OK, so in the day I raced Toyota Atlantic. I've got a lot of time playing with intake trumpets on a 4A-GE. You definitely want the trumpet on the end, but I understand why you tested like you did. We would run as short as 50mm trumpets on ovals when we were only concerned about 9200-10000 rev range. We would run from 80 to 150mm trumpets depending on if we were looking for more high end or more mid range. In the last few years, the best road course / street course setup was with cylinders 1-4 about 30mm shorter than cylinders 2-3. A high rev track might be 80/110 while a street circuit might be 130/100. It seemed to help mid-range without hurting the top end. We would also run a tri-y header on everything except the ovals, which was 4-into-1. The airbox sizing is very important. You're missing something if you haven't gone down that bunny-hole.
Interesting, thanks. Numbers sound similar if consider the difference in the itb section length and I'm targeting a lower rpm. Goal for this engine was always to have good top end while having most power possible through the middle. It did have a more aggressive cam combo at one staged that made it more similar to Atlantic power curve (the one i had on my dyno at least) but now runs a smaller cam combo and vvti which is a game changer for a street car
Another good movie and a lot of work. Best begards. In motorcycles, the higher the runner rpm, the actuator shortens it. In motorcycle engines, if the producers wanted high torque in low parts, they used a narrow and long intake to accelerate the air flow and cylinder filling, but if these were high-performance engines where the lower parts were not as important as the high rpm parts, a short intake was used. To increase the torque at the bottom of these engines, at first a throttle was used in the exhaust system, and later a variable intake system was added. In the cases I have encountered, double fuel injection is used with variable intake. One injector in the intake and the other above the runner. Finally, one important thing that is used in all these engines and affects the operation of the engine is the airbox.
Well done for conclusively dismissing the calls for variable intake length (and yes I was one of the ones curious about it)! Its almost like the designer of the G4A knows what he / she was doing.
The bike variable inlet where the longer trumpets are lifted and have shorter trumpets underneath seems easiest to implement in this case. And with less disturbance for max power.
That makes sense when the bike is used in the 7000 to 14000 rpm range. This engine only needs to cover 5000 to 9000 approximately. The difference in runner length is probably too small to make that work, and the shortest length isn't that short. A different way is to use resonances. Many car intake manifolds employ this approach. One of the most complex is that which was fitted to the Ferrari 458, itself an evolution and complication of the earlier engines. But the style they adopted for their V12s might give some inspiration, and is more appropriate here. The idea is that a valve opens up in the manifold to an extra chamber at one end, increasing the volume of the plenum. This changes its resonance(s) and hence cylinder filling in the same way runner length can. I don't know if this engine normally uses a plenum over the bellmouths, but it's generally good practice even with ITBs. "Unfortunately" it's another thing to tweak and tune. We've sort of seen this effect with the K24 EP already.
Yes this design works well, proven, reliable. But i think a sliding single trumpet is just best. it is the no compromise solution. As far as area under the horsepower curve, we are talking a few % at most either way. So having a good cost effective setup like the double trumpets makes plenty of sense.
we could test it but the goal is to reach the second harmonic… if we use a bike such as the yamaha r1 for an example and i put in some vague non accurate numbers because i dont know what yamaha has (and because yamaha knows air - beams engines toyota engines ford sho v6 engines etc) 2007 yamaha r1 - has variable itb. 13,700rpm rev limit… usable power i dont know, google a dyno graph. second harmonic wave is the goal. divide that by the rpm you want your second harmonic to come on song… and thats your required total runner length in inches from the beginning of the opening to the end of the nearest side valve face (ill use valve face because i cbf figuring out if they use the outer side or average it or another part of the end of the valve area) so… 132,000 - the second harmonic, divided by… lets say the rpm of the yamaha - 13,700 9.6 inches. the yamaha itb looks like its around 6 inches total, with the rest of the length in the port to the valve… probably (this is all just to get things in perspective, nothings accurate) so you target that, then go longer or shorter and test and test and test… until you get what you want. this is where the question lies with the 4age and any other car… can u make a variable itb 13.2 inches long, or longer? because the length gets a lot longer if you want your varying itb to generate second harmonic torque from 5000rpm… 26.4 inches… will it fit? will it work? or will it ake uhhh… 4ages for anything to happen? if you go shorter, for the 3rd harmonic… can you make your long early torque generating variable itb shrink down to the right length at high rpm? will they hit anything if you pop the long runners off like how it is in the yamaha? the r26b quad rotor has all the space… when i wanted to try that in a 2jz toyota chaser.. i realised i had no space. but then theres real world… weve seen dyno tests already with ultra long runners tested here in another video, cut down bit by bit… and the results didn’t come anywhere near the power of the g4a itb setup when trying to aim for the second harmonic, even at 10 inches. (and they were over 1 metre long to begin with - in an attempt to catch the first harmonic wave - the car didn’t want to rev out in some cases) so what gives? runner diameter (and therefore, itb size) another factor perhaps? do cam specs need to match? does it need vtec and dual vvti? does the header need to be just as long to work with the runner length???
