Intake manifold and porting...
Discussion
Sardonicus said:
spitfire4v8 said:
sequential injection doesn't really help here does it unless you have load/rev based individual trimming to go with it .. you need to fuel for the individual cylinder filling. Not that I think it would make enough difference to make it worthwhile.
Thing is you have un-timed injection with batch fire not exactly ideal especially running cams with long duration where you need all the help you can get 
have that little squirt of fuel synced with cam duration/valve opening can only be a good thing, we are trying to keep cylinder firing/condition equal are we not? well best we can anyway 
Surely simply getting a stable air/fuel mix across all 8 pots should be objective #1 from any viewpoint? If its simply a case of trimming injectors to suit the average load/rpm (think of 8 single cylinders rather the one v8) that must tend to a much better situation than all 8 the same (even if the injectors REALLY are matched to start with!). Albeit you may find mapping cylinder by cylinder to get a sweet spot more tedious (temps or lambda per header...?).
PS so basically Joo it's people you've spoken to, so how many of those experiences were specifically about V8's with horrible inlet manifolds like the TVR/RV8? Could understand that response if they were looking at throttle bodies or a simpler 4cyl straight manifold, in which case the argument would just be that timing of the injection isn't so critical, but still doesn't really account for the affects of being able to trim the injectors to suit each pot for the variance in air fill? IYSWIM. I'm fairly convinced that this is one of the reasons the RV8 responds particularly well to forced induction BTW, as that simply overwhelms the inlet manifold differences.
Any chance to try that plenum yet BTW?
I'm not sure any of the peole I've spoken too are rv8 specific, but plenty of them see plenty of engines (dave at emerald must see the biggest range of engines of anybody?) and their experience is what I draw on.
Likewise though, how many engines have you run on the dyno and how many of those have you applied individual cylinder trimming on?
I fully appreciate, understand and agree with what you say about cylinder filling differences as a result of the contortions of the inlet manifold, but have you actually measured its effect on power? or measured it at all (flow bench?) and applied that to any modelling of what might happen if you improved things?
Until anyone has empirical evidence we are all just guessing .. educated (more or less) guessing but guessing none the less.
I'm supposing that the biggest evidence we have is that 1) people like V8D who are known rv8 tuners still sell ported inlets and don't have anything else on the market except:
2) people like JE and jenvey have a completely different inlet approach with individual runners, but which in my experience (on the dyno) aren't a massive jump in hp terms over a ported rv8 inlet.
On that basis my thinking at the moment is that the inlet manifold discrepencies isn't a big deal.
The final thing is that the engines themselves seem to be very mixture tolerant, in that you can change the mixture considerably and still make the power .. what is the flow discrepency of a high power RV8 once the inlet is on the head, port to port? Is it more than 10%? because you'd need that much to influence the mixture enough to reduce power over how you'd map it conventionally I'm guessing (based on power outputs of cars with varying mixtures, and how they react once the mixture is corrected) *ie a drop from lambda 0.85 to lambda 0.95 would drop power.
Likewise though, how many engines have you run on the dyno and how many of those have you applied individual cylinder trimming on?
I fully appreciate, understand and agree with what you say about cylinder filling differences as a result of the contortions of the inlet manifold, but have you actually measured its effect on power? or measured it at all (flow bench?) and applied that to any modelling of what might happen if you improved things?
Until anyone has empirical evidence we are all just guessing .. educated (more or less) guessing but guessing none the less.
I'm supposing that the biggest evidence we have is that 1) people like V8D who are known rv8 tuners still sell ported inlets and don't have anything else on the market except:
2) people like JE and jenvey have a completely different inlet approach with individual runners, but which in my experience (on the dyno) aren't a massive jump in hp terms over a ported rv8 inlet.
On that basis my thinking at the moment is that the inlet manifold discrepencies isn't a big deal.
The final thing is that the engines themselves seem to be very mixture tolerant, in that you can change the mixture considerably and still make the power .. what is the flow discrepency of a high power RV8 once the inlet is on the head, port to port? Is it more than 10%? because you'd need that much to influence the mixture enough to reduce power over how you'd map it conventionally I'm guessing (based on power outputs of cars with varying mixtures, and how they react once the mixture is corrected) *ie a drop from lambda 0.85 to lambda 0.95 would drop power.
