Fred's Land Rover Project

Trials and Tribulations of restoring a 1965 Land Rover Series 2a 88" Deluxe Station Wagon.

(Click photos for larger version.)

New Subdash

I needed more room on the dash to mount some desireable instruments, so built a subdash in the left glovebox.

It was bent to somewhat resemble the slope of the original centre dash. Behind the aluminum is 1/4" thick plywood to give the instrument clamps something to grip.
It is held on with 4 machine screws through holes originally in the bulkhead.
It is fitted with a tachometer from a Triumph Spitfire, a quartz clock from a Lada and 4-way hazard flasher switch and flasher from a 1963 Dodge truck with indicator lamp from 'The Source'.

The 4-way hazard wiring was very simple to install.

One green/white and one green/red wire were extended from the bullet connectors on the original flasher switch. A brown wire extended from the back of the ammeter for power, and a red wire fitted to the hazard indicator lamp.

Rear Door

The rear door suffered from a rusty frame, bent in two places at the bottom, as well as pounds of body filler to hide the dents and twists. It was time for a rebuild.

First step was to de-skin the door.
Two small tools were made to gently lift the folded aluminum skin from the steel frame. They are each about 2" wide, bent into an L shape with the short leg about 3/8" long on one and about 1/4" on the other. The short leg is then tapered and smoothed.
The short leg of the tool is worked gently under the folded aluminum and levered back and forth along the length of the fold to gradually lift the aluminum. There was no hammering involved.

The bottom skin tucks up under the side pieces, so these side pieces had to be removed before the bottom skin could be removed.
Once the edges had been bent up to about 90 degrees around the bottom and sides, and to a lesser amount in the window opening, the skin could be lifted off the frame.

The cracks where the door had been bent, and the rusty holes show up now.

The bottom part off the frame was cut off with an angle grinder. I cut the welds at the mitred corners, and beside the vertical braces. I also cut around the door retainer slide box saving it in place.
To fabricate a replacement frame piece, I found a metal shop in Turner Valley.
The aluminum skin has now been stripped using both paint stripper and orbital sander. The next step is to anneal the edges. To that end, I bought a product called Tempilstik, rated at 500^oF. This looks like a thick blue chalk pencil. It is rubbed on the area to be annealed and when the metal is heated to the designated temperature, the chalk turns black and burns off. Easier said than done! The chalk pencil was so hard it barely stuck to the aluminum.
As well, heating the edge of the large panel caused it to warp and buckle. Here's hoping it flattens out when it is re crimped.

The door frame was sandblasted clean.
The new frame piece was cut to length to fit the remaining frame and MIG welded into place. Holes were drilled for the spare tire mount and the inner door panel screws. As well, improved drainage slots were cut at the lowest point of each end to ensure that any water trapped in the frame will run out, rather than sitting in the lowest area - as before. Even Land Rovers can be improved!

This is the cross section of the door frame. The replacement part was very accurate, but about 1/8" (2mm) lower profile.

The previous owner had cut off the original door check rod and left the remnants inside the door! A new check rod was made from 5/16" drill rod, brazed to a heavy washer and tested for fit. Having the outer skin off the door certainly made that task easier.
The 5/16" bolt with washers will be fitted up through the floor (as was original), the check rod placed over the bolt and the pin fitted through the hole in the bolt. This way the rear door can be removed by simply removing the 2 hinge bolts and the pin from the check rod.

The primed aluminum skin was fitted back onto the frame, also primed on all sides. The idea is to isolate the aluminum from the steel to prevent galvanic action and corrosion of the aluminum.
The first stage of re crimping the skin was refolding the crimp and hammering it down with a wooden block.

The second step was to tighten the crimp using locking pliers. The plastic strip underneath is to prevent denting the aluminum on the outside.
Because the skin had warped slightly from annealing the edges, and probably from previous dents, several spots on the door gave the 'oilcan' effect when pressed. To overcome this, I glued the skin to the frame in several places with body seam sealer/adhesive.
Also, due to corrosion and age, the skin fatiqued and cracked along the edge in some spots. Here I also glued the crimped edge to the frame. May not be original, but it's got to hold together!

