Where's the 50 BHP gone?

[Nick Mann]

Following on from my Spark Plug thread, I seem to be uncovering more issues with my car.

My mpg seems to be lower than normal at the moment - I don't seem to be getting to 200 miles on a tank at all now, whereas I managed it at least once out of every three tanks before.

A rolling road session last weekend threw up a loss of 50 bhp at the wheels from where I have been measured before.

I am getting occaisonal fuel cut - I have an HKS FCD, which I assume has not been played with and has been there the last couple of times the cars Unichip has been tuned. I have never had fuel cut on this car until late last year.

The last time I had the unichip tuned was in october, and I had it done because the car semed to be running rich and I was seeing some knock when using Evoscan. The boost module was not seeing the correct boost pressure, so a replacement boost module and £300 later, the car was sorted.

I have bought new spark plugs and fitted them this afternoon as a result of the tuners suggestion last weekend that my car needed new plugs, and that was the reason behind the loss of power. However, I then went out for a quick blast with Evoscan and once again the car is showing knock. It is also showing octane at 91%. Is octane rating measured in %? If so, then 91 is bad!! It is BP ultimate in there at the moment, so it should be a lot more than that!

In the daylight tomorrow I intend to check for boost leaks (although it is holding boost and vacuum so I don't expect to find any major problems) and check the boost connection to the unichip. I will log some runs with the evoscan, and hopefully get a couple of runs on the local dyno. I would like to reset the ecu between runs and log both on the evoscan.

Is there anything obvious that I'm not thinking of? Why does my car keep seeing knock and pulling power? Or is it not happening that way? Ideas would be more than welcome!

[Nutter_John]

Hmmm - place holder whilst I ponder for a little while

[Turbo_Steve]

I'm already suspicious of the MAF and/or Lambda sensors.
Given that it's down on power on a dyno run, I'd suggest you try someone elses MAF, or at least get some comparisons on the output with other peoples.
Also, is your airfilter ok? It's possible it's become befouled and isn't flowing properly. Final thought is cat-breakdown blocking up the exhaust.

[Nick Mann]

It's a K&N panel filter. I'll check it, but it was alright a few months back. I have an HKS foam panel to put in, so maybe a good time to swap anyway.

The Lambda seems to be working okay from the Evoscan results. It oscillates on cruise/idle and throws itself towards rich on WOT.

There is no cat at the moment! (SShhhhh.......)

Maf...... Hmmmm. I'll check Evoscan. I have a feeling it can record the frequency (?) reading from the maf. What does a maf read? I could potentially borrow one, but not this weekend.

[bradc]

to me it sounds like a fuel air mixture thing too.

So it'd be MAF, O2 sensor, air filter, fuel pressure perhaps?

[Rossco Type-S]

The guy who brought my first VR4 had the same issue when he removed the Cat, I think it was the O2 sensor.

[The Vee]

to me it sounds like a fuel air mixture thing too.

fuel pressure perhaps?

Yeah, is the pump and/or regulator OK?

[AncientOfMu]

Something to check easily after the problems I had with mine.... Voltage at the fuel pump at WOF...

[Kieran]

If the octane setting is showing 91 that's bad!! The ECU will be throwing shedloads of petrol in and probably cutting timing to try and keep things under control - I guess the big question is wether the 'knock' you're seeing is real or not. How many counts are you getting?

Personally, I think the clue is in the question. I think you need to ascertain the functionality of the Unichip you have fitted. The way to do this is to lob it into your local canal. If it floats, keep it!

Seriously.... Can you post a recent Evoscan datalog? It might offer some clues, though the if the Unichip is bastardising the readings, you might be logging false data - It depends what the unichip fiddles with I guess.

[Gowf]

As K has said, if you are getting knock then the ecu will retard the timing for safey's sake. The lamda doesnt do anything at WOT and as long as its going into open loop then it should be fine.
What readings of knock are you actualy getting, knock count or knock sum? The ecu will only start to retard once it sees a knock count of 5 or above.

It does seem like the uni chip may be your issue. If they control timing, it is possible that the timing is wound up far too much causing knock, or that the mixture is far too lean. Either way, its the first place i would look.

Just out of curiousity, what values of octane have others had when using evoscan?

[Kieran]

My values are always '100' for octane. I've seen a EVOscan log from a standard car on a cold day where a little knock creeps in at high load/low rpm (just a couple of counts) and the octane value has dropped to '97' quite quickly, but then recovered just as quickly.

[Nick Mann]

Hmmm......

I have just been out for a run. To Evoscan and try and work it out. Previous to going out, I installed the HKS panel filter, and checked/tightened all the boost pipes, including one from the boost controller that was split at a connection. I'm not sure if it was leaking, but I thought I'd chop it back anyway.

Now I am overboosting by a considerable amount!

