Why, Oh Why?

[Julian64]

My Cerbs having its timely service, I tend to do the service when I feel it needs it so its not at a particular interval. But it is a tappet service I'm doing as the front left side of the engine sounded like it needed it.

However the gist of the thread, is that when I do this, I can't help looking at a few things I would glady change about the car, and because its supposed to be a highly advanced, highly thought out race engine, I always ask myself three questions.

When I'm looking into the throttle bodies, why oh why does the injector sit on front of the butterfly and even on the opening edge, why not behind, at the very least why not put it on the edge of the butterfly that opens away from you not toward you. Anyone thought of turning their throttle bodies upside down or even remounting the injector behind the butterfly?

Two, why oh why, on an engine designed to get the last little bit out of the bang am I looking at a dirty great big poorly machined ridge just under the butterfly. Has anyone ground this down to actually allow the air to flow without the chicaine?

And lastly what the bloody hell got into someone when they decided where to place the alternator and starter motor.

Does anyone else get miffed by this stuff, or are you all too busy driving your cars at the moment?

[robertjp]

I could also add, from recent experience, why is it so easy for airbox securing bolts to fall off inside the Airbox and dissappear?!

Having needed to change coil packs i noticed, that previously the bolts had been 'stuck' in with mastic...and one had dropped out. I hadnt been near this airbox so hadnt spotted it previously.

Hot sweat time, where is the bolt? Sucked in? surely not??!!

High tech...not, race engine maybe, but designed to scare / frustrate you in many ways.

Anyway, borrowed a boroscope and checked it all out...looks OK, no valve damage, no bolt sat in throttle bodies.

[phil green griff]

Perhaps the bolt in question wasn't there in the first place?

>> Edited by phil green griff on Sunday 28th May 17:00

>> Edited by phil green griff on Sunday 28th May 17:00

[robertjp]

phil green griff said:

Perhaps the bolt in question wasn't there in the first place? >> Edited by phil green griff on Sunday 28th May 17:00 >> Edited by phil green griff on Sunday 28th May 17:00

Could be, but i also noticed new chaffing on the airbox at the same time, i would have noticed that i am sure, but i could have easily missed the bolt i suppose. Made me think it only happened on my last drive? Also found the backing washer in the box too which made me fear the worst!

[longbow]

When I stripped my engine down I found a nut and half of a stud rattling around in the sump!! I think the problem is cost cutting on materials and production methods - the design is so simple its beautiful. By the way, the 4.2 injectors are indeed aft of the butterflies and yes, I've ground down all of the casting ridges in the intakes and polished them while I was at it. I've also noticed that the port matching on the exhaust headers is non existent so thats another area the Dremel will see some action!

>> Edited by longbow on Sunday 28th May 17:28

[Toady1]

ah...the good old dremmel!

[cerbhead]

Julian64 said:

................... When I'm looking into the throttle bodies, why oh why does the injector sit on front of the butterfly and even on the opening edge, why not behind, at the very least why not put it on the edge of the butterfly that opens away from you not toward you. Anyone thought of turning their throttle bodies upside down or even remounting the injector behind the butterfly? ...............................

Julian64 – I know what you mean! I made exactly the same complaint re position of injectors immediately above the butterflies. Also you will notice the butterflies open in different direction on each bank. So on one bank the injectors blow through an increasing opening as the plate opens, while on the other the blow against an approaching plate. Bizarre! Totally illogical - surely the injectors should be below the butterflies, even though at WOT there is a clear path for the injected spray? However I came across the answer not so long ago and it is all to do with small throttle fuel mixing. (a good reference is Internal Combustion Engine Fundamentals by Heywood)

At low speed and small throttle openings, there is a high vacuum in the port, but hardly any airflow upstream of the throttle and little airflow past the inlet valve. This situation provides a poor fuel mixing capability in the port. However, through the two small gaps between throttle plate (TP) and throttle body bore there is extremely fast airflow. In fact two jets form through each gap whose velocities approach the speed of sound. These jets remain separate downstream of the TP and between them is a relatively stagnant region where counter-rotating vortices form.

The sophisticated approach – on non-highly thought out race engines like BMW etc - is to inject fuel downstream of the TP into the jets and onto the back of the hot valve head so that as much as possible vaporises and mixes in the throat immediately prior to being drawn into the cylinder. However to work effectively this requires very careful injection placement and timing. Injector firing must occur after the exhaust blowback period so the vaporised fuel can be inducted before it cools and condenses out. Also it should occur at about the point of maximum piston velocity to make best use of those air jets to mix the non-vaporised fuel. These points vary with engine speed and other operating parameters and need confirming through measurement (research lab) or prolonged rolling road trial and error. Unfortunately the AJP8, being a highly thought out race engine, cannot do this as all its injectors on a bank fire at the same time (not sequential injection as the brochure says). This means most of the time the injected fuel would just wet the closed valve and port walls giving miserable mixing and poor low speed running and power. The 4.2 suffers from this.

While on the face of it the injector placement on the 4.5 seems wrong, it is actually very good for part-throttle, low speed fuelling. That is because the fuel lying on the back of the throttle plate is sheared by the intense air jets through the TP gaps and is violently mixed. Also any fuel aerosol and 'splashing' round the TP caused by the close proximity of the injector will also be ripped into the port by these jets. Having counter-opening TPs on each bank also doesn’t matter that much in this situation.

