Snob !!!!!

Agree it would be better to keep sensor away from heat. Is it possible to put it where it starts curving backwards and further round so it points slightly up ? You can also put the boss at a slight angle to the pipe and fill the gap if it helps.

Many aftermarket downpipes do have bungs which end up with the sensor pointing slightly downwards from the pipe so I guess it's not a huge problem. My sensor did fail after a couple of years like that though.

_________________
If at first you don't suck seed, try drier grain.

Bringing this more up to date in case anyone else has any bright ideas

Last week Chris, Phil and I put our collective wit together to try and resolve the issues. Between us we were able to successfully make coffee but sadly the root problem has eluded us

The Problem - Symptoms
When lifting off the loud pedal slowly from WOT to cruising level the car kangaroos violently

The Problem - Cause
From datalogging it can be seen that the injectors are being shut off by the ECU.
You can see in the plots (maybe Phil can post one up) that while TPS input tails off over a second or so boost remains reasonably constant (remember this is from WOT to partially open throttle, not WOT to throttle closed) and decays over a second or two. At the same time the injector duty cycle appears to cut off in line with TPS NOT boost. Ignition timing is also doing some odd things at this point
So you end up with a period of very lean running which causes a major hiccup in mixture and running, not to mention a Nibbles' forhead shaped bulge in the windscreen

Other Observations
The ECU in question is a genuine Apexi PFC and a very early one at that - it is serial number 003 running an early revision 12 firmware
There appears to be issues writing complete maps/setting files to the unit - it throws a warning about compatability issues and refuses. Individual maps (base, IGN and fuel) can be written one at a time and all the settings can be changed
I tried loading a very old map (From Dave H's Greddy equiped car for them folks as remember the past) as that was the oldest map I could think of. Using this it's possible to do a "write all" without issue.
Prior to being fitted and mapped in Phils car it was in another car that apparently ran without issue

What We Tried
Numerous combinations but essentially
Phil's ECU with Phil's map - Visit from Skippy and bulging windscreen
Phil's ECU with my map from my car - As above
My ECU with my map - Issue is resolved, runs smooth as the proverbial baby's upper leg area
My ECU with Phil's map - as above

So Phil's maps, while not perfect, do not exhibit the issue on my ECU and my maps, also not perfect, do exhibit the issue on Phils ECU

We came to the conclusion that the issue has to be somewhere in the ECU
Since everything else seems to be in order it seems unlikely that there is an actual hardware issue with Phil's ECU. So we're left with a software issue of some description :-
Either a FC-Edit -> early genuine Apexi software issue
or
A genuine firmware fault in the (early) 12 build firmware loaded into that ECU
A setting in the ECU that we cannot get at that has somehow been changed

One thing that Diceman pointed out is that we were using the "new" universal Datalogit interface module - the black one. Stupidly while I have the 3S-GTE specific version I never thought to try it out

Open questions
Does anyone know what software Ryan (2Bar Tuning) uses to map the PFC?
- It's possible that he uses software that allows access to hidden settings

Does anyone know for sure how to identify a PFC-Pro version?
- The symptoms present a little bit like a partially activated anti-lag kicking in but I think the unit might be a bit too early

Does anyone have a phone number for AP Engineering?
- Support from Apexi seems unlikely as they've joined the great Tokyo drift in the sky

So does anyone else have any bright ideas as to what the cause of the issue may be. There's a few possible workarounds but it would be nice to hear any other ideas as to the root cause of the issue

Ryan uses Dataloggit.

Have you had your car mapped?

Stephen

_________________
1993 Carlos Sainz 2939, Forged, Apexi RX6B turbo with external wastegate. MOTEC M4, AFPR and 800cc injects. 3" exhaust, +1mm valves.ST205 cams. Rays 8.5" x 17". 205 rear subframe. Bilstein coilovers. 205 brakes.TRD diff

I mapped my own car

Phil's car was mapped by Ryan

Maybe you should do the same as Phil?

_________________
1993 Carlos Sainz 2939, Forged, Apexi RX6B turbo with external wastegate. MOTEC M4, AFPR and 800cc injects. 3" exhaust, +1mm valves.ST205 cams. Rays 8.5" x 17". 205 rear subframe. Bilstein coilovers. 205 brakes.TRD diff

The other issue on Phils car is turbo surge, especially noticable at 3000 RPM in 5th. It was also bringing up the knock warning at this point so I retarded the timing a couple of degrees at this point. The surge still happens, but the knock is gone.

