This is a follow-up to update observations made on this thread:
http://www.triumphrat.net/twins-technical-talk/136072-airbox-intake-air-temperature-free-power.html
Where I commented on my concerns regarding airbox-inside temperatures and their theoretical and real effect on power output and driveability. I'll repeat the old rule of thumb that states that for every 10ºF increase in intake air temperature the engine loses 1% power.
Further tests with a temperature probe inserted deep inside the throttle body have shown me that once you're on the move over 20-30kmh the air actually ingested by the engine is at only a slightly higher temperature that the ambient air. It just doesn't get time to heat-up inside the engine-heat-soaked airbox like I originally supposed.
However the IAT (intake temperature sensor), which is fitted to the far wall opposite the airfilter lid and tucked away deep inside the box, remains far too hot due to heat soak and thermal inertia. It doesn't really benefit from the cooling airflow on its way to the engine.
This was determined by riding around with the OBDII scanner permanently connected to the bike. This scanner reads various parameters in real time and real units, and shows that the IAT sensor is nothing more than a thermometer informing the ECM of the incoming air temperature.
![Image]()
On speed-density systems such as our Keihin EFI, this sensor together with the MAP (manifold absolute pressure) sensors, serve to work out the amount and density of air being drawn into the engine at any one time. This, as well as other inputs from other sensors, sets the amount of fuel injected and ignition advance.
The IAT sensor is vital in this role as it can also retard the ignition timing in high intake air temperature conditions: If it senses the air is hot it reduces the strengh of the mixture and can retard the timing to avoid detonation. Warm air makes the mixture more prone to detonation. Weakening the mixture and retarding the timing hurts power.
It follows that the reading it sends to the ECM must be accurate and in its present position it is not. The air inside the airbox and the sensor itself, even in cool ambient of 60-70ºF is at more than 30ºF higher with the OEM snorkel fitted, and it reduces to 20ºF higher with the snorkel removed and a Polaris bellmouth in place. This helps lower the temperature of the ingested air, but the sensor itself remains at a high temperature, telling the ECM lies.
The problem of heat soak and hot air flow from the engine towards the airbox area is so acute that even experimentally installing the IAT sensor on a bracket outside the airbox leaves it reading 84ºF when the ambient air is at 70ºF like on my last test. Hot engine air just gathers inside the right hand cover.
![Image]()
Based on a well-trodden path taken by car tuners where the IAT sensor can atain tremendously high temperatures when trundling through dense traffic, I decided on what I thought was the obvious: Re-locate the IAT sensor to a place where it measures the real air temperature as it goes into the airbox.
The mod was quite easy. An extended wiring loom has to be made and routed to the new position, and the sensor was attached to an existing plastic tab on the airbox filter lid. The original holes left by the sensor on the airbox wall have to be plugged, of course. The sensor element itself is totally encased in plastic and quite impervious to water and dust, by the way.
The readings the IAT is now sending to the ECM are more or less accurate. This richens the mixture accordingly and results in more "pep" from the engine.
![Image]()
I can't prove this on a dyno as there are none around here, but the gains can be felt in driveability, hill-climbing and top-gear acceleration, as well as clean pulling from as low as 1500 rpm.
The calibrated potentiometer that can be seen in one of the photos was an experiment to manually correct the temperature readings by inserting varying amounts of resistance in series with the IAT. This works well, but you have to keep changing these values with different ambient temperatures to get the full benefit.
With the IAT in its final position we don't need it, and the sensor retains its ability to compensate and adjust for varying temperatures as normal.
If you wanted to retain the snorkel, then a hole could be drilled in it to enable the sensor tip to fit inside and monitor the temperature there rather than at the back of the airbox.
Of course all this trouble could be avoided by removing the airbox altogether, and I suspect a great deal of the power gains by doing that are atributable to the cooler air, and on EFI models by keeping the sensor cool as well, but I'm as stubborn as a Spanish mule...
This mod was followed by the removal and by-pass of the O2 sensors from the system. The engine no longer operates in closed-loop with the sensors interfering with the fuelling. I'm so happy with this: No snatch or low speed surging and bucking, and, together with the chain adjusted to the latest factory specs (15-30mm), the bike feels just like a well-carburetted one...
The procedure for O2 sensor bypass and removal is a bit more involved and I'd rather not go into it here for fear of being accused of single-handedly causing Global warming by the sanctimonious, holier-than-thou crowd....