The next step is filling the dents with body filler and final sanding before painting.
There'll be more!

Trailer Hitch

The old discarded chassis had a form of trailer hitch welded to the rear crossmember but that went to the recyclers, and of course the new Marsland chassis did not have a hitch on it. I've always liked the look of the Dixon-Bates style hitches used in England, because they are height-adjustable. So I set about building a reasonable facsimile.
The parts cost around $15 CAD, plus cutting wheels, welding rod, paint and scrounged scrap iron.

The tow ball part was bought for £2 at a boot-sale in the UK several years ago. It has a 50mm ball, which is close enough to the 2" standard used in Canada.

The carrier for the tow ball was made out of 2 pieces of 2 1/2" X 2 1/2" angle iron, each about 6" long. It was reinforced with a 6" wide piece of 1/4" plate welded across the lower half. This was drilled for the two 5/8" bolts to hold the tow ball on, and a pair of 3/4" holes in the sides for the hitch pin.

Back View. The mounting piece to be bolted to the rear crossmember was made out of heavy gauge channel iron. It was originally 8" wide by 2 1/2" deep. Since 8" is too wide, it was sliced down the middle and about 2" was removed. It was then welded back together creating a channel 5 7/8" wide. It is now just wide enough for the carrier to fit snugly over it.
Four 3/8" holes were drilled in a 4" square pattern to fit the holes on the rear crossmember. A reinforcement made of 1 1/4" angle iron was fabricated and welded to the back of the mounting piece.
The ends were cut, bent down and welded to give a flat surface at the ends. Holes in the ends allow 3/8" bolts to screw up into threaded holes under the crossmember.
Four pairs of 3/4" holes were drilled in the sides of the mounting piece to accept the 3/4" X 6" hitch pin.
Side View.

Two pieces of strap iron the same length as the crossmember height were made to act as spacers. Because of the tub mounting tab and the PTO hole in the crossmember, the hitch mounting piece would not sit flat against the crossmember. These pieces are drilled with 3/8" holes 4" apart.

All the parts ready to mount on the Land Rover - fabricated bits plus
two grade 8 bolts 5/8" X 2",
four grade 8 bolts 3/8" X 4",
two grade 8 bolts (with washers attached) 3/8" X 1",
and the 3/4" X 6 1/4" hitch pin.

All mounted up and ready for work. The ball height is adjustable at 4 heights: 15 5/8", 18", 20 3/8", and 23 1/8".
Next to be fitted is the wiring socket and safety chain fastenings.

Fuel Gauge

The fuel gauge was a puzzle. It originally never read more than half full!

It seems that the type of meter used in the fuel gauge of older Land Rovers is a form of moving iron vane meter. A simple description of a moving iron vane meter from an internet page shows a single coil to attract the iron vane.

It would appear that the meter used in Land Rover fuel gauges is a variation of this type. The diagram below shows the physical layout of the inside of the fuel gauge, or meter, with the addition of the fuel tank sender unit and battery, below it.

This time, there are two coils, A and B, each able to attract the iron vane with the pointer attached. Theoretically, varying the amount the current through each coil will vary the amount it can attract the vane. Notice that Coil A is earthed to the gauge case, therefore requires that the gauge, and instrument cluster, be correctly earthed to the vehicle chassis and wiring harness to complete the circuit.

Physically this is what a burnt out one looks like inside.
52 turns of hair-fine resistance wire was wound around the broken fibreboard piece, second from the top, left side to form a shunt resistor.

It seems this shunt allows the meter to handle 12V directly off the vehicle battery. This shunt uses approximately 16’ of .006” resistance wire which is wound with cloth insulation. The resistance is between 133 and 155 Ohms in my three samples. This gives a resistance of approximately 9 ohms per foot, much higher than pure copper wire. There is no voltage stabilizer or regulator in this meter circuit.

