1500 Efi

[Triumph-V8]

I've read that the PI's high pressure results in better fuel atomisation than EFI. Don't know if that's still true? If so, that would explain why Triumphs produce more power on PI than EFI. Presumably modern inlet manifolds are designed to locate the EFI injector in the optimal location for their spray pattern, whereas stuffing an EFI injector into a PI manifold wouldn't work so well. All just guesswork.

 

I've just started looking at different styles of injector. One nozzle, four... Maybe I should build a rig to see what their spray patterns look like.

 

Better atomisation needs proper injectors first that spray perfect.

Not every old TR6 has that. Many let droplets out.

Second idea is what happens if the atomisation is bad?

You will need more fuel because droplets can not participate in the burning process that nice.

In my opinion that will affect more fuel consumption and exhaust gas quality than power.

 

The more power of the PI against EFI is a rumour that can not be cured.

Fact is that in theory it is better to spray one shot of fuel at the right time

than to have a more or less constant fuel stream all the time.

 

But:

Idle requieres to give the fuel long time before the inlet opens for better atomisation.

Power requieres to give the shot direct into the

airstream at the right time and that is opposite to nice idle.

So in many cases the mechanical injection is setup for proper idle.

 

The difference of constant fuel to shooting at the right time will be around only one HP.

The main difference in power might come from the different construction.

Nearly all switches to EFI from PI deal with a single throttle body.

This limits cams due to lumpy idle.

But on should not compare a EFI with 260 degree cam with a PI and 280 degree.

There can be around 20HP between.

As far as I know I am the first to establish EFI with six throttle plates like PI.

Many people thought that this is not possible and that is the reason that EFI is blamed for less power.

In fact you can go with six butterflies to 300 degree cam with acceptable idle and have nice power than.

 

The modern injectors have a suitable place in the PI manifold

no need to change something, it works nice.

Could be a bit closer to the valves but its okay

and can be cured with a bit more care with the accel fuel setup.

 

The position and spray pattern deals mostly with the acceleration fuel and reaction on pressing the pedal.

If the injector is far away from inlet valves or pattern sprays against the inlet walls

a lot of fuel condenses at throttle opening weakening the mixture more than necessary.

That is why race engines prefer injectors far away for the price of difficulties when throttle is opened

and get more power and all day users prefer not wetting the inlet walls with fuel because that is

condensing and going to gas always on throttle opening change.

One more reason that this does not affect race engines that much is that they have lower vacuum

having not that difference betweeen opened and closed throttle.

 

The most effort should be spent on the right size of injectors:

It must be enough fuel availiable to feed the engine at high revs where time is short and

a lot of opening time is needed to let fuel in.

Also at idle the injector should have a long opening time

that little differences in thousands of a second do not make such a huge difference in fuel amount.

At idle the injectors open only in the range of 4 milliseconds what is pretty short.

One millisecond difference which in fact is not a long time makes a difference of 25% there!

[James]

Me too

[pomwah]

Gentlemen,

 

I've been following this thread with some interest. Part of putting a 4A-GE into my Spitfire is converting a FWD engine setup into a RWD configuration, involving the "reversal" of the intake manifold, hence my interest.

 

Thinking about what has been raised/discussed above a couple of questions occurred to me wrt manifold design. Laminar flow is a problem inside the plenum so why are they essentially smooth, why not dimple the interior like a golf ball (or even a shark's skin) to create some turbulence at the interface between the airflow and the plenum etc? Secondly, why do many plenums (plena?) feature curved tubes, is it to increase overall length or to create vortices, as the air velocity is higher nearer the large radius, to mix the fuel/air more effectively/efficiently? If so, why not rifle a straight tube to create a vortex?

 

Just thinking.

 

Cheers,

James

Edited March 8, 2013 by pomwah

[GT]

Better atomisation needs proper injectors first that spray perfect.

Not every old TR6 has that.

Many let droplets out.

 

You will need more fuel because droplets can not participate in the burning process that nice.

In my opinion that will affect more fuel consumption and exhaust gas quality than power.

 

The more power of the PI against EFI is a rumour that can not be cured.

Fact is that in theory it is better to spray one shot of fuel at the right time

than to have a more or less constant fuel stream all the time.

 

But:

Idle requieres to give the fuel long time before the inlet opens for better atomisation.

Power requieres to give the shot direct into the

airstream at the right time and that is opposite to nice idle.

So in many cases the mechanical injection is setup for proper idle.

