Intake Trumpets: The Theory
I hate to start sounding like the “Old Soldier” but… 50 years ago I read a feature that fascinated me at the time. It was all about intake trumpets and how the overall intake length altered the power output of the engine. Try as I might I could not get my 15 year old head around how it worked. Now I am not suggesting that I have been pondering that feature for the last 50 years, (it hasn’t kept me awake at nights), but I can still remember how the text explained that a parallel intake trumpet worked better than a tapered one. Theory today suggests that this might actually be the case.
When G.P. cars were forced to give up their superchargers after WW11, engine designers started looking at pulse tuning the inlet length as a way of making more power from an aspirated engine. By the early sixties this pulse tuning of the intake system was well established but you had to have a set of carbs with one choke per cylinder to take advantage of it and a dyno to set it all up. We mere mortals had no access to Dynos, and rolling roads did not exist back then, so pulse tuning the inlet lengths remained the domain of the well-heeled race teams. We go-faster boys had to make do with removing the air filter and maybe putting some little bell-mouths on our SU carbs. Now, with a rolling road on every street corner (almost) that is certainly not the case these days.
To understand what is going on with pulse tuning start with a picture in your head of a bottle of wine. As you begin to pull the cork (it’s a posh wine obviously - not having a screw top) a partial vacuum builds up under the cork. When it clears the bottle neck you hear a “pop” as the air rushes in to fill the partial vacuum. Think of the piston going down the bore as the cork and the inlet valve opens at the point of the cork clearing the neck of the bottle. That pressure wave gets reflected up the intake runner (bounces if you like) and it changes sign when it expands into the atmosphere. This means that if a negative wave comes out into the air box (or atmosphere) it returns down the inlet as a positive wave – that if timed correctly can be used to cram a bit more mixture into the cylinder.
Think of it as a slug of air moving up a tube. When it exits the end it leaves a bit of a vacuum behind which is then filled by the surrounding air. Conversely if that air going down the intake hits a closed valve, it bounces back but still as a positive pressure wave. These pressure waves bounce backwards and forwards until they run out of steam; unless of course you get another inlet stroke to start it all happening again. Theoretically you can calculate these lengths given esome fundamental data about the said engine but in my world you just experiment on the dyno and see what the engine likes.
Now theoretically you get the strongest pulse with the sharpest change in section of the intake system. In other words; if the intake has a sudden expansion into atmosphere you get a stronger pulse than if it had a slow taper out to the air box. A taper takes the edge off the pulse, so if we want the maximum pulse effect why do we all run tapered trumpets and not parallel ones? A good question and one that I decided needed investigating now that I am no longer a spotty fifteen year old and I have the means to run some real world tests.
The one question I get asked more than any other from people buying throttle bodies is: “What length inlet trumpets do I need?” My answer is always the same: “I don’t know”. What I tell people is that they need to consider what they want from their engine in terms of torque curve shape and what they intend to use the car/engine for. The ideal situation is to experiment on the rolling road. Cam timing and the exhaust system both play their part in the engine pulse tuning equation but are practically quite difficult to "adjust" easily whilst on the rollers. For this reason I generally keep a selection of trumpets here, (from 40m up to 120mm). If people are not going to experiment then I usually recommend 90mm because you would wouldn’t you? … It’s about half way! One major drawback with this type of experimenting is that you have to keep four or five sets of trumpets in stock at a cost of up to £135 per set. That’s a big investment when you are only ever going to end up using one set on your engine.
Trying to kill two birds with one stone I dreamt up a system with a parallel intake trumpet fitted into an adapter that would allow the length to be altered by sliding it down inside the throttle body, or extending it out into the adapter. Obviously this means having some restriction in the original throttle body’s’ bore but since, in my experience, most throttle bodies are oversized I reasoned that this might not be the drawback it first seems.
I decided to get some tube and start playing on the flow bench. My first test was to take a 45mm bore parallel intake trumpet (from a carburettor) and flow test this alongside a same length tapered trumpet. I wanted to see if the taper gave any flow advantage over the parallel intake shape. The result was that I could not measure any difference between the two trumpets with a constant air-speed flow test. Next I made up a trumpet with a slightly smaller than 45mm outside diameter so that it would fit down inside a 45 mm Jenvey throttle body. This showed a flow restriction but only 7 per cent less than the full 45mm trumpet (it now being a 42mm I/D).
Next I happened to have a Jaguar AJ6 head on the flow bench fitted with Jenvey 45mm bodies. The bare head flowed 140 cfm at 10 inches of water test pressure and fitting the bodies (with tapered trumpets) did not drop the flow at all. Removing the tapered trumpets I fitted my 42mm I/D trumpets and repeated the flow test. It was still 140 cfm so there was no restriction to flow in the real world.
The flow bench is all well and good but it is only a tool and the next step was to power test the adjustable trumpets back-to-back with the tapered ones. For this we used a 180 bhp, 1600cc, Ford Zetec SE engine in a Fiesta. What we were looking for was a real world gain, not a tiny percentage that could be argued about for or against, or test methods or corrections questioned etc. If we couldn’t see a serious change then the adjustable parallel trumpets would be a failure. Tests were also done on an “ABA” basis. That is to say we ran test A, followed by test B, and then test A was repeated to see if any change was real world or just a fluke. We then repeated it to make sure that our method was…well repeatable.
Having established a base line we swapped the trumpets for parallel ones of exactly the same external length. (Jenvey 40 x Emr 40) Straight away we had a small change. The torque curve had a more distinct wave to it, slightly bigger dips and peaks so the pulse did appear to be stronger, albeit not necessarily in the right places! We extended the trumpets to 50 mm and did another series of tests. (Jenvey 40 x Emr 50) Now we were seeing a really strong pulse/wave form in the torque curve with big dips and peaks.
