^^^^

Ah, yep remember that thread..

Silent, if possible, try and get it on another dyno and yes, post up plots..

 

Here are the three pages of results against rpm. First one is the standard plot of 2 consecutive runs.


Second one is flywheel power, wheels power, torque and air-fuel ratio on separate plots.


Last one is a plot of boost.

 

bttt

 

bttt again

 

That does seem low given the boost etc that is run, the loss debate is endless, on a Dyno Dynamics rolling road assuming the paw was the same you would have factored 22% loss into it.

Even a UK car with a TD04 is going to get 280 ish at the fly running at 1.1 bar redline, if you are holding 1.4 bar at the redline and the car's got some ignition in it then suggest a rollers setup problem given its a new cell and all that. Easy to happen and not a reflection on the operators.

edited to add if the afr is true reading that is way too lean as Paul says.

bob

 

I know some of the other guys that ran on this day and none of them reported AFR's that lean. Admittedly most of them wouldnt know what they were supposed to be but some would. I wouldnt have thought it was the dyno operator's wideband.

This has got to be the issue with robbing you some power - it must have been struggling to remain det free up there.

 

Very interesting so far. It doesn't have any noticeable det problems. The only time I've had any real knocks popping up was on track from constant thrashing and banging into the limiter, cured by keeping the revs below 6500; or long high speed drags. Normal fast driving doesn't seem to trouble it at all, ever since I did all the heat management work around the engine bay. Perhaps keeping all the temperatures well under control is masking another power-robbing problem which is underlying it. I wonder if it's my FMIC that's the cause? I know I've had audible surge problems in the midrange with both my VF35 and my TD05 since fitting the FMIC that were not there with a v4 STi TMIC. The pipework isn't the best, but could it be causing that much of a restriction up to the throttle body and still allow an easy 1.45 bar to show at the manifold? And surely if the actual compressor was working a lot harder and way out of efficiency, the FMIC would lose control of charge temps? Or maybe I've just got a **** engine?

 

bttt

 

As has been said.. if that AFR reading is correct it is way too lean.. suggest you get it checked with a wideband.. if the AFR is really that lean then you can go down the route of finding out why (fuel pump failing, fuel filter blocked, fpr faulty - vac pipe come off the fuel pressure reg).. if it however shows up that the fuelling is not that lean then the AFR recorded on the RR was wrong so how can you believe the rest of it..
I am assuming all the current mods are mapped for?

You will get high knock readings if you bounce off the limiter.. it is not advisable to do so, but better than over revving it.

Simon

 

This is when an electrical fuel pressure gauge would help. It's running fine at idle - 3.5 bar without the vacuum which is where I set it just before it's remap. I fitted a Walbro last year and it's not given any problems so far to my knowledge. Had a new fuel filter about a year/10k miles ago as well.

As far as the rev limiter bouncing goes, I tried it out on track running 1.0 bar boost at first because it was wet, but all I ended up doing was thrashing it to the redline to try and wring some power out of it. Switching back to the full house 1.45 bar setting was a million times better, a much lazier drive not needing to cane the motor, and easily eating up Caterhams around Cadwell Park.

 

regardless of what the problem is you need to get the fuelling checked to establish if there is a problem for sure.

Simon

 

OK I'll get the AFR checked over first. Where I live isn't exactly crawling with tuners with wideband AFR monitoring capability - in fact the only one within 100 miles is where I got the rolling road test done in the first place. So I'll see if I can view the lambda signal from the Power FC switch check screen and get a voltage off that at full load. If it's as steady as on the plot then it should be easy enough to read off. I don't have a lookup table of lambda sensor voltages against AFRs - anyone know of one? Or alternatively, target voltage to get me in the 11:1-12:1 zone.

 

where is home?

lambda around 0.89 ~ 0.95 volts.. depending on how accurate sensor is etc.. they tend to lean off as they get hotter through a blast but should help.

Simon

 

Well it's a genuine Subaru lambda sensor that I bought new and fitted only a year ago. Voltage is showing 0.78v on full load 4th gear, although that's only at 1 bar not the full 1.45. So from the sounds of it, it's definitely running too lean.

 

need it checking with a wideband..

 

Well I've got it under cover and I've just been taking the whole area apart and I've found the problem. The inlet pipe had come off slightly right underneath where it was impossible to see properly. Glad it was something easy to track down. Not sure how I can do a permanent fix though as the problem has occurred because I've got my manifold on 10mm spacers which misalign the inlet hose with the turbo and because it's a machined TD05, there's not a lot of flange to attach to, it's just about the width of a jubilee clip and no more. Any ideas?

PS anyone got the short, slightly zig zagged black rubber hose that joins the metal dump valve return pipe (v4) to the side of the inlet pipe just before the turbo? I need to replace this as well.

Cheers!