 

I didn't intend to start a debate in this thread in reality.

 

In my statement I was merely saying, theorists apart and all, I have yet to see an EFI system that gives as much power as the Lucas system. It becomes even less of a margin advantage when you use megajolt and map the ignition.

 

The Lucas system depends entirely on a linear fuel demand from the engine.

That takes good engine component design & work.

 

I suspect that is the only major barrier to getting good results, and even to remove that characteristic "BIG DIP" in the torque curve at 3000rpm, so characteristic of this engine.

I even managed to do that with springs, bob weights and mechanical injection.

[GT]

But on should not compare a EFI with 260 degree cam with a PI and 280 degree.

There can be around 20HP between.

You would be suprised to see how little difference there really is between a 260 degree camshaft with lots of lift and the 280 CP cam which has (relatively) hardly any.

Once you modify the head the difference is very far from clear cut.

Relatively speaking it's difficult to get more inlet flow under 0.360" lift, but far easier to get that proportional increase between 0.360-0.420".

 

On the exhaust valve it's very much the reverse.

You can get a lot more flow out of the exhaust port up to about 0.380" and then it all flattens off and it's very much law of diminishing returns after that.

 

So you have the old arguments of Duration v Lift.

The CP camshaft has relatively good valving on the exhaust, but quite poor dirty valving on the inlet, giving it the rather "detuned character and bad emissions.

 

It's all very clear cut below 3500-4000rpm, short duration wins hands down.

After about 4500rpm the short duration camshaft falls down and the torque drops like a stone even if you get more flow from the head.

All you get with the 260 is massive lorry torque from 1000-3000rpm.

 

Martti in Finland did excellent work with (Hestec) EFI and tested pretty much all this stuff out.

 

Check it out.

 

http://www.kolumbus.fi/triumph.tr6pi/dyno-page.html

[Triumph-V8]

With no doubt the degrees of the cam is not a sufficient information.

Americans do better and quote their cams with extended valve clearence of 1.25mm.

Anyway Triumph cams 123HP and 143HP can be compared that way

showing that 1HP difference per degree and with the Bastuck family cams

I am just exploring the same thing as a rule of thumb.

Also often one can do with cams family of some other manufacturers.

 

What I wanted to point out was that if you never found a similar powerful EFI like PI

the reason might be hidden in the rest of the engine.

But you quoted the opposite already with the link above.

The tested Hestec EFI showed a much more powerful engine

than with the former employed PI system.

 

The flow results are predicted in the books of Ludwig Apfelbeck or Helmut Hütten.

If a 4 valve engine reaches 25% lift of valve diametre

and a 2 valve engine reaches 30%

the flow with more lift is predicted not to increase.

Thats what happened with the exhaust precisely.

[GT]

With no doubt the degrees of the cam is not a sufficient information.

 

Americans do better and quote their cams with extended valve clearence of 1.25mm.

Anyway Triumph cams 123HP and 143HP can be compared that way

showing that 1HP difference per degree and with the Bastuck family cams

 

The tested Hestec EFI showed a much more powerful engine

than with the former employed PI system.

 

The flow results are predicted in the books of Ludwig Apfelbeck or Helmut Hütten.

If a 4 valve engine reaches 25% lift of valve diametre

and a 2 valve engine reaches 30%

the flow with more lift is predicted not to increase.

Thats what happened with the exhaust precisely.

 

Oh grief!

This is full of errors.

 

Bastuck doesn't have a "family of cams".

He nicked the cam profile from me and can't even quote the duration right.

This cam he sells is a Holbay thing, which I stopped using because of the extended flat spot in the mid range.

 

All the other stuff he sells comes from either Kent or cribbed off someone else.

 

As for the Americans doing better.

Sorry, this is utter rubbish.

They have been incapable of making a proper race cam for the Spitfire since the very beginning....

(the Le Mans Spitfire camshaft was yet another Holbay design, I know this because John Reid took one look at it and said "oh I think I've seen that before!").

I think it's far more important to be familiar with the lobe shape and acceleration figures, as well as critical figures such as lift at overlap.

Martti also quotes his power figures as far too high.

In reality the 165bhp DIn figures is only 145.

TR6s only lose 20bhp in the transmission not 40.

 

It's true to say the majority of CP TR6 engines only make about 135bhp DIN. (I tested loads myself).

The CR engine has a MUCH milder camshaft, and the engine will NOT make more than 125, quite simply because the torque peaks much earlier.