In an attempt to explain the workings of this meter, the following 3 diagrams are used. By using Ohm’s Law, Kirchhoff’s Law, and probably even Murphy’s Law, the following should take place.

Diagram 2 shows the fuel tank at Empty.

The fuel sender arm is at the bottom of the tank, therefore no resistance in the sender unit. This means Coil A is effectively ‘shorted out’, leaving 12 Volts across the Shunt and Coil B in parallel. This allows 200 mA of current to flow, of which 118 mA is through Coil B. This current causes a strong magnetic field to attract the iron vane in the movement, and the needle swings to Empty. So far, this is what happens in real life.

In Diagram 3, below, the tank sender has moved halfway up, therefore the resistance will be about 45 Ohms, half of the 90 Ohm resistor.

This changes the total resistance of the circuit, and the amount of current flowing through various elements of it. Now, only 132 mA flows, of which 41 mA flows through Coil A, and 78 mA flows through Coil B. Theoretically, each coil should create equal magnetic force on the iron vane of the movement, holding the needle halfway between the two coils. This is where practicality does not always follow theory. In reality, the meter originally read only about 1/3 of a tank.

Each of the coils is mounted in a slot, allowing it to be moved closer or further from the pivot point of the iron vane, so it should be possible to move them enough to cause the meter needle to indicate ‘half a tank’.

In the last diagram, the tank sender is at the top, putting full resistance in the circuit.

This again changes total resistance, and current flows. Now only 112 mA flows, with 53 mA through Coil A, and 66 mA through Coil B. In order for the meter to read Full, one would expect Coil A to be the dominant one, attracting the iron vane directly to it so the needle would point to “F”. In reality, the needle barely moves above the 1/3 indicated earlier. Again, adjusting the positions of the two coils does not give the correct reading. The only way to get the meter to read Full, is to remove the wire from the Sender unit. This puts the full 112 mA current through Coil A, and the needle immediately snaps to the “F” position

So what is wrong with this circuit?

I have dismantled 3 fuel gauges and found all the coils and shunts to have similar resistance readings. I have 2 fuel sender units, and each has the same resistance. Can there be so many faulty components in one workshop?

I see two possibilities.
1. The sender unit resistance could be much higher. However experimenting with adding resistance in series with the sender does not give the correct response either.
2. The shunt resistance could be lower. However, on 2 of the 3 units inspected, the fibreboard core of the shunt was burnt through due to overheating of the resistance wire. Would lower resistance of the shunt worsen the problem?

A possible third solution is to replace the internal shunt with one of the correct resistance. In the end I did get the unit to work correctly eventually, by placing a large resistor across the terminals of the meter, in effect in parallel with the shunt.

While the fuel gauge and sender seemed to work on the bench, after installing it back into the vehicle and going for a test run - it failed again. Sometimes it would read somewhat correctly, sometimes the needle would snap rapidly the "F", sometimes it would sit on the "E". After disconnecting the sender wire to the gauge and reconnecting, the gauge might read either full or empty - very randomly. After filling the tank this last time, the gauge seemed to work, but in reverse. The further I drove, the fuller the tank got! At least it was consistant without the random snapping between "F" and "E". The general wisdom is that there is an intermittent earth problem somewhere in the circuit, but I haven't found it yet. Besides, the gauge doesn't read accurately now either.

I pulled the sender out again and checked the wiper, the resistance wire and the terminals. All seemed to be in order. I checked the voltage at the sender unit - about 10 volts without the motor running, seemed OK. The sender is earthed through the mounting screws to the tank and thence to the chassis. All seems fine there.

After reading various threads on Land Rover lists, I found one post that claimed that a vehicle converted from positive to negative ground (as indeed mine has been) would have problems with the fuel gauge reading backwards. I cannot actually see why that should be the case, though. Each of the two coils, A and B should attract a non-magnetized iron vane regardless of the polarity of the current passing through them.

