Some interesting SP6 induction trends
Discussion
OK the next project at the workshop is trying to make some gains on the midrange of the SP6 engined cars, something they are very poor at as standard.
What prompted this was the performance advantage of the 4.5 AJP in the race cerbera of Mike Saunders .. I mapped his car so knew exactly what our T350 race car was up against. At peak we were about 25hp shy (385 v 410 or thereabouts) but come down to around 5500rpm and we were 100hp down to the AJP8 .. (260 v 360 hp)
Anyway long story short I'm now looking at ways of making the SP6 more grunty in the midrange , preferably without sacrificing any top end. I'm approaching this in a suck-it-and-see manner changing the induction on a car on the dyno and measuring what happens. The car in question is a strong 3.6 T350, so it's revvy and because it's the smaller 3.6 not 4.0 it has even less midrange, so an ideal candidate to try and show gains in the midrange without losing out up top.
So what's happened so far then?
Well what i have at the moment is some hp figures for a car nominally making around 340hp, so this is representative changes in hp at around this level of peak airflow .. I will have to verify the results on a high power tuscan S or sagaris at some point but for now this is what I'm working with ..
what it shows right this moment is the standard airbox (at these hp levels) is actually very very good! In fact of the different configurations trialled so far the boggo standard airbox is so far the highest peak hp producer, which isn't what I expected at all, but keep going back to it and it keeps giving the same number : low 340hp area, say 342 as an average.
What is even more surprising is that you don't have to lengthen the inlet tract very far before hp plummets, in fact lengthening the inlet by only 3 inches reduced peak hp by over 10hp immediately and the rate of drop off after peak was like falling off a cliff.
My initial intention was to try longer inlet lengths and hope to gain some midrange without losing top end, I now found that i was right on the egde of a workable harmonic and going longer harmed the top end dramatically, the midrange gains were in the order of zero to 10hp dependant on point in rev range so not dramatic. By fluke or good judgement TVR have got an inlet length that works very well as standard and certainly the last thing it needs is to go longer ..
or does it ...?
Going further with the research I made up some even longer tubes to see if could hook onto the start of another harmonic .. the result was gains through the whole rev range up to 5500rpm then nothing, by far the worst top end so far at some 25hp down on the standard airbox. Peak hp also moved down the rev range to 6500rpm so this was not a good length for a revvy 3.6 at all at the top end.
Feeling I was maybe missing a trick at the other end of the induction lengths I made a few runs with no airbox at all .. the obvious thing happened of course in that all midrange disappeared, this was a dreadful run, however right at the top end the graph kept going, only the rev limiter stopping play at 7750 where peak had held on at 335hp and still climbing. From this i think a length of about 5 inches upstream of the throttle bodies will give the best peak figure and i will be testing that again tomorrow to see.
So where does this leave me right now:
well my dream of just having a SP6 induction mod which gives midrange gains whilst leaving the top end untouched (or best case even slightly improved) is out of the window What i need is an upstream length of maybe 12-14 inches for best midrange improvements, but an upstream length of about 5 inches for best top end ..
soo .....
Variable induction here we come
I've got a few more days of playing around, and trying to get my head around a reliable workable way of making the variable inlet tracts work so might not have any proper data from this before next week, so keep tuning in to see how it gets on.
hope this has been of some interest
Joo
What prompted this was the performance advantage of the 4.5 AJP in the race cerbera of Mike Saunders .. I mapped his car so knew exactly what our T350 race car was up against. At peak we were about 25hp shy (385 v 410 or thereabouts) but come down to around 5500rpm and we were 100hp down to the AJP8 .. (260 v 360 hp)
Anyway long story short I'm now looking at ways of making the SP6 more grunty in the midrange , preferably without sacrificing any top end. I'm approaching this in a suck-it-and-see manner changing the induction on a car on the dyno and measuring what happens. The car in question is a strong 3.6 T350, so it's revvy and because it's the smaller 3.6 not 4.0 it has even less midrange, so an ideal candidate to try and show gains in the midrange without losing out up top.
So what's happened so far then?
