We were asked to upgrade three of the exhibits for the Sherlock Holmes traveling crime lab exhibition. As part of the display, there are some interactives that show blood spatter patterns that are created by various methods. The first one is cast-off, where someone who has blood on their hands swing their hands around, and a pattern of drops flies out against the walls. The second one is arterial, where an arterial wound causes a trail of blood for every heart beat. The third one is if a body is hit with a bullet, where a finer mist of blood would emanate from the body to create a blood spatter pattern onto the wall.

Where do you start on a project such as this – clearly a balance had to be found between the ‘nasty’ and the ‘fascination’ factor. We also had to be cognizant of the cultural differences between the US, Australia and China, the next destinations for the exhibit. One of the changes we made was that we originally had a Martini glass (complete with olives, of course) to catch the run-off from the charger needle mechanism in “Cast-off”. The designer felt that the implications of using a vessel only used for alcoholic drinks was perhaps not the wisest receptacle, and so we now use a very British teacup instead. It takes a team to catch all the things large and little to make an exhibit like this work, and between the Designer and the Mad team we certainly ended up with a great result. This is another one we’re proud of!


We decided that the best way to create these patterns would be to create a mechanical device that emulated the corresponding movements and blood flows so that the resulting pattern is implicit in the function of the mechanical devices.

For ‘cast-off’, we decided that a swinging mechanism that actually ‘throws’ the blood against a wall would be a good way to emulate the source, and we started with a servo mechanism that is capable of swinging an arm around very much like a human arm would move. Then, when it came to delivering the blood, we decided on an old-fashioned rather beautiful brass syringe. We cut a hole into the top of the syringe to allow for it to be filled using a dramatic theatrical drop-by-drop filling mechanism, and mounted it to the swing arm. Sherlock’s magnifying glass is mounted to allow visitors to really see the detail of the filling mechanism. A syringe pump is then used to ensure that the amount of blood delivered for each session is extremely consistent. A syringe pump uses a syringe mounted to a lead-screw based mechanism driven by a stepper motor, so a highly accurate laboratory type of device. Since these pumps are professional devices, the MTBF (Mean Time Between Failures) is high, which will result in a high availability of the exhibit with low maintenance, and very predictable behavior. Coupling the exact quantity of ‘blood’ delivered, and a very accurate and repeatable movement, there is no reason for the pattern to ever change significantly which means that it is possible to guarantee consistent delivery.

Obviously this as a bare mechanism would be functional, but nothing beyond that, and definitely not a Mad Systems creation… so we found a collection of good looking brass pipe and fittings, and created some ‘Victorian’ looking gears on our waterjet cutter – for each of the three gears, there are three layers which are riveted together using a simple piece of brass rod. The result is then painted with a particulate brass type of paint so that it is possible to apply a patina and the result will age like real brass. The third gear has a pulley attached to it, which drives an early 1900s Singer Sowing Machine motor (we found this motor, and the original label is still on it) – but the good thing is that with all the smoke and mirrors in this exhibit, the essential components are industrial precision devices, and all the remainder is just the set dressing that has no effect on the actual functionality of the machine.

The triggering mechanism for this exhibit is a professional heavy duty ‘one armed bandit’ type of device that is mounted into the structure of the table itself so that it will take a good beating. The intent was for the movement of the triggering arm to correspond to the movement of the ‘throw’ arm so that visitors felt that they are winding up a spring that results in the mechanism from being triggered.

The video showing a few angles of the finished result is shown above. Note how we programmed the syringe pump to run just a little too long, so that on each cycle a drop or two of ‘blood’ lands into the teacup located below the filling position.


The arterial exhibit was another interesting challenge – how to create a mechanism that moves the source of the blood while an ‘arterial’ heartbeat generates blood flow.

We started off by looking at movement, and we’d recently seen a rather nice reciprocating mechanism which we re-created using the waterjet. It took a few iterations to get the quality of movement we wanted, and once the mechanism was solid we created the version you see now, with ‘Victorian’ patterned ‘ears’ on both sides of the mechanism. It certainly looks authentically complex!

The triggering mechanism is a heavily modified valve control device that is normally used in the oil industry. These valve controllers have an aperture (where you see the copper plate in the center) which is mounted over the top of a valve and then geared to open and close in-line valves. In this case, we mounted a custom waterjet cut plate into the aperture – there are holes on regular intervals, and then we use a non-contacting inductive sensor to create a very long term maintenance-free switch input. The PLC, that controls this device, looks for a few pulses within a given time period to trigger the mechanism such that visitors have to turn the wheel about 2.5 revolutions.

We then found a veterinary syringe – a stainless steel device – that was converted so that the reservoir is not used, but instead a tube is connected directly to the luer lock on the syringe. This way, there is no maintenance required. If there is ever a blockage (unlikely – over the months of testing we did not find any blockages, the mixture just remains fluid much to everybody’s chagrin as cleaning it is just an absolute messy disaster), then the syringe can just be twisted and removed to be cleaned.

The tube is then taken from the syringe, which is mounted on a Victorian patterned platform, to a barb so that in case the continuous movement of the tube causes problems, only the easily accessible ‘between’ piece of tubing will need replacement. From there it appears to be connected to a set of pipes – but in reality the tube is fed through the tubes to take a fairly short route back to the pump.

