exactly the same height as the slideway. The picture shows just how
rough the original machining is! I was also going to give the insides of
the slideways a quick lick to square them up but I didn't have enough travel
to do them in this setup. In the end, I didn't bother, it's the least important
part of the whole job.Once the machining of the upper surfaces was complete,
I then machined the underside of the slideway. This needs to be dead parallel
to the upper surface of the slideway so that the jaw catchplate rides smoothly
along the underside while the sliding jaw moves along the bedway without
jamming. The more accurate this part is, the less jaw lift there will be
||Now it was time to machine the sliding jaw
castings. To start, I bolted the existing jaw faces to my small angle plate
and machined the top to get them square to the jaw faces. These were then
clamped to the mill table and the two sliding surfaces machined followed
by re-positioning the clamping arrangement and milling the catchplate mounting
to size. I have left these a couple of thou proud of the size of the vice
body as mentioned above and will lap these down later to suit each vice.The
jaw-mounting faces were the last to be machined, the castings being mounted
in a small vice and skimmed to clean up. Then it was time to deburr all
round and give each of the machined faces a bit of a rub on some wet and
dry before offering each jaw to the main casting and checking for a nice
|It was at this point that it became obvious that the clamping-screw
hole in the sliding jaw was about forty thou lower than before and would
need bushing and each jaw was mounted back on the mill, the hole clocked
out and then the table moved to the offset position. A 5/8" dia slot drill
was then plunged down to the original depth. Meanwhile, on the lathe, a
pair of brass bushes were made from 5/8" dia bar, drilled 3/8" and parted
off. These were then loctited into the sliding jaws and left to set, followed
by drilling through the existing retaining screw holes and tapping M5. The
ends of the clamping screws were also in poor order so I cut off the front
section completely and then remachined them with flat ends and the retaining
undercut in the correct place.
|| Finally, a length of 3/4" x 3/16" ground flat stock was
cut up to provide two sets of hard jaws and two underside catchplates. The
holes in the jaws have the same side-to-side spacing as before but had to
be compensated for in the vertical plane to allow for the material removed
during milling and are held in place with M5 countersunk screws. To see
how much jaw lift I had, I nipped up a thin parallel in each vice in turn,
set a clock on the moving jaw and then tightened up fully. Originally the
deflection was in the region of three to five thou on each but after a few
sessions back and forth on the surface plate, I've got that down to about
a thou. On one, I went a tad too far and the jaw locked up but a polish
of the slideway underside freed it off. The final picture shows them being
used as a matched pair for the first time.
|Power Feed for the Milling Machine
|After sitting at the end of the mill winding
the handle for a very long time facing up a 12" length of black, flat steel
I decided that a table feed was an essential extra. Although there are feed
boxes available for my type of mill, they are around £300 each and as rare
as hen's teeth and, unless I'm missing the obvious, don't have automatic
disengagement. After spending a couple of days trawling the internet and
watching endless youtube videos, I decided to make my own based around a
windscreen wiper motor. As made, this project will work directly with a
Warco WM-16, Chester 20V, Amadeal AMA25LV and SPG SP2217-III which all use
the same 700mm x 180mm table, and is easily adaptable to mills of other
sizes. Total cost of bought-in parts was under £30, everything else came
from the scrap box. To make a useful power feed five things are needed;
a motor, a way to mount it, a power supply, some sort of speed control and
a means to engage and disengage the drive. The first item on the list was
purchased from the local car breakers and cost £8. Although it doesn't seem
to matter what vehicle it comes from, front wiper motors are larger than
rear wiper motors, probably because they drive two blades. I'm told that
mine came from an Audi A6.
||On getting it back to the workshop, I hooked
it up to my battery charger and tested it in both directions by swapping
the polarity: it worked fine and the current draw was 2A under no-load conditions.
