Duratec / Mazda L-Series MZR I4 2.0 / 2.3 / 2.5 Swap

[my8950]

Today's Duratec tid-bit.... As mentioned earlier in this thread, Esslinger sells a Duratec to SBF bellhousing adapter which allows you to use most any transmission which would bolt up to a Small block Ford.
https://esslingeracing.com/ford-4-valve ... -adapter-1

While pricey at $385, what really makes it an expensive solution overall is that it requires a special starter at $550. Or not....
yes, you can clearance the block with a carbide burr bit and die grinder. Then you can use a regular sbf starter. That's what we did
Adam Lucas Pohl

Don't forget the huge chunk that you need to bolt on the back of the crank as well to space it out, Esslinger only...

[DPDISXR4Ti]

Don't forget the huge chunk that you need to bolt on the back of the crank as well to space it out, Esslinger only...

True, but at least that's somewhat reasonably priced at $140.

https://esslingeracing.com/ford-4-valve ... nk-adapter

I'd likely only go this route if I was replacing an engine with a SBF pattern with a Duratec 2.5. I'm actually considering such a thing - replacing the 4.9 in my F150 with a 2.5. Yes, it's a crazy idea, but I like crazy ideas.

[DPDISXR4Ti]

Currently there is no bell available to mount a T-5 directly to the MZR. One did exist in the past from Quad4rods, and one may exist in the future, but today, no.

As I knew at the time, a T-5 bell for the MZR was under development, and it is now available.
https://www.rwdmotorsport.com/product/d ... llhousing/

[DPDISXR4Ti]

I had been under the impression that all the 2005-2011 2.0 Focus engines were the same, but it appears that may not be the case, as outlined here:
https://sbdmotorsport.co.uk/product/dur ... formation/

In short, the 2008+ heads have a high-port design which should flow a little better. The high port castings numbers are RF3S4G (AKA 3S4G) and RF6S4E (AKA 6S4E). The earlier (low port) head is the 1S7G. Note, this info comes from the UK and may or may not be exactly the same in the US.

Upon further review it would appear that the information provided above ^^^ is incorrect for the North American Focus. Apparently the US Focus continued to get the low port 1S7G head all the way through 2011. To the best of my (now updated) knowledge, the only thing the 3S4G head came in on US soil was the 2003-2007 Focus ST with the 2.3.

[DPDISXR4Ti]

Upon further review it would appear that the information provided above ^^^ is incorrect for the North American Focus. Apparently the US Focus continued to get the low port 1S7G head all the way through 2011. To the best of my (now updated) knowledge, the only thing the 3S4G head came in on US soil was the 2003-2007 Focus ST with the 2.3.

When I was junkyarding recently I tried to see if you can see the head casting# with the engine installed in the car. Since the number is on the exhaust side of the casting, I'm thinking it's pretty near impossible with a FWD car like the Focus. Maybe with a mirror?

BUT, perhaps it can be determined by the sticker on the front of the engine. Below is a picture of said sticker, one from a 2.3 Focus and the other from the much more common 2.0. We know the 2.3 has the 3S4G head and the sticker has "3G" as the start of the litany of numbers. The sticker on the 2.0 does not have "1S" anywhere on it, but perhaps that doesn't matter; maybe the 2.0 engines with the 3S4G heads will have "3G" like the 2.3? Sadly, with the info I have now though, it would seem that engine may only have existed in Europe. You could "create" one easily enough in the States by plunking a 3S4G head onto a Focus 2.0 engine.

FocusDuratecHead-2.jpg

FocusDuratecHead-1.jpg

[DPDISXR4Ti]

Time for a little data-point.... If you're looking at doing a 2.5 swap, the most common donor is the 2010 - 2020 Ford Fusion. Millions of them were produced and the engines pretty much last forever, so there's not much demand. You may recall from economics class, large supply, little demand equals small price. Finding low-mileage engines for less than $300 is easy.

But is there any difference in the engines over the course of that ten year span? For the most part no, but one thing I recently learned is that the 2013+ engines use a dual 4-rib crank pulley vs single 6-rib on the earlier engines. What that tells me is that if you're at all considering a supercharged engine, you'll probably want to start with the 2010-12 engine. Actually, you might want the earlier pulley regardless, as it might take up a little less depth (not sure of that though - just speculating). The pulleys are swap-able however, so putting a 2010 pulley on a 2013+ engine is do-able. It seems however that the pulleys are not keyed, so you do need to time them correctly if swapping.

