Hard rev limit all of a sudden

[914531]

Running MSII, V3.0 2.891, Car ran fine up to 7,000 RPM consistantly for many races until hitting the rev limiter, now engine hits a hard rev limit at 6k RPM and will not go past, will miss once around 5500 RPM before going to 6k. Using MSD 6AL, MSD coil and electronic module in distributor. Have replaced all ignition components except MS. There were no setting changes in the setup before this happened, now cannot get it to rev higher no matter what I change in the setup. Any ideas on what could be causing this?

Thanks

914531

[Matt Cramer]

Could you please post a data log of the problem and a MSQ?

What's the 6AL's rev limiter set up for?

[914531]

The MSD is set at 7k and the MS rev limiter is disabled. I will post a msq and datalog from a dyno run later

[914531]

Here is the file loaded in Megasquirt and the corresponding datalog file from a dyno run

Please take a look is see if I am missing something in this setup

The engine is a flat 4 porsche 1.9L running E85 so the AFR table looks rich compared to gas but runs at a comparative 12.5 AFR from 4k RPM up

[kjones6039]

A couple of things jump out at me right away............

~340* coolant temp

Then at ~3717s it drops to 180 and goes flatline?

What's with that?

At WOT, ~6000 rpm, pw is ~11ms and AFR is 7.9

Kind of rich notwithstanding the E85, huh?

I will be interested to see what the experts have to say!

Ken

[Matt Cramer]

In addition to what kjones noted about the coolant temperature, you're losing the RPM signal at 6000 RPM. What exactly is wired to pin 24?

[914531]

Ken and Matt

This is an aircooled engine so the coolant temp is really the cyl head temp, which has no effect on anything after 160 degrees, basically it is used for WUE and thats all, so it applies for about 30 seconds. On E85, AFR can run from 7 to 8.4 for max power so at 7.9 it is right in the range. The temp sensor flat lining at 180 degrees happens in both engines we are running so we could probably change the sensor but since it is not a big part of our setup we ignore it.

pin 24 is the tach signal, I did notice the drop to zero right at the spot it breaks up and this is what the problem is. I am thinking this is a component in MS that is failing and was wondering if anyone else has experienced this issue. This particular setup has the 6k issue along with all the older setups that ran fine until this started to happen. I ran many races before with this setup running to 7k always and now this 6k rev limit issue appeared. Since I have replaced all the other components of the ignition system, it kind of points to the computer. Just wondering if anyone has had to replace any internal parts due to an issue like this.

I have also changed the fueling equation at the 6k rpm to both leaner and richer mixtures with no change. I have also changed the timing as well as moving the trigger offset point from 2 to 8 degrees and redoing the timing setup. Nothing appears to move this hard rev limit at 6k

Thanks for your help and interest

914531

[Matt Cramer]

By "pin 24 is the tach signal," do you mean it's wired to the MSD tach signal, or something else?

[914531]

I trigger the MS from the electronic ignition module (+ wire) to pin 24, this is an Accufire module that has worked well in the past. To make sure it was not the module I replaced it 3 times, twice with new ones and 1 from a running identical setup on another car. I think the next step will be to pull the board out of the box and resolder all the jumper wires used to be able to use the Hall sensor.

[Matt Cramer]

Ok, I take it this wire is also hooked up to the MSD white wire?

How exactly did you set up the RPM input circuit?

[914531]

I will have to pull the box and verify the jumpers as it has been a couple of years since we built the MS. I will let you know once I verify those. And yes, the MSD white wire is also tied to this as the MSD is the conduit for the trigger. The techs at MSD, when I thought it was the MSD that was causing this, were quite certain I was getting RF interference on the trigger wire and wanted me to make sure it was separated from any other major wires, which made sense, but had no effect on this problem once I isolated and shielded the trigger wire from the MSD

[914531]

Matt

Here are the jumpers we installed when we built the MS based on the instructions for Hall effect sensors using an MSD
IGN to D14
IGBTIN to JS10 IGN
S12c to C30 (one side)
Tach Select to XG1

I see another set of instructions for Hall sensor, Optical sensor or points that show:
XG1 to XG2
OPTOIN to TachSelect
TSEL to OPTOOUT

Which appear to be for an optical sensor. I have always had a timing mark that is never rock solid with the timing light
Thanks 914531

[Matt Cramer]

Ok. Try jumpering D1 and D2 (if installed) and remove C30. Does that make any change?

[914531]

I will do and let you know

Thanks

[914531]

C30 was not installed, we also had jumpered both D1 and D2 already

In looking at the datalog again, I noticed that when you watch the trigger + or - it stays at 0 until it gets to 6k rpm at which point it drops sharply down to a large negative number, around -40. What does this this track? Is this changing the timing from the table amount? or is this the actual timing variance to the timing table?

This seems odd that it starts to change where I am getting the hard break up, any thoughts on this

Thanks

[Matt Cramer]

That basically means it's losing the signal.

You might want to try bringing the tach signal in through the VR conditioner - this one has a bit more noise filtering and is adjustable.

[914531]

We will give that a try.

[grippo]

In looking at the datalog again, I noticed that when you watch the trigger + or - it stays at 0 until it gets to 6k rpm at which point it drops sharply down to a large negative number, around -40. What does this this track? Is this changing the timing from the table amount? or is this the actual timing variance to the timing table?
Thanks

When trigger+/- counts down it means that it is losing tooth pulses, that is,a pulse is coming in much slower than the previous two, as if there were a missing tooth, but the missing tooth is not where its supposed to be. If this only happened once in a great while, the ECU would insert a tooth and keep going, but if it happens twice in a row it declares resynch (and forces rpm to 0), then it waits to get two regular teeth, then looks for the missing tooth, etc. It works like this precisely because the timing can get very screwed up if the ecu were to keep putting in teeth, so there is a lot of error checking and if things don't look right, the safest course is to resynch. When it resynchs the timing goes back to normal, so at worst you might miss a tooth, go back to sync, miss a tooth, etc, but you won't end up 180 deg out of time.

If trigger+/- counts up, it means its seeing noise, that is extra teeth that don't really exist.

[914531]

So as the negative number increases (from 0 to -40) is the computer adjusting the timing back or taking it to 0 advance? Next question is what would cause this and is there a setting in Metatune to correct this? Since this started happening after a couple of years of no problems and not changing anything in the ignition system, did something fail? While running in the car the spark craps out at around 6k RPM, putting the MS on the stimulator and watching it in Megatune, the rpm goes to about 6800 before breaking up. Any ideas on what is the cause of this? We are using a Hall effect sensor in the distributor with an MSD ignition, and always have.

Thanks

914531

[grippo]

The ecu would only be changing timing if it were adding a tooth that wasn't really missing. But it could only do this once, and if the tooth weren't really missing, it would see a tooth miscount and set resync flag, meaning you should also see rpm flip to 0 while it is resyncing. If more than 1 consecutive tooth was missing it would immediately resync, again rpm goes to 0. But if you see rpm at a reasonable value while trigger counts down, then it means the ecu is putting in missing teeth where they should be and making things work. But this is unlikely unless there was a bad tooth, for example too short, so it gave a weak signal compared to the others.