Fuel, pumps, and injectors

[shnazzle]

Patrick's been reading again... never ends well.

Let's assume I run at 43.5psi (3bar), which I believe is the stock pressure for the mr2 as well right?
At 43.5psi using 255cc injectors, it means the max achievable HP is about 180hp, naturally aspirated.
Stock, 138hp means a pretty hefty margin of 42cc per injector.

So running boost effectively reduces the max fuel pressure your injectors put out.
43.5psi means 33.5psi at 10psi boost.

So, by the same calculations, this requires 478cc injectors (at 90% duty cycle, 538cc at 80%).

On a tune for a turbo, you'd have to pull back fueling the less boost there is. Very difficult to tune.

Essentially at 10psi, you'd need to push 53.5psi through the injectors to maintain 43.5psi. At 9psi, 52.5, etc etc.
How do we do this, a fuel pressure regulator (FPR) and a pump that can push out at least 53.5psi. What can the stock pump do max?

This way you're always pushing out a standard 43.5psi, which means tuning is a hell of a lot easier as you're not running lean at high boost.


On that note.... the SP240 has 440cc injectors (ours does anyway). No FPR, stock pump. This means that to reach 240hp, at 10psi (33.5psi fuel pressure because we've lost 10 of the stock 43.5 in boost), the car needs 478cc/min per injector.

How does this work? Surely then the SP240 can be only max 220hp at the crank with the 440cc injectors. It can't fuel fuel to reach 240! How does it get around this?

[Ardent]

I imagine your sub question is, does it?

Long way down the line now, but has anyone recently (or historically) measured the output of a 240.
We all know and acknowledge Scotts 240 was quick. That is not in question.

But, we are talking actual numbers and calculations here.
What "real" data do we have?

[1979scotte]

Sp240 kits don't put out 240bhp.
My Dyno sheet said 235.
I would take that with a pinch of salt.
Rolling road figures of fly wheel power are educated guesses not 100% accurate.
I would imagine they over estimate drive train losses for our cars.
Everything is quite light weight. It's basically a front wheel drive setup at the back of the car.
Very efficient.
I am quite keen on how the Aussies and the Yanks do it with WHP.
Remember manufacturers use an ENGINE dynamometer us guys use rolling roads.

[Ardent]

On a quick flick round the interweb.
Was 240 a ref to torque rather than BHP
Ironically one review had the torque and bhp at 240. Coincedence? I don't know.

[1979scotte]

On a quick flick round the interweb.
Was 240 a ref to torque rather than BHP
Ironically one review had the torque and bhp at 240. Coincedence? I don't know.

Didn't make 240 ft-lb of torque either.
Possibly 220 but I honestly don't remember.

[shnazzle]

Even for 235,the injectors are really being pushed to the limit. Any fouling if the nozzles and you're left with a deficit which could lead to very dangerous lean conditions.

That's at 10psi. If you see spikes to 11/12psi (we definitely do) , game over.

It almost seems like a return rail and FPR really is a smart thing to put on a car running this hot on power.

[ChrisGB]

Remember, the turbo on the SP240 kit would not be able to maintain 10psi at higher revs, it actually does off to around 8psi at peak power.

[1979scotte]

Remember, the turbo on the SP240 kit would not be able to maintain 10psi at higher revs, it actually does off to around 8psi at peak power.

This is 100% correct.
Mine would taper to 7 or 8 psi at high revs.
Hopefully that will allow your figures to add up

[lamcote]

Also, if you assume the injectors get up to around 95% duty cycle it all stacks up. However I take your point about the risk this involves, it's certainly very much on the edge.

[SteveJ]

For clarification when you say FPR I assume you mean RRFPR - Rising Rate Fuel Pressure Regulator with a boost reference fed to it. The stock fuel pump has an integral FPR set at 43ish PSI

The base PE kit had a modified stock pump that fed manifold pressure to the reference port on the regulator effectively making it  poor-mans RRFPR - hence being able to run stock injectors and no piggyback ECU

[shnazzle]

Yes indeed, 1:1 rising rate.

I'm not a fan of the idea of boosting the pressure. Although some cars obviously run 4 Bar.

I'd be happy to be 100% sure that the system was delivering 3 bar consistently.

So I'm assuming then the pump needs to be capable of pushing at least 53.5psi, to cover for the stock pressure and any backpressure from boost.

What's the stock pump rated to? A walbro 255lph is an obvious choice but 255lph seems FAR past any aspirations I would ever have for a 2

[shnazzle]

Can anybody also confirm the assumption that if a car has was mapped with a RRFPR, and then the RRFPR was removed... That the map would then no longer be suitable for the car?

