S4 intake manifold: I made some changes ... dyno results ...
#31
Rennlist Member
I think it is possible to expand the upper and lower intake manifold alot more. but It would involve cutting the stock manifold and welding or finding a very strong way to bond the modifications *probably with composite material). But, the problem would then be the throttle body. You would have to either bore it out, or buy a larger, either way you would have to tune the computers. then the intake, it is a horrible design. i could not tell if you had a supercharger, but not to much air flows into the actual intake tubes. you would have to then make the box as large as possible, remove the air filer. Make a filter and second box that goes flat in between the hood and the engine, then a 3rd that kind of warps around in front and underneath the radiator. You have to be careful not to obstruct what little air is hitting the fans.
While driving on the road (which is the design purpose of the 928), what is the percentage of time spent at full throttle? MK is obviously interested in track, and only track, and has no appreciation at all for street performance or driving pleasure.
Many (almost all) of the folks on this forum are much more interested in the pleasure of driving on the street, where throttle response and feel are most important. (SEE: "X-pipe" for example.)
While I don't plan to modify my car, I can appreciate the curiosity, initiative and enthusiasm shown by the original poster before the thread hijack.
Many (almost all) of the folks on this forum are much more interested in the pleasure of driving on the street, where throttle response and feel are most important. (SEE: "X-pipe" for example.)
While I don't plan to modify my car, I can appreciate the curiosity, initiative and enthusiasm shown by the original poster before the thread hijack.
It's funny, I think you are confusing WOT at low rpm with Street performance. Remember, the tests conducted with the mods are done at WOT. So your comment is kind of moot, because if you are not flooring it, you don't need these mods. Just gear up the throttle linkage if you want more part throttle power with less pedal deflection.
So, if this mod is for street performance at part throttle, you are chasing up the wrong tree. All the flappy function gains are for near full throttle, and the dyno runs are done at WOT. and there is 50ftlbs of gain in the low rpm (below 3400rpm) if you want to floor your car at 2000rpm on the street, you should buy an electric car. However, I agree, you are passing some truck on the freeway, in 5th gear chugging along at 70mph and you don't want to downshift, its nice to have the flappy gains, plus the extra 10hp from the hogged out intake........ but, a little downshift to 4th, gets 30%more torque and maybe double the power, so why not use the gears?? It is more fun........ track or street.............my opinion of course!
#32
Pro
Thread Starter
Hi All. Apologies for delay in returning.
I've seen some issues raised by various ppl - I'll get back to those as soon as time permits.
Mark K: I had previously seen the flappy no-flappy chart and mentally noted it as 'interesting' but now the penny has dropped. With a clang. Specially for the rev range above 5200rpm. Thanks. (Chart at bottom of post).
But let my re-iterate the reason behind these changes: I did it for no other reason than to see if the manifold would respond to improved airflows. I applied well established airflow theory to improve the shape of the bellmouths, as well as improving flow to the bellmouths.
The manifold was already out and I saw an easy and v.cheap way of implememting the changes so I said: "why the heck not? Can't hurt, and has a good chance of succeeding".
(Now if I can gather some data for Porken to use in his runner mapping there may be more life in the manifold yet).
It was an interesting experiment, which as it turns has had good results.
Below is a quickly put together pic of power and AFR with the manifold mod. Vertical lines show the rpm corresponding to those shown in the flappy no-flappy chart. (That pic is further down).
Above 5200rpm, torque and power is trailing off, but the AFR is also dropping away quickly below Lamda 0.9.
How much power is being lost running with such a rich mixture?
I've also made changes in two other locations, thse being likely to give most benefit at 5000rpm+ so I need to improve the AFR above 5k rpm to really test the changes, so how do I do this, even temporarily?
Flappy No-flappy power chart.
I've seen some issues raised by various ppl - I'll get back to those as soon as time permits.
Mark K: I had previously seen the flappy no-flappy chart and mentally noted it as 'interesting' but now the penny has dropped. With a clang. Specially for the rev range above 5200rpm. Thanks. (Chart at bottom of post).
But let my re-iterate the reason behind these changes: I did it for no other reason than to see if the manifold would respond to improved airflows. I applied well established airflow theory to improve the shape of the bellmouths, as well as improving flow to the bellmouths.
The manifold was already out and I saw an easy and v.cheap way of implememting the changes so I said: "why the heck not? Can't hurt, and has a good chance of succeeding".
