Hacksaw PDF
Hacksaw PDF
Hacksaw PDF
Index
CH 1 The bits you will need 67
2. Table model with minimum weight consistent with rigidity so that it can be transported
easily and repositioned for angled cuts etc.
1. The motor. Its important to use the variable speed brush variety (some machines use
synchronous motors). Leave 50cm wire on motor connector.
5. The drum (from which you will extract the large pulley and bearing assembly)
6. The casing which is used to guard all the moving internals of the completed saw. (this sheet
metal, can also be welded to cars)
The bearing was embedded in a fairly heavy plastic surround at the back of the drum and required
some work with a hand saw around the outside of the drum ( cut A). Then a jig saw was used to trim
around the outside of the pulley at cut B.
The spindle can usually be removed through the front of the drum once the big nut holding the
pulley has been loosened .It comes out complete with the spider which is fixed to the inner drum.
Take care not to loose any bearings.
A pair of pedals and cranks from an old pushbike, preferably steel ones. These provide good crank
bearings.
For the moving frame of the saw you will need approx 10 feet of 25mm square steel tube.
A small quantity of 30mm angle iron is needed for the sliding blade carrier. (12 inches)
For the saw table (to which the vice is bolted) some 5mm x 350x110mm plate is required.
Some conveniently available building angle was slit and welded together for this purpose.
Since the saw blade is so high an ordinary record number 1 vice was used to hold the work.
Two 9mm x 11inch coach bolts needed to act as swinging arm pivots
A length of 12mm studding (220mm) and 3 12mm nuts is needed for adjustable crank rod.
Approximately 5 inches of 30mm square section tube is needed for the crank joints.
A car timing belt idler wheel to catch the saw when it cuts through the work. (maybe you have
changed this yourself its a 60k service, garages throw them out).
Welding capability
This is quite an advanced project; a large amount of stick welding is required. Always tack things
together initially, and then when you are sure they are in the right place you can make a tight
weld. The author has lost count of the number of times he has welded something into the wrong
place. You then have to slit it with the angle grinder, sometimes in difficult places, so be warned.
Do your final welding when the machine is almost finished!
Welding distortion
This occurs because the metal expands when it is hot and then the joint cools and everything pulls
together. Its a very difficult problem in many industries. Problems like banana shaped boats and
submarines can occur unless special measures are taken.
Example
The illustration shows what happens even when two strips of metal are only tacked together on one
side. When a tack is made on the other side it evens up. The opposite ends should then be tacked in
the order shown. Six tacks should be sufficient for a short distance. Even when it is decided to make
a run its a good idea to do half of one side then half of the other. The result is less neat than a
continuous run but there is less danger of distortion.
This is probably the most difficult piece of work to get right but it does bring home the point.
Part A is the front cross beam which divides the front saw table from the rest of the saw.
Part B is set to give enough clearance from the side of the case to the cone shaped pulley. The front
vertical spar is welded into the corner formed by joint between parts A and B.
Dimension A, between the two upright spars is set to accommodate different sized bearing plates
from the donor washing machine.
A plate was butt welded to the top of each vertical spar to spread the load and give more flexibility
to the bearing plate fixing bolt position.
These fixing plates were tied together with a 1 inch bar welded as shown. The pulley must rest
against the flat surface formed by the inside of the vertical spars. Butt welds are used where flat
mounting surfaces must be preserved.
The prototype bedplate was made from 1.5 inch wide bed iron, this proved to be a bit thin and
there was appreciable flexing when the crank rotated. However thicker material may prove
satisfactory. In this case the frame was reinforced by welding in extra sections to create box
sections as shown in fig 4 note 1.
The back vertical spar was also reinforced with a diagonal brace.
Parts A were cut to size from 25 mm square tube and were equalized by filing whilst clamped
together side-by -side in the vice as shown in fig 7. The opposite ends were squared up by inverting
and repeating the process.
The 9 mm shank bolts used for the pivots had rolled on 10mm threads at one end. These threads
were therefore bigger than the shaft and would not pass through a 9 mm bearing hole. The solution
was to grind these threads off with the bench grinder. An 8 mm thread was then cut with a die when
the diameter had been reduced sufficiently.
The handle of the rat tailed file was removed and a pair of vice grips used to twist it in the manner of
a reamer to enlarge the hole. It is possible the constructor may not be able to obtain the same sized
file. It will be satisfactory as long as it is approximately the same. Experimenting with bolt insertion
will verify how much work is needed. A tight mobile fit is required. A bolt was inserted into the
first set of pivot holes prior to reaming the opposite end, to ensure perfect alignment.