Maybe it was time to start from scratch again.
I had previously ordered a new sender unit, but when it came I discovered it was for the newer system with for the voltage stabilized heated bimetal type gauge. When the float was at the bottom (ie tank empty) the resistance was 177 ohms. When it should indicate full, resistance was only 13 ohms. This is the reverse of what the old sender was (ie, empty - 10 ohms, and full - 90 ohms). So I put the sender in the junk box for the time being.

Now in frustration I dragged it out and bent its arm backwards 90 degrees so it reads correctly for the old style gauge. Note the kink near where the original bend was.

Then I found another old burnt out gauge and set up a test rig.

On the left is a 9V battery above the new sender unit. I figured that a 9V battery would work since the same voltage acts equally on both coils of the gauge no matter whether the voltage is 9V or 12V. (That's my theory, at least.)
In the centre is the gauge in a vice. It is set at the same angle as it would be in the instrument module.
On the right is a variable resister wired in parallel with the gauge. The shunt resistor was burnt out of this gauge, so I replaced it with an external resistance. The whole thing was wired up with test leads as it would be in the vehicle.

First step was to set the variable resistor so that the gauge would swing through the full range when the sender arm was moved through its full range. It turned out that a shunt resistance of 52 ohms worked. This is about 1/3 of the original resistance of the old shunt (148 ohms), but that didn't read correctly anyway! Since you cannot buy 52 ohm resistors at a reasonable price, I bought the standard values - 47 ohm and 10 ohm - and soldered them in series. 57 ohms was close enough. They were 1/4 watt carbon film type, and small enough to squeeze inside the fuel gauge.
The next step was to adjust the two coils within the gauge so they caused the needle to point to the correct position on the dial to reflect the position of the float arm of the sender.

By loosening the nuts and sliding them either closer to or farther from the centre of the gauge, the needle could be made to indicate the correct reading of the sending unit.
Note that one of the nuts has an insulating washer under it, but the other does not. The one that has no insulating washer is the earth point for coil A, and therefore must be solidly earthed to the instrument cluster and back to the chassis and wiring loom.

Float arm at the bottom - needle indicates empty.
Float arm at half way point - needle indicating half.
Float arm at the top - needle indicating full! So far so good.
Now that this seems to work on the bench, the next step is to install it in the vehicle, again, complete with a 57 ohm shunt resistsmce inside the back of the guage, and see what happens.

So I installed the new (modified) sender and the newly modified gauge in the vehicle, and...and...
Well the gauge didn't read the half full tank as being half full. It said the tank was FULL. Now what?

And now the mechanics of it

It seems there had to be some more fiddling with the sender than showed up on the bench.

After measuring the sensor unit, the length of the sensor arm and the depth of the tank I calculated the effective tank depth to be 11". This is because the fill pipe is lower than the flat top of the tank, and because the pivot point of the sensor arm is at about the same level as the fill pipe.

While testing the sensor on the bench, I had been swinging it through its whole range, and adjusting the gauge accordingly. Now I had to adjust the stop tabs on the sensor to limit the arm's swing to the useable portion of its travel inside the tank.
On top of that, the resistor on the new sensor is not linear, as the old one was. The turns of resistance wire are spaced further apart when the arm rises near the FULL range. (I know it has something to do with geometry and movement through an arc, but I don't understand it enough to delve into it further.)
I managed to bend the float arm and the two arm stop tabs to get the gauge to read correctly throughout the full swing of the sensor arm.
To test this on the bench, I used a horizontal straightedge under the sending unit and a ruler to measure the vertical movement of the float. While watching the gauge, I bent the arm and the stop tabs until the gauge read pretty accurately.
Now I'm happy!

Seat Belts

The original seat belts were only lap belts. As a result of the accident (which prompted this bebuild) my wife has had a hip replacement. Even though she had her right foot firmly on the brake pedal, the impact forced her forward until her knee hit the lower rolled lip of the dash. There is still a dent in it. The non-retracting lap belts did not prevent her from sliding forward. So before this Land Rover rolls again, it will have fully retracting lap/sash belts.