Given the limitations of the AJP batch injection set-up, surprisingly the 4.5 injector placement is pretty much optimum and a good compromise between low and high speed operation. One reason why the 4.5 is generally easy to drive around town than a 4.2. It suggests that turning the throttle body upside down (and presumably changing the position of the injector so it continues to point downstream) would probably be a bad move.

Cheers

[simonsparrow]

Interesting, certainly my 4.2 is quite lumpy around town and in traffic at times.

[350matt]

So who's going to be the first to fit sequential injection on a Cerby then? It looks to me that both engine sizes have the injector far enough up the port to get some reasonable benefits...


Matt

[Julian64]

Hmmm, interesting, I thought the injection was sequential, just shows how much I don't know. Surely there's an emerald type chap out there somewhere thats got a system that does this?

Be a shame to reinvent the wheel only to find out it didn't make diddly squat to the performance/running. I believe BMW played with sequential injection early on only to find it didn't make a difference to power only to emissions.

Strangely I'm sitting here typing this while playing with a 40meg pic chip sitting on a board with a 2meg USB connection chip. Its got more ports than the QE2 and I'm currently programming it to do a stupidly simple task. The whole board is cheap as chips. Its not quite fast enough to emulate the eprom in the MBE but is much closer than the PIC's were a year ago.

If I had the program of the current cerb MBE handed to me on a plate, and an old MBE board I could graunch, I would seriously consider replacing its brain with this PIC, duplicating the MBE program, and then developing a PIC version one specifically for the Cerb. I wonder what a fully alterable sequential injection with a complete head up display via the USB would be like to cerb power and driveability, or whether the Cerb is currently at the top of the tree with the current range of tailor made eproms.

[longbow]

Probably not worth the hassle..........

www.sdsefi.com/techseq.htm

[350matt]

As far as I'm aware the 'std' Emerald Cerby ECU doesn't have sequential , however its certainly possible on a V8 as Emerald did the first one for me on my wedge and it improved fuel economy slightly, however injector positioning is crucial to getting power out of a sequential set-up, the further up the port the better.

Most modern cars point the injector at the back of the valve for good atomisation and low emissions but on both flavours of V8 cerbs the injector is a long way up the port so they should benefit more then your ' normal' car. once I've paid for my wedding I'll definitly be trying it

Matt

[cerbhead]

Dug this old thread out as I've been looking at the throttle bodies as part of the engine rebuild. I have made some pics and notes for anyone interested. Unfortunately there are more probs than just the ridges to overcome. On my car, each cylinder's TB is different - eg different height ridges in different places - as the maching is not correctly aligned in the casting. The even and odd bank TBs are also slightly different. All in, some pretty big flaws.......

http://i40.photobucket.com/albums/e23
http://i40.photobucket.com/albums/e23
http://i40.photobucket.com/albums/e23
http://i40.photobucket.com/albums/e23
http://i40.photobucket.com/albums/e23

[itsallyellow]

i need to do my tappets. How hard is it julian???

[Julian64]

itsallyellow said:

i need to do my tappets. How hard is it julian???

Not too hard, although I suggest you probably want to do it in conjunction with a cam timing check, as you have to remove the cam timing pulley off the camshaft and I bet a pound to a penny that there are few Cerbs running with the correct cam timing after a tappet check.

If you're doing it yourself be prepared for the car to be off road for the length of time it takes to get your hands on the shims. I have a few swapsies if needed and always willing to lend a hand if you get stuck.

I found the biggest problem was strangely knowing when the first cylinder on each bank is at exact top dead centre. A problem worthy of its own thread. Started off with a bit of wire which worked very badly, thought I could use the ecu to see when the crank sensor fired, and finally used a compressor on the cylinder to do a leak down test which has the added advantage of telling you if you are at top dead centre once you pressurise the cylinder with about 20psi . Strangely all systems disagreed with each other but I vote for the pressure test being a practical fart.

I wonder how the pro's tell top dead centre on the car to an acuracy of 1-2 degrees?

[ridds]

DTI guage.

Only real way to set it to TDC.

[byker28i]

Julian64 said:

I wonder how the pro's tell top dead centre on the car to an acuracy of 1-2 degrees?

You'd use a dial gauge surely. It's the only accurate way to measure tdc

[jesfirth]

and-going back the original thread why did they put the coil packs deep in the rear of the V of the engine where they heat up nicely and crack on the base.

[ridds]

Where else would you put them?

[Howitzer]

A DTI gauge although reasonably accurate isn't accurate enough. On the diesels we used to get about 4 degrees of crank movement while the piston sat motionaless at TDC.

The best way is to again use a DTI gauge.

1 - Make a pointer for the front pulley, sharp and accurate.
2 - Using a Magnetic base or clamp position the DTI.
3 - Bring the piston up to TDC and then using tippex mark about a 90° section of the pulley with the tippex.
4 - Bring the piston back down till you have 1 full revolution on the DTI and using a sharp pencil, mark a thin line on the pulley, over the dried Tippex.
5 - Turning the engine a full revolution in the opposite direction bring the piston back up to TDC until again 1 full revolution is seen on the DTI, again mark it.
5 - Measure between the 2 marks and then half this measurement, that is your perfectly accurate TDC mark for that cylinder.

Dave!