This will probably rise to somewhere between six and ten amps when driving
the table, especially if the table locks are lightly nipped up, and a power
supply will be chosen or built once this figure is known. Meanwhile, a car
battery and ammeter will suffice to get started. The next thing to buy or
make is a pulse width modulator (PWM) whose job it is to act like a dimmer
switch and slow down the feed rate. This is just a simple device which generates
a square-wave signal with adjustable mark-space ratio and applies this to
one or more power transistors that provide power to the motor. These are
easily made using the long-standing NE555 timer chip but I found some ready-built
ones on Amazon for £2.83 with free postage so two were ordered. They are
not worth building at this price!
|Before designing the motor mounting, I had
to choose what sort of clutch arrangement I wanted and decided to use a
sliding drive-dog layout that could be automatically disengaged by the machine
using stops. The advantage of this is that if my attention gets diverted
for any reason while a cut is running, the stops will ensure there is no
disaster. After much internet research, and not finding exactly what I wanted,
I decided to design my own system using a sliding brass sleeve arrangement
that would work in both directions with a centre-off position. The framework
for the motor and clutch would be mounted at the left-hand end of the table
and the operating controls at the right-hand end where I always stand to
wind the "X" axis handle. To get started, I removed the M8 nut and washer
holding the right-hand handle followed by the handle itself. This revealed
a key set into the shaft which was also removed and put away safely with
the other parts. The next item on the shaft is a thrust bearing and this
needs to be retained in place and a way found to adjust the pressure on
it. This function was previously undertaken by the Nylock nut on the end
of the shaft.
||Measuring the shaft showed it to be 10mm dia
and a sleeve was designed to fit this and become both the bearing adjuster
and the drive mechanism for the table. It is a 40mm length of 1/2" hexagon
mild steel which is drilled and reamed 10mm to a depth of 20mm and the balance
drilled and tapped M8. The sleeve is screwed onto the shaft and adjusted
until it just locks the table, then backed off a tiny amount and a grub
screw fitted in the end of the sleeve acts as a locking device. The wiper
motor has a knurled and angled face with an M8 threaded section so this
was dissassembled and put in the lathe so that a 10mm dia spigot could be
machined on it. For this, the sleeve is 25mm long with a 10mm dia hole reamed
to a depth of 10mm and the rest tapped M8. There is also an 8mm wide undercut
on the O/D machined to the root of the hexagon and the reason for this will
become obvious later. The grub screw in the end is still needed to ensure
that the sleeve doesn't unscrew itself from the shaft when going in reverse.
| The motor mounting was now started and the
main requirement was to get the ends of the two shafts in line and about
half a millimeter apart, the idea being that a sliding collar can connect
the two. After working out the PCD for the motor mounting holes, a piece
of aluminium plate was sawn and then drilled to suit followed by the bolting
on of the motor. This assembly was then offered up to get some idea of spacings.
It obviously needed some way of being connected to the table and this meant
that some fixing holes had to be placed in the table end-plate. A piece
of 25mm x 3mm steel angle was chosen as the starting point and a 150mm length
was drilled with a pair of 5mm clearance bolt holes and a relief section
milled away to clear the leadscrew boss on the end-plate. The end-plate
itself, which is made of cast iron, was drilled and tapped with two M5 holes
in corresponding positions.
||At this point, I decided to discard the first
motor mounting plate and, instead, made a frame from some more of the steel
angle. Hole positions for joining the pieces together were first calculated
and these were drilled M5 clearance or drilled and tapped M5. The end piece
also had a pair of M6 clearance holes drilled to mount the motor. Everything
was assembled and it was obvious that some adjustment in all three planes
was neccessary and most of the clearance holes were made into slots instead.
|| Now it became relatively easy to adjust everything and
get the two hexagons exactly in line. The only downside at the moment is
that the motor has to be mounted as shown in the next picture but I believe
there are other motors that are the opposite hand, which would place the
motor rearwards, and I may try and find one. Now I needed to make the hexagon
sleeve and a piece of 1" diameter brass was faced to 48mm length and then
set up vertically on the mill table. Using the bolt hole circle formulae
from the Zeus book, and allowing for the size of the drill, a series of
six holes were drilled through the block using a long-series 2.5mm drill.