Reason being that the fueling would have to be adjusted up per level of boost, as the effective injector pressure would go down as boost went up.

The expected effects might be a very rich condition off boost and low boost and possibly lean at full boost?

[lamcote]

The map would certainly not be suitable any more.

I would expect AFR off boost still to be OK (because the fuel pressure isn't increased by the RRFPR with no boost, so pressure, and therefore flow, would be the same in both cases) but it would go progressively more lean as boost increases.

There is reference on Spyderchat to getting 300whp using the standard fuel pump. It doesn't matter whether that is achieved from big injectors, higher pressure or a mixture of both (the effect is the same), if it can pump enough fuel for that much power you're fine however you achieve it.

[Micha]

Sorry for my english but if i can help...

Injector will flow some fuel if there is some lower pressure atthe nozzle in comparison with fuel rail pressure. If the manifold pressure is higher than the furl rail pressure there is no injector which will flow any fuel.
Than is why turbo engines have fuel pressure regulator and thus constant differential pressure between the manifold and the fuel rail. That is not the case of the Toyota MRS because we have pump integrated pressure regulator which do not account for the manifold pressure.

From physics and Bernoulli we can say that the flow is evolving as square root of ratio of pressures at the nozle.
So the only whay to make it work is to keep the fuel pressure superior to the manifold pressure and choose the injectors which will not be opened 100% of the time of an engine cycle and calibrate the fuel table in order to reach desired AFR or lambda.

In order to choose the injectors I use a common method which is to say :
- Majorated brake specific fuel consumption at maximal power (from experience of what I have seen on the engine bench for PFI engines) : 320g/kWh but lets see it large 320 +20 = 340 g/kWh
- Target power : 240 [bhp] / 1.36 = 176 kW
--> 60kg/h
- 4 injectors
--> 15kg/h/cyl
- density : ~720 kg/m^3
--> 20.8x1e-3 m^3/h or 347cc/min/cyl for 100% duty
- 80% max duty
--> 434 cc/min/cyl at specified differential pressure (fuel / environment)

On the other hand considering the injector duty cycle:
At 6800rev/min -->3400 cyc/min --> 2.941x1e-4 min/cycle
If we consider 80% duty cycle as the maximum  target --> 2.353x1e-4 min/cycle maxi avalable time to inject
We need 15kg/h/cylinder --> 0.25kg/min/cylinder which is  7.353x1e-5 kg/cylilnder/cycle at 6800rev/min or 1021.2x1e-4cc/cylinder/cycle

Now we can gather those two information in order to get injector specification : 1021.2x1e-4[cc/cylinder/cycle] / 2.353x1e-4 [min/cycle] = 434 cc/min.

The pressure influence... Injectors are given for a specified pressure and are delivered with measured fuel flow in order to calculate dead time and linearize the flow = f(time).
If you know what is your targt manifold pressure to reach desired BMEP or torque (which you know as you have alleardy used those information in order to choose your compressor); and you don,t have fuel pressure regulator, then you can multiply the injector rating found previosly with :
k = ((fuel rail pressure/atmo pressure) / (manifold pressure at power point/atmo pressure))^(-0.5)

Others can have other methods, and as far as we know what are the assumptions we have made it is OK  s:D s:D

[shnazzle]

So a few more numbers confirms there that 440cc is pushing the limits a fair bit. But, again that's assuming 240hp. But we now know that the SP240 doesn't run 240 and also it tails off boost at full load high rpm. So, that's how it is kept within spec of the injectors I guess.

So, with the addition of a RRFPR and associated mapping, you're looking at a more comfortable/safe 240hp capability. With regards to fueling that is. Torque effects on the gearbox and engine internals aside

[Micha]

I agree with shnazzle.

Where I can mistake is with the 1ZZ efficiency estimation at the power point.
My assumption of 340g/kWh is made of base efficiency at 280g/kWh (lambda 1) and 0.82 lambda (enriched) or 12 : 1 AFR.
I have never seen 1ZZ engine on a bench, and 320g/kWh is a common BSFC for turbo engines. Still, in the SAE article about 1ZZ development we can see 310 g/kWh.

However, low pressure 1ZZ is a long stroke, 10.5 : 1 compression ration engine which has good efficiency. If you use bigger turbine with low induces engine backpressure it could be better at the power point than what I assume.