Lastly, there does seem to be some difference between a 2013 - 16 engine and the 2017-20. I'm not quite sure what that difference is, and if it matters at all for our purposes. Addendum... Apparently, there were some minor updates to the 2017+ engines. One of those updates is that they went away with the friction washers in these years since they're not needed with the newer design sprockets.

Duratec25-2017+.jpg

<edited 7/5/2025 with corrected info>

[my8950]

Time for a little data-point.... If you're looking at doing a 2.5 swap, the most common donor is the 2010 - 2020 Ford Fusion. Millions of them were produced and the engines pretty much last forever, so there's not much demand. You may recall from economics class, large supply, little demand equals small price. Finding low-mileage engines for less than $300 is easy.

But is there any difference in the engines over the course of that ten year span? For the most part no, but one thing I recently learned is that the 2013+ engines use a dual 4-rib crank pulley vs single 6-rib on the earlier engines. What that tells me is that if you're at all considering a supercharged engine, you'll probably want to start with the 2010-12 engine. Actually, you might want the earlier pulley regardless, as it might take up a little less depth (not sure of that though - just speculating). The pulleys are swap-able however, so putting a 2010 pulley on a 2013+ engine is do-able. It seems however that the pulleys are not keyed, so you do need to time them correctly if swapping.

For me, I still want it boosted. Trying to determine how to do all that as well without a complete rebuild. Wondering if a 2.0 ecoboost, or your method mentioned here and then boosting it is better approach.

[DPDISXR4Ti]

For me, I still want it boosted. Trying to determine how to do all that as well without a complete rebuild. Wondering if a 2.0 ecoboost, or your method mentioned here and then boosting it is better approach.

The nice thing about the factory N/A 2.5 long-block is that it can much more readily take boost pressure than the N/A 2.0 or 2.3 long blocks. It's just built with better parts. And the nice thing about super-charging is that you can leave all your N/A hot-side stuff alone. So indeed, you can go ahead with a 2.5 N/A install today and throw a super-charger at it somewhere down the road.

I've shifted my thinking on what would be my first Duratec swap. Initially it was the Scorpio, but now I've turned my attention to my '91 F-150. Sounds crazy, but my thinking is to swap out the old 4.9 with a 2.5 and utilize the existing drivetrain via the Esslinger bell adapter. Need more torque? Throw a Rotex Supercharger at it without unscrewing a single bolt on on the exhaust side.

[my8950]

Brad, are you aware of any exhaust manifold that would be turbocharger ready for the 2.5?

[DPDISXR4Ti]

Brad, are you aware of any exhaust manifold that would be turbocharger ready for the 2.5?

Probably best to look in the Mazda MZR world. I see them listed for the 2.0/2.3 - not sure if they could be used as-is on the 2.5; maybe if the ports were opened up? Chip would know.

[DPDISXR4Ti]

I didn't realize until recently that the MZR crank is not keyed, like what we're used to on "old fashioned engines". It would seem it was just a way to take some cost out of the engine, but it adds some time and effort for those of us who would work on them. Fortunately, the timing process is pretty well documented and not that difficult. Here's a couple quotes from the "RWD Duratec World" page on FB...

David Long
the order they do things is TDC is locked with the calibrated setting bolt against the crank, the pulley is mounted with the 6mm alignment bolt to hold it in alignment while the nut is tightened, the sensor is then set centred on the white tooth with the alignment tooth. The senior adjustment is to accommodate any minor slippage of the pulley while tightening. But the sensor should be set with engine TDC locking tool and not the pulley alignment bolt.



Paul Kemp
I’ll add some more detail, the crank is set at TDC by a bolt that goes in the side of the block. The cam timing is set by a bar that goes into offset slots in the back end of the cams. At this stage the cam drive chain pulley is loose as is the auxiliary drive pulley with the crank sensor toothed wheel.
A small bolt aligns the aux pulley and now the crank sensor can be set using the setting tool or by eye with a tooth in the middle of the sensor. The toothed wheel is a 36-1 wheel that is to say that one of the teeth has been removed and for a 4 cylinder engine the missing tooth will be 90 degrees away from the sensor. Most have the missing tooth 90 degrees before TDC i.e. it passes the sensor before TDC and the ECU counts the teeth after the gap.
The crank pulley and the aux pulley have diamond encrusted friction washers to proved added friction as do the cam pulleys.
This has all been done to make engine building at the factory quick and simple.
If your going big power then keying the crank cam drive and pegging the cam drive is recommended


Tim Spencer
Diamond dusted washer between the crank and timing sprocket, bolt hole in front cover for the trigger wheel location, slotted cams in the back to use a tool to hold them, and finally a crankshaft pin hole. Make sure you set TDC and from there, assemble.