The only power run I did, the boost controller started beeping as it went past the 1.4 bar safety warning. I glanced down as I let the throttle off to see the boost controller reading 1.75 bar! /pan

Looking at the logs afterwards, the knock sum rose to 18 on WOT, but hit 23 as I lifted off. If this is over-fuelling-pseudo-knock, then that makes sense, as I have a VTA BOV, so it will overfuel even more on lift off.

I'll check the boost connections again, try another run and report back.

[Nutter_John]

1.75 bar , ouch are your actuators stuck???

[Wodjno]

sounds like u got a boost leak. Make sure pipe hasn't come off or leaking where it's T'd into ? or at any point along the piping before the boost solenoid..

[Nick Mann]

Okay - found the leak in the control between the boost controller and the actuators. It was on the same pipe as the split I found, but at the other end, so I guess I damaged that when fixing the other problem. Everything is now re-seated, so time for another run!

[Wodjno]

Good luck..

It does sound like your MAF might be goosed.. But maybe it's a fault with the uni-chip also.. Hope your log shows up summat specific..

[Nick Mann]

Here's the graph.

The speed is not reading right - I think the evoscan is converting to mph from kmph, even though it is already converted! The boost is now holding as it should, and I have fixed a couple of other minor leaks in the control tubes along the way.

[Nick Mann]

You can see the knock levels that I am getting - positively scary! When it got to 35ish I was in 4th gear, still at WOT. At this point the car spluttered a bit. Not like full on fuel cut, but a much more minor loss of power feeling. You can see that the octane rating started at 100, and dropped to less than 90 during my time on WOT. It recovers to 100 when driving normally, i.e. when it doesn't detect knock.

[Kieran]

Hmmm.........

Did you log air temperature and MAF readings Nick?

[Nick Mann]

Air temp started at 67-68 degrees, and dropped to 63 whilst on WOT. It jumped back on the throttle closing.


The air flow (measured in hz, I think?) varied between just over 40 and 1100. It seems to be logical for the TPS and RPM readings - i.e. higher at higher load.

I have attached the raw data zipped in xls format.

[Nutter_John]

looking at that graph it looks like the unichip is playing with you timing as i have a datalogger on mine and with 0 timming added on wot it moves the timming to 32 degrees , where as yours is going up to about 37

very strange that the octane rating is dropping to 90 when you booting it , I will connect mine up and see if I can get an octane rating .

When was the last time you changed the fuel filter ?

and deffo that level of knock sum is starfish cringe time , when mine firt went on the rollers last year to tune the map2 it was spiking at 9-11 and we pulled it off as thats scary stuff , be carefull Nik does not look to good at those levels

[Nick Mann]

Yep - that is why I'm worried. However, I *think* the car is running rich, which means that the knock is not due to lean conditions. It is still not good, but as I understand it, it's a whole lot better that lean condition knock.

This is reason number one why I need to get a wideband!

And any comparison you can make will be appreciated!

[Turbo_Steve]

Edited to remove post...I'm thinking of the Nissan.

[stuartturbo]

Obviously everyone has missed the most likley
Judging by the 300bhp/rally saxo's etc


HAS a Max Power sticker fell off/pan



or in your became attached:thinking: :inquisiti /lol

[Dan_G]

bloody eck, thats alot of knock .... i poop myself when i see a value of 6.

Be careful you dont kill the engine, you definatley need to know what is going on with your afr's...

i'd stay away from WOT until you have some more answers...

[Nice But]

Guys sorry to intrude on this over my head thread....... but what is WOT ?

[Nutter_John]

derrr wide open throtle :P

[Kieran]

derrr wide open throtle :P

Hey, you can't talk - terms such as 'Clean' or 'shiny' are still alien to you! /pan /pan /pan

[Nutter_John]

Hey, you can't talk - terms such as 'Clean' or 'shiny' are still alien to you! /pan /pan /pan

Ha ha ha , can't argue on that point

[Nick Mann]

Oi!

/Hijack

[Nice But]

Sorry Nick.......I've been told.....


BACK ON TOPIC PEASANTS :scholar:

[AncientOfMu]

Hey, you can't talk - terms such as 'Clean' or 'shiny' are still alien to you! /pan /pan /pan


:inquisiti

[Nick Mann]

Following the arrival of my toys (http://www.clubvr4.com/forum/showthread.php?t=30074) I have been watching my AFRs as I am driving. I haven't had a chance to datalog with Evoscan at the same time yet (I won't be able to log everything until the MAP2 gets fitted next month) but I already have some interesting observations:

At very low throttle, mid speed and low revs, the AFR is bouncing at around 14.5-14.7. This seems to be doing it all fine.
On removing my foot from the throttle, the AFR jumps to 20, which is the leanest reading it can display. So the injectors are switching off as they should be as far as I can tell.
On wide open throttle, at high revs I get an AFR of around 11.5-12.
On medium throttle at almost any rev range the AFR drops to 10 and stays there. That is the richest reading it can display.
On wide open throttle at medium revs, the AFR drops to 10 but recovers back to 11.5-12 by high revs, most of the time. Sometimes it stays at 10 and the car loses a lot of power.

