Hydraulic foot press for soil bricks

In his professional career, Rudolph Draaisma constructed a soil brick press. It worked fairly well, but could be improved upon. As it happened to be, the company he worked for abandoned the project, and the press was not perfected. In later years, there was an interest in a manually operated soil brick press, for use in rural areas of developing countries. Rudolph Draaisma then constructed the hydraulic foot press for soil bricks. Drawings and information pertaining to the construction of the press is now available free for anyone to use as they wish.

– Curator


The machine consists of 6 main parts:

Frame:                dwg. HFP / 001-10
Mould:                 dwg. HFP / 001-20
Press block         dwg. HFP / 001-70
Fill tray:               dwg. HFP / 001-40
Foot pedals:        dwg. HFP / 001-60
Hydraulic system:dwgs HFP / 001-90 and 91

The complete set of drawings, supporting drawings, and manufacturing instructions are available for free download.


The required accuracy in manufacturing is as to normal standards, with the exception of certain parts in the mould, the press plates and alignment of guiders. As this is the first machine of this type to be build, I have not set out any tolerances. Instead, critical parts are made to fit on each other, for which reason a certain order in manufacturing of what parts, is required.  While doing, appropriate tolerances will be achieved and MUST be documented (!), in order to make ‘as-built’ drawings later.

The first parts to be manufactured are the mould, in assembly with the press block, as this will enable to build the frame’s superstructure (dwg. HFP 001 / 001-104) in whatever way it may deviate from the drawings to fit. In addition, the correct build-in size of the assembled(!) main press cylinder, as shown in dwg. HFP / 001-011, item 13, must be known (measure it).

As the mould will be the discriminator for all other parts and be used as a template for the press block as well, there are no specific requirements to strict size-accuracies. However, it is absolutely imperative that the lining (dwg. HFP/001-21, item 2) fits smoothly, straight and rectangular and have the inside sizes as given (= brick size). Hence, start with making the linings and build the rest of the mould around it.
Secondly, as obvious it is, the guider rods (dwg. HFP/001-20, item 5) must be absolutely parallel aligned with the lining.

Less obvious and thirdly, is the length of the mechanical stop (dwg. HFP/001-24). Make this part first when the mould and the press block can be assembled, as it determines the height of the brick.  It’s nominal length, 130 mm, can and must be adjusted to assure a brick height of 100 mm (see also the functional description below ).

Fourthly, do not place the holes for mounting the cover-lever system (dwg. HFP/001-23) yet. Wait until the cover and the levers are made, which can be done after the top press plate (dwg. HFP/001-80) is available for fitting it in.

Use, either the guider rods inserted in the mould, as a template to mark the center of the bushings, or make separate center plugs to achieve the same. Furthermore, accuracy of the bushings to be rectangular aligned on the surface of the press block body, is required.

At this stage, the bottom press plate (dwg. HFP/001-81) is needed in order to assure its centered position in the mould.  Although the positions of the holes for the press rods (dwg. HFP/001-71) are given, I would recommend to make 6 centerplugs, screw them in the according holes of the press plate and use this as a template, while inserted in the mould, to mark the positions of the according holes in the press block.  The position of the large hole,  45, nor its size is critical.

No special requirements on accuracy here, apart from enough clearance to let it slide smoothly and without jamming between its guiders on the tray table (dwg. HFP/001-11, item 2).  Adjust the width of the fill tray accordingly, if required.

Prior to making the frame base structure (dwg. HFP/001-103), which should be done before the superstructure is made (HFP/001-104), the build-in size of the hydraulic cylinders, as shown in dwg. HFP/001-90),  must be known by assembling them on the base mount (item 5 of base structure). The length of this base mount (1007 mm) should be adapted to whatever size is required. The effective stroke of all cylinders is 160 mm (the design stroke can be larger, but not less). The cylinders are single acting, with no spring return. The piston diameters are not critical on the mm – for standard cylinders, use the nearest size available (  20 mm can be up to 1″,   40 mm can be between 1 ½  and 1¾”). Try to maintain the ratio 1:2 between the chosen diameters as close as possible.  Larger diameters will reduce the press force on the press block and smaller diameters reduce its stroke per pump step  (longer time needed to press a brick).  The given diameters, 20 and 40 mm, give the best compromise I could find. With these sizes a brick can be pressed with 7 pedal steps and ejected with 6 steps, at pressing forces of between 600 and 2500 kgf. The maximum calculated hydraulic pressure will not exceed 25 bar (~ 370 psi ). See also functional description further on.

