CLT and IAT mysteriously incorrect

[ZetecVan]

Hi all

Megasqurt: MS-I V2.2
Firmware: 029q2

I've been running my Megasquirt quite successfully in my Ford Anglia van running a two litre silver top Zetec and Jenvey throttle bodies for the last 10 or so years until recently. Once it had been rolling roaded back in (about) 2007 I'd not touched any of the settings. Then one day, it was awkard to start. I noticed the warm up LED wasn't on. I waited on the drive with it ticking over for a few minutes before setting off. It was running a bit lumpy. Then after a mile the warm up LED came on. By this point, the engine was actually warmed up. It was running too rich so I came home.

On the Stim (basic stim, not a JimStim) the CLT temperature reads either 102 degrees, or when it's fully clockwise 77 degrees. There is no inbetween. The IAT reads either 102 or 16. Again, there's no inbetween.

With the main chip out, I've measured the voltage at pin 25 (CLT) and 31 (VREF) and I'm getting between 3.54v (with CLT on the stim fully anticlockwise) and 4.96v (fully clockwise), and a similar figure between 24 and 31 for the IAT, plus all the values in between.

The firmware version is 029q2. When I installed the engine, I can't remember changing any settings for the coolant sensor, and I do remember buying one off ebay, so it's probably a 'recommended' sensor.

I've reflashed the chip with 029q2 again, plus more recent versions just to see if they work, but I still get the same 102 or 77 degrees result. I'm back with 029q2.

Can anyone point me in the right direction?

I've attached the MSQ. Note that the warm up enrichment settings have been played with since the failure.

Cheers

Mike

[arran]

If the signals seem to be getting to the central processor correctly when the MS is on the work bench, is it not a possibility that the problem might be with the vehicle wiring?
Those two sensors share a sensor ground so I'd be looking at that

[ZetecVan]

The first thing I checked when it was in the car was the wiring. I thought it might have been shorting out, or had come broken but it's all fine.

You're right in saying that the signals are getting to the CPU when it's on the bench. But when it's on the bench, the CPU doesn't seem to be interpreting them correctly, because I only get the two values in Megatune, so I'm thinking the problem is with the CPU. I've tested it with a spare CPU I had, but I get the same result.