Tuning Link G4X on Ls3 Twin turbo

I Have a sand car that I used to run Autronic injection .so I thought I would update with a little bit more modern injection. I just switched over to Link storm G4X. Originally Ben Strader tuned the Autronic . I have not found any tuners that know link in my area (Las Vegas Nevada). I have been trying to tune it myself. As a novice tuner it is getting close. But it does not run anything like it used to. This is an 416 cubic inch ls3 twin turbo with big Air to air intercooler. ID 1300 injectors . The Autronic system was tuned with Ignition on Map and Fuel on TPS. and I was Running 500 CDI with with M&W coils in wasted spark. Now with link I am running 8 Gm coils with both fuel and ignition tuned off the map. I posted a pic of my old timing map vs the new one. cause they look very different. But when i look at the areas that really matter they only seem to be 1-2 degrees of each other . except old map used a lot less load cells. The car has way more turbo lag now. I used to be able to smash the throttle in any gear and it would spool up and spin the tire. on A hub dyno it did make good power. 960 hp with a mild tune at 16 lbs of boost. That was on Vp 110. it just seems to have nothing on the bottom end. could this be because it used mapped with fuel being throttle or tps and now fuel is mapped on load? Hope This makes since what I am trying to ask. I would post up my tune with a data log. But it will not let attach that file type. Can any one recommend any good tuners in the Las Vegas Area? Thanks Ken Buys

Might be a silly question, but did you confirm the ECU and physical TDC lined up?

Looks like you have no advance for much of the range - that is going to kill your engine's performance?

yes timing on engine and the Ecu do match. The timing I am running is conservative. But the timing in my old injection which does not have as many load cells. (the first pic) Ran really good it would spool and build boost instantly. had very little lag. It is not even close to running like it used too.

Hi ken,

its obvious the previous ignition table has more degrees advance across all the zones than your new one. For example the 5000 rpm 100kpa has 6 degrees difference, if your set up is the same and is not knock limited, i suggest you add the missing load points off the load axis from the previous table and copy the numbers with reference to their rpm and then interpolate between the copied timings. I believe that will get it spooling and responsive as before.

since he was using less cells and higher timing numbers, the ecu was interpolating between these high numbers which makes it logicaly higher from your new table all across

hope that helps.

yes ok i will give it a try and see if it helps Thank you for looking at it. Ken Buys

@ Ken, any chance you can give us a log of some driving and a copy of the tune? A single screenshot only paints a very small part of the picture.

FYI, you could set up the Link ign table axes to match the autronic exactly if you wished. But the Link one doesnt look a million miles off to me, the 100 & 120Kpa rows are probably where the biggest differences are between the two, but if anything the Autronic one looks a bit on the aggressive side around that MAP anyhow. At least my experience with aspirated LS3's, they dont generally want more than 22-24deg at 100kpa for most of the RPM range regardless of fuel. Im not sure how much influence your potentially lower compression ratio has on that observation but I have tuned a few different variations now and that seems to be a common theme for me.

How different are all the other ign compensations between the Autronic & Link - for example IAT trim, ECT trim, transient retard etc?

Have you confirmed the timing matches at both low RPM and high RPM?

Is the accel enrichment set up and working correctly?

How significant is the delay in MAP Vs TP when you stab the throttle open?

How well does lambda track target under heavy acceleration?

How was the ignition calibrated? - Was it steady state tuned to MBT in at least the lower load/RPM area's?

Adam, Thank You for the reply. when I try to upload my tune or a log on this forum it gives me an error that file type is not allowed. Perhaps I could email it to you.

Yes the ignition timing does seem to match at low and high rpm. I just copied my old ignition map exactly and was going to it a try this weekend.

lambda does seem to track under heavy acceleration for the most part. But I do have short exhaust pipes coming off the turbos. so when off the throttle it does show lean. It goes through lambda sensors very fast. I think I am already on number 7 or 8. Not sure if this is from the leaded fuel or the short exhaust letting cold air in.

Perhaps it is in the accel enrichment. I have not played with that much. Like i said I am a novice tuner and know enough to be dangerous.

I really would like for you to take a look at my logs. Another Issue it does have. If i try to do a full throttle pass from a stand still It does fall on its face. but then it comes back and runs fine. I was thinking it was a trigger issue. so I added my old 4x MSD pick up that i used to run with my Autronic as a test. and it did exactly the same thing.

I dont think The Autronic was set up with many Ignition modifiers as far as IAT or ECT but i could be wrong. I will need to open my Autronic tune and take a look at it.

Let me know the best way to upload my tune and log. Thank You again Ken Buys

If you're using a high octane leaded fuel, the lead will kill the lambda - some say it starts to show in as little as 2 hours, with them being U/S after a couple of hundred - but that'll depend on the lead levels.

What fuel are you using, there may be lead-free alternatives?

This forum is a bit funny with some attachments, try putting your files inside a zip folder.

Gord I am sure it is the lead in the fuel. I am running VP 110 . unleaded race gas gets real expensive when I use 55 gallons a weekend

Adam I think I got my tune and one log zipped and attached.