In order to mount the cylinders, the glider support (dwg. HFP/001-90) must be made. I have not strictly dimensioned it, as sizes depend on piston shaft diameters and center height over the base mount. Apart from connecting the pump pedal’s actuator (item1, dwg, HFP/001-60) with cylinders,  items 3 and 4, its important function is to avoid vertical side forces acting on the piston rods (“unknown” height in the drawing) – horizontal side forces are not anticipated.  The basic shape however, should be as shown in the drawing. Not shown is a grease-nipple, needed to lubricate the glider support on the base mount. Its length of the glider support should allow space for a through-hole between the inserted piston rod ends, as to connect the pump pedal actuator with a M10 bolt and nut, or an according pin-shaft with end-securing clip-washers (not shown).

With the cylinder assembly at hand, the frame’s base structure can be made. All shown (length) sizes in dwg HFP/001-103 are nominal, not critical and can be adjusted. Just assure that the figuration as shown in dwg HFP/001-90 is obtained,  as such that all moving parts go free. Provided the build-in dimensions of the pump pedals are as shown, the sizes 568, 300 and 29 in dwg HFP/001-90, as well as size 722 in dwgs HFP/001-103 and 10, should apply. If not, the only important thing is the position of the supports (items 6 in dwg HFP/001-103), in relation to the glider support (preventing the cylinder base mount to bend through) and being directly under the HE-B beam of the superstructure (item 7  dwg. HFP/001-104), as shown in dwg HFP/001-90 as well. The total length of the frame (1580 mm), extending to the left in the drawing, can be anything it gets. The frame’s width ( 596 mm) should be kept, as it is the same for the superstructure and its constituent parts. The pump pedals are assembled and mounted with through-going rods through according holes in the frame work, as indicated in dwg HFP/001-103. Also here, I did not go into detail, as it is simpler and better to design it on sight (but, do document the as-built data !).

The frame superstructure (dwg. HFP/001-104) follows from the drawing and no greater accuracy is required, apart from the width (596 mm) and being straight and rectangular. However, there is the length of part items 1 and 4, which depends on the actual build-in size of the main press cylinder – see dwg. HFP/001-011, item 13. If this build-in size is smaller than anticipated (438 mm), no big harm is done, as it can be compensated by inserting shim(s). If it is larger however, the lengths should be adjusted accordingly (with the value “dcz” ). The purpose of the shim is to adjust the filling depth of the mould; it does not effect the stroke lengths for pressing and ejecting bricks. The diameter of the main press cylinder is 120 mm and can be chosen to 5″. Also this cylinder is single acting with no spring return. Though the operational stroke length is appr. 175 mm, but its design length is 240 mm; 10″ can be accepted (see functional description below).  The mould is mounted with 4 screws in the superstructure. To prevent it from moving upwards, it is obstructed by the mould stops on the tray table (item 6) – assure proper welding for strength. It may be a good idea to drill the screw holes with fitting the mould in place as a template, especially, if the total frame width deviates from 596 mm (likely).


The hydraulic system has a low and a high pressure stage.  The force needed to press a brick increases exponentially with the rate of compression. Hence, at the beginning of the press stroke, only a moderate force is needed.  The product of press force and stroke length is constant, naturally, because it is equal to the energy applied, which is given by the body weight and muscle power of the operator, stepping on the pump pedals.  In order to press a brick reasonably fast, we do not want to apply a higher force than necessary, because the according smaller stroke length would take a longer time to press the brick. It also would exhaust the operator unnecessarily, as he/she has to make more steps.  Hence, we want to start with a lower press force, giving speed, and switch to a higher press force, when needed during the last 40% of the total press stroke.

