Mazda Mx5 M52TU Turbo swap

[_nikki]

So I've been planning this for well over a year now and have been working on it for well over 6 months and I have this project documented elsewhere but thought I'd throw up a project summary here.

Back in February 2020 I bought this stbox golden nugget mx5 as I needed a new daily after selling my WRX. Immediately after getting it I threw a turbo setup together to get by whilst I planned and got prepared for the main project.



The plan was to drop in a BMW straight 6 and gearbox, originally I wanted use an N54 engine but after doing a good bit of research I decided that I couldn't be bothered to mess around with direct injection or can bus when the rest of the car makes a fax machine look up to date! So after some more research I settled on its older cousin the M52TU/M54 engine as they can be had relatively cheap and are far simpler than the N54, although they lack the turbo. I started getting a plan together and originally I intended on getting hold of a complete stbox to use as a mock up vehicle so that I could still keep my turbo setup whilst I was working on this however for some reason when I was looking the prices had gone through the damn roof. Instead I opted to work directly on the golden nugget and daily my other stbox (a 2002 ford ranger, its a reliable old goat but sounds like its about to explode all the time).

Anyway I started to get the wheels in motion, secured a 2.5L version of the M52TU with about 100k miles and matching 5 speed for £270 shipped. After some more research I also opted to get a matching BMW diff as the gearbox ratio's are tad on the short side and there is no overdrive, resulting in 70MPH with the mx5 4.3:1 would be about 4200RPM.

First things first, the M52TU/M54 are high compression engines and don't like being boosted past about 8PSI, but as far as I'm concerned if you're only going to 8PSI you may as well not bother at all. So I did as the Europeans do I installed a stainless decompression plate in the head gasket lowering the compression down to about 8.5ish:1, I also time-certed the head bolt threads in the block (as 2 of them stripped out whilst torqueing down the head) and installed ARP head studs.

I pulled the Mx5 engine, gearbox and diff and sold them as pre turbo'd package to a chap whose 1.8 rods decided they wanted to join NASA and get the astronaut wings. With the engine out the first order of business was seeing to the "minor" rust that had been pointed out in the previous MOT about 3 months prior and all I can say is it was anything but minor.



I wasn't entirely surprised as I had noticed the piss poor attempt at patching the rust when I looked over the car when buying it and it's not the first time I've had to fix rust that's just had steel welded over the top. Well I cut out all the rust and cut some new steel patches on the plasma table and welded it all back together along with some additional internal reinforcements for where the new ARB mounts would go.



well with some persuasion involving some hammers, angle grinders and welders I was able to get the engine in position without having to dissect to much of the car whilst retaining the correct lateral and longitudinal angles and as you can see I managed to sneak in a tidy low mount turbo in the form of a GT3071R. The manifold is a custom stainless log manifold as there is no room for anything else and I paired it with a GFB EX44 wastegate. I threw in some 875cc injectors and a AEM 400LPH fuel pump along with full custom Teflon stainless braided lines and AN fittings. Next up was the rad and intercooler mounting.



I got the exhaust all sorted which was 83 welds of Pie cut nonsense/awesomeness on 3 inch stainless resulting in about 20M of back purged welding which if I remember rightly took an entire day! The result was well worth the work. You can also see the BMW diff mounted into the Mazda subframe and how tight some of the clearances for the exhaust where. Now you might be thinking the 83 pie cut welds would be more than enough for one project but i decided that as my time is free it would be far cheaper to buy straight aluminium pipe and pie cut the intercooler piping than to buy mandrel bends and weld them all together. so another 48 pie cuts later only this time on aluminium and I had the intercooler piping sorted out.




I have indeed skipped over quite a lot of the nitty gritty and crap that went on to this engine in the car but as I said It's already on the various Mx5/Miata forums. so by this point I was heading into the final assembly stage so I pulled the engine out, welded the oil return fitting into the oil pan, cleaned the engine a bit and wrapped and reinstalled the manifold. I reinstalled the engine, wrapped and reinstalled the downpipe and got everything plumbed up ready to for wiring.




