Smiths Fuel Quantity Indicator 2901 km 01
For fun I bought a Smiths Industries dual fuel quantity indicator. This is an avionics instrument for indication of the fuel amount from 0...300 kilograms of fuel. This means the instrument is not intended for an airliner and not for a very small aircraft. I bought the instrument unseen and as is and it turned out to be broken for half of it. As seen, the left fuel tanks is always full even is the instrument is not connected. ;-) The front plate is marked with a 'P' and a 'S'. I expect that this is intended for Primary and Secondary.
The instrument seen from the rear.
I bought the device from TDM Electronics in Poland. I bought multiple devices from TDM and are a happy customer. The price is right, well packed and fast shipping. And in this case the price was great. This instrument is rather cheap. A new instrument would coast $500+ these days.
There's a type plate on the instrument as seen below. Apparently it's a '2901 KID 01' fuel quantity indicator. As mentioned the driving voltage of each meter is 2...7 mA. The device is made by the well known fabricator Smiths Industries Limited in England.
The first thing noticed after inspecting the instrument is that one recessed bolt was heavily damaged. Apparently someone tried 'something' here. After some fiddling with a plier the rather loose screw came out suprisingly fast! I replaced the screw with a good one. There's some paint damage, but that's not bothering me. The goot thing is that the screw came out quite easily.
Surgically removed screw...
After the inside of the device is removed, the contents of the device can be seen as shown below including detail photos. At the front end is the dial plate including lighting place. Directly behind the dial plate is the 'right meter' positioned. After the right meter is the 'left meter' placed. At the end is a printed circuit board for holding the two resistors. All the wires are connected to the connector at the rear end.
Top view if the inside contents.
Right meter detail.
Bottom view of the instrument showing the wiring.
Well, reverse engineering is maybe overrated since it's a rather simple device, but technically it's a reverse engineering job to find out the original design... After removing three screws from the rear end, four from the shell and one (damaged) recessed screw from the front part (holding the dial plate) the device is opened and the contents of the instruments could be pulled via the rear end. The instrument consists of three parallel placed lightbulbs and two meters. That's all. There are two resistors in series with the meters to acts as a current limiter. I used a piece of paper and a pencil redrawing the functional contents of the instrument as seen below. Later on I redrew the schematic as shown in the next paragraph.
Reverse engineering notes.
Based on the reverse engineering drawing I redrew the schematic as shown below. There are three lightbulbs in parallel connected to pins [F] and [G]. Both meters are identical except that the the left meter rotates clockwise to achieve full scale and the right meter goes counter-clockwise to achieve full scale. I measured 69,63 Ohms as the resistor value, bu I assume that the E12 values are used so a 68 Ohms resistor would be logical. It must me noted that the white wire has a blue ring assuming this is an indication of number 6. In avionics instruments is rather common to use the resistor table colour coding to indicate a wire number. And blue represents number six. Therefore the white wire is indicated as [WT 6].
To drive a dial, a DC voltage has to be applied. Pins [C] and [A] are the positive connections. The scale is from 2...7 mA. By measurement this means that 258 mA is 2 mA is zero scale and 903 mV is 7 mA and is full scale.
The rear view of the device is shown below.
The connector with some bent pins. This had seem better days...
The gauge is equipped with thee light bulbs. Since the device is rather old, the light source are three lightbulbs instead of nowadays common led lighting.
Lightbulbs do wear due to ageing of the filament and are sensitive for shocks. To prevent the loss of light during operation, three lightbulbs are installed in parallel. If one lightbulb fails, there's still 66% of the light available.
The light colour of bulbs is rather 'reddish'. To obtain white light, a blue filter is used to obtain white light. The red part of the light is filtered by the blue filter to obtain a natural white colour. The blue filter is a blue flexible plastic cap placed over each bulb. The three blue caps can be seen on the image below. In normal use there's a cap placed over the bulbs that acts as a reflector. For the photo, the cap is removed.
The original voltage information is unknown. Usually 5 Volts is rather common for dial lighting, but in this case the light amount is not that great. I don't want to destroy the bulbs so 5 Volts is good enough for me. I asked Smiths Industries to send information about the voltages, but I haven't received any information about them...
Dial plate with lighting.
As mentioned earlier, the left meter is broken. As seen on the image below, one of the two resistors failed due to overheating. Since the coil of a meter is much more delicate, this is no good sign... And after measuring, the meter coil is indeed burned. I assume that there was some error in the fuel sensor sending too much voltage (more than 903 mA) to the sensor or there was a short circuit somewhere. A resistor can be replaced, but reviving the meter coil is not an option. The overload was so big that the spring got deformed and is now in 'default' full position. Repairing is virtually impossible, so I'll leave the meter in original, partially broken, condition.
It's rather odd that the meter is not protected to higher voltages. It would be logical to use two diodes as a voltage limiter. A silicon diode starts conducting above 700 mV and a germanium diode has a voltage drop of 300 mV. If two diodes are placed antiparallel 'over' the meter, the voltage would never be more than 0,3 Volts (in germanium diodes are used) and protecting the meter. The meter will receive some overload, but probably not so much it would destroy the meter. Usually avionics instruments have multiple safety features inside, but this rather common and simple meter protection is not included... Strange...