For this reason there are two actuator cylinders, as shown in the hydraulic scheme of dwg HFP/001-91.  For low pressure, actuator cylinder LP is used, by closing off the outlet of the high pressure actuator cylinder HP.  Now this cylinder is blocked and its piston cannot move inwards. The pump pedal then only can move the piston of the LP cylinder, which gives the press block cylinder a stroke of appr. 18 mm per single pedal step. There are furthermore two ‘pedal return’ cylinders, that reset a pressed down pedal to the upward position, when the other pedal is moving down and vice versa. They are simply interconnected, in order to exchange oil between them; they are hydraulically isolated from the rest of the system.  When the operator feels that pressing goes “heavy”, he/she sets an actuator valve in the HP position, by which the LP cylinders get blocked and the HP cylinders  are released. Now the press block cylinder moves at appr. 4.5 mm per single pedal step.

When the mould is filled, the situation is as shown in SKETCH 1. The press block can move upwards 57 mm, until it makes contact with the mechanical stops, that themselves can move 3 more mm upwards. Then the press block cannot move any higher and the brisk is pressed on its ready size. This situation is shown in SKETCH2.  By pushing the handles of the mechanical stops upwards, they swing out and when left free, fall back, while moving down over the previous 3 mm clearance. Now they rest against the press block, without obstructing it any longer. This situation is shown in SKETCH 3.  The operator switches back to LP, opens the mould cover and start pumping again, by which the brick is ejected for demoulding. This takes about 6 single pump steps. The situation is shown in SKETCH 4.

During the press stroke, trapped air above the piston of the press block cylinder has been compressed to appr. 3 bar ( 45 psi),  acting on the piston with a force of around 225 kgf (the hydraulic press force was over 2000 kgf ). It is for this reason that the design stroke of the press block is 240 instead of 175 mm.  The compresses trapped air in the cylinder will assure that the press block moves down and fast, when the operator drains the cylinder by opening a second actuator valve (see dwg HFP/001-91).  To rely on gravity alone, would not be reliable, especially as the press plate may jam a little in the mould. With this compressed air feature, there should be no doubt that the press block moves down and fast, for a new filling to commence.

The mould cover is mounted on a glider in side slots, that are connect to a kind of canty-levers.  When closed, the top press plate sticks in the mould, so the cover cannot move horizontally.  When the operator pulls in the handle to open the mould, the cover can only move upwards, as shown in dwg HFP/001-23. In this way, the top press plate is freed from the pressed brick and when it fully exits the mould, it will swing out and remains in a horizontal position.  On closing, the reversed procedure occurs. When the press plate is over the opening in the mould, the cover cannot move further horizontally (it stops against the fill tray).. and so it can only move down.  To secure it during pressing, the operator moves the two lock pins forward, that grips in according holes.

The short lining plates in the mould can be taken out and re-inserted mirrorwise. This allows to press “end bricks” that have no visible slot(s) in the end face(s).  The press plates must than also be replaced, or plugged where the obsolete holes are – later design to be made.

* * *

(if something can go wrong..it will go wrong !)

1) The lock pins on the mould cover should be vertically somewhat off-center from their counter holes, as to assure minimum friction through line-contact –  they might jam when the brick is pressed under against the cover.  This force is not known; it can be high due to friction between the pressed brick and the walls of the mould. Of course, the press cylinder should be drained just a little, as to remove any press force on the brick and on the mechanical stops under the mould. Nevertheless, the brick can stick hard enough in the mould, to maintain a high force against the cover? It’s a unsure thing and hence subject to Murphy’s Law. Maybe an idea to make the pins a little oval, so they get a clearance in the holes by turning them before pulling out? Basically, the whole cover opening- and closing mechanism is “murphyous”; it has to be tried out with correct fitting of the moving parts.

2)  Draining the press cylinder “just a little”, can be too much for the mechanical stops, if the press block moves down more than 3 mm.  A “press-release” button (?) should be considered, that can be “hit” momentarily and only drains a very, very small amount of oil, that would be enough to release the pressure in the system, but not to move the press block visibly. Alternatively, a very small leak by-pass over the drain actuator valve…but Murphy would either let it leak too much or let it clog up instead?