lovestriton wrote:the problem only appears once the weather starts to get over 25 degrees
Discounting mods and chips that can mess with the readings from the sensor, this is the clue. It's not the sensor that is setting off the error condition but something downstream. RHK is kind of on the right track but you need to inspect, very carefully, the wiring between the sensor and the ECU, as well as any connections. Look for any signs of damage or anything that could cause an internal short or voltage fluctuation, especially where the wiring goes near to something hot or even is retained a bit tight or goes through a tight hole. Even something like an accessory wire that has been pushed through a grommet and has damaged the wire in some way.
A bit weird that it's triggering a limp but not recording an error but I suspect that the boost gauge has proven that it's not the actual boost condition that is causing the problem. Do you have a Scangauge or similar where you can monitor the actual signal? I'd be looking for boost conditions that don't match what's actually happening - like boost increases when you go around a corner with foot off the throttle - you know, weird stuff. You can shake or move the harness too while watching the boost, looking for something that changes the apparent condition.