Well what i have at the moment is some hp figures for a car nominally making around 340hp, so this is representative changes in hp at around this level of peak airflow .. I will have to verify the results on a high power tuscan S or sagaris at some point but for now this is what I'm working with ..
what it shows right this moment is the standard airbox (at these hp levels) is actually very very good! In fact of the different configurations trialled so far the boggo standard airbox is so far the highest peak hp producer, which isn't what I expected at all, but keep going back to it and it keeps giving the same number : low 340hp area, say 342 as an average.
What is even more surprising is that you don't have to lengthen the inlet tract very far before hp plummets, in fact lengthening the inlet by only 3 inches reduced peak hp by over 10hp immediately and the rate of drop off after peak was like falling off a cliff.
My initial intention was to try longer inlet lengths and hope to gain some midrange without losing top end, I now found that i was right on the egde of a workable harmonic and going longer harmed the top end dramatically, the midrange gains were in the order of zero to 10hp dependant on point in rev range so not dramatic. By fluke or good judgement TVR have got an inlet length that works very well as standard and certainly the last thing it needs is to go longer ..
or does it ...?
Going further with the research I made up some even longer tubes to see if could hook onto the start of another harmonic .. the result was gains through the whole rev range up to 5500rpm then nothing, by far the worst top end so far at some 25hp down on the standard airbox. Peak hp also moved down the rev range to 6500rpm so this was not a good length for a revvy 3.6 at all at the top end.
Feeling I was maybe missing a trick at the other end of the induction lengths I made a few runs with no airbox at all .. the obvious thing happened of course in that all midrange disappeared, this was a dreadful run, however right at the top end the graph kept going, only the rev limiter stopping play at 7750 where peak had held on at 335hp and still climbing. From this i think a length of about 5 inches upstream of the throttle bodies will give the best peak figure and i will be testing that again tomorrow to see.
So where does this leave me right now:
well my dream of just having a SP6 induction mod which gives midrange gains whilst leaving the top end untouched (or best case even slightly improved) is out of the window
What i need is an upstream length of maybe 12-14 inches for best midrange improvements, but an upstream length of about 5 inches for best top end ..soo .....
Variable induction here we come
I've got a few more days of playing around, and trying to get my head around a reliable workable way of making the variable inlet tracts work so might not have any proper data from this before next week, so keep tuning in to see how it gets on.
hope this has been of some interest
Joo
very interesting reading, whilst not a TVR or speed six owner, this is great R&D work, interesting they were very close on runner lengths off the bat. here is something that could be of interest to you, an SAE paper on the mazda 787B engine which used a telescopic intake manifold to make excellent mid-range and peak power.
http://protonet.org/doc/Mazda_R26B_A4.pdf
http://protonet.org/doc/Mazda_R26B_A4.pdf
I've wondered about a sliding trumpet like the mazda, the only problem i have is that of control system, and cost! I suspect I'm going to go down a mechanical flap valve actuation to switch between to runner lengths, indeed that is what i'm sketching up right now... something solonoid / air valve enabled .. I think there might be a ducati bike that moves it's trumpets too .. I had thought about moving 6 trumpets away from and towards the shorter length to make it longer, but the logistics of shifting 6 pipes of about 9 inches in length isn't really on !
At the moment I have two pipes side by side, one short and one long, with a simple flap valve separating the two .. I'm imagining the hp curve will be "vtec just cut in, yo!" in appearance as the flap does it's stuff but a seamless stepless inlet variation like that Mazda is a bit beyond the control systems we have .. nice article though, cheers for that
At the moment I have two pipes side by side, one short and one long, with a simple flap valve separating the two .. I'm imagining the hp curve will be "vtec just cut in, yo!" in appearance as the flap does it's stuff but a seamless stepless inlet variation like that Mazda is a bit beyond the control systems we have .. nice article though, cheers for that
I think you are right that moving direct from the long to short tracts through the actuator will create more of a step in the curves unlike a seamless move, however as you said with control systems available a seamless system could be pretty hard to achieve. will you be tying the actuator into the ECU as an rpm actuated flap or as a standalone system to start with? keep us up to date with this, im doing very similar things with intake and exhaust lengths but on a simulation level using ricardo engineering software on honda CBR600 engine at the moment so it is always interesting to hear real world application of theoretical ideas.