Related, but different topic - cam timing...
Cam Timing Alignment Tools Compatible with Ford/Mazda Zetec 2.0L 2.3L 2.5L MZR/Duratec Engines
https://www.amazon.com/Mekar-Timing-Ali ... kaNM5Whkuw

[DPDISXR4Ti]

I know "everyone" wants to run an after-market PCM when swapping a newer engine into an older vehicle. Myself, I'd prefer to run a factory PCM, but one that's had things like EGR, catalyst monitors, and PATS/Immo disabled. There seems to be very little support for Ford PCM's in this regard, but a fair amount of support for Miata MX-5 (NC) 2.0 PCM's. I found this today, which might mean you could use a more common Mazda (3, 6, CX-7) PCM & harness while still using the knowledge available in the NC Mazda world...

The Denso ECU used in the NC Mazda Miata (2006–2015) was also used in several other Mazda and Ford vehicles with similar powertrains. Some known applications include:
1. Mazda 3 (BK/BL) 2.0L & 2.3L – The early Mazda 3 models used a similar Denso ECU, particularly for the 2.3L MZR engine.
2. Mazda 6 (GG/GH) 2.3L – Some versions of the first-gen Mazda 6 with the 2.3L naturally aspirated engine shared ECU architecture.
3. Mazda CX-7 2.3L Turbo – Though the turbocharged model had different maps and control strategies, it still used a variant of the Denso ECU.
These ECUs are closely related, but they have different calibrations, firmware, and sometimes different pinouts. However, the core architecture remains largely the same.

[my8950]

Brad, are you aware of any exhaust manifold that would be turbocharger ready for the 2.5?

Probably best to look in the Mazda MZR world. I see them listed for the 2.0/2.3 - not sure if they could be used as-is on the 2.5; maybe if the ports were opened up? Chip would know.

I am still researching this, rather following your work so I don't have to do as much. But, I did ask to Chip, and the other guy who is running the Fusion 2.5 in his Mustang, his response was, "some eBay manifold", or buy from Esslinger.
I don't really want to buy from eBay and just pick something and hope it works, I don't want to buy from Esslinger because I am not rich, and I don't really think their the best like some people do. Anyhow, just putting it out there, still trying to find an option to use the 2.5.

I did pick up a 2.0 EcoBoost from an 2020 Edge since it was local and quite cheap, also probably crazy tired, but it is complete, and I can use it to mock up mounts and whatever else, see how far it gets and put it in a car at some time maybe. So, I have this engine if any measurements or whatever are needed.

[DPDISXR4Ti]

Some tidbits from Brian Axup at Esslinger on building a junkyard 2.5...
...if you're going to go Junkyard Dog on it, start with a 2.5L and swap out the front timing cover for an early VCT 2.3L from a 2006-09 Fusion, then put the necessary crank and cam sensors on for your ECU as well as the right coolant necks, alternator, 2.3L Ecoboost waterpump, and the intake manifold & throttlebody that best suit your swap.

...the early 2.3L VCT timing cover has external access to the detent for the tensioner. If you ever need to remove the cam gears with the engine in the car, having this access point to the tensioner is priceless.

Regarding using a factory PCM of some sort...
The only swaps on an OEM computer are MX5 guys for the most part. You would need to swap the crank pulley and cam/ crank sensors over to use the MX5 ECU. You would also need to add a Mazda coolant temp sensor as well as the Tmap, throttlebody, and pedal
(Note my comment in the post just above where some of the other more common Mazdas <3 and 6> may use a Nippondenso PCM nearly identical to the one used in the MX5)

[DPDISXR4Ti]

Related to my recent posts above, if wanting to run a Mazda PCM, you need the Mazda Crank Pulley/harmonic balancer and matching sensor. Why? Mazda uses a 36-2-2-2 toothed wheel rather than the 36-1 that Ford uses. Unless you have welding skills that allow you to do this...

Justin Garvin
on the ford engines, I simply weld in the missing tooth on the harmonic balancer, then cut out the 3 to make it resemble a Mazda balancer
That way i don't have to remove the balancer. I've done this several times with 100 percent success