More tinkering this week... From playing around with / researching into various things, it looks like the car has under 80bhp at the rear hubs. This is ~20bhp down on the race car, and explains the poor top end speed. I was really hoping the lack of screen would free things up a bit here, so the quest for more power on the cheap and without touching the engine mechanically began...!
I plan to make an exhaust in the near future so I will tackle that later, but for now first up was to investigate the manifold and downpipe. These are welded internally so it does produce a bit of a restriction, and doing the maths showed up to a 15% reduction in cross sectional area due to the welds! Of course this wouldn't translate into a power hike of the same amount, but every little helps as they say... This modification is also permitted in the race series, so as a test run before doing it on the race car it would be a useful experience.
Here's the downpipe at the manifold end, approx cross sectional area of 1385mm^2:
And after attacking with the die grinder, approx 1520mm^2, an increase of ~10%:
The head end of the manifold was a similar story, but with around 730mm^2 per header to start with (therefore double the area in total of the downpipe) I didn't go as extreme on these:
After:
(They did get neatened up a bit after the pic!)
So, while that lot was off I took the chance to weld in an extra lambda sensor boss in the downpipe and fit an AFR gauge. There's been a bit of a theory for a while that it's hard to N/A tune the 1.6 engine, but BLiNK have shown that you can make some pretty impressive gains just by tinkering with the Air Flow Meter (AFM) and timing - about 17% hub power over standard!
http://www.mx5nutz.com/forum/index.php?showtopic=112653
With the MX-5 using a very simple Throttle Position Sensor (TPS) which is 'on / off' rather than variable, combined with the AFM setup, you can get a reasonable idea of what's going on with the Air / Fuel Ratio (AFR) without having to do Wide Open Throttle (WOT) pulls.
As standard the AFR of the MX-5 seems to take a dive towards rich (more fuel, less air) at around 4000rpm, and can get as rich as 10:1 at the red line - I guess this is to keep things safe as standard and allow for wear on fuel pump, injectors etc. However, it makes best power at around 13.2-13.5:1, so we can tweak a few things to try and sort that out...
The standard modification is to change the timing from 10 degrees BTDC (before top dead centre) to 14 degrees, which does help perk up the car everywhere by making use of the higher RON fuel we have here compared with the 90 RON fuel in Japan when the car was built. The optimal timing advance seems to vary from car to car, somewhere between 12 and 16 degrees, but it's impossible to tell without a dyno so I've had to stick mine at 14.
From there, you play with the AFM. Note you do NEED a way of metering the AFR to play with stuff like this, or you can cause serious engine damage!
This is the silver unit that sits on top of the air box, and has a door that flaps open against a light spring when air flows through it. More air flowing through = flap opens further = more fuel is put in. One of the standard modifications here is to tighten up the spring, so with the same amount of air going through as before the car puts less fuel in because the flap isn't open as far. This works to a point - BLiNK show gains of around 5bhp in the usable range for a track car - but obviously is then restricting the flow of air, not ideal.
Following some reading, I learned about a few other tweaks you could do to help fool the AFM into doing what you want it to:
http://forum.miata.net/vb/showthread.php?t=265452
The 3mm allen screw can be used to adjust the relative position of the wiper to the flapper and the whole board itself can be shifted around too via loosening the three phillips headed screws (the holes in the board aren't a perfect fit to the screw thread). A fourth adjustment is to rotate the whole AFM, so gravity takes over and either helps or prevents the flapper from opening up.
So, a bit of testing... First up was to get the car warm and run it through the revs with the standard setup. This gave an AFR of around 11.9-12:1 at 5-5.5k rpm. This was slightly different to the numbers on the graphs produced by BLiNK, but their AFRs are read from the tailpipe and that tends to give a reading of approx 0.5 points higher due to ambient air surrounding the gauge. The values I found weren't horrendous, but definitely some gains to be made.
Next up I swapped the AFM for one from another car I had, and rigged it up with a 'Frankintake' made from another MX-5 intake pipe and the airbox from a 1.8 16v K-Series.
This yielded a better AFR of around 12.6-12.7:1 under the same conditions - I'm not sure if this was down to the induction setup or the AFM, but the main thing I was going to use this contraption for was to test how rotating the whole lot affected AFR.
With the unit rotated so the door would fall open under gravity, AFR richened up by a whole point to 11.6:1, and going the other way it leaned out to 13.5:1 - interesting! Having the AFM in this orientation would also help to remove any effect caused by lateral g force - but it's not legal in our race series (we have to use the standard airbox in the standard location), so I didn't investigate this any further.
So, back to the standard airbox and the original AFM, I decided to try and get the flap opening further which would require a softer spring, but still lean out the top end - which would require tweaks to the wiper arm and shifting the board.
A bit of iteration ended up with the board shifted to the left as shown on the photo above, and the arm moved around clockwise - meaning with the flapper open further than before the wiper would still be less far around the track, and putting less fuel in = leaner running.
After some tweaking, I ended up with an AFR of 13.2:1 at 5k rpm, dropping to around 11.6:1 at the limiter. I decided this was safe enough not to go any further, plus at this point I had forgotten the car had no fan and the coolant started to boil, haha. I'll need to rig up the AFR gauge to my datalogger and get some proper logs, but for now it seems promising, should give at least 10% more bhp and I've learned a few bits in the process :smile: Just goes to show that there's no 'black art' to stuff like this, just a willingness to apply a little bit of thought and realise there is the potential to break stuff if you screw it up!