A set of belts was salvaged from a late model Toyota pickup truck. The belts had the retractors, anchors and upper restraints in all the right places to fit into the Land Rover.

The next task was to make brackets to mount the retractors.

4mm thick steel plate was bent to form an 'L', then drilled to fit onto the bolts that hold the reinforcement between the B pillar and the front of the tub. The two small holes near the edge are for this purpose. The notch allows the bracket to fit into the reinforcement. Originally I had planned to put the bracket behind the reinforcement (between it and the tub), but it forced the B pillar out of alignment affecting the door strike plate. These are only 1/4" bolts through a single layer of aluminum, so more serious anchorage was needed.

Therefore the large hole in the short end of the 'L'. A hole through the lower edge of the tub, behind the seat base accepts a 7/16" bolt, which also passes through the other steel plate underneath. This plate is 2 3/4" X 4" so should provide plenty of strength. (I should have brazed captive nuts to these plates also.) In an accident most of the force will be upward, taken by the 7/16" bolt and steel reinforcing plate below the tub.

A 7/16" captive nut is brazed onto the backside of the 'L' to accept the retractor mounting bolt. The small hole above it is for the locating pin on the retractor, to keep it vertical.

The left hand brackets in the photo are for the left hand side of the vehicle, the other pieces are for the right side. The extra cutouts are to clear the fuel vent pipe.

The retractor being positioned toward the plate. It will be turned 90 degrees to bolt on.

The locating pin is visible on the retractor and will fit into the small hole in the black plate. The mounting bolt hole is just below it, and the corresponding hole on the retractor is just visible below the seat cushion.

This is the right hand side, and the retractor has to be mounted high enough to clear the fuel vent hose, but low enough not to interfere with the seat squab when folded upright.

On the left hand side, the retractor is mounted a bit lower, but is just high enough for the retractor mounting bolt to clear the edge of the seat base. Unfortunately they both are just in line with the seat squab mount, making fitting the bolt a little tricky.

Just behind the B pillar (and out of site in this photo) are the two bolts that hold the reinforcment and now also this mounting plate.
Also out of site at the bottom is the 7/16" bolt down through the mounting plate, the forward edge of the tub and the steel strengthening plate. It is just behind the lip of the seat base mountings.

The upper anchor. This is a genuine Land Rover part, available from many parts suppliers.

Although the retractor is mounted such that the belt is at right angles to this anchor, it is immediately below, so the belt does run freely without chafing or binding.

The lower end of this bracket is held to the aluminum channel of the door frame by a 1/4" bolt. But the upper end is held by 2 bolts (1/4") which pass through the aluminum side panel plus the galvanized steel roof frame (not seen in this photo). In an accident the force would pull this bracket forward, so most of the strength is provided by the roof frame and the 2 bolts.

The buckle ends of the seatbelts are attached to the original mounting points on the forward wall of the tub. These have 7/16" bolts and 3" X 4" steel plates as well as steel angle brackets inside the tub. As this is the standard Land Rover setup, I felt it didn't warrant extra photos or explainations.

Tub Repair

The biggest problem was the damage to the front left corner of the tub. It seems the previous owner had an accident that badly crumpled the bulkhead and tore the outer skin. As a crude repair, the spot welds were cut between the outer skin and the top of the inner fender and a patch forced between them, held in place with pop-rivets and hidden with Bondo. After all the dirt, paint and most of the filler had been removed this is the result.

With the tub turned over, and the patch removed, this is what it looked like inside.

Front left corner of tub - laying upside down.

Note the badly wrinkled bulkhead. This is even after I have straightened it somewhat.

This shows the patch piece from the donor tub removed, and the remains of the corner of the outer skin drilled and removed from the front corner of the tub.

Tub corner and patch piece.

Another view of the wrinkled bulkhead. I have also removed the stiffener that went along the bottom edge (top as shown here) of the inner fender/bulkhead between the outer skin and the inner fender.

