Acura RSX, ILX and Honda EP3 Forum banner
1 - 20 of 65 Posts

·
Internet Crack Dealer
Joined
·
34,438 Posts
Discussion Starter · #1 ·
Most Horsepower from a stock fuel pump

There has been a question raised about the max horsepower of a stock full pump, can anyone answer this?
 

·
Registered
Joined
·
226 Posts
Nope - a question about the stock fuel pump fitted to the Lotus elise.

Its a quick and easy answer - it will flow enough for 220-230 BHP (simiulated crank figure) (some a tiny bit more, some a bit less and it drops with age).

Basically not enough to produce a claimed 280BHP at the crank.

SEOT.
 

·
Internet Crack Dealer
Joined
·
34,438 Posts
Discussion Starter · #3 ·
Whos why of testing, the one from the UK? From my understanding you guys use a very different way of testing, thus our numbers will always be different. Fill me in on your way because I actually care to learn.

If the testing is different then how can you really compare because you will always develop different numbers? If I take a car that has 170 here and 155 there, doesn't that mean that if I gain 15hp in both places the claims are real just different starting and ending numbers? If your system is showing to be 15hp off of ours (just making up a number for arguements sake) then how can you say the gains are not the same? It is a flawed arguement that you can never win. The gains will be the same, it just depends on the starting point and how different the testing is.
 

·
ILikeFastCars&FastWomen
Joined
·
733 Posts
really said:
Nope - a question about the stock fuel pump fitted to the Lotus elise.

Its a quick and easy answer - it will flow enough for 220-230 BHP (simiulated crank figure) (some a tiny bit more, some a bit less and it drops with age).

Basically not enough to produce a claimed 280BHP at the crank.

SEOT.
How much energy does gasoline contain per unit and how much variance in efficiency is there between different gasoline engines using a given amount of fuel to make power? What I'm saying is that there is a difference in how much power is made with a given amount of fuel and it can depend on how efficiently the engine burns the mixture and turns that energy into rotational torque. Not all engines will make the same amount of power with a given amount of fuel.
 

·
Registered
Joined
·
586 Posts
At what pressure and BSFC? You can't simply say it will flow enough for a certain amount of hp without qualifying that statement. I'll bet you can see where this is going too, can't you? ;)

SC
 

·
Registered
Joined
·
230 Posts
OK I've been following the 'other' thread from the get go. I have refrained from posting up to this point. But since fuel issues are being discussed I suppose I have something to *contribute*.

Background - I have a PR conversion i.e. JDM K20A in back of my Elise.
The Lotus fuel pump is regulated to deliver 3 bar to a 118hp Rover engine. The Honda requires 3.5 bar so the ultimate fuel rate of the pump will go down as a result of increasing the fuel rail pressure.

Contribution
My car is fitted with a fuel pressure gauge. The gauge shows that under simulated load - (car jacked up pressure on the brakes), my fuel pressure dropped from 3.5 bar at idle to 2.5 bar under load. Adjustable regulator was fitted and under load fuel pressure increased to 2.8 bar. The manual specifies a minimum pressure of 3.3 bar. I have now had to fit a swirl pot and secondary fuel pump in order to maintain fuel pressure. The car has been dynoed (prior to fitting the secondary pump) and achieved 223hp at the crank - this is in line with the 204hp at the hubs seen on another PR conversion.

This would seem to back up claims that the Lotus fuel pump is not capable of flowing sufficient fuel to achieve 247hp at the hubs. One could argue my pump is getting on in years (it's six years old) - and maybe it's getting a little tired. But the figure of 220 - 230hp max. has been bandied around for many years - certainly well before the Honda conversion ever saw the light of day.

Since fitting the new fuel pump - I have not noticed any meaningful improvement in performance that would lead me to believe that fuel pressure was holding back the car to the tune of 50hp.

I've attached a photo showing the new layout, just in case there are any doubters. Pressure gauge in the foreground. In addition I have also attached a graph showing a number of dyno runs for various engine combo's in the Elise. My car is the JDM Honda run. All hp expressed is at the crank and all runs were performed on the same Dyno (Emerald) - Honda Euro is a European Honda K20A2 as found in the Civic Type R (the Elise it's in has a standard ECU and relatively simple Header).