(Now if I can gather some data for Porken to use in his runner mapping there may be more life in the manifold yet).
It was an interesting experiment, which as it turns has had good results.
Below is a quickly put together pic of power and AFR with the manifold mod. Vertical lines show the rpm corresponding to those shown in the flappy no-flappy chart. (That pic is further down).
Above 5200rpm, torque and power is trailing off, but the AFR is also dropping away quickly below Lamda 0.9.
How much power is being lost running with such a rich mixture?
I've also made changes in two other locations, thse being likely to give most benefit at 5000rpm+ so I need to improve the AFR above 5k rpm to really test the changes, so how do I do this, even temporarily?
Flappy No-flappy power chart.
#33
Rennlist Member
Hi All. Apologies for delay in returning.
I've seen some issues raised by various ppl - I'll get back to those as soon as time permits.
Mark K: I had previously seen the flappy no-flappy chart and mentally noted it as 'interesting' but now the penny has dropped. With a clang. Specially for the rev range above 5200rpm. Thanks. (Chart at bottom of post).
But let my re-iterate the reason behind these changes: I did it for no other reason than to see if the manifold would respond to improved airflows. I applied well established airflow theory to improve the shape of the bellmouths, as well as improving flow to the bellmouths.
The manifold was already out and I saw an easy and v.cheap way of implememting the changes so I said: "why the heck not? Can't hurt, and has a good chance of succeeding".
(Now if I can gather some data for Porken to use in his runner mapping there may be more life in the manifold yet).
It was an interesting experiment, which as it turns has had good results.
Below is a quickly put together pic of power and AFR with the manifold mod. Vertical lines show the rpm corresponding to those shown in the flappy no-flappy chart. (That pic is further down).
Above 5200rpm, torque and power is trailing off, but the AFR is also dropping away quickly below Lamda 0.9.
How much power is being lost running with such a rich mixture?
I've also made changes in two other locations, thse being likely to give most benefit at 5000rpm+ so I need to improve the AFR above 5k rpm to really test the changes, so how do I do this, even temporarily?
Flappy No-flappy power chart.
I've seen some issues raised by various ppl - I'll get back to those as soon as time permits.
Mark K: I had previously seen the flappy no-flappy chart and mentally noted it as 'interesting' but now the penny has dropped. With a clang. Specially for the rev range above 5200rpm. Thanks. (Chart at bottom of post).
But let my re-iterate the reason behind these changes: I did it for no other reason than to see if the manifold would respond to improved airflows. I applied well established airflow theory to improve the shape of the bellmouths, as well as improving flow to the bellmouths.
The manifold was already out and I saw an easy and v.cheap way of implememting the changes so I said: "why the heck not? Can't hurt, and has a good chance of succeeding".
(Now if I can gather some data for Porken to use in his runner mapping there may be more life in the manifold yet).
It was an interesting experiment, which as it turns has had good results.
Below is a quickly put together pic of power and AFR with the manifold mod. Vertical lines show the rpm corresponding to those shown in the flappy no-flappy chart. (That pic is further down).
Above 5200rpm, torque and power is trailing off, but the AFR is also dropping away quickly below Lamda 0.9.
How much power is being lost running with such a rich mixture?
I've also made changes in two other locations, thse being likely to give most benefit at 5000rpm+ so I need to improve the AFR above 5k rpm to really test the changes, so how do I do this, even temporarily?
Flappy No-flappy power chart.
the lamda is not far off optimal for optimal power. slightly rich mixture engines at 11:1 don't make that much max power vs ones at optimal ranges in the 12.5 fuel to air ratio range. .8 is just a little rich of optimal by the way as its around 11.8:1, which is well in the safe range and max power range of most tunes.
what you have here is a test of low end power, where things are a little unstable and, you have seen some gains to 3400rpm. this is the quite unusable range. then, we have to ignor the 3400 to 5200rpm range because flappy wasn't working in prior dyno run, and then guess that from 5200rpm and higher , it really wouldn't matter if flappy is open or stuck closed as flappy doesn't effect that RPM range that much.
what I get from the test is that what you did works, and it works in the range where it counts the most.... 5200rpm and higher.
so, if you did some more high flow beneficial mods, lets by all means see them in a new dyno run.
