When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
This was one of my latest project's, my boss has had some bad experiences with MS, so when it came time to order an ECU for his brother's 1.6 turbo, he said "anything but MS". Therefore, I had the opportunity to install Link's supposed "plug and play" 1.6 ECU.
First impressions: I LOVE that they use the stock ECU case. OEM mounting, elevated off the floor and well supported. Love Singular's solution, but by the time you bolt an ECU to a bracket, it sits a little high, and it comes with brackets from the factory, why not use them.
Secondly, their software PCLink is much different than TS, and after tuning a dozen or so different platforms, I'm ok with that. It's much more like an Excel spread sheet, rather than tunerstudio's vague user interface that's left me frustrated more than a few times. I've now messed around with an E36, FD S15, and this Miata on Link, and really love it's versatility, programming freedom, and so far am amazed at it's drivability.
Now on to Link's claim of being PNP. Yes, it is. If you don't want TPS, a wideband, and are ok leaving your FP on all the time, yes, it's PNP. I've attached some wiring suggestions on how to fix those issues.
If anyone has any questions on the above or anything else with this PNP, feel free to ask me here, I'll answer them to the best of my ability. Link has a base map that'll start a stock car, however I'll post the map I ended with using ID1050xs, modeled fuel calculation (similar to what we're used to with MS), among other improvements from their base map.
So I'm going through this installation process again, now on my own car. Definitely makes all the little annoyances with the MX5 link all that more annoying when you're not getting paid to fix them. I'd really like a board with 12v and gnd on the CAN plug, a fuel pump that works without re-pinning the factory AFM input to an available output and then jumping it at the AFM plug, has 5v pinned to the factory TPS, and has a tach output pinned to 2I.
I don't feel like I'm the only 1.6 owner who's deleted the factory ignitor, AFM, and non-VTPS. I'm talking to a friend who's built some PNP Link adapter boards for other models, so I'll get exactly what I want, along with an onboard Spartan 2 wideband. That being said, at least all my sequential wiring from my MS is all still functional.
I also learned from the Link forums (which is probably where I should be discussing all this) that there is an issue with the 1.6 sensor grounding method. Mazda grounded them at both the engine and the ECU. Which explains why I've often seen coolant and air temps increase with RPM so consistently. Maybe this is a known issue, but it's made it past my inspections for all these years.
Long ago I converted my harness to an NB one, at least the connectors and routing, so all my grounds are at the throttle body, and I'm guessing I transferred this issue to there. To test my setup, I plan on unplugging the ECU, and I shouldn't have continuity to ground on my TPS, cam sensor, crank sensor, IAT, ECT, or wideband.
If your coils are working in test mode, that means your wiring is correct. If they're not working while cranking, that means the ECU isn't triggering them.
I see in your Link forum thread you've tested trigger inputs. I'm wondering if it has something to do with the start input. The start position on DI4 didn't work on the first '89 Link I installed (even came with a smooth diff cover), I've used this virtual output ever since, and it's worked reliably, maybe give it a try, but I'm not sure if it'll change anything. Once you set this as a GP output, go to your digital inputs, start position, and change DI4 to this GP output instead. My theory is it's waiting for this input to begin injection and ignition, but never getting it. Although I believe that first '89 was running before I started using this, just starting poorly.
You have to remember, Link produced this ECU for whatever reason to bolt into a bone stock 90-93 and have every single factory ECU I/O function. Therefore, your AFM still needs to trigger the fuel pump, you have no TPS, no knock input, etc, etc. Keep this in mind when getting these ECUs up and running.
If your coils are working in test mode, that means your wiring is correct. If they're not working while cranking, that means the ECU isn't triggering them.
I see in your Link forum thread you've tested trigger inputs. I'm wondering if it has something to do with the start input. The start position on DI4 didn't work on the first '89 Link I installed (even came with a smooth diff cover), I've used this virtual output ever since, and it's worked reliably, maybe give it a try, but I'm not sure if it'll change anything. Once you set this as a GP output, go to your digital inputs, start position, and change DI4 to this GP output instead. My theory is it's waiting for this input to begin injection and ignition, but never getting it. Although I believe that first '89 was running before I started using this, just starting poorly.
You have to remember, Link produced this ECU for whatever reason to bolt into a bone stock 90-93 and have every single factory ECU I/O function. Therefore, your AFM still needs to trigger the fuel pump, you have no TPS, no knock input, etc, etc. Keep this in mind when getting these ECUs up and running.
Thanks, that's.. that's smart. I'll try that as soon as I can. I've got a VTPS that's working, and the FP is hotwired via the old AFM connector.
It's idling at perfect stoich now. Wasn't enough voltage to do the non-injection timing step - sparks too weak with not enough dwell, not enough power to run the timing gun and pick up such weak sparks.
contact link for info on how to get knock out of the g4x - they can provide instructions
there are TWO knock inputs on the board, but they are not marked and not wired up by default.
I mean, on the NAs there never was a knock sensor out the box so I guess, it's pnp ?
Anyway, that's that.
Last edited by atlex; 10-20-2022 at 02:37 PM.
Reason: contact link for info on how to get knock out of the g4x - they can provide instructions
Good to know, I’ve got to do this on my own. Very annoying, but yeah, they don’t have it from factory, so why bother! At least I know the pins now, appreciate it.
why knock 1, knock 2, and sensor ground? Is one being used as a shield?
Good to know, I’ve got to do this on my own. Very annoying, but yeah, they don’t have it from factory, so why bother! At least I know the pins now, appreciate it.
why knock 1, knock 2, and sensor ground? Is one being used as a shield?
There are two inputs for knock if you have two knock sensors.
The sensor ground is to be used for ground for the sensor, as well as the shield.
IMO they should have given the option to run it through the DI system. I mean, what is the expansion loom for, right ?
But yes, you should run shielded, and I don't think Link sell shielded cable and the looms for the expansion ports aren't shielded so.. it vaguely makes sense that it's up to us to do it right.
07-26-2021, 10:41 PM
How to make a Link G4X PNP...PNP
4
