Dual Wideband Datalog
Turbo Mini Forums

About ignition amplifiers: I found Bosch dual ignition amplifiers very fragile. I burned out two of them while debugging my ignition code. Now I'm using 2 single channel amplifiers.

AFR:

I also think AFR difference comes from the wall wetting and not "charge stealing".

According to Heywood, wetting is mainly dependent on the wall temperature. For every temperature the film thickness ratio is constant.

The travel speed and evaporation Tau of the fuel film are very slow, so while both valves are closed the film evaporates and also goes closer to the valves. Once open the inner valve gets all the evaporated fuel, which compensates all the newly sucked (from carburator) fuel that is trapped to restore the wall film. The outer valve does not get evaporated extras, while the wall film is still being restored.

This assumption could be easily verified on a very hot MPI engine: with a very thin or even none fuel film all cylinders should run the same AFR. This can not be verified on a single carb engine, in such setups there is always at least 10-20% of fuel traveling on intake walls.

I believe injection must try to maintain the wall wetting level constant. This is why the choice of proper injectors is very important. Today we can buy injectors with plenty of spray shapes. I think (and I bought) MPi needs wide spray injectors, if placed close to the head. Because a tied spray will send most of fuel on both valves and there won't be enough time for the fuel trapped in the inner cavity to reach the outer valve. A wide spray will send more fuel on walls, thus maintaining the wall wetting. And this will behave like in usual engines with a balance between evaporation and film trapping.

May be I'm wrong. In this case I lost $250 I payed for four 450cc EV6.

And I'm still writing my injection code :)

Comment about logs: I think (I may be wrong) your engine isn't properly adjusted. At low MAP your AFR is around 15-16. Which means your idle is certainly lean. I think idle must be slightly rich, especially with carbs. That's why we have so many problems with idle CO level. If you connect a CO meter on any well adjusted engine you'll notice the CO falls as soon as RPM increases. Because combustion is better, so no need to be rich.

Edited by alpa on 7th Dec, 2007.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

If this data holds up after the instruments are sorted, it is very interesting.
If in fact we are getting lean on outer cylinders even with mild cams, it makes me wonder if this kind of logging would show scatter patterns solving that issue.

1275 with Back Door Turbo

Is the inlet manifold heated?

Took another couple of logs this morning.

First calibrated the sensors. Although still fitted to the car, it has not ran for 5 days so should have been reading free air.

No change.

Second log attached was with the sensors daisy chained using the serial ports

LM-1 -> LMA-3 -> LC-1 -> Laptop.

I also retained the 0-5v reading of AFR from the LC-1 for verification purposes.

MAP logged in PSIa this time for some reason, so a slight difference to the graph.

However, results much the same.

This afternoon's job is to swap the sensors over and see what happens.

Attachments:

24.35kb

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

well done paUL .
i wonder if the theory about rpm related inertia causing that change over as the revs rise ,and the charge dashing down the outside ports despite robbing is correct ...

,personally im of the mind that the charge is possibly being thrown to the outside port ,by the manifold design in this instance ,but also that the velocity in the port causes the robbing to become innefective in denuding the outer port because the gas gets there fast enough to fill the void caused by the robbing , so that it is not noticed by the outer cyl .
very interested ot see if swapping the sensors has an effect .
im also wondering about the comments on other fori about the sensor temp affecting readings ,and wonder if your sensors are running at different temps sufficient to create a different reading on the same a/r , im sceptical that this could be true ,because if so we would all be changing readings simply by revving the engine .

i would like to see a steady state high rpm test ,say uphill with the handbrake on !! to compare to the same rpm on a transient reading ,accelerating through that point say 5000 rpm in the same gear ,and see if they read the same .

another good test would be on an injection car to see how it copes ,eg a spi ,to see how the manifold changes all this ,or perhaps a su on a turbo manifold .

Edited by robert on 7th Dec, 2007.

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Paul, Can you tell me where you fitted both wb sensors in the ex manifold and also what plugs and leads you have fitted.
Regards
Dave

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

After an hour or so of skinning my knuckles, breaking my back transfering the sensors, nothing appears to have significantly changed.

The latest log is attached. The LC-1 is now on the inner cylinders and the LM-1 is on the outers.

The relationship between the inner and outer cylinders is still the same as the first log, so I'm assuming the readings to be correct.

Robert, I like your ideas but time is of the essence and I can now get on fitting the EFI knowing that the instrumentation is correct.

There is one obvious reason why the outer cylinders are running lean, that I have not checked. Can anyone guess what it is? Clue is in the picture above.

Attachments:

28.7kb

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

exhaust leak?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

This is terrific work.

What about the manifold heat question Sprocket asked?

I've got mine wrapped up with heat blanket around the exhaust and the intake manifolds, largely because the downdraft Weber inlet manifold features an uphill climb from the plenum bottom to the inlet ports on the head.

If the evaporation off inlet manifold walls speeds up, maybe this goes away.
I seem to recall something in Vizard's book about the stock manifold exhaust head being too much but water heating the manifold improving performance versus no heat. Maybe imagining it, will have to look tonight.

1275 with Back Door Turbo

Attachments:

18.31kb

Edited by Paul S on 7th Dec, 2007.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

What sort of wideband sensors are those? they look mega different to the ones I have.

I have found the center cylinders to run almost 100c less than the outers. I too have to swap things around to to find out if its the hardware. I have a suspission its the hardware as I zapped the second EGT chip and had to replace it.

Is the inlet manifold the cast iron jobbie or the ally one and if its the ally one is it heated? The heaing is important as it stablises the manifold temperature.

The SU manifold has many sharp turns in relation to the SPi manifold. The SPi manifold has long straight runners. This shape manifold Vizard shows to flow well also, that is the downdraght webber manifolds, and its almost identical.

Will be interesting to compare the results between the SU manifold and the SPi manifold.

Top Job!!

Paul,
The fuel evaporation rate increases with lower pressures. Could you try to log stable and transient RPM modes with partially closed throttle, say at 60kPa ?
I'm not sure it would allow a better understanding as evaporation rate is very slow, but may be we would see something interesting.

Thanks for your work.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

Attachments:

40.37kb

Edited by Paul S on 10th Dec, 2007.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Perhaps a rubbish idea, but surely one way to prove the instrumentation would be to fit a second boss to one of the downpipes, and then log a decent length run with both sensors in the same downpipe. Obvioulsy if there is any drift occuring over time or due to temperature this would then show up as differing readings between sensors?