Mazda MX-5 Miata banner
661 - 667 of 667 Posts

·
Registered
Joined
·
32 Posts
Just read through this, what a mess! It's one of the things that has put me off getting one of my cars turbo charged by a company to be honest! I have owned a couple of turbo mx5's but these were already proven working etc.
 

·
Banish a little rust a day people
Joined
·
5,237 Posts
Discussion Starter · #663 ·
Looking over some settings on the tune and I'm not sure some of them are right, opinions please just to check im not going mad -
In the ECU I'm sure I set the trigger angle to 282, measured with a light, before it went for tuning but they've set it to 280. It has a 36-1 protege wheel, the miataturbo below thread states the missing tooth is 80 degreed btdc, which would be 280 atdc, but the timing points are 3 degrees wide so as the ECU is set to trigger on the falling edge you would add 2 degrees making it 282 wouldnt you?

I dont have a timing light, and the person who helped me last time has covid atm.
 

·
Banish a little rust a day people
Joined
·
5,237 Posts
Discussion Starter · #664 · (Edited)
I'm back.

I wanted to focus on getting to the bottom of all the issues on the car and getting it actually working and MOT’d before posting again, and I’m now at that point (sort of).

To keep it relatively brief I found someone knowledgeable on the Mx5 Supercharged Facebook group who offered to help look over the map and correct a lot of the issues I was left with. It was a lot easier said than done, there were a lot of very obvious initial problems and oddities in the map which we corrected without much hassle but the main problems were that we couldn’t get the car idling right. AFRs were way to rich and fluctuating widely, and it had a bad misfire but no matter what we tried it wouldn’t improve. We spent probably 2 months of weekends trying to work this out assuming it was a physical issue with the car, and I ran all the checks I could think of:
  • I checked and corrected the incorrectly set base timing (as I referenced in my last post it was out, and why it was changed from what I set it as with a light I have no idea).
  • I checked the injector seals a number of times
  • tried new injectors
  • tested for vac leaks using a smoke machine I built
  • checked compression (all good, exactly the same across all cylinders and peak values the same as my mk2.5)
  • swapped the cam sensor
  • tried different spark plugs
  • tested HT leads
  • new coil pack
  • etc

None of these revealed the issue and we were at a loss, suspecting the ECU was faulty. Eventually we decided to look over some of the most basic settings on the tune and found 2 issues which were the cause of all the headache. The first - the wideband had been setup wrong in the ECU, the configuration for a different model had been used….. I loaded the right configuration, tweaked the fuel map and the AFRs immediately responded and could be bought inline to something sensible rather than throwing raw fuel out the exhaust. The random misfires were the result of an equally ridiculous error in the setup, the injector angles had been set to 65000 degrees…. no that wasn’t a typo, they had been set to sixty five thousand degrees…. correcting these to a much more sane 355 degrees as per the default recommended Speedunio values immediately solved the misfire. With these issues sorted and the car actually mappable it took very little time to generate some logs, tweak the map (the previous 18:1 AFRs in some places and off the scale rich in others not being partiality desirable), and repeat until we got it to where it needs to be.

After that I threw an MOT on it (no advisories) and made a few small setup tweaks such as adding a Dayco 89269 autotensioner, a heat shield for the underside of the charger just to be safe, and a slight revision to the intake setup as it was possible there was a slight restriction (which turned out not to be the case but it also improved the mounting so was worth it regardless for the £10 cost).

Motor vehicle Hood Automotive tire Automotive design Automotive fuel system


Motor vehicle Brickwork Brick Wood Automotive exterior


Vehicle Motor vehicle Automotive design Automotive air manifold Automotive exterior




Now the map wasnt so horrendous it was throwing raw fuel out the exhaust I wanted to refit the cat, but in the on going tale of discovering fresh incompetence related issues after struggling for 2 hours with the v bands I couldn’t manage it, so I decided to put it and the decat side by side and took some measurements. Not only has cat been made up to the wrong length, but it’s not even at the right angle.

Tableware Drinkware Brickwork Wood Brick


Parking meter Communication Device Audio equipment Gadget Bicycle accessory


Gadget Audio equipment Gas Sports equipment Camera accessory


Bumper Automotive tire Automotive exterior Gas Automotive wheel system






This is as close as it gets when things are rotated to be as close as possible (bolts wont go any tighter either, and the header is under significant strain to pull in to even that). So yeh, also need to find someone to fix that.

