john banks

Over 890mV from the lambda sensor at WOT - I don't know how much over and I don't care - it is pig rich - probably double figures % CO. Charge temps peak at 30 C on the hardest back road blast I am capable of, ambient 15. I put down power as much as possible to try and get the temps to go up.

At similar boost (not sure of the accuracy of the gauge on that car) on track after about 10 laps of Knockhill the charge temps reached 52 C ambient 10 C but they cooled very quickly and the driver was REALLY hammering it and keeping up with much more impressive machinery.

At the same level of boost on a Select Monitor no sign of retarding and now I run 2 RON extra. I actually think the boost could go higher for road use, but I want to leave a margin. I will probably lean it out a bit subject to the usual safety parameters.

So it is quite conservative really Any other objections?

Paul_H

read the thread

Scotsman

iTrader: (3)

I'm sure I'm like a lot of people reading these threads - very interested and excited about some cost effective mods to the scoob but wishing I actually understood what everything means! (at least to a basic level).

Any pointers to web sites that can help us innocents understand where and what things such as actuators, restrictors, solenoids, fuel cut lifter, etc are?

Reading the various threads lets me make some obvious assumptions but it would be nice to have a better understanding.

Cheers
Richard.

john banks

http://www.google.com and Scoobynet taught me all I know, but I had a basic understanding of electronics before - used to fiddle with an electronics kit as a boy.

Scotsman

iTrader: (3)

Away searching

Pity the search feature on scoobynet is still down and out.

Richard.

James_PowerMad

Those sums I promised...

(Vout - Vz) = ((Vin - Vz) * R2 / (R1 + R2)) + Vz ### when Vin >= Vz

Vout = Vin ### when Vin < Vz

Where Vz is the Zener diode voltage


I would recomend using a voltage drop (zener diode) which is as close to 0 psi as possible (slightly above 0 psi), this will ensure the ECU sees atmospheric pressure correctly (2.3 volts).

Then choose (R1 + R2) / R2 to be the fraction by which you want to increase the pressure target and fuel cut by (with respect to 0 psi - zener voltage).

E.g. Your original target is 14 psi, and you want 18 psi, then...

(R1 + R2) / R2 = 18/14 = 1.29

So, some good values could be:

R2 = 1k
R1 = 290 ohms

I am particularly interested to find out whether this mod alone will cause the ECU to increase boost after some 're-learning' time, or whether actuator tightening is necessary.

[Edited by James_PowerMad - 4/3/2002 9:53:18 AM]

[Edited by James_PowerMad - 4/3/2002 7:56:36 PM]

James_PowerMad

Oh, yes...

The above calculations assume that the circuit has been buffered with an op-amp before, and after (non-inverting, unity gain stages).

We would need to know the sensor and ECU input characteristics in order to get away without using buffering.

john banks

A neat solution James. You could use a dual op amp for the buffer in a single 8 pin DIL package for simplicity (eg OP296GP Maplin NP22Y). It also doesn't need to be a fancy one either. A 2.7V Zener diode can be had from Maplin QH00A.

Assembly would be simpler with the single chip, and we could have one resistor to adjust slope. I just wonder if this is what the HKS FCD does?

john banks

HarryBoy

John,
If you wish to run under the standard ECU fuel cut (MY00 802 ECU) then, I hope this is not a stupid question, do you need to use the FCD? If so you would only need to adjust the wastegate actuator arm and play with the restrictor size.......

Harry

James_PowerMad

John,

Yup, I agree with your values - well the electronic component ones anyway! ;-)

So, are you going to try it out?

I will try it when I get the time... As I mentioned, I will be interested to find out whether the ECU will slowly 'learn' to hit its new boost target, or whether restrictor / actuator mods are necessary.

john banks

Need to find time to get the parts first, but I am quite keen to try it.

I have a sneaky suspicion that the MY99/00 doesn't learn duty cycles and adjust over time from watching it at play. Possibly the others are different. I think it just drops the duty cycle if you go over target, so yes Harry you do need to use the FCD to stop the ECU holding just 14 PSI even if you only want to hold say 16.