The bearing is mild steel to mild steel, not ideal, but there is nothing to stop the end of the square
section from being filled up with bronze or white metal at some future time to give better wear
resistance (soft-to-hard metal).
Referring to fig 6 at the start of the section we now need parts B through G, be sure to make parts D
equal using the methods already described.
Part G was tacked into position when parts A were held in position threaded by a pivot bolt and part
F clamped with two G clamps. Part B was also clamped into position with spacers D. Just to
make doubly sure measurements of squareness were taken and a straight edge used. When satisfied
G, F and E were tacked into position. The top swinging arm should be true due to the squareness
and uniformity tube components used.
Using the hacksaw a cut was made holding the part in the vice at 45 degrees. This allowed the best
control of the operation. Then the work was turned over and another cut made at 45 degrees. The
cut was squared up when both sides have been done.
If the slit is not parallel with the sides of part H then the saw will deviate from the vertical. Part H
itself must also be vertical.
The blade is retained in this slit by a pin made from a piece of welding rod.
Referring to fig 6 part H was tacked onto part B square with parts D.
Then the part can be slightly adjusted with a hammer then tacked on the other side. It should not be
welded up solid until it has been tacked on all sides and thoroughly checked for squareness.
This time there are two sets of pivot holes to line up. A and B. Pilot holes were drilled using the drill
press, this gives approximate accuracy and squareness.
Then one set of holes were enlarged using the process already described with the rat-tailed file.
Finally the bolt is inserted to align this end and the other set of holes at the opposite end are
enlarged.
Again this process encourages equality and when parts C are separated and parallel the bearings
should run true.
With parts C joined to the top swinging arm using the 9mm bolt the cross braces were clamped onto
opposite sides of the square section.
These were welded on squarely when Parts C were made parallel by equalizing distances A and B.
Cross braces were welded onto opposite sides of parts C and are connected at the center with a
small piece of 25 mm square tube.
The holes for pivot B which go through supports D should also be aligned before drilling either by
clamping a steel plate to both square sections or by clamping both in the vice.
Otherwise twisting could occur when supports D are fixed to the back of the base plate which will
tighten the bearing surface.
Another vertical spar which carries a slit for the blade was welded to the fabricated section. A
straight edge was used to check for squareness.
The 8 mm nut welded to the end cap allows a radio spares 8 mm stud knob to be used to tension
the blade. This bears against a blanking plate welded to the end of part B.
There were a lot of unknowns here so it was decided to make everything as adjust able as possible.
A clamp was fabricated on the flat end of the drum shaft left when the spider was broken off. .
Be sure to remove the shaft from the bearing before any welding, the grease will fry and plastic may
melt.
The shaft of the sawn -off pedal is held in a square hole by a set screw. This allows the crank
throw to be adjusted.
The pedal bearing was prepared by removing much of the flat pedal leaving only the sleeve
Note how the connecting rod consists of a 12mm threaded rod passing through a c shaped piece
made from a piece of 30mm square section tube.
This section is welded to a complete tube which surrounds the pedal bearing. The hole for the 12
mm studding was drilled so that the studding is in contact with the side of the tube.
Two 12 mm nuts retain this connecting rod and also enable adjustment. A jubilee clip may be
required to retain the connecting rod end on the pedal sleeve.
Adjustment of the 12mm nuts allows the saw arm position in relation to the crank to be changed.
.The remaining pedal sleeve was used as a swinging arm connecting rod bearing.
The bearing retainer nut was used in conjunction with another bracket for a more positive fixing.
Again the pedal sleeve was surrounded by a piece of 30mm square tubing.
A 12mm nut was welded to this tubing to permit linking with the 12 mm studding used for the
connecting rod.
The above diagram shows underneath the back of the top swinging arm of the saw. Dimensions
may be varied to suit the pedals available.
On the prototype the two motor mountings on the face were bolted to the base angle section which
had to be cut away to give clearance so that the belt would run parallel and stay on the pulley.
The drum bearing backing plate was clamped into an approximate position and the three mounting
holes drilled. Elongated holes in the metal brackets allow further adjustment to suit the belt used.
Care is needed so that the large pulley does not foul the inside of front vertical 45 deg guard.
The length of the existing belt can be shortened once the motor has been mounted and the
approximate position of the drum bearing is known.
Custom made belts are likely to be expensive in small quantities. Fortunately these belts can be
shortened.
Its quite surprising but an overlap joint made with super-glue will actually hold. The joint should
be 0.5 inch with the edges beveled on a bench grinder.
According to the manufacturers instructions the glue should be applied to one surface then
the joint should be clamped for a few minutes in a vice.