Good stuff Joolz, I take it you've read up on similar experiments by the Jap boys in with IL6's in Skylines et al. One thing the tapered plenum does seem to do is give a lot of air to No6 which can make it run a lot hotter due to it being leaner, check each primary temp with a IR reader. I think I've posted before some of the ECU's they run allow them to trim each injector to balance this out. Best you can achieve as standard is to flow check the injectors and put the highest flowing one in No6. Another aspect worth experimenting with is a deflector to try getting more air into the early cylinders. You've played with pulse plates before is there room to fit one within the air box to give the longer pulse reflections and boost mid range that way. I assume your early experiments proved teh air box volume is spot on, or is it as this will effect where it delivers best mid or high revs 
. I suppose you've covered or thought of all these, best of luck and keep us posted, there may be a tweak that can be transferred into the V8 out of it 
.
. I suppose you've covered or thought of all these, best of luck and keep us posted, there may be a tweak that can be transferred into the V8 out of it .What i'm limited by is enabling any spare outputs in the TVR versions of the MBE ecu, if i can't i'm going to have to have a speed-signal-reading PIC controller which is no biggy really.
Harry .. i was supposed to have already tried this trialling on an ajp8 but the car i was scheduled to use developed a problem which has put that research back a little bit, but the SP6 research will come in handy. I've already tried some different lengths on the AJP8 but for some reason that engine seems particularly overly-sensitive to induction resonances leading to those pronounced humps in the power curve .. also the ajp8 isn't really short of midrange (360lb.ft is massive compared to the weedy 290/300 of the SP6) so the ajp8 was ideal to shift everything up the rev range .. the remit on the SP6 is almost exactly the opposite!
Harry .. i was supposed to have already tried this trialling on an ajp8 but the car i was scheduled to use developed a problem which has put that research back a little bit, but the SP6 research will come in handy. I've already tried some different lengths on the AJP8 but for some reason that engine seems particularly overly-sensitive to induction resonances leading to those pronounced humps in the power curve .. also the ajp8 isn't really short of midrange (360lb.ft is massive compared to the weedy 290/300 of the SP6) so the ajp8 was ideal to shift everything up the rev range .. the remit on the SP6 is almost exactly the opposite!
Don1 said:
Interesting reading. I seem to remember that the old Caterham R500 had rollers, but I'm fairly sure they were throttle bodies?
Yep, roller throttle bodies, rather than butterflies. They avoid having a great big lump of brass disturbing the airflow in each intake, especially at full throttle. I think...Right a bit more information . as predicted the shorter inlet has released more top end hp .. up to 348hp now .. however it's at the expense of a drop everywhere from about 4000rpm up to just under 7000rpm, the shorter pipe only benefits in the last 750-1000rpm which is a shame as the longer i make the long induction length the further it pushes the crossover point down the rev range, meaning there's a whole couple of thousand rpm in the mid/upper range that loses out to the std airbox.
I'm also thinking that when I test this on a 4 litre the peak power is going to be further down the rev range anyway as that's not such a revvy engine so really the short induction length should be closer to the std tvr length I feel for those engines. Damn TVR for getting the SP6 so near right grrrrr!!
Anyway more bits on order, going to see what is the practical longest induction lengths I can realistically fit into this T350 engine bay (handy having a T350 as it's more crowded than a tuscan engine bay with the steering linkage being different and the shape of the inner wing, so at least I'm working with worst case space limitations) .. Once i've got a handle on how long i can go and what it means in hp terms I can think about what system would be best, and more importantly is it enough better than std to make it worthwhile. At the moment I have gains right at the very top of 6-10hp depending what graphs i compare and gains through the lower midrange of 5-15hp depending what graphs I compare and different parts of the rev range .. all I need now is to fill in that upper midrange and we're somewhere close to gains right through.
One point to note though is that all of these tests are so far on a std mapped car, no remapping taken place yet, which gives me hope because even though remapping the SP6 doesn't give great gains, if i can more than offset the losses in the upper midrange and make more from the gains in the lower mid and very top of the rev range then just like the cerbera short induction it will make a package which gains everywhere (hopefully) ..