I got the engine wired up pretty darn quick so much so that the custom axles I had ordered still hadn't come in from the manufacturer. So I had to get the propshaft shortened slightly in order for it fit in the MX5, I think I started with a z3 propshaft but even that was about 100mm to long. This is nice and straight forward to resolve so I got this sent of to a specialist and it was back within a week. The axles were less straight forward, a lot less! The problem lies in the different ways in which BMW and Mazda allowed axial movement in the respective driveshafts. Mazda use a sliding inner CV joint whilst BMW use a sliding outer joint. This probably has something to with the fact the vast majority of Mazda's cars are front wheel drive and therefor they re-used parts from there existing parts bin resulting in a FWD style axle in a RWD car (I could be wrong here its essentially just my take on the matter). What this did mean however is that I could not just use the inner BMW joint and the outer Mazda joint with a custom axle shaft as that would have resulted in a fixed length axle and it would have exploded one of the joints as soon as the suspension actuated for the first time.

Instead after some head scratching and internet searching I decided to use MX5 outer joints with Audi S4 inner joint along with an adapter plate setup to connect them to the diff. When everything finally arrived I got the axles painted and assembled and they fit like a god damn glove and just because I colour matched the axles to the body.



I threw in a core4 Speeduino ecu for an e46 and after some messing around I got the car running and here is a little clip of it



I was trying to do this on a tight(ish) budget and in the end to get the car on the road and driving it cost a grand total of £5577.90 without taking into account any labour costs and there would have been a lot of labour costs. So the car was/is currently running on wastegate pressure at 10psi and I don't plan on changing that until I've managed to get the last bits dialed in and things like the brakes and clutch upgraded. Once that's done I'll get it on adyno and start cranking up the boost until I get to something I'm happy with (or something breaks)

unfortunately I haven't got any videos of it driving at the moment and unfortunately it's back in the workshop after I had a run in with a speed bump resulting in a my rad getting destroyed. Hopefully next weekend I can get a new rad and get it all sorted again

If people are interested in the full right up, let me know and I can either write it up here or point you in the right direction.

Cheers,
Nikki

Edited by _nikki on Monday 27th September 16:12

[clarkson22]

Yes mate full write up please and lots of pics!!! Good work mate looks fun!

[trevalvole]

Blimey, that's impressive! What you've done is far beyond my capabilities.

What led you to put a BMW engine in an MX-5, rather than, say, starting with a BMW roadster?

[_nikki]

trevalvole said:

Blimey, that's impressive! What you've done is far beyond my capabilities.

What led you to put a BMW engine in an MX-5, rather than, say, starting with a BMW roadster?

I originally considered doing this years ago when I had my first MX5 but decided against it at the time for a number of reasons, so when I got rid my WRX and wanted a new daily I decided that I was gonna try it this time as I had time, spare cash and an essentially empty workshop (because of Covid lockdown).

I did briefly consider the Z3 and Z4 but in all honesty I don't really like the Z3 or Z4 convertible and the Z4 coupe at the time was still pretty expensive not to mention that they didn't come with the N54 engine that I originally planned to use so it made no real sense. Also I'm kind of a JDM fangirl and I've had several MK1 and MK2 MX5's before so I know the platform pretty much inside out.

EDIT: to clarify I only looked at the e85/e86 chassis Z4's, the e89 chassis z4 which did come with the N54 were still like £10k-£20k at the time.

Edited by _nikki on Tuesday 28th September 10:26

[geeks]

What a fantastic thing, loving your work!

[trevalvole]

A friend of mine has an E36 with that BMW engine, and always says that BMW deliberately restricted it, iirc on the inlet side, presumably to increase the gap between it and the 2.8. This makes sense to me as I recall that the two valves per cylinder, non-VANOS engine in the E30 325i produced the same (170ish) power.

Assuming this to be the case, has your work plumbing in the turbo etc. removed this restriction?

[rampageturke]

Looking good, will have another read through later for some inspiration to get something swapped into my mk1.

Is the other documented thread the same? Or does it have more in it?

[_nikki]

trevalvole said:

A friend of mine has an E36 with that BMW engine, and always says that BMW deliberately restricted it, iirc on the inlet side, presumably to increase the gap between it and the 2.8. This makes sense to me as I recall that the two valves per cylinder, non-VANOS engine in the E30 325i produced the same (170ish) power.

Assuming this to be the case, has your work plumbing in the turbo etc. removed this restriction?