The belt was modified when both motor and drum bearing plate had been mounted so that its
approximate length could be determined.
Then using the elongated holes the belt tension was set just enough to prevent slippage.
Setting the belt onto the inner grooves on the motor pulley will pull the belt into the large pulley.
These adjustments were made so that that belt just stayed on the large pulley as both were turned by
hand.
The blade also needs to be lifted out of the way when releasing the vice. Here these functions are
combined in one.
A car timing belt tensioner is ideal for this purpose, having a heavy duty bearing incorporated.
Its wheel runs against the left hand square section swinging arm member when the saw cuts
through the work. The rim of the wheel shown in fig 20 was ground off so that it presented a flat
bearing surface
It is mounted on a sliding joint with the front vertical drum angle section. This is used to raise and
lower the saw blade.
A peg was welded to this sliding section this allows a jockey wheel to be raised or lowered by a
lever pivoted on the back drum pulley upright.
The idler wheel is spring loaded to absorb the impact of the saw arm dropping on it when the work
is done.
Fig 22 shows the metal strips carrying the lifting pin and the jockey wheel which are welded to the
outer surface of the sliding angle section.
This was done so that it will slide unimpeded on the side of the front vertical spar supporting the
crank bearing casing.
A web was welded across the vee when it has been proved that this lever clears the side of the
saw and any long work which is likely to be sawn particularly when the vice is set to 45 degrees.
The sharpened saw retaining latch is butt welded in the position shown to give the correct spacing
in the saw-up position.
The full-sized diagram shows how the lever and jockey wheel carrier fit together on the side of the
frame. To retain the lever in the upper position a chamfered section engages with a flange on the
slide retainer.
The next page shows how the blade lifter angle is sandwiched between a piece of angle section
welded to the outside of the front vertical spar with the appropriate spacing piece in place. This
allows it to slide when the lifting lever is operated. There is a lip which then retains the lever in the
upper position. Then lifting the lever and moving to the right clears the lip allowing the blade to be
lowered.
When the vice rotates on its fixing/pivot bolt it must not foul the front upright (on fig 31
where the 0.2 inch clearance is indicated).
Also the position must be set so that the saw blade is clear of the vice jaws for all angle
settings.
When the vice is set for 45 degrees holding a long bar there has to be enough clearance
with the large pulley wheel guard.
Only when all these conditions are satisfied should a pivot hole be drilled and tapped into
the bedplate.
Moving the vice position towards the front of the saw is permissible as the blade stroke is
large.
FIG 25 22 inch
Cutting the washing
machine case.
29 inch
17 inch
Cut Cut
here here
Fig 24 shows how the washing machine case was cut with an angle grinder to obtain
material from which to make the saw casing. The side panels were used to fabricate the
case and the bends already part of the washing machine case were also used.
Edge 180
degree bend
Top of washing
FIG 26 Bending the machine lid recess
crimp
left side of the case
90 degree
bend
Edge 90
degree bend
Case length 17 inches
The back has an internal 180 deg bend to get rid of the sharp edge and 90 degree bends
which go inside the back and are welded to the sides and top. There is a cut-out at the front
left side to allow clearance for work cut at 45 degrees.
Knowledge of electronics is needed to complete this section if not a ready made controller
will be available from the author to suit demand. This can be wired in by a qualified
electrician or the user if he feels confident.
This item is essential for the operation,
had it not been available the saw could not have been constructed. It uses a TDA1085C
Universal Motor Speed Controller I.C together with the original triac and snubber network
from the donor washing machine board. It has been included at this point in the project
because its helpful to get things moving at this stage. Although the motor can also be
powered up with a variac variable transformer.
A1 A2 G
BTB16 TRIAC
Mounted on a metal
heat sink FIG 29
Snubber network
100 100nF
In operation the triac operates as a switch controlling the current going to the motor, the
snubber prevents damage to the triac by absorbing surges. The snubber capacitor will be a
fairly high voltage type.
A dropper resistor of 6.8k is used to derive the 15v supply for the IC and this should be a
ceramic high wattage type. Constructors from USA may need to use half this resistance.
Two series silicon diodes give the potential on pin 5 to set the speed; this could be made
variable up to 15000rpm for other applications.
The 50m shunt which limits the motor current under fault conditions was made up from
resistance wire using a 4 terminal method to determine its value.
The board was designed to fit inside a small plastic box available from Maplin electronics,
the heat-sink needs to be in the open so only one half of the box was used. There is a danger
of electric shock if the heat sink is touched. However the plastic box is facing towards the
outside of the saw so the possibility of the other side being touched is remote.