These form the points of the hexagon.
|| The block was then returned to the lathe and the centre
was removed using a 1/2" drill that I reground to have a 90 degree point
angle. This was to ensure that the drill followed the centre of the workpiece
and didn't get deflected by the six corner holes. At the same time, a 1/4"
wide undercut was made in the bore 1/4" in from the front and to just below
root diameter. A piece of the hexagon bar was cut about 2" long and a 1/2"
dia spigot put on the front, the idea being to use this as a broach in the
vice. It worked - just - but wasn't very good and the hole was cleaned up
with files instead. The broach came in handy as a gauge, though.
||Everything was now assembled, with the motor mount adjusted
until the hexagon sleeve slid smoothly along the drive and driven shafts
and the PWM was wired up to provide variable speed. As there was no mechanical
means of moving the sleeve at this stage, it was engaged by hand. Everything
worked as expected and I was able to control the feed from a dead stop to
the equivalent of a fast turn by hand. Making all the mechanical linkages
was straightforward and the only new holes on the milling machine have been
confined to the table endplates. The first items to be made were the main
actuating rod and the two brackets to hold it in place. This required a
hole at the top of each endplate, which I have tapped M5, two pieces of
the 25mm x 3mm steel angle cut to suit and a length of 6mm dia mild steel
rod. A spare brake-rod eye has also been modified and fitted to the end
of the actuating rod.
|| The next items I made were the operating lever and it's
support bracket, which has been bolted onto the right-hand table endplate
using a longer M6 Allen bolt in the existing hole. The bracket was another
piece of 25mm x 3mm angle and the lever is made from an offcut of 1/2" x
3/16" flat mild steel which was turned down at one end and threaded to allow
a plastic handle to be screwed on. It is fixed to the bracket with an M5
shouldered bolt and to the actuating rod eye with a clevis pin. To be able
to auto-stop the feed, I then had to make a new block to replace the one
on the front of the table. This was drilled to allow the actuating rod to
pass through, and acts as a centre support as well. I have milled an angle
on the top so that I can still use the table rule if required. A pair of
stop collars were also made and loaded onto the actuating rod.
|| To be able to move the driving sleeve from side to side,
a gear selector fork arrangemant was made using a piece of steel box section
which has a 4mm wall thickness. A pair of pivots were made on the lathe
from some 5/8" dia mild steel and these have had tongues milled onto the
top to fit the groove that I have now put in driving sleeve. This was then
bolted to a legth of 3/8" square mild steel which becomes the operating
lever at this end. A bracket was now made to create the anchor point for
this sub-assembly and was bolted to the front of the motor frame.
||Then the operating lever was offered up and the approximate
centre point marked to create the pivot point, followed by drilling and
reaming 6mm. An M6 shoulder bolt was made and the whole lot was then assembled.
This left only the coupling piece from the main actuating rod to the operating
lever left to make and this was machined from another piece of 3/8" square
mild steel. The bottom was turned to create a shoulder bolt once more, since
this allows the operating lever to pivot, and the top was drilled 6mm dia
and a cross hole drilled and tapped in the end. This allows adjustment of
the relative positions of the two levers at final setup.
||Although the electrics are still jury-rigged together
at the moment, the whole lot was now tested for function and it threw up
a small problem. In "neutral", for want of a better word, and with no power
to the motor, the lever mechanism work really well. Applying power and engaging
the feed also works as expected. The bugbear is in the disengagement of
the feed. Under load, it is quite difficult to return the lever to the central
position and it seems that the problem is in the amount of friction between
the sleeve and the driving hexagon because slight forward pressure on the
winding handle is enough to free it instantly. I have made the fit quite
loose, polished everything up and applied lashings of grease but it still
sticks so I will now look at reducing the contact are during engagement.
||I have remade the bracket that supports the sliding sleeve
fork and moved the pivot point nearer to the sleeve. This has gained me
a 2:1 mechanical advantage and is a great improvement on the previous pivot
point which was actually a 20% mechanical disadvantage, the pivot point
being nearer the actuating rod than the sleeve. So far, it appears to be
working quite well and the feed disengages at the stop point each time.