James_PowerMad

You won't need a FCD and my proposed circuit at the same time, though Harry.

The idea behind my design is simply to scale down the positive pressure that the ECU 'sees'. Thereby the ECU thinks that the pressure is OK when you want it to! The effect on the Fuel cut point is the same (moves it up by the same proportion).

HarryBoy

Trying to order from the Farnell Web site is a nightmare. As an alternative to the LMC7111 the LMC6462 (RS 1822047)is available from RS which is much easier to order from!

It's an 8 PIN dual OP Amp package.

Harry

James_PowerMad

The zener diode works by only beginning to conduct current in the reverse direction when the voltage across it (in the reverse direction) is the zener voltage.

If when the input voltage got up to the zener voltage, a large current was conducted, then the voltage across the zener would drop below the zener voltage (no conduction).

In practice the current conducted will be the right amount to keep the voltage across the zener very close to the zener voltage (unless the current drawn is too much for the diode).

As far as I can imagine, the effect you mention would only happen if there was some hysteresis in the diode. E.g the diode starts conducting at V1, and stops conducting at the lower voltage V2. From what I know about zeners, this effect is insignificant (or is not part of the zener characteristic).

john banks

Makes sense James. I guess we'll have to try it.

Harry please note if you are building my circuit you need two op amps in SEPARATE packages as one powers the other - this is how it clamps. Also make sure it tolerates input voltages above supply - otherwise it won't work.

If we get James' circuit to work I think it will be much better. If we get the slope right the ECU will even regulate held boost for us and stop it going too high - say on the track - it will just peak more rather than run more.

Chris.Palmer

Guys,

I'm in the process of building a MAP display guage for MY96.

Either my calculations are incorrect or the earlier MAP Sensors are scaled slightly differently.

kPa mmHg PSI Bar mv
-26.7 -200 -3.87 -0.267 2100
-21.36 -160 -3.096 -0.2136 2200
-16.02 -120 -2.322 -0.1602 2300
-10.68 -80 -1.548 -0.1068 2400
-5.34 -40 -0.774 -0.0534 2500
0 0 0 0 2600
5.34 40 0.774 0.0534 2700
10.68 80 1.548 0.1068 2800
16.02 120 2.322 0.1602 2900
21.36 160 3.096 0.2136 3000
26.7 200 3.87 0.267 3100
32.04 240 4.644 0.3204 3200
37.38 280 5.418 0.3738 3300
42.72 320 6.192 0.4272 3400
48.06 360 6.966 0.4806 3500
53.4 400 7.74 0.534 3600
58.74 440 8.514 0.5874 3700
64.08 480 9.288 0.6408 3800
69.42 520 10.062 0.6942 3900
74.76 560 10.836 0.7476 4000
80.1 600 11.61 0.801 4100
85.44 640 12.384 0.8544 4200
90.78 680 13.158 0.9078 4300
96.12 720 13.932 0.9612 4400
101.46 760 14.706 1.0146 4500
106.8 800 15.48 1.068 4600
112.14 840 16.254 1.1214 4700
117.48 880 17.028 1.1748 4800
122.82 920 17.802 1.2282 4900
128.16 960 18.576 1.2816 5000

Any Comments?

Regards,

Chris

{edited to add}
You'll probably have to edit this to see it properly

Also put in correct figures - DOH!


[Edited by Chris.Palmer - 4/3/2002 5:36:35 PM]

HarryBoy

Two Packages! Ahh....missed that, luckily I ordered two and yes it can tolerate i/p voltage above supply rail....

Harry

john banks

Certainly is scaled differently, although it seems a bit more than a 2 bar absolute MAP sensor?

James_PowerMad

I know why there was confusion about the sums earlier, John...

I left out a rather important part of the equation in my earlier post. The circuit's functionality is still the same as I said earlier, and John's resistor calculations are still correct, but I probably confused some of you by leaving out the + Vz from the original equation.

E.g

Vout = ((Vin - Vz) * R2 / (R1 + R2)) + Vz ### when Vin >= Vz

This gives the correct continuous output curve for the circuit.