It also has injection timed at 25deg ATDC with D shaped inlet manifolds, which all reduce emissions, and emission variations between cylinders.

That is what they did it all for.

What you have to realise is, the TR5 with that lighter flywheel and a good well built engine has much more margin for power.

A properly blueprinted one will make as high as 153-158, especially if you derestrict the air filter.

The CR engine will sit there whatever you might try at 125, so you see the CP engine has 25% more power when properly built.

 

The difference between the 145 shown on EFI, and 153-155 on Luca injection is literally day and night.

The power delivery is completely different.

 

I would say Martti's engine was very far from a production engine.

It's almost blueprint standard, full balanced and has had some head work done, yet it STILL doesn't make the power of the equivalent Lucas Pi power plant.

 

The TR5 CP original engine spins right up to the red line when properly built and behaves like a true sports car engine, with a fair deal of punch above 3000rpm.

I've driven loads of them, because we had almost nothing else in France.

 

The Hestec unit you see is quite the reverse. It's like the CR unit, really flat torque curve, but very little top end.

It's what he set out to do, and he says so.

 

We are all very curious to see the power difference swopping from P...nix exhaust to mine, but it's not yet spring up in our part of the world, as you may remember!

[CHRIS211083]

I think Tim would love to test your exhaust for you GT and in turn you could test the EFI manifold.

 

Chris.

[Gt6s]

 

Nearly all switches to EFI from PI deal with a single throttle body.

 

As far as I know I am the first to establish EFI with six throttle plates like PI.

 

 

Not unless you did it before 2002. thats when I did mine.

 

 

Laurence

[GT6 Nick]

Hmmm, I'm learning a lot here. In theory, you'd think that if an EFI injector could deliver similar amounts of fuel to a PI injector, at the same speed and similarly atomised, then their performance potential should be the same. As I understand it (as that most dangerous of beasts, an enthusiastic amateur) the differences between the two systems are related to single vs individual throttle bodies?

 

Anyway, I just wandered off and read Martti Ojanen's TR6 website: http://www.kolumbus.fi/triumph.tr6pi/

Nice car, nice website :-)

[esxefi]

Not unless you did it before 2002. thats when I did mine.

 

 

Laurence

 

and i must be somewhere near a close second

[Gt6s]

Hmmm, I'm learning a lot here. In theory, you'd think that if an EFI injector could deliver similar amounts of fuel to a PI injector, at the same speed and similarly atomised, then their performance potential should be the same

 

I agree completly. Tthink the confusion is compairing 6 PI throttles with a single throttle body on EFI, comparisons should be made like for like EFI through 6 PI throttles v PI.

 

Laurence

[Gt6s]

Just as a by the way. Was at a car club meet the other night, one of the guys is putting EFI on a 1500 Spit using stripped out weber 40's as throttle bodies. The injectors will be mounted on the head above the inlet ports. Just as Ford did when they EFI'd the Pinto.

[spitfire6]

Also 6TB and Emerald EFI here. 2008 I tink

 

Also shite MPG compared to PI. 33-37MPG to 16-22MPG LOL.

[Triumph-V8]

Just as a by the way. Was at a car club meet the other night, one of the guys is putting EFI on a 1500 Spit using stripped out weber 40's as throttle bodies. The injectors will be mounted on the head above the inlet ports. Just as Ford did when they EFI'd the Pinto.

 

Need not to strip Weber DCOE because Weber had seen the signs of time

in its last days and invented nice throttle bodies.

Later several companies continued production and for example dbilas dynamic in Germany sells these items.

They are based on the shape and pattern of the DCOE and use a lot of those supply.

Fuel rail and all that stuff can be obtained there ready to use

and for those who are not familiar with metall work

this might be an option to built an EFI just with combining and mounting.

 

Nice to read that others also use 6 individual flappers.

Would like to know more details about cams and headwork and so on.

How are the engines used, how does it work?

Price for one naked body is about 250 Euro.

[GT]

Yes it's Iain's car, which I HAVE TESTED, convinced me more that all this theorising is what it is.

After spending the really serious money, I had the task of trying to put AT LEAST. some of the worst aspects right.

The money had been spent and it didn't work.

The engine was even fragile & couldn't hold oil pressure.

 

To give you some idea of the scale of what I think, that car was mapped and set up by no less than Dave Walker on his own Sun RAM 12 himself. Nobody should have done a better job than that.

I have the torque curves for the engine.

They are nothing short of catastrophic.