I have heard about this before and I'm certainly no expert on the matter but it is generally accepted that the M50 intake flows slightly better than the m52, M52tu and M54 manifold however the effort the install the m50 manifold on the newer engine is just not worth it. You are indeed correct that the power barely changes between the e30 and the early e46 but the torque delivery is the key difference. on the M50 peak torque is at 4700rpm, the M50tu (single vanos) peak torque is 4200rpm, the m52 (single vanos) peak torque is 3950rpm and the m52tu (dual vanos) peak torque is at 3500rpm. (Also the peak power output drops further down the rev range with the newer engines, m50 - 6000rpm, m50tu - 5900rpm, m52 - 5500rpm, m52tu - 5500rpm)

some of the change in torque output came about from adding Vanos and some of it came about from changing the design of the intake manifold (I believe the primary change is to the length of the intake runners). So there is no restriction as such its just that the newer manifolds are setup for better torque down low at the expense of peak power. Since this should still apply even with the addition of turbo, I'd rather have a few a more NM of torque than an additional few BHP.

EDIT: I forgot to say that the 2.5 and 2.8 both use the same manifolds so if there was a restriction it would inherently affect the 2.8 and hinder its performance not increase the gap

Edited by _nikki on Tuesday 28th September 15:07

[_nikki]

rampageturke said:

Looking good, will have another read through later for some inspiration to get something swapped into my mk1.

Is the other documented thread the same? Or does it have more in it?

Thanks! Have you got any ideas so far? Turbo K-Swap is a good option as well

The are a couple of other threads on other forums that I wrote as I was building the car so they have the full details of everything but I will probably do a complete write up here as well when I have a free evening.

[rampageturke]

_nikki said:

Thanks! Have you got any ideas so far? Turbo K-Swap is a good option as well

The are a couple of other threads on other forums that I wrote as I was building the car so they have the full details of everything but I will probably do a complete write up here as well when I have a free evening.

I'd love to go rotary, love them too much and seeing as I've been eternally priced out of RX7s, feels like a good compromise. But the RX8 engine sucks and if I swap a 13B in there I might as just get an RX7. So don't know really. I have a '95 1.6 so might in the end just get a cheap 1.8 engine and diff and turbo it and call it a day, but this BMW lump is interesting.

[_nikki]

rampageturke said:

_nikki said:

Thanks! Have you got any ideas so far? Turbo K-Swap is a good option as well

The are a couple of other threads on other forums that I wrote as I was building the car so they have the full details of everything but I will probably do a complete write up here as well when I have a free evening.

I'd love to go rotary, love them too much and seeing as I've been eternally priced out of RX7s, feels like a good compromise. But the RX8 engine sucks and if I swap a 13B in there I might as just get an RX7. So don't know really. I have a '95 1.6 so might in the end just get a cheap 1.8 engine and diff and turbo it and call it a day, but this BMW lump is interesting.

Yeah rotaries are awesome, I used to have an RX8 before I got my first MX5 and yeah they are wk and unreliable as hell. I can see the benefits of a 13b MX5 the main benefit I can see is that it would likely cost less than half of what people are selling decent RX7's for at present.

[trevalvole]

_nikki said:

trevalvole said:

A friend of mine has an E36 with that BMW engine, and always says that BMW deliberately restricted it, iirc on the inlet side, presumably to increase the gap between it and the 2.8. This makes sense to me as I recall that the two valves per cylinder, non-VANOS engine in the E30 325i produced the same (170ish) power.

Assuming this to be the case, has your work plumbing in the turbo etc. removed this restriction?

I have heard about this before and I'm certainly no expert on the matter but it is generally accepted that the M50 intake flows slightly better than the m52, M52tu and M54 manifold however the effort the install the m50 manifold on the newer engine is just not worth it. You are indeed correct that the power barely changes between the e30 and the early e46 but the torque delivery is the key difference. on the M50 peak torque is at 4700rpm, the M50tu (single vanos) peak torque is 4200rpm, the m52 (single vanos) peak torque is 3950rpm and the m52tu (dual vanos) peak torque is at 3500rpm. (Also the peak power output drops further down the rev range with the newer engines, m50 - 6000rpm, m50tu - 5900rpm, m52 - 5500rpm, m52tu - 5500rpm)

some of the change in torque output came about from adding Vanos and some of it came about from changing the design of the intake manifold (I believe the primary change is to the length of the intake runners). So there is no restriction as such its just that the newer manifolds are setup for better torque down low at the expense of peak power. Since this should still apply even with the addition of turbo, I'd rather have a few a more NM of torque than an additional few BHP.

EDIT: I forgot to say that the 2.5 and 2.8 both use the same manifolds so if there was a restriction it would inherently affect the 2.8 and hinder its performance not increase the gap

Edited by _nikki on Tuesday 28th September 15:07

Thanks for the comprehensive response, which I accept. It does still seem odd to me that they would go to the effort of adding 4 valves per cylinder to an engine, presumably for better high-revs gas flow and then have it make no more power.

