Configuring Base Ignition Timing from Scratch

Hi guys, I am very new to tuning, this is my first project and I am a little confused as to what to do with my ignition map.

I have an L28et which I have added an intercooler to, otherwise I think its stock and stock 5-6psi boost. I calibrated the base timing to 10 deg (Or as close to it as I could see on the pulley) and I have my timing set at 4 degrees in idle - I have no IACV connected to this and was the only was I could stop the rpm increasing out of my prefferred speed.

Now the rest of my map is set at 10 degrees, I have the fueling to an acceptable point, the car drives well and from what I can tell with my Phormula audio knock detection, I can't hear any knock up to redline, or in steady state while tuning the fuel map. I'm driving up to 6500rpm wot and my knock levels are really low - I have the gain @ 7 Or 8 in the pc link software and my knock levels are around 30-50 at idle and max I'm getting is around 70-80 whilst doing a ramp run - and only 1 or 2 instances in total of this. I'm not sure if these numbers seem right, but I set the gain so I have some numbers whilst idling. I'm also not getting the growth of noise you would expect (30-40 idle, 300-400 6000rpm) it is basically super flat and low knock levels in the ecu knock detection.

So I guess my questions are - how do I start to build my ignition map other than wot ramp runs considering I have no base numbers to start from? (As in no safe starting map)

And should my noise levels be greater than what I'm seeing or is this a case of my timing being so flat that a super low knock detection is normal?

Thanks for reading and your help!

Hi Guys, I have found a few L28et ignition maps online to go off - I basically located a stockish map and retarded the timing everywhere.

Now while the car is driving and able to boost, im still really uncertain about the knock levels I'm getting.

I can set the gain to something to give me some numbers in the idle area of knock of say 20 to 50, but I find when I'm doing a WOT ramp run I am pretty much getting those numbers right through the rev range - and 2 or 3 instances of a random cylender knocking up to 200-400, whilst the rest of the knock being really low still.

Firstly I am not seeing the relative growth of noise to engine speed/load, which is making it difficult to set a threshold to increase my timing against.

I also have a Phormula knock analyser, and I am not hearing any knock whilst doing the ramp runs - and when the knock is spiking I am also not hearing that through the headset.

I am road tuning which is making it a little tricky, I'm worried to increase my timing as I'm unsure if there is an issue that I haven't figured out.

Any advice would be appreciated.

Can you post a PC log and a copy of your current tune. It doesnt sound right to have such low knock levels. What do you have for a knock sensor?

Hi Adam, thanks for your interest, hopefully you can see by my logs, I'm a little hesitant to go a full rpm sweep.

I have a Bosch donut sensor bolted to the block in line with cylinder 5 or 6, I have a second Bosch sensor around cylinder 3 on the block for my audio knock detection. Both are bolted right near the top of the block.

Feel free to pick me apart, I'm learning, but really struggling with a known good base map to work off for ignition.

Adam, I also notice the ECU is not pulling timing when I get an event over my knock threshold, any help would be appreciated.

Running 98 octane unleaded.

The ECU is not pulling timing because you dont have any Knock ignition trim tables set up. Since you have direct spark you can control each cylinder independently so I suggest you set them up like below.

You need much more overall gain. Try it about double where you are at now, we want to see knock levels ranging up to about 300-500 during a normal pull. Once you have reasonable knock levels coming through, you may need to adjust the individual cylinder gains so you are getting balanced levels from each cylinder.

Your ignition table doesnt look good. Typically you would have max advance at about 100Kpa MAP, (0kpa MGP), then taper it off as boost increases, yours is realatively retarded at 100kpa then increases with more boost.

I suggest you import the ignition map out of our Evo 4-8 base map as that has a much better shape in the off boost areas but is still very conservative on boost for your type of engine, it will be a much better starting point and it will drive much better down low. You can right click on any table to do the import/export.

I have incorporated the EVO4-8 ignition map, turned the trim tables on and increased the knock gain so I am getting some higher numbers now.

I have noticed that even when I give the ECU 0 degrees of timing to play with for knock control, I still seem to have the trims pulling timing which is confusing me. It even looks like I'm having timing pulled for no apparent reason here and there, when I'm well below the knock threshold.