 

There is a case study** which has slowly formed over the time between the birth of my little daughter Alexandra (yes it sounds improbable but it is weirdly relevant), and a certain person in Sutton Coldfield who ripped me off to the tune of 10K, to prove he could make an engine on EFI which not only would make more power than anyone else's but it would be ALL high tech, using other people's bits....

DITTO. Same result.

Catastrophically bad torque curve, and again MEGA budget, all steel engine etc.

 

Both those engines have ZILCH torque low down, then a power band 800rpm wide between 5000-5800rpm, then NOTHING.

 

Another case study.

Another mega budget engine (15k+), same story, Weber Alpha EFI, all put together by a "dedicated team" at Revington.

Same result, same total absence of a decent result, and not even the power of a decent CP engine.

I just took this guy out in the car you have in the video YESTERDAY afternoon.

 

The air was full of expletives.......and then "I WANT THAT!", yet the car I'm driving has a low budget recon engine, without even a set of forged pistons, and runs on 95 octane fuel.

Nowhere does it have less torque than the original engine, except the previous peak at 3000rpm arrives at 2300 instead, then builds into a mountain at 3500-4000.

The metering unit has been specially calibrated not monkeyed around with, like the one that came on the car from "Up yours" Richard

 

Driving it on a mixture of A roads and motorway, I've NEVER been in one of these motors that's this quick.

Once it's on cam around 3500-4000, it's literally devastatingly fast all the way to 6500rpm, and it does 30mpg.

 

I'm a doubting T.

I believe what I see.

 

What I see made from those "crap bits" from Triumph, with original cranks and stuff, & the right mixture of engine components is a zillion times more convincing than this PAINFUL** experience rammed up my ass from some Ego tripper (to achieve exactly what)?

[Nick Jones]

Frankly I don't see why there should be any significant difference between well calibrated PI and well calibrated EFI when used on the same inlet tract. I've got my doubts whether it makes much difference to WOT performance whether you use the original butterflies or single TB on the plenum. Clearly there will be behavioural differences between the two throttle arrangements at part throttle and transition, each with positives and negatives.

 

An awful lot is due to setup. I had a very interesting discussion with a TR5 owner at a show last year. He bought his car with a Weber Alpha system on it, fitted by a well known TR specialist. Flat-spots, rough running and epic thirst. It came with two 4 figure receipts for RR tuning from a Weber agent. He was quoted similar for them to have another go (and only Weber agents have the software) so he bought a Megasquirt ECU for alot less, fitted it himself and had the car running far better than it ever had before within 2 hours of getting it running, just by driving round on Tuner Studio auto-tune. This only a few days before the show. He was going to get it RR'd locally (we discussed places to go and I suggested Tipton garage) but I don't know the final outcome.

 

Overall, I'd expect an electronically managed system to be able to manage better fuel economy in road use (mostly due to 3D ignition control). I also expect that it would be easier to set up, not least because you wouldn't need a PI-savvy witch-dotor to fettle the MU.

 

Alternative induction tracts obviously throw other factors in (as do headwork, cams exhaust et al) but I'm still pretty happy with my scrap-bin special on the Vitesse.....

 

Nick

[GT]

I had a very interesting discussion with a TR5 owner...He bought his car with a Weber Alpha system on it, fitted by a well known TR specialist.

Flat-spots, rough running and epic thirst.

 

.......I'd expect an electronically managed system to be able to manage better fuel economy in road use (mostly due to 3D ignition control). I also expect that it would be easier to set up, not least because you wouldn't need a PI-savvy witch-dotor to fettle the MU.

 

Alternative induction tracts obviously throw other factors in (as do headwork, cams exhaust et al)

 

Doesn't that illustrate the point perfectly?

TBQH, Dave Walker is regarded by many as being up with Vizard and Baker for engine tuning.

I have met the bloke.

 

He is manifestly incapable of tuning his own EFI on a very simple pushrod 2V engine, and that engine has no torque.

We did a little extra work on the SUN RR, and some little tweaking to the fuel map which improved it no end, but it still had poor torque particularly low down and mid range, and that was with a p..nix 6-3-1 and "flowed head".

 

If DW can't tune his own darn EFI, or get it to perform like you say, what chance for the bloke in the street?

 

As for the "witch doctor" aspect to lucas Pi, I don't think I've ever had to use any more than a little intuition and "common sense" in conjunction with a Hartridge machine.

It all comes down to measurements, curves and replication.