It is interesting that an MX-5 owner prefers more torque (and that's before turbocharging) than buzzy high-end power.

[_nikki]

trevalvole said:

Thanks for the comprehensive response, which I accept. It does still seem odd to me that they would go to the effort of adding 4 valves per cylinder to an engine, presumably for better high-revs gas flow and then have it make no more power.

It is interesting that an MX-5 owner prefers more torque (and that's before turbocharging) than buzzy high-end power.

I'd completely missed that you said it was the 2 valve per cylinder e30 engine (I was thinking about e36's), so the M20 engine made 169HP and 226NM of torque. When BMW introduced the M50 4 valve per cylinder engine which replaced the M20, it made 189HP and 245NM of torque. That's a pretty good jump in both peak power and torque so the added valves did make a big difference but after that the changes to M50/M52 engines just moved the peak power and torque around slightly.

Well high end peak power is great if your car is a dyno queen and you're just trying get the highest numbers possible on your dyno sheet, but in reality I'd rather have more torque and more area under the curve where it is actually usable without upsetting the old bill. That's just my 2 pence anyway

Edited by _nikki on Tuesday 28th September 18:40

Edited by _nikki on Tuesday 28th September 18:41

[samoht]

Awesome project, bet it'll be fun, will be interesting to hear driving impressions once you've got radiator #2 installed. Exhaust work looks great, incredibly tight clearance.

Did you ever consider the Jag V6 like Rocketeer are doing? I guess inline six is easier to turbo though.

Is that a new front diff mount in there, given you're not using the PPF ?

[_nikki]

samoht said:

Awesome project, bet it'll be fun, will be interesting to hear driving impressions once you've got radiator #2 installed. Exhaust work looks great, incredibly tight clearance.

Did you ever consider the Jag V6 like Rocketeer are doing? I guess inline six is easier to turbo though.

Is that a new front diff mount in there, given you're not using the PPF ?

I did about 400miles in it before I destroyed the rad and so far its been absolutely awesome.

When I first heard/saw the rocketeer kit I thought it was a pretty cool idea but It's not my cup of tea, I've never really been a fan of v6 engines, I don't like how unbalanced they are and in general most of them sound crap. I also kinda feel that if it comes with instructions and you can just bolt it all together it's not really a project, its no different to ikea furniture. All of that aside the price is just insane.

Yeah I had to make a custom gearbox mount and completely modify the subframe so that it had all 3 mounting points for the BMW diff.

[trails]

Loads of work, looks very cool...will be interesting to hear what it's like to drive with quite a chunky lump in the front. GT30 and 2.5 gotta be good for 450\500bhp, do you have a target?

Totally stealing your tool charger mount idea for my workshop, I guess the Dewalts are newish as the leads aren't hidden

[_nikki]

trails said:

Loads of work, looks very cool...will be interesting to hear what it's like to drive with quite a chunky lump in the front. GT30 and 2.5 gotta be good for 450\500bhp, do you have a target?

Totally stealing your tool charger mount idea for my workshop, I guess the Dewalts are newish as the leads aren't hidden

Yeah I'm shooting for 400 WHP but if I can get more I'll take it as long its not detrimental to anything else. I've got a lot to do before I can actually start cranking the power up though, still need to do brakes, suspension and some more rust issues to resolve.

Haha I think at the time the pictures where taken I'd just moved them from the other part of workshop so it was still a bit of a mess

[trails]

_nikki said:

Yeah I'm shooting for 400 WHP but if I can get more I'll take it as long its not detrimental to anything else. I've got a lot to do before I can actually start cranking the power up though, still need to do brakes, suspension and some more rust issues to resolve.

Haha I think at the time the pictures where taken I'd just moved them from the other part of workshop so it was still a bit of a mess

Will be quite a thing once it's done, looking forward to the next update

[trevalvole]

Is it running again?

[_nikki]

trevalvole said:

Is it running again?

not just yet unfortunately but its getting there. I ended up using a different radiator this time so it required some modification to get it to fit and work correctly. Along with that I ended up needing to redesign the lower rad mounts and move the intercooler forward slightly as well as angling it forward at the bottom (same as the rad). I finished welding the rad and intercooler piping after modifying them so hopefully tomorrow after work I can finish up the last few bits and get it back on the road.