Will have more info over friday and the weekend I hope
I'm also thinking that when I test this on a 4 litre the peak power is going to be further down the rev range anyway as that's not such a revvy engine so really the short induction length should be closer to the std tvr length I feel for those engines. Damn TVR for getting the SP6 so near right grrrrr!!
Anyway more bits on order, going to see what is the practical longest induction lengths I can realistically fit into this T350 engine bay (handy having a T350 as it's more crowded than a tuscan engine bay with the steering linkage being different and the shape of the inner wing, so at least I'm working with worst case space limitations) .. Once i've got a handle on how long i can go and what it means in hp terms I can think about what system would be best, and more importantly is it enough better than std to make it worthwhile. At the moment I have gains right at the very top of 6-10hp depending what graphs i compare and gains through the lower midrange of 5-15hp depending what graphs I compare and different parts of the rev range .. all I need now is to fill in that upper midrange and we're somewhere close to gains right through.
One point to note though is that all of these tests are so far on a std mapped car, no remapping taken place yet, which gives me hope because even though remapping the SP6 doesn't give great gains, if i can more than offset the losses in the upper midrange and make more from the gains in the lower mid and very top of the rev range then just like the cerbera short induction it will make a package which gains everywhere (hopefully) ..
Will have more info over friday and the weekend I hope
trackcar said:
handy having a T350 as it's more crowded than a tuscan engine bay with the steering linkage being different and the shape of the inner wing, so at least I'm working with worst case space limitations
Good reading all this....but what is the space like in the Cerb engine bay compared to the T350? more? less?Interesting stuff, and great to see details of the exploration being shared with us all. A different & refreshing sales approach. 
So in same spirit, bit of info on mine. I deliberately specced the 4-litre 'sprint' engine with Dulfords for greatest torque over 2000rpm up to 6500, with cam profiles etc and the long-trumpet ACT box. Though it will rev above 7500 quite happily and safely.
However, as always we never had time to try variations, to see if we'd gone a bit too far and could win back some of top end. My torque ramps up rapidly to peak at a lowly 4400rpm, then falls very gently as you'd expect, but with a much faster fall above 6600rpm.
This was what I wanted, but still, an easy step with my hacksaw could be tempting...
So in same spirit, bit of info on mine. I deliberately specced the 4-litre 'sprint' engine with Dulfords for greatest torque over 2000rpm up to 6500, with cam profiles etc and the long-trumpet ACT box. Though it will rev above 7500 quite happily and safely.
However, as always we never had time to try variations, to see if we'd gone a bit too far and could win back some of top end. My torque ramps up rapidly to peak at a lowly 4400rpm, then falls very gently as you'd expect, but with a much faster fall above 6600rpm.
This was what I wanted, but still, an easy step with my hacksaw could be tempting...
Paul .. here is the comparison graph of the three inductions trialled first.
the blue line is the std airbox
the red line is a modified aftermarket airbox with slightly longer trumpets .. note it only loses out above 6000rpm to the std airbox, this is representative of the trend of the hp in your car i would imagine, meaning what you have is probably very good for your intended use with the extra midrange gains it gives.
the purple line is my long pipes as discussed earlier in the thread. its the best between 3250 and 5000 (the bit i'm interested in) but loses out massively above 5500rpm.
Plasticman has come up with a design for moveable trumpets which is close to what I had considered before, and then discounted, but he's a canny guy so if he thinks it would work it's very probably worthy of further investigation .. expect a visit one day DaveH to discuss options!
graph :
the blue line is the std airbox
the red line is a modified aftermarket airbox with slightly longer trumpets .. note it only loses out above 6000rpm to the std airbox, this is representative of the trend of the hp in your car i would imagine, meaning what you have is probably very good for your intended use with the extra midrange gains it gives.
the purple line is my long pipes as discussed earlier in the thread. its the best between 3250 and 5000 (the bit i'm interested in) but loses out massively above 5500rpm.
Plasticman has come up with a design for moveable trumpets which is close to what I had considered before, and then discounted, but he's a canny guy so if he thinks it would work it's very probably worthy of further investigation .. expect a visit one day DaveH to discuss options!
graph :
Ok all testing on lengths finished, the results are in
Have just been playing around with some very long lengths .. 22 inches from butterfly and the midrange gains are massive, but starting at the bottom of the rev range first ..