It's made like a bit of meccano, and couldn't be easier to work on.

The fact is, it's pointless to talk about results, unless you have measured the torque figure on some form of Dyno, as well as the HC/CO levels flat and part throttle.

 

I've done that, from bog standard GT6 engines with Stroms, to Lucas Pi on the same, to EFI, to all sorts of set ups.

SO FAR not a single EFI set-up has ever come close to the Lucas mechanical system, especially if you combine the Lucas system with an ignition map/multi coil crank sensed programmable ignition system.

All I am doing is stating what I have seen.

 

I would love to have someone come along & prove the reverse, really would.

[Gt6s]

Doesn't that illustrate the point perfectly?

TBQH, Dave Walker is regarded by many as being up with Vizard and Baker for engine tuning.

I have met the bloke.

 

He is manifestly incapable of tuning his own EFI on a very simple pushrod 2V engine, and that engine has no torque.

We did a little extra work on the SUN RR, and some little tweaking to the fuel map which improved it no end, but it still had poor torque particularly low down and mid range, and that was with a p..nix 6-3-1 and "flowed head".

 

If DW can't tune his own darn EFI, or get it to perform like you say, what chance for the bloke in the street?

 

When I was researching which UK made ECU to buy I spoke to Dave Walker and asked him loaded questions which I allready knew the answers to. His answers convinced me he is not the genius he properts to be.

 

On the other hand I also asked similarly loaded questions to Alan Warburton at DTA even specifics about modifying PI throttle bodies. His competent answers convinced me to buy DTA.

 

Laurence

[Gt6s]

Need not to strip Weber DCOE because Weber had seen the signs of time

in its last days and invented nice throttle bodies.

 

Nice to read that others also use 6 individual flappers.

Would like to know more details about cams and headwork and so on.

How are the engines used, how does it work?

Price for one naked body is about 250 Euro.

 

It is paying for throttle bodies is the issue.

 

My mill is a budget 2.6 so as to use stock head gasket. Lightened / balanced, running to 7000 Rpm (sees that regular) I modded the head, modded tubular manifold (did'nt like it so I cut it up and recreated it), Revington TR sprint cam. 37 73 / 73 37 290 duration power band 4000 to 7000, DTA engine management. Idles nicely pulls very strong from nothing but REALLY comes on at 4000. Engine in a stripped out Gt6 MK3. 0-60 4.7 sec measured on a Race Technology performance meter and confirmed by ECU data log on same run. Shift light set at 5800 = 140 mph in OD top which it does VERY EASILY infact she is begging for another gear at that.

Had her once on the hard cut limiter which is set at 7000 rpm in OD top (What's that vibration ?). 3.63 diff 28 % OD oversize tyres 205/50/15 = 24 mph per 1000 rpm. (Enough info there to work that out).

As for use, used on the road (totally tractable) track days and autotests. Currently undergoing repair (welding) and attempt to get her to meet regs for sprints and hill climbs this season.

 

 

Laurence

[Triumph-V8]

I used the PI manifolds where a set in GB was just under 100GBP complete.

I like doing some work on these items with ball bearings and new homemade spindles.

But if somebody claims that an EFI does not reach the mechanical PI systems

I see two main reasons for that (in my eyes) false opinion:

 

First is that the mechanical setup must be okay as sensors have to deliver proper signals,

fuel must be availiable constant and with the right pressure and throttles must be synchron.

 

Second the EFI can only be as good as the one who does the setup.

 

Nearly every modern race car, formula one or touring cars use EFI

it is more than expected that this is thebest way.

 

To enable people who are not that familiar with making their own hardware

I wanted to show an easy way to get good results without spending a fortune to a tuner.

 

In Germany we have a guy that takes 3500GBP for a complete system and if

that is the budgeta normally trained person will come to success with a lot less

when using throttle bodies and can fit them just right out of the box.

 

If somebody is more trained (and you need not to bea mechnaical engineer)

he can follow easily the recommendations at mintylamb with similar results.

[Gt6s]

I used the PI manifolds where a set in GB was just under 100GBP complete.

 

My PI bodies came with two PI engines for £ 45.

[Triumph-V8]

Like paradise....... :-)

In Germany parts for Triumph got expensive.

They tried to rip me with a naked head no valves for 250Euros.

[James]

How nice to be mentioned

[Nick Jones]

Ah yes, that Minty Lamb fella - lots of useful stuff on his website

 

Are you back on the Island......

 

Nick