Thank You for the help Ken Buys

Ok, I see a few potential issues, not all of them are related to your issue but stuff you want to look at anyhow... And some random comments in no particular order:

1. Lambda 2 is not working, I suspect this is because you have 2 CAN channels assigned to CAN Lambda. You actually only need 1 assigned and that will receive data from up to 8 lambda devices. Im pretty sure these 2 streams will be overwriting each other so is probably the cause of no lambda 2. So switch off the CAN channel 1 which you have set to ID 951, then check what lambda 2 status and error runtimes are reporting.

2. You have some dropouts in Lambda 1, this could be caused by the same issue #1. So once that is fixed, keep an eye on Lambda 1 status and error runtimes to confirm that is ok. The Link CAN lambdas perform a lot more sensor validation tests during operation than most other wideband controllers, some of these tests are quite sensitive to large voltage spikes or inductive noise on the power supply, so there could potentially be a noise issue causing false error conditions which will cause the lambda to be reported as 0. You will typically see error 16, 33 or 34 reported if there is noise upsetting them. I would suggest adding lambda 1/2 status, error and temperature to the ecu logging set.

3. Under cruise and light acceleration conditions your lambda is very lean but its hard to know how much of that is due to the short exhausts. Have you tried richening it up in those areas just to confirm it is not actually as lean as it suggests?

4. You have a tiny amount of accel fuel being dumped in all the time due to noise on the TPS signal. TPS delta was not logged in your log so have a look at that next during the next run and make your accel enrichment deadband a bit higher than the normal noise level. Around 0.6%/100ms is typical and yours is at 0.1 so its fairly sensitive at present.

5. Your Accel enrichment is possibly a bit excessive, you can see in the pic below when the throttle is opened your injector PW is about doubled and shortly later the lambda goes very rich, you cant always rely on the lambda feedback in these situations but when you do have too much accel fuel it can make them feel pretty unresponsive. Your sensitivity numbers are much larger than typical. You have values of 20, our LS3 race car has 2.0. I would also suggest enabling asynchronous injection, set the min async PW to 0.5ms, this greatly reduces the requirement for accel enrichment, try your sens numbers down around 2.0-5.0 if you enable that.

6. This one is not related to your issue, but I did notice your RPM limit has a long decay on it, im not sure if that is on purpose or if you have experimented with it much, but my general experience is this is good for stock road cars but can make a powerful engine feel like its completely dead for eternity after you hit the limiter. A more typical cut and trim decay would be ~ 50ms, compared to your current 500ms.

7. In many places in your log you will see the ignition angle is flatlined at 15deg so that is a long way off what your ign table is commanding. There are two causes of this, one is quite obvious but the other is not. So the one I cant confidently answer is in a lot of the cruise areas where the ign angle has flatlined, im pretty sure this is the idle speed control kicking in so your advance will be locked to the target idle ign angle until the dashpot function is finished. But from the data in this log it appears in some areas that it is kicking in when not all lockout conditions are met. I suspect some settings have been changed in the map Vs what was in the ecu when the log was done? Drop the AP'TP lockout to a more normal 0.5% will likely solve that. Now the other reason for the flatlined ign was during WOT so this one is clearly not idle ign related. But I can see the cause and will use the pic below to help explain. You can see the ign angle has flatlined between about -1:14 and -1:13, this one im pretty sure is the idle ign as described above. But then again at about -1:12.6 you see it flatline and it stays there. You will see at that point the inj PW briefly dropped to zero, battery voltage spiked up a little and the trigger 1 error counter incremented. So what has happened is something has caused a trigger error, when the ecu loses position it sets RPM to zero, cuts all spark and fuel until it resyncs from scratch. We should have seen the RPM drop to zero but this has been missed due to the logging rate. Since the RPM has dropped to zero and then recovered the ecu has applied the start up ign target and holds it for the start up hold and start up decay times (idle control settings). We have changed the way this works in later firmware since it was potentially dangerous so updating your firmware will stop that happening again. You still potentially have a trigger issue but there is only one trigger error in the whole log so it is going to be difficult to diagnose that unless it happens more regularly. So just keep an eye out for trigger errors for now and investigate when/if you have more examples.

Adam, That is quite the diagnoses. Thank You Very Much. Can I send you some money for your time? I think as far as the Number 2 wide band not working. I stopped replacing both wide bands. it goes through them so fast and I am not running it in closed loop. I have just been using #1 for data logging. I will try to completely remove the number 2 and plug it off. Hope that will fix the conflicts.

Good catch of the timing being what it is how it is locked at 15 degrees. im not sure why it looses trigger when leaving from a standstill it only does once. then it is fine. I went as far as changing it out. I installed my old MSD 4 x magnetic crank pick up. and it still did the the same thing. could the cam sensor also cause the same rpm to drop to zero? But I will update my firm ware and try switching to asynchronous injection to on. If it would help I have more logs I could upload. But I am not asking for you spend your time for free. Thank You again. I think with this info its a great start. Ken Buys

I will try to completely remove the number 2 and plug it off. Hope that will fix the conflicts.

It wont, you just need to turn off channel 1 as you have clashing ID's at present.

I went as far as changing it out. I installed my old MSD 4 x magnetic crank pick up. and it still did the the same thing. could the cam sensor also cause the same rpm to drop to zero?

Yes, the cam sensor is just as important for a multi-tooth crank wheel. Your cam edge must also be reasonably centred between two crank teeth, so check that on the scope.