Nevertheless, when I calibrated my engine which is equiped with TD0616G that I bought from loadswine, I kept his injectors also which are 525 cc/min because of the previous message calculation. And if I remeber well (I calibrated couple of other turbo 1ZZ and I can missmatch the engines), I had something just between 70 and 80% duty cycle at 6900rpm (0.5 to 0.6 bar relative boost) with 11.5 AFR. I have to check but I belive that those numbers are from my engine.

[shnazzle]

Oh wow! Granted it's a td06 (which looks like a very good fit for the mr2 as it seems) buy 525cc running at 80% makes me think even more that 470cc injectors should be a minimum for 240hp.

Was the td06 running on a rising rate fpr?

[Micha]

It is a very special turbo from Greddy because it runs relatively slowly in comparison with equivalent turbos. Turbine wheel is good from inertia point of view because without backplate. It is a kind of mixt-flow wheel with radial turbine housing. A bit special...

I am using it with OEM fuel system as I run 7 to 8.5 PSI of boost. However, my fuel table is not following volumetric efficiency shape at higher speeds to compensate decreasing differential pressure at the nozzle.
Next engine I am working on will have modified fuel system with rail regulator and return line because I would like to go up to 14PSI.

The problem with OEM fuel system is that :
- it is boost limited
- needs relatively larger injectors to compensate and thus will probably work roughly at idle and low load operation
- increasing the fuel pump pressure is half way solution

I have to confirm you that 70/80% duty was on my engine with 525cc/min. I don't look often my PFC as I don't have to. Just driving every day   s:D s:D   Near 50 000 miles since I converted it.

loadswine must be happy to know his parts are still alive and well pampered   s:D s:D  Nigel has very high knowledge on 1ZZ engine and on the overall car.

[m1tch]

Would you not just run a returnless system and a RRFPR so that the mapping will be consistent?

[shnazzle]

That's how I see it. There's absolutely nothing wrong with just going for big injectors to compensate. But why go through the hassle of tuning fueling at different rates per boost when you can just tune it in a scaled fashion. I'm sure that's part of the reason why Lee has been so successful tuning his own

[lamcote]

Yes, I agree in theory. The only thing is, I keep reading that RRFPRs can be a bit temperamental (and expensive), but I have no personal experience.

Also, unless you can set the rising fuel pressure to match the increased demand for fuel EXACTLY (and I'm not sure how easy it is to do that), you will still be tuning fuel rates per boost anyway, just by less, so it might not be worth the added cost?? In practise it might be better just to increase the base pressure (to avoid the need to use very big injectors which lead to poor idle) and tune for boost in the fuel map? Not sure, I'm just speculating.

[m1tch]

Yes, I agree in theory. The only thing is, I keep reading that RRFPRs can be a bit temperamental (and expensive), but I have no personal experience.

Also, unless you can set the rising fuel pressure to match the increased demand for fuel EXACTLY (and I'm not sure how easy it is to do that), you will still be tuning fuel rates per boost anyway, just by less, so it might not be worth the added cost?? In practise it might be better just to increase the base pressure (to avoid the need to use very big injectors which lead to poor idle) and tune for boost in the fuel map? Not sure, I'm just speculating.

An Aeromotive RRFPR is probably a few hundred, better to go with a decent RR FPR vs the cost of a new engine.

The new Injector dynamics 1000cc injectors (or larger) still give good idle - injector technology has come a fair way since I started to get into cars!

[shnazzle]

The whole purpose of the RRFPR is that you don't have to map to boost. Granted, as applies to everything, pay peanuts get monkeys. The aeromotive or sytec ones seem to be very well regarded.
You set initial stock pressure at idle (43.5psi?) without the manifold pressure reference attached and then attach the reference. The fuel pressure will be stable, taking one very important variable out of the equation when deciding on fueling. Basically all you need to worry about is pulse.

[lamcote]

That still leaves the issue of fuel demand v rising rate. If the fuel pressure increases at 1psi per 1psi of boost, does this mean that the injector will automatically be flowing the exact amount of fuel required without any tuning?

If not, I don't see it is improving anything, other than ensuring fuel pressure doesn't fall too low at high boost, which is also achieved by simply increasing base pressure.

[lamcote]

But if you are tuning the pulse anyway, is it any more difficult to adjust it to allow for fuel pressure?  I genuinely don't know.

I thought that when you tuned the map you just altered the fuel pulse in any given cell to get the AFR you wanted. Does it matter whether any change arises due to fuel pressure variance, air flow variance, or both?

Maybe a fixed fuel pressure helps more with DIY tuning rather than on a dyno?