EA888 swap into an older chassis

Hey gang, I have a potential customer who is asking me to wire his EA888 engine into an older chassis. Pretty unique build, its essentially a VW beetle on a custom chassis with a longitudinally mounted 2.0T mated to a Audi RWD transaxle of some kind all in a custom chassis. He wants it to remain somewhat period correct looking and OEM in function. I am more a USDM and JDM guy so VAG stuff is kinda foreign to me.

I am exploring swapping to port injection and standalone vs keeping the DI and standalone vs using the stock ECU and finding out what problems will arise. I am more of a standalone guy, but this project was started by another wiring guy who claimed he knew how to get it running on the stock ECU and the costs obviously increase with each other option. Syvecs is the only clear option for keeping the DI, as they have a PNP for these cars and my local tuner of choice is quite familiar with both Syvecs and the 2.0T on a Syvecs.

He has the ECU, stock engine wiring, bunch of body modules in a box, more wiring in a bag and the key and immobilizer unit. When the engine was sourced they pulled it out of a wrecked but running 2015-16 Golf I believe based on my research using the ECU part number which is 06K 907 425B

I have researched for a bit now and found other people have swapped these into MK1 and MK2 golfs, but little information other then that was provided. Being that this is a customer and not a personal project, there's a mindset shift from "I'll figure it out and deal with the consequences" to "how do I deliver the goals at a predictable cost?". If the stock ECU will run the car and be able to be tuned and not having the rest of the CAN data its expecting to see can be programmed around/out and he gets a turn key car, then great. If its a nightmare of "its doing this and no one can figure out why" then standalone is the route. I do not possess the skills to create a device to imitate stock vehicle CAN messages, so that is not an option either.

From what I've learned thus far, I'll need the immobilizer bypassed and possibly VSS of some kind fed back to the ECU. Gauges can be fed via OBD2. I'll be rewiring a portion of the harness to feed the ecu, injectors, coils, starter etc power and have the ECU control the fuel pump relay, fans etc. I do not know the best software/dongle to use for this either.

Has anyone completed this swap? is there caveats I need to be aware of? Any other context or insider info into this is appreciated

The CAN stuff can be such a nightmare on VAG cars, I likely wouldn't take this project on for a customer personally. Just too many variables like you've pointed out.

If you do end up keeping the stock ECU there is an open source software called Simos Tools that would have more options than most custom tuning software for altering CAN stuff. A pretty active forum/discord group of devs maintaining it who have the capability to provide file patches to change stuff like that. Between Simos Tools and VCDS/VAGCOM I think everything could be done with OEM ECU but I definitely understand your hesitations.

With enough time and money I'm certain you can sort it out on stock ECU or standalone, but it sounds like you'd be handed a pile of parts which may or may not be from the same vehicle, might not be all you need to keep the stock ECU happy, and then yes you could use OBD2 data for basics for a display, but if they want more than what's available on generic PIDs, this becomes not only more work, you'd likely want a donor car to sniff to sort things out. I'd ask the customer if they had a budget in mind to get some feel for how deep they're comfortable getting in.

Hmmm, I can understand the client wanting to keep the OEM ECU, to "save money" or "make it easier', but in 99% of cases this can be WAY more expensive when one takes into account the time and labour costs chasing down details and sorting out problems.

I'd suggest changing straight over to an after-market ECU that is an already known factor, and is intended to be, relatively, easy to install and configure. Selling the OEM may help recoup some of the expense, which could help.

I'd strongly recommend keeping the DI injectors operating, at least at lower rpm/boost and suppliment them with port injection for additional fuelling, with the DI running around 95% their nominal maximum in OEM tune - if you don't keep them cool by passing fuel through them you're going to melt the tips and cause all sorts of other issues. This means the ECU will need to have the ability to run two sets of injectors, which, fortunately, most can.

I do believe all the modules are from the same car, as they dismantled it.

But yeah, you guys are echoing what I've been told by two other local to me VW tuners. They all agree that it may work, but nobody knows how much effort or caveats are involved. Found a youtube channel that did a similar swap into a MK3 jetta but their solutions for wiring related issues were not acceptable as a final product in my opinion. But it did run and drive.

The customers power goal is very conservative, adding port isn't necessary outside of deleting DI and running a cheaper standalone. But the parts list to delete the DI and add port pretty much negate the cost savings of the ECU lol. I am also uncomfortable leaving the DI injector in there.

I looked into Simos tuner a little and its open source, which doesn't inspire confidence as something I can sell. Open source stuff great for personal projects but not ideal in this context.

As long as the fuelling demand is within the ability of the DI injectors to provide, within an acceptable 'window', you're probably going to be best to keep using them. I would have them cleaned/serviced, though.

I'm far from being a tuning expert, but I've been wrenching on my own VW's for about 15 years. As long as the ECU was from a manual transmission car, I would imagine that you'd be able to run the factory ECU and reflash it if you/the customer wants more power out of it. I would imagine that deleting the DI and going PI would be more trouble than it would be worth, but that's outside my expertise.

If you do decide to try the reflash tuning option, Eurodyne Maestro is an option as is HP Tuners. I've done a little tuning on older VWs (2003 AWP 1.8t) and had some pretty good luck with Eurodyne, but it has some oddities in it for sure in terms of some tables not even affecting anything.

I'm working on learning HP Tuners software right now for the MK6 (2013 GTI CBFA 2.0t) and the MED17 ECU is very different than the older AWP ECU. HP Tuners does seem to have nearly every option you could think of which could allow you to work around the CAN communications between other modules. I wish I had more practical application knowledge for you, but hopefully this at least gives you a little more info to work with.

Thanks for the insight Brian,

I was able to contact somebody locally who claims to have done this setup already. He's been working with a local VW tuner called TuneZilla. He also directed me to a CAN converter that can put wheel speed on the bus, as that seems to be the most critical data it won't have.

In theory making what he has work is the 'simplest' solution, but we'll find out. I believe the donor engine was from an AWD golf wagon, and it was automatic.

The discussion I am going to have with the customer when I accept this job is no guarantee of final product functionality, charging hourly, and I'm not tasked with tuning it/programming CAN. I'll be working with the tuner certainly, but it's out of my control in some aspects.

Glad you've got some local resources that seem like they'll work out well for you.

If you're keeping track of this build somewhere I'd be interested to keep an eye on it. Sounds like it should be pretty cool when it's done and I'm interested to see how you move forward with it. I would think that you could adapt the wheel speed sensors from the donor platform assuming you can get the sensor wheel for the hall-effect sensor mounted up. Lots of custom stuff for this build for sure!

Good luck!