I for one would give my right nut to get the same power figures as produced in the US and there is no reason why the car shouldn't be producing identical results (give or take a few hp) esp. given the fact I have removed the only potential bottleneck in my car the fuel system. I am the only Honda Elise owner as far as I know that has fitted a secondary pump - including PR and the only one to have actually measured the fuel rail pressure.





 

·
Registered
Joined
·
586 Posts
jwilson,

Your information is appreciated, but we're missing the ultimate determinant - fuel pump flow rates at various pressures. You can run the Honda engine at 2.5 bar or 3.5 bar - the injectors have plenty of duty cycle left in either case.

Furthermore, did your kit include the Hondata reflash? Since you did not rev the engine to 8800+ rpm, I'll assume it did not for now. The Hondata reflash runs the engine much leaner than stock. Approximately 10% leaner at peak power which results in a much lower BSFC. That's the beauty of the Hondata reflash - it uses less fuel to make more power (at least on a relatively stock engine ;))

This is not to say that the Lotus fuel pump didn't limit your particular car - but it doesn't mean that it _has_ to be the limiting factor with the right tuning combination.

Oh, and will someone please post unfettered dyno results from Emerald instead of these 'expletive' estimated crank numbers? And do let us know if you measured coolant and intake temperatures (and how about gear ratio, ramp time, etc.)

SC
 

·
Registered
Joined
·
230 Posts
Thanks for the reply Shawn. A little more info, yes I have a Hondata ECU.

My reason for adding an additional fuel pump was one of safety. To me it was a small investment to make to be sure I wasn't going to trash the engine. The car gets *alot* of track time at constant high rpm.

I presumed the injector's were coping somewhat since the engine hadn't gone pop in the previous 7 months.

Regarding your comment on fuel rail pressures. I should have made myself clearer. Although the injectors may have sufficient headroom you have to wonder about a pump that cannot maintain it's regulated pressure i.e if the pump was setup to deliver fuel at 2.5 bar and it did so from idle to max load then fine. However you have to worry when a pump is setup to deliver 3.5 bar from idle to max load but is unable to do so. Am I right in thinking that the K20A has a wideband primary O2 sensor that monitors and corrects the A/F ratio even in open loop conditions. The reason I ask is if the ECU believes there is n fuel pressure and fuels accordingly isn't there a risk of the engine running too lean if the pressure drops off as the load increases or does the ECU have fuel pressure sensor that is able to compensate?

As far as Emerald goes, I know you have a problem with simulated crank figures - however they have an excellent track record in that stock Elise’s generally always produce results with a couple of hp of manufacturers quoted numbers. When I did my run I was with another Elise with a Rover rated at 156hp he produced a number with 2 or 3 hp of this figure. I know it's not perfect *but* it's unlikely to be over 50hp out. When you add my number to the number achieved by another PR car on a Dynapack with Doug present (204hp at the hubs) and if you allow for driveline losses for my car the two numbers are very close to one another. Agreed?

I have the raw data from Dynapack happy to publish it if you want.

p.s. You can call me Ian :D
 

·
Registered
Joined
·
226 Posts
ChurchAutoTest said:
jwilson,
Furthermore, did your kit include the Hondata reflash? Since you did not rev the engine to 8800+ rpm, I'll assume it did not for now.
SC
Thats OK as the graph in question (The PTR one avaliable here: http://www.prototyperacing.com/dynok20b.htm) also tops out at 8500. I believe Ian has had lots of hondata input to his ECU by now but thats a bit beside the point as its a question about how could the graph be accurate (with the 8500 rpm top out as you mention and perhaps without the benefit of later flashes) when its a moot point as to the lotus sourced fuel pumps ability to flow enough fuel.
 

·
Registered
Joined
·
226 Posts
Its also worth mentioning that the simulated crank figure shows an uncanny ability to get to similar figures to proper engine dynos - at least on the k-series plots I have seen.