I will now go and eat Quiche in shame... ;-)

James_PowerMad

bttt, so John doesn't miss it.

john banks

Yes I saw the Quiche in shame bit.

Don't be shamed!

GOOD NEWS - Xavier has found that with a 1mm restrictor and tightened actuator on his (I think MY01) he can in cabin adjust from 1 to 1.25 bar by altering the setting on his FCD! I have asked him whether the 1 bar setting is very peaky and await his reply. His FCD changes the slope of the whole MAP line like the one James posted but from 0V rather than from 2.7V. Probably the same effect.

Now this effect may be the clever MY01 ECU having better closed loop boost control. It might work on MY99/00 - only one way to find out.

Chris.Palmer

For My96 I believe the max reading (Fuel Cut) is set at

PSI Bar mv
15.48 1.068 4600

Above this I am led to believe the sensor misreads (maybe no longer linear?)


According to the Workshop Manual -

At Idle - vaccuum should be approx :

PSI Bar mv
-9.288 -0.6408 1400
-8.514 -0.5874 1500
-7.74 -0.534 1600

Ignition on Engine not running - Zero (ie atmos pressure) :

PSI Bar mv
-40 -0.774 -0.0534 2500
0 0 0 2600
40 0.774 0.0534 2700


PS - The table of figures in my earlier post were produced by extropolation based on the WorkShop Manual Figures shown above. I did this for comparison - sorry if I mislead.

Chris

James_PowerMad

I don't think it is valid to change the slope of the whole line, because you will confuse the ECU as to what pressure the Atmosphere is at!

I guess the ECU measures atmos pressure occasionally (otherwise it would not need a pressure sources switching solenoid).

We don't want the ECU thinking that it is on the top of Everest! ;-)

john banks

I agree James, your circuit is superior, I was just posting his findings. His is a MY01 and I think that measures atmospheric at the ECU rather than with the switching solenoid. Also not sure if ours will just peak if the restrictor is small but the slope is not adjusted enough, but a range of adjustment in-cabin might be possible.

So Chris yours is effectively a 2 bar absolute MAP sensor.

[Edited by john banks - 4/4/2002 10:13:57 AM]

Chris.Palmer

John,

I believe so - but cannot find official confirmation.

The My96 MAP certainly appears to be accurate up to the fuel cut - so could possibly run boost up to that limit & lift the defend slightly higher to prevent overboost.

Think the early MYs also have other weaknesses too that would suggest 15.48 PSI a sensible absolute Max even with a cut lifter.

The main reason for starting my MAP display project was to see exactly what the ECU is seeing not rely on an inaccurate pneumatic boost guage.

Chris

GavinP

Chris,

Think the early MYs also have other weaknesses too that would suggest 15.48 PSI a sensible absolute Max even with a cut lifter.

It is not all one-way traffic (i.e. newer is not all better) - early cars may have closed deck blocks and under-piston oil sprays which don't appear on later cars.

Sorry for wandering off-topic there...

Thanks

Gavin

Chris.Palmer

Gavin,

I agree, my comments are really with respect to Pre 97 UK models I don't have much experience or knowledge of the early WRXs.

As far as I'm aware early WRX owners have the pleasure of closed decks, UK owners are lumbered with open decks.

I believe the early WRXs may also have better pipework & MAP sensors & therefore they may tolerate more boost than the early UK models. This I'm not sure of though.

Chris


[Edited by Chris.Palmer - 4/4/2002 1:09:36 PM]

GavinP

Chris,

I think both UK and JDM early cars are very similar - obviously later cars get forged pistons, better intake pipework and larger injectors among other things.

I think the main concern is intercooling on early cars and the 380cc injectors prove to be a limitation to significantly higher boost.

Thanks

Gavin

john banks

No problem with the off-topic guys.

Coming back to topic, I think James' method will be better because you can set your held boost to what you want with a restrictor, and then adjust the gradient so that the held boost does not then rise above this. In thrash conditions/cold weather you will get a larger peak but the ECU should lock onto your boost target. On the MY01 it seems that is even takes away the peak for you and adjusts boost just by changing the FCD. But the ECU is a bit cleverer in the MY01.