I'll get through a couple of tankfuls to see what effect all this has on fuel consumption and report back.
http://www.triumphrat.net/twins-technical-talk/136072-airbox-intake-air-temperature-free-power.html
Where I commented on my concerns regarding airbox-inside temperatures and their theoretical and real effect on power output and driveability. I'll repeat the old rule of thumb that states that for every 10ºF increase in intake air temperature the engine loses 1% power.
Further tests with a temperature probe inserted deep inside the throttle body have shown me that once you're on the move over 20-30kmh the air actually ingested by the engine is at only a slightly higher temperature that the ambient air. It just doesn't get time to heat-up inside the engine-heat-soaked airbox like I originally supposed.
However the IAT (intake temperature sensor), which is fitted to the far wall opposite the airfilter lid and tucked away deep inside the box, remains far too hot due to heat soak and thermal inertia. It doesn't really benefit from the cooling airflow on its way to the engine.
This was determined by riding around with the OBDII scanner permanently connected to the bike. This scanner reads various parameters in real time and real units, and shows that the IAT sensor is nothing more than a thermometer informing the ECM of the incoming air temperature.
On speed-density systems such as our Keihin EFI, this sensor together with the MAP (manifold absolute pressure) sensors, serve to work out the amount and density of air being drawn into the engine at any one time. This, as well as other inputs from other sensors, sets the amount of fuel injected and ignition advance.
The IAT sensor is vital in this role as it can also retard the ignition timing in high intake air temperature conditions: If it senses the air is hot it reduces the strengh of the mixture and can retard the timing to avoid detonation. Warm air makes the mixture more prone to detonation. Weakening the mixture and retarding the timing hurts power.
It follows that the reading it sends to the ECM must be accurate and in its present position it is not. The air inside the airbox and the sensor itself, even in cool ambient of 60-70ºF is at more than 30ºF higher with the OEM snorkel fitted, and it reduces to 20ºF higher with the snorkel removed and a Polaris bellmouth in place. This helps lower the temperature of the ingested air, but the sensor itself remains at a high temperature, telling the ECM lies.
The problem of heat soak and hot air flow from the engine towards the airbox area is so acute that even experimentally installing the IAT sensor on a bracket outside the airbox leaves it reading 84ºF when the ambient air is at 70ºF like on my last test. Hot engine air just gathers inside the right hand cover.
Based on a well-trodden path taken by car tuners where the IAT sensor can atain tremendously high temperatures when trundling through dense traffic, I decided on what I thought was the obvious: Re-locate the IAT sensor to a place where it measures the real air temperature as it goes into the airbox.
The mod was quite easy. An extended wiring loom has to be made and routed to the new position, and the sensor was attached to an existing plastic tab on the airbox filter lid. The original holes left by the sensor on the airbox wall have to be plugged, of course. The sensor element itself is totally encased in plastic and quite impervious to water and dust, by the way.
The readings the IAT is now sending to the ECM are more or less accurate. This richens the mixture accordingly and results in more "pep" from the engine.
I can't prove this on a dyno as there are none around here, but the gains can be felt in driveability, hill-climbing and top-gear acceleration, as well as clean pulling from as low as 1500 rpm.
The calibrated potentiometer that can be seen in one of the photos was an experiment to manually correct the temperature readings by inserting varying amounts of resistance in series with the IAT. This works well, but you have to keep changing these values with different ambient temperatures to get the full benefit.
With the IAT in its final position we don't need it, and the sensor retains its ability to compensate and adjust for varying temperatures as normal.
If you wanted to retain the snorkel, then a hole could be drilled in it to enable the sensor tip to fit inside and monitor the temperature there rather than at the back of the airbox.
Of course all this trouble could be avoided by removing the airbox altogether, and I suspect a great deal of the power gains by doing that are atributable to the cooler air, and on EFI models by keeping the sensor cool as well, but I'm as stubborn as a Spanish mule...
This mod was followed by the removal and by-pass of the O2 sensors from the system. The engine no longer operates in closed-loop with the sensors interfering with the fuelling. I'm so happy with this: No snatch or low speed surging and bucking, and, together with the chain adjusted to the latest factory specs (15-30mm), the bike feels just like a well-carburetted one...
The procedure for O2 sensor bypass and removal is a bit more involved and I'd rather not go into it here for fear of being accused of single-handedly causing Global warming by the sanctimonious, holier-than-thou crowd...
.I'll get through a couple of tankfuls to see what effect all this has on fuel consumption and report back.