FIELD
TACHO
ARMATURE
TRIAC TO MOTOR
TACHO
TACHO
REVERSE
GROUND ALL METAL
FORWARD
NEUTRAL TO BOARD
LIVE TO BOARD
GROUND
NEUTRAL
FIG 33 Motor wiring
LIVE VIA SWTCH
LIVE
9 WAY TERM BLOCK
CONTROL BOX
The diagram shows wiring for the authors washing machine motor looking into the six
way connector. Most of the time wires can be traced visually, backing up with ohm meter
measurements. Across the tacho the resistance is high compared with either armature or the
field.
Since the original connecting plug was retained with a reasonable length of wire attached it
was a simple matter to make all the connections into a terminal strip bolted to the controller
plastic box. The motor armature and field windings are wired in series and the direction of
rotation can be changed by swapping the red wires.
The washing machine single pole push button switch was used to switch the live into the
filter. Its other side goes to LIVE VIA SWTCH on the controller. Neutral goes straight
through the mains filter to NEUTRAL on the controller.
For the sake of safety the three pin plug should be fitted with a 5 amp fuse. The mains cable
should be securely clamped to the frame with a P clip.
As shown in fig 34 the electronic controller inside a 120x 80 cm Maplin box was mounted
onto the main frame with the exposed parts towards the centre of the saw. Mounting the
mains switch and the filter completes the electrical items.
MAINS
FILTER
TACHO
GROUND
NEUTRAL
LIVE
ON/OFF
SWITCH
Now all the components necessary to run the saw on the blade lift are assembled for a safe trial.
Parts D (fig 12), the lower swinging arm bearing pillars were clamped to the rim of the angle
section to the rear of the saw.
1 The rear frame does not run outside the edge of the bedplate on RH side (so that the case will not
foul any moving parts).
2 The right top swinging arm rail (looking from the back) runs against the blade lift wheel.
3 When the connecting stud rod is screwed in place it runs approximately in the centre of the
swinging arms.
4 The saw blade clears the vice jaws when the vice is set in the straight position.
Running the partially completed saw will reveal any faults and the crank adjustment for throw and
position can be tried.
For instance it is possible to have the vertical swinging arm making contact with the back vertical
pulley wheel support or at the other extreme the inside of the case.
In the prototype a jubilee clip was used to keep the connecting rod in the centre of the pedal sleeve.
Silicon rubber bathroom sealant could also be used here both to dampen vibration and seal in
position. Here a fixing which allows for misalignment is needed.
The speed controller was checked with the belt disconnected. (the saw will shake itself to bits if its
faulty and runs at 1000rpm).
With the two diodes connected across pins 5 and 6 of the TDA1085C the speed should be OK.
Then the saw was run with the belt in place. In the prototype it was noticed that the speed was prone
to reduce when sawing a steel bar. Although some belt slippage was noticed the main problem
proved to be electronic.
The 47K resistor connected to pin 3 started off as 100K but this limited the motor current so it
was reduced. In the event of a fault the 50m measures the motor current and the TDA1085C
keeps it at a safe value.
The first long cut was made into a soft wood block (its easier to assess deviation with a long cut).
If the rear swinging arm bearings are loosened sufficiently they may be adjust by tapping with a
hammer. So long as there is negligible twist in the blade a reasonable straight cut should be
achieved.
Blade twist would be caused by either the two blade slits and/or carriers not being parallel.
The remedy is to partially slit some welds and adjust with the hammer. Then cautiously re-welding
to avoid distortion.
The rear bearing carriers may be tack welded for extra stability.
The saw was set to work cutting through so 3mm x 30mm square tube it cut through the flat section
successfully but was laboring on the narrow sections. This is not unusual for any type of metal saw
since there is more pressure on the blade with narrow work. Even proprietary metal cutting band
saws have trouble in this repect. A spray of WD40 enabled the cut to proceed more smoothly.
No extra weight was required on the saw arm to cut at a reasonable speed.
When the saw is cutting at 45 degrees the cut is made nearer to the end of the saw arm so less
leverage is operating, the saw still cut at a reasonable speed.
To sum up provided the blade is kept lubricated there should be no trouble cutting any normal steel
stock.
Possible improvements
As mentioned earlier the swinging arm bearing are plane steel-to-steel which will wear particularly
if lubrication is neglected. This will lead to a sloppy action.
However the open tubes which make up these bearings can be filled up with lead-bearing bronze or
similar and the holes drilled out again. This would create a longer lasting bearing surface.
References
http://www.nucleus.com/~harlan/saw.html Yes it can be done
http://www.calenterprises.co.uk My site
[email protected] My E-mail