Up to 2500rpm it doesn't seem to matter what induction length you have there is virtually nothing to pick between any of the systems tested .. it may be that with something say 40 inches long you might shift the graph at the bottom end but to all practical purposes whether I fitted long, medium or no pipes at all the results all pretty much overlay each other from every graph to within a few hp. The difference between the best and the worst is no more than 10hp tops across such a wide variety of lengths .. the worst being the std airbox and no airbox at all, the best being the mid length air pipes.
After 2500rpm the 22inch inch pipes show to be massively the best tested. All the mid length pipes come out roughly equal in comparison again with no more than 10-15 hp to pick between the best and the worst. The best being the shortest mid length pipes and the worst again being the std airbox. This now sounds like the std airbox is pretty rubbish, but the graphs are so close i'm really talking about tiny hp differences here.
At 5500rpm we start to see the first time that the 22inch length gets beaten, indeed from 5500 onwards as long as you have some pipe on there at all you beat the 22inch performance. From 5500 to 6500 any of the mid length pipes are good, the graphs overlay apart from a few hp here and there, but if anything a length of about 7-8 inches is best.
Above 6500rpm the short 5 inch length is the clear winner once again.
So what does this tell us?
Well really the engine needs 3 separate induction lengths through the whole rev range: 22inch, 8 inch and 5 inch coming in at 5500 then 6500 as the crossovers to get the best spread of power without going to a steplessly variable version like in the mazda link above.
You could get away with 22inch and then 5 inch and you'd only be throwing away realistic gains between 5500 and 6000rpm which is the only time this length shows anything like an advantage.
Only problem now is a 22inch length needs a U bend like a toilet to package into the engine bay
In the graph below are the results of interest. again all tested same car same day so all are benchmarked against each other.
Have just been playing around with some very long lengths .. 22 inches from butterfly and the midrange gains are massive, but starting at the bottom of the rev range first ..
Up to 2500rpm it doesn't seem to matter what induction length you have there is virtually nothing to pick between any of the systems tested .. it may be that with something say 40 inches long you might shift the graph at the bottom end but to all practical purposes whether I fitted long, medium or no pipes at all the results all pretty much overlay each other from every graph to within a few hp. The difference between the best and the worst is no more than 10hp tops across such a wide variety of lengths .. the worst being the std airbox and no airbox at all, the best being the mid length air pipes.
After 2500rpm the 22inch inch pipes show to be massively the best tested. All the mid length pipes come out roughly equal in comparison again with no more than 10-15 hp to pick between the best and the worst. The best being the shortest mid length pipes and the worst again being the std airbox. This now sounds like the std airbox is pretty rubbish, but the graphs are so close i'm really talking about tiny hp differences here.
At 5500rpm we start to see the first time that the 22inch length gets beaten, indeed from 5500 onwards as long as you have some pipe on there at all you beat the 22inch performance. From 5500 to 6500 any of the mid length pipes are good, the graphs overlay apart from a few hp here and there, but if anything a length of about 7-8 inches is best.
Above 6500rpm the short 5 inch length is the clear winner once again.
So what does this tell us?
Well really the engine needs 3 separate induction lengths through the whole rev range: 22inch, 8 inch and 5 inch coming in at 5500 then 6500 as the crossovers to get the best spread of power without going to a steplessly variable version like in the mazda link above.
You could get away with 22inch and then 5 inch and you'd only be throwing away realistic gains between 5500 and 6000rpm which is the only time this length shows anything like an advantage.
Only problem now is a 22inch length needs a U bend like a toilet to package into the engine bay
In the graph below are the results of interest. again all tested same car same day so all are benchmarked against each other.
Edited by trackcar on Friday 30th January 15:49
Joo are these with or without airbox fitted, I'm assuming with . I only say that as you may be able to with airbox volume and hence resonance tune it to the mid range whilst leaving the intake throats tuned to the higher revs and get a better balance between the two (not quite that simple in practice I know). Likewise whilst the 8" only just loose out to the 5" at the very top end they are long enough with pulse plates give you a pulse reflection as if the intake was 20" assuming the P plate was 2" away. Just a thought.