However, a little more work is neccessary before I will feel fully confident
of it. Last night, I made some bushes on the lathe while leaving the mill
to clean the face of some black bar, about twenty minutes a pass so rather
boring (and tiring) to just sit there winding the handle.
|| I have used it quite a bit lately and thing are loosening
up nicely now. The feed disengages when a stop hits the centre support but
I'm going to need to put some sort of spring bias on the lever because it
jumps out of drive occasionally in one direction but the other way is fine
- for now. I have added the microswitches that control motor direction and
these function as expected. Moving the lever in the opposite direction reverses
the motor and pressing the button makes it go flat out. However, it's more
like a mobility scooter than a Porsche and I can wind it faster by hand.
Still, it works. I was wondering where I could get a power supply to replace
the battery charger and my Dremel power supply has been sitting practically
on the end of my nose the whole time. 13.8 volts smoothed and regulated,
and good for 5 amps which is more than I need. I thought this motor might
need more but it doesn't. The other thing of note is that this windscreen
wiper is a permanent magnet motor and is a great swarf attractor. So not
such a smart choice after all. And I needed to get a new 10k pot from Maplins
because this Chinese one on the speed controller gave up the ghost.
|Large Fixed Steady
||I now have my smokebox material and it's a rather large
lump of metal which needs a few machining ops done on it. I am, however,
a bit light on tooling and this presents a challenge. I need some support
at the tailstock end but a large pipe centre won't solve all the problems.
So I decided to have a go at making a decent-sized fixed steady for the
lathe using whatever was in the scrap box. A fixed steady is usually a sturdy
lump of cast iron but as long as one is not greedy with the cuts it doesn't
need to be particularly strong and rigid. Pretty much all the effort is
outwards from the radial forces but there is very little lateral force involved
so this can simplify the design. I made mine from an old car disc brake
and this is how I went about making it. There were no drawings involved
with this, winging it with back-of-a-fag-packet sketches as I went along.
The first job was to make the bedway clamp and for this a lump of 3" x 3/4"
flat mild steel was sawn off 4" long. After finishing the size to fit between
the leading arms of the saddle, I made the angled recess to locate on the
rear way. I set up expecting to mill out the clearance slot first with a
small end mill and follow this with a large 45-degree cutter.
||However, the cutter was nowhere to be found so I had to
change over to the angle vice and use a standard endmill instead. This is
a 16mm one I bought for a quid at the Midlands show, so it's paid for itself
already. Because of the large overhang, I had to do this with twenty thou
cuts down in the leading direction and twenty thou cuts out in the trailing
direction. Size-wise, I tried to attain the same width as the tailstock
vee. After checking that it sat flat and square across the ways, I cleaned
it up a bit and also made the underside clamp from some black steel. So
that was the easy bit done.
||Machining the disc needed some thought. There was no accuracy
required here but I needed the ability to hold the part and once separated
from the hub, this would be nigh on impossible without a faceplate. Holding
on the hub, I skimmed the O/D to clean up and took a couple of cuts down
the face, then turned it round and took a cleaning cut down the opposite
side. I'm running at about 100 RPM and using the power feed to keep a steady
cut. I also cleaned up the bore of the hub to facilitate holding later on
and, because I didn't fancy trepanning this, started to bore out to the
finish I/D using facing cuts rather than sliding cuts. But before I cut
right through, I needed to get some milling done while I could still hold
it. So it was over to the mill and set up the rotary table, allowing me
to mill some guide slots for the bearing supports, or fingers, which I will
make later. I started milling with a 12mm cutter but it rang like a bell
because of the large overhang so I changed to a 6mm cutter which allowed
faster speeds and a better finish.
||I also spotted the various drilled holes and the splitting
lines before returning the workpiece to the lathe and separating the flange
from the hub with more facing cuts. When there was about twenty thou to
go, I saw a witness of the cut start to appear on the rear of the workpiece
so stopped the lathe and gently broke the flange away from the hub with
a hide mallet. The bore was cleaned up with files and the part then split.