Wrinkled bulkhead.

I was able to panel beat the bulkhead back fairly flat. I made a new stiffener and riveted it into place.

The next step was to make another patch piece, larger than the repair, then cut the outer skin back to remove the many pop-rivet holes. For this I have some expert help from an old style body man, trained at Aston Martin-Lagonda in England. The patch was then welded with oxy/acetylene torch into the original outer skin.

Then the patch piece was held onto the turned up lip of the top of the inner fender and the B pillar with countersunk aircraft rivets. It seems plug welding aluminum is much more difficult that doing it in steel panels.

The many countersunk pop-rivet holes!

Notice the coarse grinder grooves left by the previous repairer.

The inside of the tub, paint stripped and hand sanded.

The outside of the tub. The three steel cross-wise supports have been removed and re-galvanized. I even found some old canvas/rubber belting to cut up as pads between the supports and the frame.

You can see where the repair was needed at the upper right corner of the photo.

The tub came back from the bodyman. He knew how to handle aluminum.

The patch was salvaged from one of the crumpled front fenders. I made a cardboard pattern and he cut and bent and shaped it to suit. He then butt-welded it in using an oxy-acetylene torch and aluminum rod. Not a bad job! No major wrinkles and a little body filler smoothed the seam over.

Even the inside and bulkhead were repaired and patched where needed. Countersunk aircraft-style rivets were used to hold the corners and reinforcing pieces together. The bodyman said trying to do plug welds would drive him crazy!

A bit of body putty, a coat of etching primer and two coats of paint hide the repair pretty well. I did not attempt to remove every wrinkle from the tub - after all it is a Land Rover and should look like it has done some work!

Bonnet Repair

Both the outer aluminum skin and the steel framework were bent as a result of the accident.
The bonnet had to be dismantled to be straightened. In the process I discovered a lot of rust in the two lengthwise braces. So the rusty bits were cut out and new ones made.

Here the pieces are ready to be welded in place. This time I put drain holes in them so they won't trap water in future.

Underseat Battery/tool Box

This is how my old steel box looked.

And the worst of it isn't even visible in this photo. The bottom had rusted out at each seam.

So a new box was needed. I thought I'd have a go at making one out of aluminum, to prevent the corrosive action against the aluminum seat base, and for lightness. It may not be totally original in a Series 2a, but I did find aluminum boxes under a Series 2.

So here it is in 14ga aluminum. The corners are rivetted, and sealed with body sealant. It will rivet back into the seat base, as the original did. However, I didn't put the depression in the bottom, as I doubt a battery will ever need to go under the seat, and it makes for less seams to leak.

So on to the next problem - whatever that may be!

Fred Griffiths

Back to Home Page.	Back to Spitfire Hangar.

This page updated November 7, 2008.