I would like to get evoscan running at the same time and see if I can work out what the min/max throttle position is to get an AFR of 10. It does look like the Unichip is seriously overfuelling under certain conditions......

[Kieran]

Hmmm - I wonder if they've only mapped the unichip for 80%+ throttle position and the rest of the time the standard ECU is left to cope as best it can with all the mods you have?

Reason I'm wondering is that with no fuelling control, I bottomed out the AFR readings at our last dyno day - Mind you that was at WOT.:thinking:

[Eurospec]

Sounds like the map on WOT is a bit lean. You *might* be running into knock at a 12afr. I run a 12afr on 109 octane race fuel!

On a turbo car, on road fuel, really an 11.5 is as far as you'd want to go. An 11 would be nice. a 10.x would be safer.

The mid throttle run being richer than the WOT is interesting. The afr is normally a 10 flatmor even a 9 on full chat with these cars, so it may be over boosting, reaching 100% IDC, running lean and knocking a bit on WOT.

If its knock the dyno graph will be all nasty- full of drops and spikes- as the ecu pulls timing, kills the knock and goes for it again.

Hope that helps,

Cheers,

Ben.

[Nick Mann]

Cheers, Ben. I'll watch more closely over the next few days. I am beginning to get used to getting readings with a quick glance at the gauge - staring at it with your right foot planted is a bit scary! I really need to get it mounted properly.........

[bradc]

perhaps it sees 12:1 and has a bit of knock, so it goes to 10:1, then a little bit later on tries for the lean map again?

My car runs at abouit 12.3:1 from 4-6000rpm, from 6-7000 it drops down to 11:1, and I don't have any knock at all on 98 octane. They can run around the 12:1 region without any big problems it would seem.

[Eurospec]

Interesting brad. There will be a lot of power coming from that kind of an afr.

If its gonna knock, 4000 to 5500 seems to be the nasty bit.

Cheers,

Ben.

[bradc]

well you've seen my graphs and it does indeed make a lot of power at those engine speeds. I do have a bloody big 600x300x100 intercooler though, it always stays cold to touch :)

[Nick Mann]

Okay - following this for a few weeks, I am beginning to see patterns......

If I am on the motorway, cruising, the AFR is usually sitting at around 12. Uphill it drops and on steep hills it drops below the bottom of the gauge. Downhill it can go to 14.5 oscillating, or even completely lean if the hill is steep enough.

If I squeeze the throttle, I can always make the AFR drop to 10 or further, and if I hold it there I get power loss and can get soot in a trail behind the car. If I slam the throttle, the car goes to around 12.5 AFR (I have seen 13 once!!) and then slowly richens up until it hits 10 and power loss/soot occurrs. If I slam the throttle, then ease off to medium throttle, I can hold the AFR at around 10.5-11.5, which is the only time I get good clean acceleration at motorway speeds.

The more I look at it, the more I think it must be the unichip. The fact I can dictate the AFR by the way I drive consistently must mean that the unichip is doing what it thinks it should be? The next stage is to remove the unichip and see what the AFRs do. I am pretty sure they will go even richer, but it's worth finding out.

Roll on the MAP ECU so I can try and set the fekker up properly!!

[Davezj]

ht i would mention this.i don't want to do the eggs and granny sucking but i thougt i would mention this.

Are you are using 98 ron fuel always.

i did a couple of evoscan runs when i first got it, to compare 95 to V-power fuel and i was horrified to se the result.

on 95 and V-power pootling about town didn't make much difference or steady on the motorway. But when you boot it, it makes a massive difference.

On 95:-
on a 0-110ish run i got a 60-80ish knock value which is quite worrying and a 32ish degree retardation on the timing.

ran the tank right down and put about 55L in.

v-power:-
same run 0-110ish
knock of 1 and no retardation of the timing.

i couldn't believe the difference.

should be V-power all the way.

[bradc]

either that or the unichip has forgotten it's settings and thats what the stock ecu wants to do with your rather modified setup

[Nick Mann]

Yep - it's on V-Power. Occaisonally someone elses super, but 9/10 times it's on V-Power.

Brad - you could be right. If so, removing the unichip should make no difference.

[Kenneth]

do you have the standard intake setup?

If so, what happens if you disconnect the unichip? If you do this and all is well, then that is a good indication the unichip is to blame.

[Nick Mann]

Ken - I have the standard air box with an HKS panel filter. The intercooler is an ARC unit and the heads have been ported/polished.

I think the main issue with running on the standard ECU is the fuel sytem. I have a Walbro 255, with a rising rate FPR. (IIRC the fuel pressure rises faster than the boost pressure.) I suspected the FPR for a while, but the lean/rich issue seems to be connected to TPS more than boost pressure. With a small change on the throttle pedal, the boost stays pretty much the same, but the AFR changes significantly.