Fix what I have or start anew.

I've been trying to road-tune my BEAMS 3S-GE AE86. To clarify — I haven’t touched the main maps; those were professionally tuned in California a few years ago. I only tried adjusting a few settings to fix cold start/cold correction issues, but I went down a rabbit hole of different adjustments and now I’m not sure if my problem is mechanical or if I accidentally enabled something that wasn’t active before and threw the tune off. If anyone is wondering I did a compression test and my compression is normal.

Since my initial tune, I’ve added a CAN-Lambda wideband, oil temp and pressure sensors, and an IAT sensor.

Current issue: Around 5400 RPM at WOT, the throttle fluctuates — it sounds like I’m hitting redline early and I lose power. The log shows a TPS delta of about 27%.

I’ve attached a link to today’s log and my ECU file if anyone wants to take a look or offer insight.

https://gofile.me/79pDV/JJUFUhuFg

Sorry for the slow reply, I dont visit here very often. If you want quicker help with a Link ECU next time you will be better to post on the Link forum.

Anyhow, your issue at ~5400RPM is a trigger error. I would say this is probably because your trigger 1 filter level is set to the max level so you will probably be filtering out real edges at high RPM. I generally wouldnt use anymore than level 2 filtering for a 36T wheel, level 1 is the norm.

A couple of other issues to look at:

Your VVT is not working. The tooth offsets are very odd so Im not sure what has caused that, I'm suspicious it may not have ever been calibrated by the original tuner?.

Your lambda is reading very lean - too lean to be realistic, the lambda controller is also reporting excess pump current, so it either has a big air into the exhaust, or possibly the sensor is faulty.

Acceleration enrichment is zeroed out at coolant temps above 40C, so that is likely affecting drivability.

There is some other odd settings such as a 2nd fuel table and a 4D fuel table set up and enabled but with no input set to activate them so they arent actually doing any harm, just potential to cause confusion or errors in future.

Adam,

Firstly, thank you for your reply. I did submit my post to the Link forum on Facebook, as well as to several AE86 and BEAMS groups, though most are still pending admin approval.

After looking deeper into the setup, I also noticed that the VVTI wasn’t configured properly. I don’t think the tuner realized this was a BEAMS Blacktop and not a 4AGE Blacktop. That said, the BEAMS 3SGE uses dual solenoids, but it was set up for a single solenoid configuration. That’s the default cam control mode out of the box. I changed these settings prior to the latest run, but couldn’t figure out how to assign the trigger to exhaust bank 1 or 2. The solenoid itself is definitely plugged in—I’m just unsure how it should be configured in the software. Should I run a calibration on the Cam control? If so, how do I do that?

The tuner also wasn’t very familiar with Link software. The acceleration enrichment being zeroed out, however, was me experimenting. I filled in values based on the HP Academy course, but the behavior didn’t change, so I reverted to test if that was the cause. I can easily restore those settings.

As for the fuel pump overcurrent issue, I wondered if it could be related to the fuel pressure regulator. It’s set at 62.5 psi (431 kPa), which should be correct for this engine. When it was originally tuned, I had an exhaust leak, but I’ve since replaced the gaskets and resolved that. I’m also considering whether the exhaust solenoid might be triggered at the wrong time—opening the exhaust valves when there isn’t enough inlet lift. I remember from the VVTI course that this is a continuous vvt system, but with the trigger not set up it could be telling the valves to open too soon?

I believe the 2D/4D fuel tables being active was also my own doing. I’m not entirely sure where that input should be defined, but the engine ran like that for quite some time without errors. It doesn’t help that the HP Academy Link training uses an older version of the software, but I’m getting more comfortable with it every day. I’ve been experimenting with different settings since April 2024, and I now have around 40 versions of this tune. I recently started the HP Academy program after exhausting most of the YouTube material on Link and tuning in general.

Best,

Timothy

Here is the awesome link support forum:

https://forums.linkecu.com/

"That said, the BEAMS 3SGE uses dual solenoids, but it was set up for a single solenoid configuration. That’s the default cam control mode out of the box. I changed these settings prior to the latest run, but couldn’t figure out how to assign the trigger to exhaust bank 1 or 2. The solenoid itself is definitely plugged in—I’m just unsure how it should be configured in the software. Should I run a calibration on the Cam control? If so, how do I do that?"

No, the VVT control on this engine is what we call "single solenoid", the "dual solenoid" option is for systems such as the BMW high pressure vanos that uses 2 solenoids per camshaft - one for advance, one for retard. The settings in our Altezza base map are correct if you need any reference.

After you have restored the settings back to the correct values then with the engine warmed up and held at a fast steady idle (ie above the 1400rpm lockout), set the cam angle test to "calibrate", it should then auto calibrate the tooth offsets and turn itself off after a couple of seconds. Check the cam position follows target after that.

"As for the fuel pump overcurrent issue, I wondered if it could be related to the fuel pressure regulator. It’s set at 62.5 psi (431 kPa), which should be correct for this engine."

Not the fuel pump. "Error 54 - Excess Pump Current", is being reported by the lambda controller. This is related to the oxygen ion pump inside the lambda sensor, excess pump current means there is either more O2 molecules than there should be, or the sensor is damaged.

Adam Thank you for the advice. I got it running like it was before with your help. I think the main thing was the reluctor sensor being set to high. I set it back to low and also reset to solenoid setting to single solenoid, and calibrated the cam. I still have issues with cold start but that is another issue. I am going to make a new thread for that one in the webinar forum and the link forum, because I have it set to carge temperature approximation table.