Offering up to the lathe it appeared that the base of the ring needed to
be 7mm above the bed clamp and because I needed to make a piece of angle
to fix them together I made this 7mm thick to get the required lift. The
lower section sits on the angle and is bolted through with three M5 cap
screws. In the photo the top section is shown.
||A quick rummage in the "bearings" drawer produced three
626 bearings which are 19mm x 6mm x 6mm and this determined the size of
the fingers, which were made from 3/4" x 1/2" black bar, cleaned up all
round and milled to be a snug fit in the 16mm slots. A 6mm cross-hole was
drilled and reamed and the slot cut away with a 6mm carbide endmill to take
the bearings. To create the adjusting slot in the fingers, I stitch-drilled
the full length of the slot by centre-drilling at 5mmm intervals and using
a 4.7mm drill to clear the material followed by 5.1mm. The only suitable
slot drills I have are 3/16" HSS and, because this material was a bit rough
and abrasive, it was kinder to the cutter.
||The slots were finished at 5.2mm wide by 40mm long which
will give a reasonable range at these large diameters. Different fingers
can be made as and when required for other diameters. That left the rear
hinge and the front clamp to make. The hinge was made from 1" x 1/2" bar
with a 1/2" section milled away from each piece and a pivot hole through
the middle and the clamp was made from 1" x 1" x 1/8" angle, held together
with a nut and bolt when in use. Here is the beast bolted together and on
the lathe with the smokebox held rather precariously in the chuck and resting
in the steady. Im running at about 100rpm again but not taking any cuts,
just making sure it runs OK.
|General Purpose Tapping Fixture
||I've always been comfortable with freehand tapping but
after breaking a 12BA on the third hole, and with thirty three still to
go, I decided that it was time to make a tapping fixture. Nowadays, my hands
aren't steady enough to keep these delicate little taps perfectly upright
whilst twisting the wrench. Once again, all these parts were made from surplus
materials except the little Jacobs 5/32" chuck which I have a pair of and
will take taps with 4mm or 5/32" shanks. I wont describe this in detail
because it's very simple; a 3" square baseplate, a 5/8" diameter mild steel
column and a head from the same material as the base. The shaft is 1/4"
diameter mild steel with a Jacobs No. 0 taper turned on one end. The bearings
are roller bearings salvaged from an old printer 7/16" O/D x 1/4" I/D x
7/16" long and the handle is turned from a lump or Delrin, fixed to the
shaft with a 3/32" taper pin. The locking screw on the side is M8 nylon
and is only to hold the chuck up whilst loading a tap. The head height can
be adjusted to suit the workpiece and the alignment is done simply by lowering
the chuck jaws down to the hole in the baseplate.
|The workpiece doesn't have to sit on the little
table, either. Just swing the head round and use the base to rest on any
flat surface thus keeping the tap vertical. By holding the fixture in the
bench vice it can be used in the horizontal mode, and I found this the most
comfortable working position. Tapping the other thirty three holes was achieved
inside half an hour and no further breakages occurred. The fixture took
a few hours to make but was well worth the time spent. These small taps
ain't cheap! And, luckily, I was able to remove the broken tap from the
third hole in the workpiece without damage so not quite a disaster.
||I've never been particularly happy with any
of the methods for producing a dimpled effect on loco and tender steps.
I have tried using a punch directly onto the underside of the tread with
a backing block of wood, and lightly punching onto brass shimstock to produce
an inlay for gluing or soldering into place. I wanted something simple that
was repeatable, easy to use, and didn't distort the tread. Most punch actions
make the tread curve in the direction of the punching action and any effort
to flatten the tread out again flattens the dimple pattern. The other problem
is that these all work better if one has three hands, the third hand supporting
the work. My solution was to convert a heavy-duty leather hole-punch into
an indenting tool. The first job was to remove the brass anvil, which just
punched out, and the rotating punch holder which was slightly more problematic.