They are never going to be the same as a dyno cell is always going to have different temps and is unlikely to be running the same inputs and outputs, but its food for thought.
 

·
Banned
Joined
·
1,123 Posts
ChurchAutoTest said:
Your information is appreciated, but we're missing the ultimate determinant - fuel pump flow rates at various pressures. You can run the Honda engine at 2.5 bar or 3.5 bar - the injectors have plenty of duty cycle left in either case.
Point taken, however, the engine (as in Hondata flash) I assume was done with a relatively stable fuel rail pressure?

The pump used in the Elise/Exige is actually a AC one made in the US (used in a Chevy Cavalier?)

at 3.5 bar it's flow rate is almost zero, at 3 bar it can make ~1L/min when new.

Clearly how many BHP you can get from 1L/Min is going to depend on the engine's efficenct etc, but what has been seen so far is that ~220 is that point (at 3Bar), this has been found on MANY engines from highly tuned NA to forced induction.

Now, how this realtes to the other (now closed) thread is this:

if PTR's Exige (with it's stock pump) can make 247.7 at the flanges then even assuming that the gearbox looses are minimal, that's still way more than the 220 or so that the pump can cope with at 3Bar, let alone 3.5.

Other examples of this configuation (exhaust/induction/ECU flash) have been independantly tested here in the UK, and can not get close to replicating this figure, they come in at ~204 (at the flanges) or ~223 (simulated Flywheel).

Then, we have another engine (Derek's) that comes up with very simmilar numbers on the very same dynapack (all be it slightly higher).

now, you can see, when presented with all this, what's the logical assumption?
 

·
Registered
Joined
·
586 Posts
I don't recall anyone ever checking fuel rail pressure on Prototype's car when tested over here. That would be a question for them. Suffice it to say that there was no difficulty in getting it to run the appropriate AFR as modifications were added.

1L/min, or about 16 US gal/hr is approximately 100 lbs/fuel hour. At a BSFC of 0.5 (most FI engines run 0.55 or so at maximum boost, 0.5 is a relatively normal production engine BSFC for normally aspirated engines), such a delivery rate would be sufficient for 200 crank hp. At 0.55 BSFC, 180 hp. At 0.4 BSFC (which is what we approach with a highly tuned normally aspirated engine, 250 hp.

As you can see, a variance of even 5% in the delivery of a pump can make a significant difference in total hp capacity in this case - and we haven't even addressed the potential for higher flows at lower pressures.

What I find interesting is that you're telling me that Lotus uses a fuel pump which would ostensibly find itself near capacity on its own 180+ hp engines when run at typical production car conservative fuel specifications? Just for reference, my 108 hp CRX had a fuel pump classed to deliver 28 gph, which is far more normal for a manufacturer (always have excess capacity). Someone really needs to have a pump flowed across multiple pressures to see the delivery rate (as Hondata have done for most stock Honda pumps). If you're telling me that various sources have produced a consistent 220-230 hp maximum be it with FI or NA engines, and using that as an indicator of maximum capacity, I'd have to take issue with that. Outside of the BSFC argument, an FI engine will require higher rail pressures to achieve the same relative delivery as an NA engine at lower pressures - why? Because you must regulate the pressure differential across the injector. Delivering the same fuel as a 3 bar setup on an NA engine will require 3.5 bar rail pressure on a 0.5 bar boosted engine. Clearly then there should be no fuel flowing at all!

Perhaps the real issue here Scuffers, one which has been lost in all the recriminations going back and forth between Prototype and the "screaming for vengance" set of the Lotus crowd, is fuel pump variance. Clearly others like Stan Wan have posted that they were able to achieve significant hp gains (in the UK with their oh so accurate engine dyno replicating chassis dynos - sorry, that still a load of tripe) on K-motors with a similar modification set. So why haven't some Lotus owners been able to? Perhaps the fuel pump is the key.

