The Roadster NA Journey Begins

gslender · Postby **gslender** » Mon Oct 19, 2015 1:04 pm

Also odd that the solution is to drill out the stock sensor and install a Haltech sensor!!!! And people complain about Megasquirt being hard to use

Sent from my iPhone using Tapatalk

Magpie · Postby **Magpie** » Mon Oct 19, 2015 1:10 pm

That solution, to drill out the old one was mine and mine alone. There are other ways, I just decided to get some aggravation out on the sensor. A bit of aluminium could have been used for the mounting bracket.

gslender · Postby **gslender** » Mon Oct 19, 2015 7:32 pm

Magpie wrote:That solution, to drill out the old one was mine and mine alone. There are other ways, I just decided to get some aggravation out on the sensor. A bit of aluminium could have been used for the mounting bracket.

Yeah, but the solution being to not use the OEM sensor seems poor regardless, and I'm just surprised given that you've paid so much to people whom you'd think would have been there and done that.

Sent from my iPad using Tapatalk

Postby **NitroDann** » Mon Oct 19, 2015 7:37 pm

Id probably +1 that.

Magpie · Postby **Magpie** » Mon Oct 19, 2015 8:37 pm

How many MX5's/Miata's running a PS1000 with OEM sensors with no issues have been built? This is a rhetorical question!

My engine is not a typical build and hence what works in a 'normal' build may not be the best solution. However, when things goes wrong how is it best to find a solution? Blame other people, complain about what has been paid? Yes this COULD be done but does it solve it? I also could have done what everybody else has done and built a Gary Stewart motor or go forced induction. I did not go the path most trodden and the reasons for this are mine aloneand not even up for discussion.

The Haltech tuner has NEVER heard of a similar issue, so why my build? All the steps taken to resolve are what the Haltech tuner would have done.

Why change from an OEM sensor if it has worked in the past?

However, the Haltech tuner did say that in all their high performance builds they don't use too many of the OEM sensors.

Consider this... http://www.superstreetonline.com/how-to/engine/htup-1106-ignition-breakthrough-tech/

Another important component of the C.O.P. conversion is the trigger system for the cam and crankshaft pickup. This is a magnetic Hall effect sensor that tells the ECU exactly where the components are in their respective rotation. T1 Race Development and AEM Electronics both have their own magnetic Hall effect sensor pickups. T1’s is commonly referred to as the T1 cam trigger, while the AEM unit is called the Engine Position Module, or EPM. Most aftermarket ECUs read much more precisely when using a Hall effect sensor over the OEM magnet units. Both are excellent choices, and both are simple four-wire hookups. I’ve used both the T1 and AEM systems with excellent results!

Or this, from megasquirt..... http://www.megamanual.com/ms2/pickups.htm

A tooth that is too long will 'spread out' the zero crossing, making the zero-crossing unpredictable, causing timing 'jitter'.

Gee this sounds familiar, why has the megasquirt people NOT suggested this?

Now the OEM cam angle sensor works perfectly with the OEM cam wheels. However I'm running toda adjustable cam wheels. Are these a perfect match for the OEM, Again a rhetorical question. Is my build 'normal' or 'extreme'

It is proven that the CAS DOES fail in my build. Not up for discussion.

The wiring has been re done to remove this from the equation.

If this does not work it will be me contacting Haltech and working through the issue with them directly. Simply it would then HAVE to be within the software not the hardware. However I could just post that a solution has been found irrespective of the actual result.

Corey · Postby **Corey** » Mon Oct 19, 2015 9:37 pm

Magpie and I spotted at WTAC!! Probably talking about all of Danny's beautiful fabrication work if I recall correctly.

Postby **NitroDann** » Mon Oct 19, 2015 10:00 pm

One feature of haltech ecu's are how the current software is setup, an example here is some of the triggering settings.

It works like this
Choose triggering filter level: 1, 2 or 3.

Pro: Easy to setup, your tuner wont tell you that your car needed 4 more hours tuning because he couldnt figure it out.

Con: 1, 2 and 3 are major adjustments to a complicated hidden system, and sometimes the system itself needs configuring, but the software doesnt allow it.

Its a Mac vs PC type argument.

Dann

PS: The cam angle sensor isnt that important when you have a great sensor on the crank, which is the thing you are trying to measure. Can you disable the CAS altogether?

Magpie · Postby **Magpie** » Tue Oct 20, 2015 6:25 am

Dann the filtering has already been tried, however only when the sensor has failed not before it fails.

An option would be to go back to the previous optical CAS.

Not sure about disabling the cam angle sensor, this by itself would not allow the car to run. A cam angle sensor signal is required from somewhere.

Corey good to meet and chat.

Postby **NitroDann** » Tue Oct 20, 2015 6:48 am

A few ideas to consider:

Magpie wrote:Dann the filtering has already been tried, however only when the sensor has failed not before it fails.
Its exceptionally unlikely that the sensor has failed, if I wave a screwdriver past it with it hooked to an oscilloscope I am sure a signal will be generated.

An option would be to go back to the previous optical CAS.
Optical is very accurate and reliable.
Not sure about disabling the cam angle sensor, this by itself would not allow the car to run. A cam angle sensor signal is required from somewhere.
The cam sensor doesnt do anything unless you have closed loop VVT. Otherwise the sensors are both trying to measure crank position and the cam sensor is wildly inaccurate compared to the crank sensor.
Corey good to meet and chat.

Magpie · Postby **Magpie** » Tue Oct 20, 2015 7:40 am

Thanks Dann. Agree the sensor has not failed, it is possibly the quality of the signal has. I will seriously consider your advice. I will email Haltech and ask is this possible. The software only allows a choice between certain triggers. From memory mine is set on Mazda B.

If it works initially then slowly develops a miss this would suggest a loss of quality in the signal or something is getting out of syncs between the crank/cam signals to the ECU.

This could be heat in the sensor. After one session I took the temp of the cam angle sensor (with a contact type) and recorded 40 deg c. This sensor had not failed. Now I should have done this each time I got back into the pits however on day 2 I had a fan issue so needed to do a cool down run after each session.

I will post the video tonight of when the car shut down on the track during the last session of day 1. It would not restart on the track but once back in the pits fired first time.

Dann I was in Newcastle (passing through) on Sunday shame your workshop was not open.

Postby **NitroDann** » Tue Oct 20, 2015 8:59 am

Sounds like signal degradation due to heat which is common for these kinds of sensors. Why the haltech filters dont handle it as well as other ECU's on this particular sensor/engine combo I dont know.

Magpie · Postby **Magpie** » Tue Oct 20, 2015 9:32 am

Agreed NitroDann.

Steampunk · Postby **Steampunk** » Tue Oct 20, 2015 10:14 am

Has anyone suggested excessive vibration as a possible cause?

Magpie · Postby **Magpie** » Tue Oct 20, 2015 10:16 am

Nope

MattR · Postby **MattR** » Tue Oct 20, 2015 10:30 am

Ok then......

Mark, it may be worth checking that vibration is a possible cause for your issues.

There someone did suggest vibration.

Seriously, I hope you find the issue soon. I know how frustrating niggling little issues are to diagnose and fix on any build especially modified and competition motors.

And when the issue is found, be prepared for it to be a "doh" moment.