#34
Nordschleife Master
Hey mad scientist, you might be onto something! ;-)
Whether or not you are losing power and how much if so depends on whether you're at 0.84 lambda in every cylinder or at say lambda 0.79 in cylinders 1 (90-degree exhaust blowdown interference) and 5 (curvy runner) and at 0.89 in cylinder 2 (no exhaust interference and a straight runner). If it's lambda 0.84 in every cylinder at 6000 rpm it's probably close enough for government work. If it's 0.79 in some cylinders and 0.89 in some cylinders then there's some kW's left on the table.
Now, be a good sport and pull of the driver side plenum cover and shoot us some photos!
Whether or not you are losing power and how much if so depends on whether you're at 0.84 lambda in every cylinder or at say lambda 0.79 in cylinders 1 (90-degree exhaust blowdown interference) and 5 (curvy runner) and at 0.89 in cylinder 2 (no exhaust interference and a straight runner). If it's lambda 0.84 in every cylinder at 6000 rpm it's probably close enough for government work. If it's 0.79 in some cylinders and 0.89 in some cylinders then there's some kW's left on the table.
Now, be a good sport and pull of the driver side plenum cover and shoot us some photos!
#35
Pro
Thread Starter
Is there a quick way of adjusting the AFR, even temporarily?
BTW - when driving the car it is so much more lively than U would expect from the look of the chart.
.
#40
Pro
Thread Starter
I've just returned from doing a few errands, with all over my face.
SOTP difference is as obvious as a pie in the face. LOL.
Air con on and my 30kg (60lb) dog on board and it's still better than an empty car with A/C off and using original manifold.
.
#41
Pro
Thread Starter
I just wondered if the may have been a simpler way, like defeating the WOT switch function temporarily.
.
#42
Pro
Thread Starter
#43
Archive Gatekeeper
Rennlist Member
Rennlist Member
I have taken to clicking on the little red 'report' triangle on every one of these godforsaken spambot posts and reporting them as such, in the hopes that someone at IB will be annoyed enough to do something about them. Hopefully that something will not be to ban me for being a tattletale.
#44
Pro
Thread Starter
Hey mad scientist, you might be onto something! ;-)
Whether or not you are losing power and how much if so depends on whether you're at 0.84 lambda in every cylinder or at say lambda 0.79 in cylinders 1 (90-degree exhaust blowdown interference) and 5 (curvy runner) and at 0.89 in cylinder 2 (no exhaust interference and a straight runner). If it's lambda 0.84 in every cylinder at 6000 rpm it's probably close enough for government work. If it's 0.79 in some cylinders and 0.89 in some cylinders then there's some kW's left on the table.
Now, be a good sport and pull of the driver side plenum cover and shoot us some photos!
Whether or not you are losing power and how much if so depends on whether you're at 0.84 lambda in every cylinder or at say lambda 0.79 in cylinders 1 (90-degree exhaust blowdown interference) and 5 (curvy runner) and at 0.89 in cylinder 2 (no exhaust interference and a straight runner). If it's lambda 0.84 in every cylinder at 6000 rpm it's probably close enough for government work. If it's 0.79 in some cylinders and 0.89 in some cylinders then there's some kW's left on the table.
Now, be a good sport and pull of the driver side plenum cover and shoot us some photos!
The mod process was a bit evolutionary in that as I prepared to make a planned change I'd think it through (with all the CFD pics) and then sometimes do it slightly differently. Starting fresh the next morning it was occasionally a case of "what was I thinking?". LOL.
Manifold LHS - part way thru.
In engine bay. Looking very bling. Yes - that is gold paint - a happy accident. That filled area (shown above) can be seen (at the LHS bellmouth).
#45
Rennlist Member
You can tweak the AFR a little with the resistance pot if you are running the no cat option. Interesting that you are looking for an AFR of 13.3. Stock tuning produces about 12 to 12.5 when the WOT map is engaged whereas the stock cruise map alone will give something leaner at high loads but not sure what. I presume you do not have a RRFPR? You could test for this by defeating the WOT contact switch but I suspect you would end up too lean for max power.
For folks like yourself you really need the ST2 and a dyno to optimise total gains from all the various [marginal?] improvements possible.
Loue [OTT] always used to advise many years ago that the 928 motor makes max power at around 13.5. Just shows how far ahead of the game he was. He also had dyno curves showing the impact of AFR against max power. Over enrichment does drop the power but nothing compared to a lean mix of course.
Look forward to reading more about your exploits epsecially if you can integrate your work with that being done by Professor Ken.
Regards
Fred