A couple of workarounds I have thought of:

1: restrict the voltage of TPS with a resistor and zener. The problem is worse when over 1/2 throttle so maybe restrict to about 2.5V (may be why Ryan tweaked the TPS adjustment)

2: Ditch the TPS signal and feed the MAp sensor signal into TPS input as well. This will give accelerator enrichment from rising pressure rather than throttle - not a biggie I suspect.

_________________
If at first you don't suck seed, try drier grain.

Interesting post Steve.

My Apexi PFC is: 'Model Serial Number PFC ST205 9806-0011 Produced by Apexi'. Going by the serial number I believe this to be a very early ST205 ECU. The AP Engineering versions have a sticker describing them as such.

When you say Phil's is numbered 003 are the rest of the numbers the same as mine?

I don't know what version of firmware is installed on mine as I've never plugged it in.

It would be interesting to see if mine displayed the same symptoms as Phil's. More than happy to lend it out if you want to try it.

_________________
Don
GT4DC Chairman
1994 Toyota Celica GT-Four ST205WRC JDM 269bhp @ 0.9bar
1994 Toyota Celica GT-Four Special GT 590bhp @ 1.8bar
1989 Van Diemen RF88/89 Formula Ford 1600
2008 Nissan Patrol GU 3.0L ZD30DDTi 154bhp

The day I bin my map with hundreds of hours (poetic license in the post) invested in it in favour of a one hour wonder map will be the same day Satan is seen driving a snowplough to work


Just to be clear, my map on my ecu in my car drove without issue and made (according to datalogging) 267hp @ the wheels @1.2 bar with very conservative fuel and timing calibration

Chris,

That top plot is a new one. The first hump is very odd with the fueling wanging about like that on a very slow liftoff. It appears that any change in dTPS/DT causes a major fuel change

The second hump in the second pic shows exactly what I was trying to say in words. TPS goes off @52.2 taking the fuel with it while boost stays constant giving a very weak mixture. Looks like a proper head but @52.7

Still don't have any clue what all the subsequent ignition timing is about

BTW folks,
Analysis of overall fuel (basexINJ) shows that the expected injector pulse is nowhere near the logged value

Apologies if this has already been covered: low voltage can really screw with the ECU inputs and outputs. Has this been checked on Phil's car? As the PFC is working OK on Steve's that would be a good baseline for comparison. Just a thought.....

_________________
Don
GT4DC Chairman
1994 Toyota Celica GT-Four ST205WRC JDM 269bhp @ 0.9bar
1994 Toyota Celica GT-Four Special GT 590bhp @ 1.8bar
1989 Van Diemen RF88/89 Formula Ford 1600
2008 Nissan Patrol GU 3.0L ZD30DDTi 154bhp

We didn't try Phils ecu on another car. Steve's ECU was fine on phils car running either Steves or Phils map. The standard ECU was absolutely spot on with fueling, it was how I would want to map an aftermarket one !!!

I measured the voltage on the ECU at 13.7, the ECU reads as 13.5 and is stable and smooth (relative to Earth at the ECU). I measured the alternator at 14.4 and battery at 14.2 so there is some drop through wiring, relays etc.

There are some small spikes on the Earth and power lines when measured relative to the engine Earth but this is to be expected with the inductance of the wires from ECU to manifold Earth and switching pulses for the injectors. I was surprised to find all ECU wires identical as if connected together locally at the ECU, I would have expected the high power to be run separately.

_________________
If at first you don't suck seed, try drier grain.

Wow, I have missed a load of posts, seems that the "View posts since last visit" is not working?

Anyway, running the stock ECU until I get chance to put my map onto Dans ECU and see how things are, If I can load it?

Not sure what this issue with versions is about?

I am trying to get hold of Ryan to see what SW he uses.

_________________
[color=#000080]ST205 WRC 1994
Hybrid turbo
Apexi AVC-R
Apexi Power FC (Not being used ATM)
Apexi Power commander (Not being used ATM)

Hi Phil,
Check the top of the web page. If it says Log Out [] then you need to log out and back in again

There's a gremlin somewhere in the forum software that crops up occasionally

That's sorted it, thanks

_________________
[color=#000080]ST205 WRC 1994
Hybrid turbo
Apexi AVC-R
Apexi Power FC (Not being used ATM)
Apexi Power commander (Not being used ATM)