. I only say that as you may be able to with airbox volume and hence resonance tune it to the mid range whilst leaving the intake throats tuned to the higher revs and get a better balance between the two (not quite that simple in practice I know). Likewise whilst the 8" only just loose out to the 5" at the very top end they are long enough with pulse plates give you a pulse reflection as if the intake was 20" assuming the P plate was 2" away. Just a thought.Anything that was short enough to go into an airbox was done with, but no filter as the pipes were longer than the depth to the filter.
The 22inch pipes were just facing forwards in the open engine bay with a fan blowing cooler air over them, but may still have been subject to warm air so those results are very pessimistic.
There's so little opportunity to play with airbox volumes that I'll just go with whatever is biggest I can fit in and hope that's good enough
Adding pulse plates to the ends have made no difference at all which is od as I'd have thought they would have worked much better than on the cerbera as the SP6 inlet is direct line of sight, however I must say i didn't investigate this very far as I know that from the cerb research that they only work over a tiny rev range, say a few hundred rpm anyway.
Hi Pod .. the single exhaust was made by clivef to win the V8 noise competition at SclubHeaven .. no silencers just a 3 inch pipe front to back.
It was loudest V8 but they wouldn't give me the prize because it's a 5 litre, go figure! .. what they'd have done if an owner of one of the 2 factory 5 litre V8S models had turned up i have no idea.
This year will (hopefully) be 4 litre, but turbo with antilag .. let's see what they do then
The 22inch pipes were just facing forwards in the open engine bay with a fan blowing cooler air over them, but may still have been subject to warm air so those results are very pessimistic.
There's so little opportunity to play with airbox volumes that I'll just go with whatever is biggest I can fit in and hope that's good enough
Adding pulse plates to the ends have made no difference at all which is od as I'd have thought they would have worked much better than on the cerbera as the SP6 inlet is direct line of sight, however I must say i didn't investigate this very far as I know that from the cerb research that they only work over a tiny rev range, say a few hundred rpm anyway.Hi Pod .. the single exhaust was made by clivef to win the V8 noise competition at SclubHeaven .. no silencers just a 3 inch pipe front to back.
It was loudest V8 but they wouldn't give me the prize because it's a 5 litre, go figure! .. what they'd have done if an owner of one of the 2 factory 5 litre V8S models had turned up i have no idea.
This year will (hopefully) be 4 litre, but turbo with antilag .. let's see what they do then
trackcar said:
Hi Pod .. the single exhaust was made by clivef to win the V8 noise competition at SclubHeaven .. no silencers just a 3 inch pipe front to back.
It was loudest V8 but they wouldn't give me the prize because it's a 5 litre, go figure! .. what they'd have done if an owner of one of the 2 factory 5 litre V8S models had turned up i have no idea.
This year will (hopefully) be 4 litre, but turbo with antilag .. let's see what they do then
It was loudest V8 but they wouldn't give me the prize because it's a 5 litre, go figure! .. what they'd have done if an owner of one of the 2 factory 5 litre V8S models had turned up i have no idea.
This year will (hopefully) be 4 litre, but turbo with antilag .. let's see what they do then
Just make sure you fit a flameguard at the back... we don't want a melted rear end...
hi Jools
interesting reading as ever, just a thought but it might be worth trying a couple of 'splitter' plates inside the air box to try and spread the mid range gains a bit further.
These would be simple ally plates between the trumpets on either cylinders 3 & 4 or 2 & 3 and 4 & 5 depending on how the speed six has its exhaust manifold configured.
The idea is to prevent charge robbing between the tuned cylinder sets , the height of these plates will introduce another tuning effect as well - just to keep it all interesting...
Have you thought of trying a stepped exhaust manifold as well?
Matt
interesting reading as ever, just a thought but it might be worth trying a couple of 'splitter' plates inside the air box to try and spread the mid range gains a bit further.
These would be simple ally plates between the trumpets on either cylinders 3 & 4 or 2 & 3 and 4 & 5 depending on how the speed six has its exhaust manifold configured.
The idea is to prevent charge robbing between the tuned cylinder sets , the height of these plates will introduce another tuning effect as well - just to keep it all interesting...
Have you thought of trying a stepped exhaust manifold as well?
Matt
Gassing Station | Speed Six Engine | Top of Page | What's New | My Stuff