It clearly _is not_ an issue with the abilities of the K20A engine (in particular). In the case of the Prototype Exige, you're talking about an engine rated to deliver 220 hp stock - a rating derived with all accessories attached. Simply removing the power steering pump alone is worth 5 hp. Optimizing (leaning the mixture, advancing the ignition and altering the cam timing) is worth at least 10 hp more. And freeing up the intake and header/exhaust provide further gains. Why is it so hard to believe in 250+ hp?

SC
 

·
Ur Post Count Means Nada!
Joined
·
4,510 Posts
We know that the K20A2 can make 240hp on a dynapack.
We know that the K20A can also make 240hp on a dynapack.
We know for a fact that they can produce those numbers with bolt-ons, cams, ecu tuning. Both engines having used their originally accessories allowing them to achieve that power. When i say accessories i mean Fuel Pump, Fuel Rail, blah blah.
If a K20A or K20A2 was swapped into a Lotus Exige, and used the Lotus Fuel Pump which was intended for use on the stock engine that came with the car then how could you expect your K20A/K20A2 to make the same power it made in a DC5/EP3/EG?

I am not trying to talk out of my ass. I am just going by what info this thread has provided as of now.
If i had a Lotus and performed a K swap, i would make sure i had ALL the accessories needed for the K to work as it did in the car it came out of.

Please if what i just posted if a bunch of crap then shut me up and i will just read for the remainder of the thread.

Regardless, i just wanted to tell Sean Church that i have learned alot from reading the info you post on these boards. I appreciate it man!
-Jalal
 

·
Registered
Joined
·
226 Posts
ChurchAutoTest said:
. So why haven't some Lotus owners been able to? Perhaps the fuel pump is the key.

SC
If the fuel pump is the key how on earth would the same model pump produce the goods back in the states? Its too much of a varience to be accounted for by a varience in abilities of the same model fuel pump.
 

·
Registered
Joined
·
226 Posts
MJ23FE said:
If a K20A or K20A2 was swapped into a Lotus Exige, and used the Lotus Fuel Pump which was intended for use on the stock engine that came with the car then how could you expect your K20A/K20A2 to make the same power it made in a DC5/EP3/EG?

I am not trying to talk out of my ass. I am just going by what info this thread has provided as of now.
If i had a Lotus and performed a K swap, i would make sure i had ALL the accessories needed for the K to work as it did in the car it came out of.

Absolutely. If you want the sort of power claimed, and possibly avaliable with the k20, then you need to make sure you provide it with the juice to do so. The PTR claim was made of the power using the stock lotus sourced fuel pump. Imagine this claim in an environment were people had been head scratching at dyno plots showing the pump giving up and trying to source alternative parts that would fit in the tank (don't think one has been found yet, most chose to run an inline system to compensate).
 

·
Banned
Joined
·
1,123 Posts
ChurchAutoTest said:
I don't recall anyone ever checking fuel rail pressure on Prototype's car when tested over here. That would be a question for them. Suffice it to say that there was no difficulty in getting it to run the appropriate AFR as modifications were added.

What I find interesting is that you're telling me that Lotus uses a fuel pump which would ostensibly find itself near capacity on its own 180+ hp engines when run at typical production car conservative fuel specifications? Just for reference, my 108 hp CRX had a fuel pump classed to deliver 28 gph, which is far more normal for a manufacturer (always have excess capacity). Someone really needs to have a pump flowed across multiple pressures to see the delivery rate (as Hondata have done for most stock Honda pumps). If you're telling me that various sources have produced a consistent 220-230 hp maximum be it with FI or NA engines, and using that as an indicator of maximum capacity, I'd have to take issue with that. Outside of the BSFC argument, an FI engine will require higher rail pressures to achieve the same relative delivery as an NA engine at lower pressures - why? Because you must regulate the pressure differential across the injector. Delivering the same fuel as a 3 bar setup on an NA engine will require 3.5 bar rail pressure on a 0.5 bar boosted engine. Clearly then there should be no fuel flowing at all!

Perhaps the real issue here Scuffers, one which has been lost in all the recriminations going back and forth between Prototype and the "screaming for vengance" set of the Lotus crowd, is fuel pump variance. Clearly others like Stan Wan have posted that they were able to achieve significant hp gains (in the UK with their oh so accurate engine dyno replicating chassis dynos - sorry, that still a load of tripe) on K-motors with a similar modification set. So why haven't some Lotus owners been able to? Perhaps the fuel pump is the key.