After a fairly long wait due to availability it then went for mapping somewhere competent. The first try looked pretty good, the fuel was dialed in slowly and we started really gentle and safe and it made 185hp/150whp but promptly blew the supercharger gasket out cutting the day short. I’ve wondered about this before as the plate for the outlet doesnt match the plate on the charger very well at all, it’s a design that could use significant improvement IMO. To correct this I added some material to the corners to give the gasket more surface to seal against, had the flange milled flat and returned two weeks later. After some tweaking it made 185bhp/150whp with 10psi peak boost… again. This obviously wasn’t right, with 260bhp/220whp the ball park figure that was expected. The tuner pulled up a dyno graph of a 1.8 mk1 with an M45 kit on it and things started looking even stranger. Compared to the M45 mk1 it was making about 30hp less in the low rpm range at 3 psi more pressure. Something wasn’t right, we checked for leaks again and found none, the belt wasn’t slipping, and given it will make 10psi we are confident the charger isnt to blame (confirmed with another visual check of the blades and housing).

I again had to wait weeks for availability (a good sign the tuner is reputable I guess) and had a diagnostic run down with another compression test and a leak down test. Results were <4%, 2%, 0%, 0% leak down and 175psi hot compression identically across all cylinders which are all well within a healthy range.

I am now at a total loss. The car is making boost, there are no leaks, the engine is healthy, the timing is fine, the crank key hasn’t slipped, adding timing on the dyno makes no difference, we took some logs with the charger disconnected (2” IC pipe and IC still connected so a little restrictive for NA) and virtual dyno estimates it’s making about 90-105whp which seems about right, it idles and drives around sweet, it’s just about 50-70whp away from where it’s meant to be.

This is a bummer, I have no idea why it’s so down on power. I’ve spoken to a tonne of people both online and at a few shows (Skuzzle, Aran Large, Chaps, etc) and everyone seems to be in agreement that everything should be fine and we can think of no reason why it’s not making the power. Someone at Japfest suggested it might be spark blowout and that they had similar with no sign of breakup, that it just also didn’t make the power and a set of COPs solved it. It seems unlikely to me this would be the case with no symptoms of hesitation, misfire, or reflected in the AFRs and I don’t want to spend the £200-300 just to test.

Any or all suggestions welcome.

In the meantime I’m shifting focus onto the track car for a short time as it’s now on the road, needs the odd touch up maintenance item, and that damn 2860r installing that’s been decorating my shelf for the last year. It looks like the blue NB will probably now sit in the garage for a bit as something mustn’t be right with it, although the engine checking out healthy does make me wonder whether I could use it safely

Just for my records the Celica had an oil and filter change at 130467 miles.
 
  • Like
Reactions: Zed.

·
layabout
Joined
·
2,640 Posts
the injector angles had been set to 65000 degrees
o_O

W⚓'s

:rolleyes:


After some tweaking it made 185bhp/150whp with 10psi peak boost… again. This obviously wasn’t right, with 260bhp/220whp the ball park figure that was expected. The tuner pulled up a dyno graph of a 1.8 mk1 with an M45 kit on it and things started looking even stranger. Compared to the M45 mk1 it was making about 30hp less in the low rpm range at 3 psi more pressure. Something wasn’t right, we checked for leaks again and found none, the belt wasn’t slipping, and given it will make 10psi we are confident the charger isnt to blame
cam / cam / crank timing?

Rich.
 

·
Banish a little rust a day people
Joined
·
5,237 Posts
Discussion Starter · #666 ·
o_O

W⚓'s

:rolleyes:




cam / cam / crank timing?

Rich.
I dont think so. It starts, idles and drives too smoothly. Disconnecting the charger so it's running without FI also nets virtual dyno figures of about 100whp, and previous logs of a virtual dyno plot with the supercharger are within 2hp of the actual dyno plot so it seems pretty accurate and im inclined to believe it.

I did pull things apart to check the timig and it all looked totally fine. The only thing I am going to double check is whether the exhaust cam gear has been rotated 120 degrees and is locating the pin on the wrong groove as I did take it out to do the oil seal. But I'd of thought I'd of noticed if it'd jumped by that much and it being a full 3rd of a rotation off wouldnt allow the car to start or run at all.
 

·
Banish a little rust a day people
Joined
·
5,237 Posts
Discussion Starter · #667 ·
Done the final checks on the supercharged car:
  • A maximum of 1 degree ECU timing latency between idle and 6k rpm was found, no where near enough to explain the lack of power.
  • Pre-charger intake restriction was measured at under 1psi at 6k rpm.

At a total loss as to what's going on.
 
661 - 667 of 667 Posts
Top