Edited by spitfire4v8 on Thursday 11th December 11:39
Bobby Shaftoe said:
In my humble opinion, going from 44mm to 45mm would be a waste of time. The best bang for the buck in your case would be replacing your mild Kent H218 camshaft with a wilder grind, and increasing the compression ratio to suit.
yep that would do it 
Not as quick or cheap, but would get you very noticeable results
spitfire4v8 said:
Thing is you'd need very large injectors or very high rail pressures to get all your fuel into the chamber within the inlet open period (not even that, within the inwards gas flow direction period ..).
It's easy to do that on light throttle/low rev situations, but on full load (where inlet manifold flow rates become the factor we are discussing) you'd have to do the usual trick of setting the injector end point and increase the opening duration forwards from that, meaning the first part of the injected period will be well before the inlet valve has opened on high power cars. Most of the people I speak to who are in the aftermarket (ie not OE emissions / economy spec side of the industry) say it's not worth it. (You could run two sets of injectors (phased) near the head face I guess to shorten the injection period to something like what you need?).
Sounds a lot of work.
But that's my point It's easy to do that on light throttle/low rev situations, but on full load (where inlet manifold flow rates become the factor we are discussing) you'd have to do the usual trick of setting the injector end point and increase the opening duration forwards from that, meaning the first part of the injected period will be well before the inlet valve has opened on high power cars. Most of the people I speak to who are in the aftermarket (ie not OE emissions / economy spec side of the industry) say it's not worth it. (You could run two sets of injectors (phased) near the head face I guess to shorten the injection period to something like what you need?).
Sounds a lot of work.
its low speed drive-abilty that your after once your on the cam/sweet spot and moving quicker the fluffy response you often get low down is not such an issue, lets face it once the motor is spinning up so long as it sees air and fuel in the correct ratio its not really fussed 
ah sorry i thought we were on about mixture variance due to the differences in flow of the inlet runners ..
there may be small improvements in part throttle from full sequential .. never driven a full seq tvr to find out though. a nicely mapped aftermarket on bacth or semi seq is a nice driving machine though, not sure how much improvement youd actually feel ? Would be nice to find out some day.
there may be small improvements in part throttle from full sequential .. never driven a full seq tvr to find out though. a nicely mapped aftermarket on bacth or semi seq is a nice driving machine though, not sure how much improvement youd actually feel ? Would be nice to find out some day.
spitfire4v8 said:
ah sorry i thought we were on about mixture variance due to the differences in flow of the inlet runners ..
there may be small improvements in part throttle from full sequential .. never driven a full seq tvr to find out though. a nicely mapped aftermarket on bacth or semi seq is a nice driving machine though, not sure how much improvement youd actually feel ? Would be nice to find out some day.
Agreed there may be small improvements in part throttle from full sequential .. never driven a full seq tvr to find out though. a nicely mapped aftermarket on bacth or semi seq is a nice driving machine though, not sure how much improvement youd actually feel ? Would be nice to find out some day.
I would hope that running full sequential fuel would be an improvement but who knows? the improvements may be minimal you cant tell with the RV8 Sardonicus said:
Thing is you have un-timed injection with batch fire not exactly ideal especially running cams with long duration where you need all the help you can get have that little squirt of fuel synced with cam duration/valve opening can only be a good thing, we are trying to keep cylinder firing/condition equal are we not? well best we can anyway
Simon I'm puzzled as to what you mean by 'un-timed injection'. I'm not an automotive expert but just out of interest I've been looking at the injector fuelling/ignition timing used in the RV8/14CUX set up, and to help with that I've produced a 'very simplified' timing relationship chart: have that little squirt of fuel synced with cam duration/valve opening can only be a good thing, we are trying to keep cylinder firing/condition equal are we not? well best we can anyway 
From this chart there appears to be a well-defined symmetry and a clear timing relationship for controlling the fuel charge into the inlet manifold runners. There are four possible ECU interrupt controlled injection fuelling periods of approx 1 to 10 mSec per two crankshafts revolutions, which are timed from the ignition spark signal. There also appears to be a defined balance between odd/even injector bank and odd/even ignition, as to whether or not this bank/ignition selection has a bearing on the respective active inlet runner I'm not sure as yet.
I guess this 'old' technology is now considered as un-timed injection but, and I may well be completely wrong, 14CUX's batch injection timing is not as random as first appears?
