The TPS (throttle position sensor) is a vital part of our EFI systems. In carburetted Bonnies it plays a relatively unimportant role in trimming the ignition timing for reasons of economy or driveability, and in fact the bike appears to run perfectly OK without it, as experienced by various posters here, but on the EFI bikes it is consulted by the ECM for lots of decisions concerning fuelling, ignition timing, idling, rate of aceleration, etc, etc.
To many it appears to be a mysterious component, and judging by the prices Triumph and others charge for them, they have acquired a mystique. In fact they're only a glorified electronic potentiometer, not a million miles away from the volume control pot on your grandads 1918 valve Radio and was in fact invented in the form of a rheostat by Sir Charles Wheatstone in 1843...
The main differences are a ruggedised contruction to withstand the rigours of working within the confines of an engine, vibration, brusque changes in temperature, exposure to the weather, etc and the materials the conductive element are made of. Whereas ordinary volume control pots have a carbon element conductive track, these TPS have conductive plastic tracks, far more resistant to wear and more accurate.
The TPS incorporates a connector plug that connects it to the bikes loom, it's under a protective PVC sleeve. This has three wires on it. Connections to the bike's ECM are as follows:
wire is the +5 volt supply to one end of the conductive track.
Pink and Black
connects the other end of the track to ground or -ve.
Green and yellow
is the connection to the cursor or wiper that gets moved by the throttle shaft and selects a different resistance value as it moves around and sends it in the form of a voltage to the ECM for processing.
As the throttle is moved the TPS supplies the ECM with a signal (volts) proportional to the amount of angular movement, and the ECM interprets this as a percentage of throttle opening, as in 0% is idle, and, on our bikes, 92% is full throttle. Why not 100%? I hear you ask...don't know, but this is the same on all other EFI systems where the full throttle position never equals 100%.
The initial adjustment is very important and the service manuals give a voltage level that the ECM has to receive from the TPS at rest or idle position. On our Bonnies this is 0.600 volts with a tolerance figure of +-0.02 volts. So this is from 0.580 to 0.620 volts.
With the oficial factory diagnostic tools or, in our case, programmes such as DealerTool, TuneBoy and the freeware TuneECU, setting of the base value is as easy as loosening the clamping bolt that holds the TPS in place, rotating it and reading the volts on the PC screen and tightening the bolt up again.
This are a couple of screen shots showing where the adjustment is shown on some diagnostic software:
If you have none of the above tools then the adjustment can be carried out as follows:
You need a digital multimeter set to the low DC volts scale (2v, 10v, etc) and provided with test leads terminated in small croc clips, a pin or other thin, sharp implement such as a hypodermic needle, and a driver or T25 Torx tool to undo the locking bolt.
Roll back the jacket to give you access to the cables. Find the Green with Yellow tracer wire and pierce the centre of it with your pin so that the pin touches the inner wire strands and makes contact. Try not to push with your hand or fingers just behind in case you slip and pierce your finger instead.
Don't worry about perforating the insulation of this wire. The low voltages involved means that the tiny breach in the wire integrity is of no consequence.
Connect the positive probe lead of your multimeter to this pin and the negative probe to a good ground point. I used one of the fins that have a polished edge and was lucky in obtaining a good connection, but you might prefer one of the engine casing bolts or other solid metal part.
Turn on the ignition but do not start engine. The meter display should now give you a reading close to the factory figure if everything is OK (0.600 volts or so). If you get nothing re-check for a good ground connection and ensure the signal wire pin has gone all the way through to the inner conductors.
If you wish to re-adjust, loosen off the fixing bolt a little bit, not completely, just enough to be able to rotate the TPS slightly, or the TPS will wobble on its shaft and the readings will go crazy. See the photo for the direction of rotation to set the volts higher (clockwise) or lower (anti-clockwise). Re-tighten bolt while observing the reading. You might find that the re-tightening action moves the reading out of tolerance, so re-try again to compensate for that until it's about right.
Once fully tightened you can actually re-adjust the reading a little bit at a time by lightly tapping the edges of the TPS in the required direction with the handle of a screwdriver or similar, it's that sensitive. Do not overtighten, by the way, the thread is tapped into a Zinc-alloy casting that strips easily.
We can now experiment with different settings of this component to try and modify the behaviour of the ECM and smooth over or even eliminate one of the causes of snatching when the throttle is closed and then re-opened at high revs, namely the DFCO (Deceleration fuel cut off), an EFI strategy to save fuel and emissions.
What they do is to cut off all fuel supply by turning off the injectors when the throttle is shut off above certain revs. On mine at least, it appears to be above 3000-3500 rpm.
When the engine revs drop to about 2000 during deceleration the injectors are turned on again to enable idling to resume.
If you re-open the throttle above that lower rpm threshold, the inlet manifolds are "dry" and have to be "re-wetted" with fuel as it were, and the power comes in with a bang. On some very powerful bikes bikes this has been a terrible driveability problem. Just ask a Yamaha FZ-1 rider about it. Yamaha solved the problem on later models but there's a device called a FCE (Fuel Cut eliminator) that you can see here
This being the last "glitch" in my otherwise fine running bike after all my mods, I've taken some steps to soften the effect. After my usual exhausting Google marathons, I find that there's a widely used mod, but it seems applicable only to turboed engines. It involves modifying the output characteristics of the MAP sensor to trick the ECU into believing the throttle is never really closed. This mod is not done to make things smoother, but to enable power-mad turbo drivers to obtain higher levels of turbo boost.
I have softened the effect to the extent where the whole bike is a lot more pleasant to drive by re-setting the TPS (Throttle position sensor) to tell the ECU that the throttle is never closed. This was done by reading the percentage of throttle opening with a OBDII scanner, which should be Zero with the grip closed and 92% with it fully open, and instead adjusting it first to 1.2% and later on to nearly 3%.
On normal more sophisticated EFI systems this shouldn't be done, as the ECU will operate the motorised IACV (Idle air control valve) continuously to settle the proper idle speeds.
On our simpler systems this valve is absent and idle speed can be re-adjusted manually by using the knob under the Throttle bodies.
(The mod makes the idle speed rise 200 or so revs).
If you have no OBDII scanner or TuneECU or DealerTool software, then the mod can be done by adjusting the TPS by voltage, rather than percentage, as shown in the service manual.
The only difference is the level. The factory advises on 0.600 Volts, and if instead you adjust it to around 0.710 Volts this is the equivalent to 1.2-1.3% throttle opening percentage.
I have now tried up to 0.770 volts which is nearly 3% and it feels even better.
Of course if you carry out the adaption procedure the ECM will assume the new voltage setting IS
0% and re-set itself to 0% at that figure.
The bike seems to also run better overall as the ECU now supplies a slightly richer mixture overall and the ON/OFF snatching at high engine revs is greatly softened and gearchanging improved.
Disadvantages are a slight reduction in engine braking (theoretical that, I haven't really noticed) and, I expect, a slightly higher fuel consumption. I was also expecting some increased popping back in the exhaust when dropping revs on deceleration from high speeds and it's there but only slight and nothing to worry about.
I've been running this mod for a couple of weeks now and I'm very pleased with it, but I would invite comments on this as I'm sure there must be some downsides to it, it can't be that easy to fool the ECM...
It's difficult to predict what effects this would have on your own bike, it's difficult to predict if the results can be repeated on a different bike with a different set of mods.