This is something I have seen many times in the past from engines with peaky cams, it had a familiar look and I was starting to get excited. Next up was the 80mm trumpet (Emr 50 Vs Emr 80). Much to my amazement the result was even better. Usually when you change length like this you get bumps in the torque curve where you had dips and dips where you had bumps. What’s happened here is the dips have been filled in with avengeance. Suddenly we had the makings of a much smoother torque curve. Moving the trumpet out to 95mm did not have a major effect but was still incremental (Emr 80 Vs Emr 95). However, what we did see was a trend towards improved torque at the lower part of the rpm range.
This more or less fits in with the old adage about longer trumpets for low speed power and short ones for higher peak bhp. (Emr 95 Vs Emr 120) Suddenly we had a torque curve looking a lot like the 50 mm trumpet gave but with much higher values. Could we “fill in” the dips by going longer again? Emr 120 Vs Emr 130 shows a trend to fill in the dips as we saw before but not enough of a change to get excited about, probably the 10 mm longer trumpet wasn’t enough of a change. We decided on a bigger change and fitted some new, much longer trumpets of 155 mm. Interestingly the peak torque is up and so is the torque at higher rpm. We were certainly seeing some interesting changes! Extending yet again to 170mm we suddenly filled in the dips in the torque curve. (Emr 155 Vs Emr 170).
Back to the beginning.
Now take the original power curve with the 40mm tapered trumpets and compare it to what you could have with 170mm parallel trumpets. Peak power is lower by about 500 rpm but the torque gains are massive! As a final test I decided to make up some “trumpets” with no trumpet radius: that is a dead straight parallel tube. Theoretically this would give the strongest pulse of all as you can’t get a more sudden section change than a straight pipe into the atmosphere. The torque was down right from the off and it really killed the top end power stone dead. At peak that little radius was worth 8lbft. We figured that this made sense as the air flow had to be down compared to the flared section trumpet.
We were all happy with that explanation – until I did a flow bench test. The flared section was only worth 2 cfm on the static flow test, dropping from 140 cfm down to 138 cfm. At this point you are probably looking for an explanation as to what was going on but I don’t have one myself. However, I did ask my good friend, and very clever chap, Paul Cronin. Paul believes that the steady state flow bench is not replicating the sort of air speeds you see with an engine so that the positive pulses would have a much higher peak air speed than my average flow rate.
Just as a bit of silliness we finally tried adding the straight tubes to the 170 mm trumpets. That gave us an overall trumpet length of 330 mm! Much to my amazement we had suddenly turned our peaky race type engine into a tractor! Looking at the two torque curves you would not think they were the same engine with the same cams, compression and exhaust. One is all top-end; the other is all bottom-end..
What price “power bands” now when selecting a camshaft? Cam manufacturers often quote power bands like: “4000 – 7000 rpm” when the reality is that if you can change the inlet length you can move the power band pretty much to anywhere you want it. We didn’t try any other lengths because it was not something you could practically fit to your car but it was certainly an interesting exercise.
Up until this point all our power testing had been done on full throttle. I now wanted to know what effect intake length had on part throttle. The answer was almost none at all. The pressure wave appears to reflect off the butterfly up until the throttle is almost wide open. We only started to see the pulse effect after about 80 per cent throttle. This is good news because it means that if you have already had your engine mapped it only needs re-mapping on the highest load sites with this adjustable length intake system, the part throttles do not appear to change. It should also be noted that during testing I did not optimise the fuelling or ignition for every run. As long as the mixture did not go dangerously weak I left it alone.
Interestingly we had an Elise in fitted with a Duratec engine recently. The intakes were very short because there is little bulkhead clearance and what trumpets it had fitted were tapered. The torque curve wasn’t so much a curve as a flat line and not really what you would expect from a 2.0 litre engine at 132ft lbs. Out came the prototype adapters and we fitted parallel trumpets as long as we could squeeze in the gap. Peak power remained the same (but at 400 rpm lower in the rev range) and peak torque gained nearly 13 ft lbs! I should add that the longer intakes caused the engine to pink badly due to the better cylinder filling so the timing had to be retarded to suit the new tuned length.
On a practical front what does all this mean? Basically if you want to get the best from the engine the intake length is a critical factor that should be addressed. Changes of just 10mm show a meaningful result. What’s more you can have two or three inlet lengths for different applications on the same engine; like one for the road or hill-climb and another for a circuit track day. With map switching in an ECU it certainly is a practical application. To make something a bit more presentable than our prototype trumpet holders we got out the CAD package and managed to draw up quite a neat bit of kit. I organised having some trumpets made which gave you the ability to go from 60mm up to 140 mm within just the one length. There is a danger that if you pass the trumpet too far down the body it will hit the butterfly and might jam it open so you need a marker showing the minimum length limit.
Getting an air filter on a long intake system can be a bit of a nightmare if you are trying to bolt your back plate direct to the throttle body. The system I came up with of having an adapter to hold the intake trumpet means you can also hold an air filter back plate. You simply put a couple of adapters on backwards and slide them up the trumpet to meet the air filter. You might be thinking by now that if this overall length tuning business is so critical, why don’t car manufactures use it on production cars? The answer is: they do! There are many successful engines that have used variable inlet lengths and current engines are stillcoming out with dual length intakes systems in an attempt to have your cake and eat it. People like BMW do not go to those extremes on a fad, pulse tuning IS critical to engine performance.
Based on the testing we have done to date we can draw certain conclusions:
1. The tapered trumpets work no better than the parallel ones.
2. It is the overall length of the induction that is the single most important feature of any throttle body system.
3. A change of just 4mm in overall length gives a measurable change in the torque curve.
4. You only need to re-map the wider throttle settings of the fuel and ignition maps when you change the induction length.