In this case, we started off with a peristaltic medical grade pump only to find that the pressure required to get a nice pattern on the glass pushed it to the edge of its capability, which sometimes resulted in stalling the stepper motor. This was obviously not something that was acceptable, and after some experiments it was decided that a fuel injector pump out of a car would be a better option – this pump was more than man enough for the task, and no matter where you are, a replacement will be relatively easy to find.

As with the other exhibits, a PLC controls the sequence of events, and the result is a nice blood spatter pattern on the window, that definitely resembles that of an arterial wound. Not for the squeamish!

Through (known here as ‘Blast’)

The ‘Through’ exhibit was most challenging to us initially. Clearly using a blast of air somehow would be a good choice, but that didn’t leave a lot of room for theatrics.

We started with the mechanism, and basically took a small compressor which pressurizes a small air tank. We wanted to ensure that we had minimal variations between ‘shots’, and so designed a mechanism with two solenoid valves with a fixed volume between. The first valve opens, the fixed volume pressurizes, the first valve closes, and the second valve then opens to ‘blast’ air. This way we can guarantee the volume of air used, and together with a quality pressure regulator the result is consistent.

A special head was machined with a small reservoir in it. Liquid is dripped in from above, and during the design we found that getting the needle into the fill aperture prevented ‘blow back’ of ‘blood’ laden mist, which the team was rather happy about. Standing behind this machine to find that it’s blowing the liquid back into your face during testing was not a great moment in the development of these interactives, although those not affected did get some pleasure seeing the results…

In order to add some theatrics, we mounted a glass tube (to emulate the actual pressure vessel which is mounted below the table and not visible) with some brass taps to emulate the functionality of the control mechanism. We found some beautiful (genuine) old air pressure indicators that are mounted in the exhibits (one indicates the pressure from the compressor, the other one the pressure of the pressure vessel so that it moves during each cycle). There is also a bleed line that pushes air into a ‘blood’ reservoir mounted high up within the exhibit, so that once the pressure is live, there is a continuous slow and nasty looking bubbling within the blood chamber. Needless to say that the blood in this reservoir does not have anything to do with the real blood flow in the system. Like the ‘Cast-off’ exhibit, this one uses a syringe pump to create a very consistent blood delivery system.

We then felt that we needed something a little more dramatic to conclude the visual appeal of the exhibit, and found a good looking manual two-way valve. The valve handle is mounted using a stiff metal wire to a simple gear motor that is activated during part of the cycle. When the handle returns to its upper position, the valve opens to create the ‘through’ blood pattern, so that visitors get a direct visual correlation between the end of the movement and the delivery of the blood spatter. Various audio effects through a hidden speaker complement the real events that are used to create the pattern on the glass.

and finally….

When creating these types of exhibits, especially traveling exhibits, it is important to remember the environment that these are likely to have to work in. Visitors are obviously going to be much like what we’re used to: damage is likely unless very robust mechanisms are chosen. Spares are a problem, so a less sophisticated result is the best possible solution. A good compromise has to be found between a mechanism that consistently produces the required results, and something that can be maintained and kept going in the field. Using less of the ‘blood’ is obviously better, and making the job of changing the blood source bottles, the clean water supply (a bucket) and the dregs bucket easy to do is one way for the designers to remain popular. These mechanisms allow for the blood mixture to be prepared in the original bottles that the liquids come in, and no pressurization or tinkering is required so that nobody needs to get home looking like they had a fake tan (our guys ended up looking suspiciously orange all over more than once, and huge quantities of various scrubs were used to return them to a state where they were willing to emerge in public after a day in the lab!).

It’s funny what you learn in the process of doing these type of projects – if you ever need to clean glass with a window wiper, make sure that the wiper is the same as the shape of the glass you’re trying to clean. If you look at these videos, you’ll see that they are not 100% clean – it took a while to get the right wipers in, and these videos were taken using some rather expensive wipers that turned out to be far from the optimal option.

Maintenance, although minimal, is something that should never be forgotten with exhibits. The harder it is to provide maintenance, the less likely it is to ever happen. To that effect, all three ‘tanks’ that surround the exhibits are mounted on hinges so that it is easy to gain access to the exhibits below. The windows are relatively easy to replace, and when the tanks are hinged to their open positions, hidden steel cables limit movement so that they get into a stable partially open position. Liquids can be replaced easily by accessing the ‘boxes’ mounted below the tables, and since there are no pressure vessels, the source bottle, clean water and dregs containers are easily filled, emptied and replaced.

One other thing that we added is a mechanism that allows for input/control device failures: if the one-armed bandit, rotating valve controller or dive valve mechanism fails, there is a switch inside of the units that allows them to be set to continuous automatic cycles so that they would not just sit still. There is also a special input that allows for a simple push button to be added in case there is a need to manually control these devices temporarily while the ‘real’ input mechanism is being fixed. Of course all there input devices were selected and designed to be fool-proof and very much hardened for visitor use, but the user interface is always one of the weak points (if someone really wants to break something and there is no supervision it’s hard to prevent it), so we like the ‘belt and braces’ approach to ensure that our client will have the higher possible exhibit availability with the least amount of maintenance required.

A huge “Thank you” to Geoffrey Curley and his team, who trusted us enough to let us get on with our ideas to create something that’s truly different from where we started. We hope never to see these exhibits again – it means that we’ve done what we set out to do!

And finally: a major thank you to the team – this was seriously messy at times, and being covered in ‘blood’ ruined a good few shirts and dates. Thanks for playing, thank you for your hard work, and thank you for a great job well done!


Share this Project

Leave a Comment