Although I used my advanced butchery skills (junior hacksaw, hammer, cold
chisel, verbal encouragement etc) it would have been simpler to just drill
straight through the carrier with a 5/8" drill. It was made of one of those
horrible aluminium-zinc alloys and not worth trying to adapt. I also discarded
the indexing spring as I only require one tool location and would like some
fine adjustment of its rotary position.
|The tool carrier was remade from mild steel
using the same dimensions as the original and a solid mild steel pin made
to mount it on. The carrier has a flat milled on it and a 6mm reamed hole
to locate the tool bit. The tool bit is from 5/16" diameter mild steel with
a 6mm back and the front tapered at ten degrees except for the point which
is forty-five degrees. The anvil mounting point was opened up to 6mm and
the anvil made from 5/16" mild steel. The M6 thread on the back is to hold
it firmly in position but allow easy replacement and the front has a 1/8"
diameter nose with a small indent formed with a No.1 centre drill. All the
parts were fitted together and checked for alignment, followed by a couple
of trial indents. This is a close-up of the tool.
||I found that the tool bit had a tendency to rotate away
from it's positon so an M5 bolt was used to lock the position after alignment.
The hole on view at the back of the tool carrier allows the tool bit to
be punched out if it get's stuck. This final picture shows three different
sets of dimples. On the left is 18swg material and the dimples are not particularly
well-formed, the middle is 20swg brass and they are much crisper. On the
right is 8 thou brass shimstock and the inset shows the graph paper that
I used as a guide. It just needs a bit of care and attention to get a perfect
pattern. Trying to dimple 16swg material failed miserably but may just need
more acurate and harder tools. Now that the principle has been proven I
am going to try different types of anvil, and may use silver steel which
can be hardened for the tool bit. The other tool bits and anvils that I
am going to experiment with are for riveting. I think it should be possible
to squeeze copper rivets in sizes up to 1/16" and also the smaller brass
sizes using this tool. Although there is limited throat depth, such a tool
could still come in handy for things like the smokebox rivets (I hope).
|Rotating Loco Stand
||Inspired by the loco stand described by Bob Shephard,
I have made a similar contraption to support my Britannia. Many of the components
are a direct copy of Bob's constuction so many thanks to him for sharing
photos of it. I started by purchasing a Clarke's 450Kg engine stand from
Machine Mart and building it up as designed. Three things became immediately
apparent; the front support was too long, the height was too low and the
fixed wheels on the rear needed to replaced with castors. I reduced the
length of the front support by 300mm to bring the front-to-back dimension
to approximately 600mm and redrilled the 12mm hole for bolting to the upright.
||Next, I sawed off the lugs that carry the
rear wheels and drilled new holes to take a pair of castors. These came
from Toolstation, part no 70389. I chose these because they can be mounted
with a single M10 bolt through the swivel base, swapping front to back to
get the stand more level. Finally, I cut the vertical tube and extended
it by bolting a pair of home-made "U" channels around the two halves with
150mm gap between. The "U" channel was prepared on the mill by cutting the
side off of some 50mm square box section using a slitting saw. A pair of
flat panels were cut from sheet steel to fill the gap between the sides
of the risers and M10 bolts were used throughout. .
| The rotating frame is made from 50x40x6 steel
angle, chosen because that is what the local forge had on the shelf, and
was bolted together followed by fully welding the corner joints. I also
made a pair of end plates which can be bolted on and will have wooden buffer
blocks fixed to them. The right-hand end is arranged so that the loco can
be wheeled on from this end. A new longer locking pin was made because the
original was far too sloppy, and the lower section of the cross-tube drilled
to take the pin right through.
|| The "rails" are covered with self-adhesive
draught excluder to protect the loco wheels. A pair of clamps have been
made that fix on top of the loco frames and to the underside of the rotating
frame. This allows the loco to be fully inverted with no risk of something
breaking and I use the axle-box spacers that I made to keep the suspension
at the normal ride-height to save damaging the springs by having them fully
compressed all the time. The frame can be rotated to eight positions and
can be stored in the vertical position when not required to save space.