Sardonicus said:
greed I would hope that running full sequential fuel would be an improvement but who knows? the improvements may be minimal you cant tell with the RV8
I guess it all depends on the state of tune. For most of us, fully sequential wouldn't make much of an impact, however if you're running a really wild cam and timing, then it may be of some benefit. I would hope that running full sequential fuel would be an improvement but who knows? the improvements may be minimal you cant tell with the RV8 davep said:
Sardonicus said:
Thing is you have un-timed injection with batch fire not exactly ideal especially running cams with long duration where you need all the help you can get have that little squirt of fuel synced with cam duration/valve opening can only be a good thing, we are trying to keep cylinder firing/condition equal are we not? well best we can anyway
Simon I'm puzzled as to what you mean by 'un-timed injection'. I'm not an automotive expert but just out of interest I've been looking at the injector fuelling/ignition timing used in the RV8/14CUX set up, and to help with that I've produced a 'very simplified' timing relationship chart: have that little squirt of fuel synced with cam duration/valve opening can only be a good thing, we are trying to keep cylinder firing/condition equal are we not? well best we can anyway From this chart there appears to be a well-defined symmetry and a clear timing relationship for controlling the fuel charge into the inlet manifold runners. There are four possible ECU interrupt controlled injection fuelling periods of approx 1 to 10 mSec per two crankshafts revolutions, which are timed from the ignition spark signal. There also appears to be a defined balance between odd/even injector bank and odd/even ignition, as to whether or not this bank/ignition selection has a bearing on the respective active inlet runner I'm not sure as yet.
I guess this 'old' technology is now considered as un-timed injection but, and I may well be completely wrong, 14CUX's batch injection timing is not as random as first appears?
I'm fairly sure you can answer your own questions approaching it from the engines POV rather than from your ECU's operational POV IYSWIM.
spitfire4v8 said:
The final thing is that the engines themselves seem to be very mixture tolerant, in that you can change the mixture considerably and still make the power .. what is the flow discrepancy of a high power RV8 once the inlet is on the head, port to port? Is it more than 10%?[/footnote]
I did flow test a manifold into bores, and observed one of the outers was roughly 65% of the inners... Using water down the trumpets & timing a known volume. Sounds Heath Robinson-ish but Pete Burgess ( whose academic & real world knowledge in this area is way beyond most of us ) seemed to think it was a fairly sound test method.Only obscene amounts of porting & weld around the outers managed to get to anywhere near a 10% divergence. 2 of the crossovers seem to effectively act like stop valves.
SILICONEKID345HP said:
What the f
k are your lot talking about ?
k are your lot talking about ?spend said:
I did flow test a manifold into bores, and observed one of the outers was roughly 65% of the inners... Using water down the trumpets & timing a known volume. Sounds Heath Robinson-ish but Pete Burgess ( whose academic & real world knowledge in this area is way beyond most of us ) seemed to think it was a fairly sound test method.
Only obscene amounts of porting & weld around the outers managed to get to anywhere near a 10% divergence. 2 of the crossovers seem to effectively act like stop valves.
So in effect what you are saying is; you could do almost no work at all on the 'good' runners and concentrate on the 'bad' runners, and you might achieve some sort of improved balance.Only obscene amounts of porting & weld around the outers managed to get to anywhere near a 10% divergence. 2 of the crossovers seem to effectively act like stop valves.
This then surely needs to be taken into account with how well each of the intake ports on the heads flow (pardon my memory, but they aren't balanced either as I recall, but I can't remember which was which
or how big the difference normally is).Clearly the overall start to finish flow has to be taken into full account.
I had no idea the standard intake manifold was so poor. All you mad modders take note; Throw that b
h in the bin and start over!Goaty Bill 2 said:
SILICONEKID345HP said:
What the fk are your lot talking about ?
k are your lot talking about ?spend said:
I did flow test a manifold into bores, and observed one of the outers was roughly 65% of the inners... Using water down the trumpets & timing a known volume. Sounds Heath Robinson-ish but Pete Burgess ( whose academic & real world knowledge in this area is way beyond most of us ) seemed to think it was a fairly sound test method.
Only obscene amounts of porting & weld around the outers managed to get to anywhere near a 10% divergence. 2 of the crossovers seem to effectively act like stop valves.
So in effect what you are saying is; you could do almost no work at all on the 'good' runners and concentrate on the 'bad' runners, and you might achieve some sort of improved balance.Only obscene amounts of porting & weld around the outers managed to get to anywhere near a 10% divergence. 2 of the crossovers seem to effectively act like stop valves.
This then surely needs to be taken into account with how well each of the intake ports on the heads flow (pardon my memory, but they aren't balanced either as I recall, but I can't remember which was which
or how big the difference normally is).Clearly the overall start to finish flow has to be taken into full account.
I had no idea the standard intake manifold was so poor. All you mad modders take note; Throw that b
h in the bin and start over!The head inlet ports are pretty well matched, if you take the time to inspect manifolds and heads the different paths the air has to take to the valve is massive in the manifold & miniscule in the heads (even just with a cursory glance???)
