Suzuki Club UK

MS has a global parameter that changes the MAP scaling.
It is possible to disable the baro update, you just change the voltage at which it sees WOT or MAP to a value that it will never reach but then when MAP exceeds baro you get a 'negative' value that the ECU sees as a very large positive value. I looked at this a few years back when I was considering supercharging the 16V in the 413K and running a Jimny ECU. I gave up on that idea after seeing too many cracked blocks. The 413K has been in storage for 3 years but will be coming here shortly as my new garage is now relatively complete. Its a pity you are miles away, I'm still planning on running the Jimny ECU so I'll be looking at the code again. The 81A11 uses a 16-bit dual core processor and has a lot of power, a turbo tune would be good to have.

Have you got an A or B block 16v?
I've not managed to crack a block yet so hopefully my luck will hold out
G16A blocks are quite easy to get hold of which is good!

Does the Jimny ECU have knock capability? Thats something I like the look of from MS, even if I have to use 2 knock sensors to get a decent reading it would be useful to have when tuning with a charger

Also you're not that far from me, on the way to Scotties anyway!

ScottieJ wrote:Hang on all Twiss does flat out is drive....everything else takes forever

I'll be impressed if this engine is done by 2015

You're going to see an update soon with a bag of components and a circuit board!
I'm going to get fuel injected before you get J20'd

I do sleep flat out as well!

The G13BB doesn't have knock sensing as standard, I think the M13A does. Last time I looked at the MS knock sensing it was next to useless anyway as it wasn't properly implemented.

From what I've been reading apparently it is functioning correctly now, including the system to automatically retard or advance to get it to stop knocking. You can pick voltage levels at which a knock is detected (dial out engine noise I guess) and set thresholds of RPM and MAP values between which the knock sensor will run.

Got to be worth a shot if it helps to not blow up an engine!

all these insults. ouch.

twiss wrote:From what I've been reading apparently it is functioning correctly now, including the system to automatically retard or advance to get it to stop knocking. You can pick voltage levels at which a knock is detected (dial out engine noise I guess) and set thresholds of RPM and MAP values between which the knock sensor will run.

Last tme I looked the signal wasn't properly windowed so the performance of it was quite poor because its integrating all the noise from the valve train.

Yeah I have heard that with MS, you have to be really careful about where you place the knock sensors to ignore the valvetrain noise... I was reading something where it was compared to putting high compression pistons in some kind of chevvy, apparently then make the combustion a lot noisier and you have to relocate the sensors on the block to get rid of the noise...

There are lots of problems with a DIY knock sensor install. Before I moved to designing engine management systems I used to design piezo-electric (aka knock) sensors and interface hardware.

You only get knock between around 10 degrees and 70 degrees ATDC (engine dependent) so a proper interface should only measure during that window. Many aftermarket interfaces measure during a fixed time period so the window gets longer as revs increase and as the window increases you pick up valvetrain noise. The bodge is to increase the knock threshold but its not an ideal solution as the knock signal actually gets smaller with RPM.
Without accurate reference equipment you will need to guess the threshold.

First off the location is quite critical, the sensors are highly directional so you need to know in what direction the knock signal is travelling. From what I remember the G16 block has a boss for the sensor so that shouldn't be too bad but it needs to machined flat. The flatness and squareness of the mounting is very critical because if its not 100% rigid then the sensor response will be all over the place, we had many issues just mounting them to the shaker table to test them. If you look up the spec. for the sensor it will give the required tolerances. The problem is that there is no way to test the response on the engine so you won't know if its working properly or not.

I did some testing a few years ago on the Jaycar interface with a Bosch sensor:

http://www.rhinopower.org/knock/knock.html

Ahhh excellent, cheers for the info on that... may or may not be worth it then... if I can get the parts for cheap it might be worth a go, even if its just to give me a rough knock indication and not to automatically change the timing...

You said you are still using the dizzy on yours... I dont suppose you know any info about how much advance the engine should get at different revs? Does the fuel injection dizzy have mechanical advance or is it just a 8* BTDC signal? (think its 8* on the 1.6?)