It clearly _is not_ an issue with the abilities of the K20A engine (in particular). In the case of the Prototype Exige, you're talking about an engine rated to deliver 220 hp stock - a rating derived with all accessories attached. Simply removing the power steering pump alone is worth 5 hp. Optimizing (leaning the mixture, advancing the ignition and altering the cam timing) is worth at least 10 hp more. And freeing up the intake and header/exhaust provide further gains. Why is it so hard to believe in 250+ hp?

SC
Quite right,

just a few points though,

the std Elise (in S1 form) came with a 118Bhp engine, thus there was no requirement for a 'better' fuel pump, highest spec S1/Exige is 177Bhp, so it was not deemed nessesary to change it (or more like they could not be bothered- this is Lotus we are talking about)

you point about Forced induction is not lost, and you are right, however, in general terms, the FI enignes have a better BSFC than NA at high outputs, which sort of explains why they seem to cop out at about the same HP.

The one thing I will take issue with you though is the idea that the 220Bhp stock rating is with any ancilaries (Power steering pump/alt/AC/etc), this is NOT how the engines are rated, they are rated with zero load from any ancillaries, (also with no induction etc.) this is the deffined standard for rating an engine (see ISO 3046-1:2002).

thus once it's in the car with all it's bit's and bobs, it's very unlikely that you will see the full 220Bhp (certainly the ones I have seen dynoed here are normally a couple of BHP short, and we don't have hydralic Power Steering pumps (EU cars have electric PS).


Now, I am not saying that it's not possible to get big numbers from the K20A, the port sizes and basic design give a shed load of scope for this, what I take issue with is that you can get these numbers with bugger all mods.

I fully accept that the std exhaust manifold is crap, too small, too short etc etc., and that running a properly proportioned one will free up not only more power but mid-range too, however, without any internal changes to the engine (speciffically being able to re-time the exhaust cam) I just can't see how you can gain 50HP at the flanges (or crank for that matter).

let's not loose sight of what we are talking about here, at these sorts of numbers, you almost at the level that the top flight factory touring car engines are putting out, and I can assure you that they are far from stock anything.

I am very interesed in what people are going to get from these engines, and I am in the middle of working though some stuff at the moment (just awaiting the install of an engine dyno cell). Be assured, when it get's mounted up I will try just about everyhting, and will document what did what, etc etc so that all can see. If I do find that without any internal mods 250+ is possible, I will be the first to say so.

Simon
 

·
Registered
Joined
·
230 Posts
Shawn, what in your opinion are the most likely causes of my car and that of Paul Golding (Calypsoelise) failing to attain the hp seen on your Dyno with the PR Exige? Given that Paul's car has been on a Dynapack and was configured identically to the PR car right down to the silencer. And in addition my car has produced similar numbers to his (as much as you dislike my numbers they do backup Paul's).

Secondly, I also have to wonder how PR with the Supercharged Exige is able to produce somewhere around 290hp at the hubs on your Dynapack while retaining the Lotus fuel pump.

To me the world is a very logical place - science gives us Law's we cannot break. I know of someone with an Elise with an Audi TT engine in it producing 260hp (crank) - he has *had* to fit an additional fuel pump and is making less hp than the Supercharged K20A Exige which utilizes only the standard Lotus pump. So in my mind I need to be able to explain this logically. Is it logical given the known limits of the pump that this pump could be good for 290hp at the hubs or well over 300 at the crank.

Answer - no it's not logical, therefore there must be another explanation.

From my point of view, I only want to know if I'm missing out on 50hp because either:

a) it never existed or
b) the Dynapack in the UK and Emerald are wrong and I already have it - but I just don't know it :laughing:

*Logically* it can only be one or the other.

Lot of stuff there but I hope you can take the time to answer it all. If I'm beating down the wrong path completely then show me the light :D
 
1 - 20 of 65 Posts
Top