Much of these problems are due to the crossplane crank and compactness of the RV8, & of course the plenum + height restriction... when you look at how best to 'partition' inlet & exhaust combinations folks always get drawn into the flat plane crank scenario.. it all gets a bit much. ....but back on topic what we are discussing here is trying to make the best silk purse we can out of the original sows ear IYSWIM.
spend said:
Personally I think thats where many folks have got the wrong end of the stick... many will say just forget the plenum & use throttle bodies going to higher tunes ~ but this also yields equal runners at the same time, however common appreciation doesn't seem to put 2 & 2 together ~ they just visualize it as bigger air at the butterfly when the probably more significant impact is the change in feed to the port (ie the manifold gets replaced with equal straight runners at the same time).
The head inlet ports are pretty well matched, if you take the time to inspect manifolds and heads the different paths the air has to take to the valve is massive in the manifold & miniscule in the heads (even just with a cursory glance???)
Much of these problems are due to the crossplane crank and compactness of the RV8, & of course the plenum + height restriction... when you look at how best to 'partition' inlet & exhaust combinations folks always get drawn into the flat plane crank scenario.. it all gets a bit much. ....but back on topic what we are discussing here is trying to make the best silk purse we can out of the original sows ear IYSWIM.
Thats just it making the best of what we got The head inlet ports are pretty well matched, if you take the time to inspect manifolds and heads the different paths the air has to take to the valve is massive in the manifold & miniscule in the heads (even just with a cursory glance???)
Much of these problems are due to the crossplane crank and compactness of the RV8, & of course the plenum + height restriction... when you look at how best to 'partition' inlet & exhaust combinations folks always get drawn into the flat plane crank scenario.. it all gets a bit much. ....but back on topic what we are discussing here is trying to make the best silk purse we can out of the original sows ear IYSWIM.
lots of engines fall down on the intake manifold front the Ford X flow's stock intake was also awful but still didn't stop mine making 128bhp out of a 1660 high CR motor on 1 downdraft Weber a decent side-draft manifold set up (carbs not TB's) on those created massive gains more so because of the even A/F distribution, you get the jist without going off topic 
Can you not equalise the length of the runners with the trumpets? Why are the original trumpets in the 4:4 configuration? If the runners/ trumpets aren't an equal length in the stock engine and if air volume in the plenum is an issue with the RV8, why have a trumpet base at all? Would it not be better to have a larger plenum fitted directly to the intake manifold or is it about gradually changing the direction of the air once it comes into the plenum to get it around the bend in the intake?
The other thing, with respect to EFI vs carbs, is that depending on the ECU, you can have more control over the fuelling and spark, however, there is a common view that carbs can yeild more power, which would suggest that it is more about air volume and the and less about precision timing of the fuel/ spark.
What about using one of these to add bike carbs? Would this likely cause emmissions issues?
The other thing, with respect to EFI vs carbs, is that depending on the ECU, you can have more control over the fuelling and spark, however, there is a common view that carbs can yeild more power, which would suggest that it is more about air volume and the and less about precision timing of the fuel/ spark.
What about using one of these to add bike carbs? Would this likely cause emmissions issues?
Edited by Chuffmeister on Saturday 13th December 15:53
You lot are the bks.
I only understand all this because I read an induction and porting book for Ford X flows when I was about 15 years of age!
I havnt been on here much the last few months, but what a brill topic.
I live and learn. Mega.
I've been thinking of buying a motorbike but maybe I could just put carbs on the Chim? !!! ?
Best of both worlds.
As usual your all an inspiration . Alun
ks. I only understand all this because I read an induction and porting book for Ford X flows when I was about 15 years of age!
I havnt been on here much the last few months, but what a brill topic.
I live and learn. Mega.
I've been thinking of buying a motorbike but maybe I could just put carbs on the Chim? !!! ?
Best of both worlds.
As usual your all an inspiration . Alun
Bobby Shaftoe said:
In my humble opinion, going from 44mm to 45mm would be a waste of time. The best bang for the buck in your case would be replacing your mild Kent H218 camshaft with a wilder grind, and increasing the compression ratio to suit.
You have a point. V8D contacted me to say that they haven't bench tested an intake manifold by itself, only the trumpet base/ manifold combo. The ECU and cam are on my list. I just wanted know the quantifiable gains from 44-45mm intakeGassing Station | Chimaera | Top of Page | What's New | My Stuff