	First step was to de-skin the door. Two small tools were made to gently lift the folded aluminum skin from the steel frame. They are each about 2" wide, bent into an L shape with the short leg about 3/8" long on one and about 1/4" on the other. The short leg is then tapered and smoothed. The short leg of the tool is worked gently under the folded aluminum and levered back and forth along the length of the fold to gradually lift the aluminum. There was no hammering involved.
	The bottom skin tucks up under the side pieces, so these side pieces had to be removed before the bottom skin could be removed. Once the edges had been bent up to about 90 degrees around the bottom and sides, and to a lesser amount in the window opening, the skin could be lifted off the frame.
	The cracks where the door had been bent, and the rusty holes show up now.
	The bottom part off the frame was cut off with an angle grinder. I cut the welds at the mitred corners, and beside the vertical braces. I also cut around the door retainer slide box saving it in place. To fabricate a replacement frame piece, I found a metal shop in Turner Valley. The aluminum skin has now been stripped using both paint stripper and orbital sander. The next step is to anneal the edges. To that end, I bought a product called Tempilstik, rated at 500^oF. This looks like a thick blue chalk pencil. It is rubbed on the area to be annealed and when the metal is heated to the designated temperature, the chalk turns black and burns off. Easier said than done! The chalk pencil was so hard it barely stuck to the aluminum. As well, heating the edge of the large panel caused it to warp and buckle. Here's hoping it flattens out when it is re crimped.
	The door frame was sandblasted clean. The new frame piece was cut to length to fit the remaining frame and MIG welded into place. Holes were drilled for the spare tire mount and the inner door panel screws. As well, improved drainage slots were cut at the lowest point of each end to ensure that any water trapped in the frame will run out, rather than sitting in the lowest area - as before. Even Land Rovers can be improved!	This is the cross section of the door frame. The replacement part was very accurate, but about 1/8" (2mm) lower profile.
	The previous owner had cut off the original door check rod and left the remnants inside the door! A new check rod was made from 5/16" drill rod, brazed to a heavy washer and tested for fit. Having the outer skin off the door certainly made that task easier. The 5/16" bolt with washers will be fitted up through the floor (as was original), the check rod placed over the bolt and the pin fitted through the hole in the bolt. This way the rear door can be removed by simply removing the 2 hinge bolts and the pin from the check rod.
	The primed aluminum skin was fitted back onto the frame, also primed on all sides. The idea is to isolate the aluminum from the steel to prevent galvanic action and corrosion of the aluminum. The first stage of re crimping the skin was refolding the crimp and hammering it down with a wooden block.
	The second step was to tighten the crimp using locking pliers. The plastic strip underneath is to prevent denting the aluminum on the outside. Because the skin had warped slightly from annealing the edges, and probably from previous dents, several spots on the door gave the 'oilcan' effect when pressed. To overcome this, I glued the skin to the frame in several places with body seam sealer/adhesive. Also, due to corrosion and age, the skin fatiqued and cracked along the edge in some spots. Here I also glued the crimped edge to the frame. May not be original, but it's got to hold together!
The next step is filling the dents with body filler and final sanding before painting. There'll be more!

	The tow ball part was bought for £2 at a boot-sale in the UK several years ago. It has a 50mm ball, which is close enough to the 2" standard used in Canada.
	The carrier for the tow ball was made out of 2 pieces of 2 1/2" X 2 1/2" angle iron, each about 6" long. It was reinforced with a 6" wide piece of 1/4" plate welded across the lower half. This was drilled for the two 5/8" bolts to hold the tow ball on, and a pair of 3/4" holes in the sides for the hitch pin.
Back View.	The mounting piece to be bolted to the rear crossmember was made out of heavy gauge channel iron. It was originally 8" wide by 2 1/2" deep. Since 8" is too wide, it was sliced down the middle and about 2" was removed. It was then welded back together creating a channel 5 7/8" wide. It is now just wide enough for the carrier to fit snugly over it. Four 3/8" holes were drilled in a 4" square pattern to fit the holes on the rear crossmember. A reinforcement made of 1 1/4" angle iron was fabricated and welded to the back of the mounting piece. The ends were cut, bent down and welded to give a flat surface at the ends. Holes in the ends allow 3/8" bolts to screw up into threaded holes under the crossmember. Four pairs of 3/4" holes were drilled in the sides of the mounting piece to accept the 3/4" X 6" hitch pin.	Side View.
	Two pieces of strap iron the same length as the crossmember height were made to act as spacers. Because of the tub mounting tab and the PTO hole in the crossmember, the hitch mounting piece would not sit flat against the crossmember. These pieces are drilled with 3/8" holes 4" apart.
	All the parts ready to mount on the Land Rover - fabricated bits plus two grade 8 bolts 5/8" X 2", four grade 8 bolts 3/8" X 4", two grade 8 bolts (with washers attached) 3/8" X 1", and the 3/4" X 6 1/4" hitch pin.
	All mounted up and ready for work. The ball height is adjustable at 4 heights: 15 5/8", 18", 20 3/8", and 23 1/8". Next to be fitted is the wiring socket and safety chain fastenings.