I also am getting really high knock around the 4000 RPM for once instance on each log file, my timing should be low enough and rich enough that knock shouldn't really be occurring...

Have a look and tell me what looks wrong.

Thanks for the help as well, I'm still learning and struggling a little bit!

The knock system is not pulling any timing, you can see the "Cylx Knk I-trim" are all zeros for the entire log. The few big spikes of knock level that you have are always occurring after you have lifted off and the engine is slowing down so this is likely something like drive-train or timing chain backlash. You can use the TP and TP delta lockout to make knock control ignore those situations. I suggest you set up the PC Logging to record everything so you have more relevant data such as TP and knock system status logged to help set these lockouts up properly (hit the "add all button" at the bottom of the logging set up screen).

Having said that, Im still not convinced about the signal level we are seeing here, something doesnt look right. What is the part number of the knock sensor you have? There are some that have the brass center boss tied to the ground pin so it can short the signal to ground if you have the polarity wrong.

Have you watched this webinar? https://www.hpacademy.com/previous-webinars/049-setting-up-knock-control-link-g4-vipec/

Hi Adam, yes I have watched the webinar again and again, because I'm so new to it all some may be going over my head - but even when I follow step by step I'm getting to the point where I can't complete a ramp run because I'm getting those large spikes which is worrying, I also haven't heard anything at all from my audio knock detection.

My noise level also isn't growing linear with my rpm so it's hard to know if something's wrong - also there isn't one cylender which seems noisy, so I dont know if I could apply the individual gain levels to flatten out the Knk levels to set my threshold tonan acceptable level. They all seem to be able to spike - it still seems to be roughly the same level regardless of rpm, but with those huge scary spikes.

That sensor should be ok, but since there are so many knock-off parts around nowadays, can you get an ohm meter and check if there is continuity between the central boss and either of the pins. Have you got any other old sensors around you can try?

Can you tell me which log/time has this "scary spike", the only spikes I see are when you lift off?

I agree though, something appears to be not right with the levels reported and the gain needing to be so high. It this a Link loom?

Hey Adam, it isn't a link loom but the HKS loom I'm guessing - I went over all the inputs and checked they were doing what they should be. The knock sensor is a genuine Bosch, is came with the Phormula KS4. I am receiving my Knock from the KS4 output to the XS knock input.

I could increase the gain in the KS4 which I think would increase the noise to the Knock input - I'm just not sure if thats what I should do.

The knock signals I'm worried about are 1000 knock and 800 knock in the two logs I sent you, I'm not sure if I lifted because of the knock or if it happened because I lifted. Do to MAP and ign numbers look sensible in the logs? I was having a look and I think those 'scary' events were happening while I was at WOT, and I'm pretty sure I had 6psi of boost, but when I looked at the log it diddnt look like that - it also had me at 33btdc which just diddnt seem right for that boost pressure, I should have been at around 26btdc I think - so I'm not sure what's going on there.

I will turn on Log All so we have as much info as possible in the future.

Is there anything else you suggest I should look at? Should I just ignore those 1-off events or drag my timing back a bit more to see if that helps..but it is already quite retarded.

Far out man, you’ve got the knock sensor connected to some other device that filters it at an unknown frequency, passes it through unknown signal processing, then outputs some completely different signal that may or may not be related to knock to the ecu? It would have been good to mention that at the start...

connect the knock sensor direct to the ecu like it is meant to be and start over.

Hey Adam, thanks for the heads up - I thought the Phormula KS4 would have just sent the raw signal back out for the ECU to receive - first time I had head that its a silly idea.Anyway, I have rectified that and connected to the ECU directly and also have logged all parameters - my maps seem safe but I still seem to be getting large spikes - have a look at this latest log, I have a massive spike @ 4700rpm but still rich, and the timing doesn't seem too advanced for 130kpa...let me know what you think.

I would not be convinced you have knock there, but thats where your audio detection comes in to play, you use that to confirm that the noise the ECU is hearing/reporting is actually knock.

I would suggest you also update the ECU to the latest firmware, there where big improvements made to knock control processing speed in V5.6.5 and you are running 5.6.4.

I would also try changing the frequency channel to 6KHz narrow as that may block out more general noise.