Wow! Thank you for this tests ❤ to some, like me who loves doing tests and dyno this is one of the best I’ve seen and most of the good ones I see are from your channel! props to you and stay awesome!
I watched all of it and was feeling sorry for that little motor by the end of it! A lot of useful info there, I have a na 1.6 mx5 motor on throttle bodies and always wondered what changing lengths would do, thanks for putting this up.
Quite interesting to see that all youre test with various trumpet designs didn´t catch up with youre original setup. What i have found was a difference when you design the whole inlet runner on a taper....quite better in top end power with little lost on low and mid range. Was a qonsequent setup change from straight to a 7° taper And for youre last test if the strut tower will make something different.... Once i hold my hand in front of the trumpets on the dyno, even with a quite short distance of 20mm you couldn´t feel THAT much force on it...wich concludes that most of the air will be sucked in from the side (why a correct bellmouth is so important) I always enjoy watching youre videos as i have seen similarity to things i have also done before.....and to often throw them in the "not worth bin"
Awesome work as always a Boet, I only prefer the 165mm runners, but also very interesting to see what the 65mm did on the graph, but hey the blue guys were cool 🤓🙌🏼 Thanks for always being our 4age mythbuster
my 45 stock double have 40mm and a bit and they have some kind of seperator in front of the venturi.. thats why it has top end and my car cant feed it due to small port stock head... my 38 4/2/1 to 45 exhaust will stay.. but no room for longer trumpets.maybe some semi diagonial like those alfa romeo junior ones will help . if you ever find a 45 carbed please try some with curve on trumpets because engine is on ke70 and its not diagonial but full horisontal so even less space at inlet side and even more at headers but made me weld handmade stainles headers due to no ready manifold is made to pass the right steering rack system and medium sandwitch with oil cooler ... was a headache but it worths work on 4a ///
Same for runner lengths with dual plane intake manifolds.. or the plastic intakes like youd find on an LS. Longer runners gives you more top end shorter ones low end. Its however you want to tilt the curve. the last generation of 5.7L Hemis had a valve system to switch between long and short, so it was the best of both worlds.
I love your videos and it's your fault i'm putting ITB's on my car. I learned alot from your videos and soon I'll be able to research a good trumpet design for my application! Thank you so much for the time and effort put on all your videos, I've been here for years now lol Hope someday I can grab some merch too! Much love from brazil!
Nice to see a fellow brazuca round these parts! I'm trying to learn how to build a proper ITB system too, and this guy sure provides a lot of useful content
I think most people are forgetting that in bikes with very wide rev ranges and sensitivity to power, smoothing of the torque with different runner lengths(Variable or different length stacks) you may have some advantage in driveability. I think the reason it worked so well on the 787B was more due to the rotary engine which behaves much like a 2 stroke in that intake and exhaust design varies the power and torque far more. Would be awesome to do a test on a rotary to see if there are benefits, sadly I dont have one anymore or id offer it up for testing
Yeah they are both a different animal. can see it working better with more rpm and yeah rotary. I'd like to do some rotary stuff, but the rotary bills put me out of that game.
Yeah the fuel bill alone made me gap, even with a crap exhaust setup 160WHP at 6900RPM is pretty impressive for a 4 port RX8 motor which are fairly cheap@@Garage4age
I'd love to see one of those old-school ccv scavenge fittings on your exhaust. If you did it properly right at your header collector when it's the narrowest it could pull very nice vacuum into your valve cover. Awesome video
electric water pump conversion with pwm control like the Davis Craig pumps, gut the mechanical pump internals so you can still run it as a pulley if needed. should pick up a few ponies in the top end.
my 2008 Yamaha R1 has IBTs VTIR and its just amazing the difference it makes compared to my 2004 R1. They almost have the same peak HP but all the way to peak the 2008 is making more power, especially down low. It feels like a whole different beast of a bike.
love this channel. My 1G has equal length headers, variable intake runner length, and vvti. When it hits peak power of 160hp oh boy 🤣 ur nuts start quaking buddy
Very nice now I don't have to worry about trying to design some crazy telescopic system in my tight engine bay. Would be very interesting to see different length runners being used at the same time like @dantevito1193 said
I like the idea of the variable intakes. Harmonics battle I guess that's why toyota used yamaha! Temperature is what affects the speed of sound in air. Harmonics are really crazy how much they affect everything. Maybe all those hippies are not crazy! haha I know there is something about harmonics and motion where you can get the same power back of the motion, by playing the right side of harmonic tones to it. Wouldn't that be cool if we could make that work on the mighty 4age haha
Even though 9k is a pretty crazy redline for a car, i don't think it's high enough to get a noticeable difference. Bikes rev much higher. R6s redline close to 18k and come with variable stacks from the factory, but the trumpet shape leaves some power on the table that can be gained via replacement stacks that retain the variable length system. Some aftermarket stacks claim from 7-14rwhp, on a 95whp engine that's pretty nuts.
Long runners are quite limited on lenght there, usualy they go over the top on short places like that. You could go extra long and short and have flap openin for short side at 5000rpm or like it is in many cars but maybe in this engine thats lives in high rps that is unnessacry :D
This engine is running pretty big overlap. Interestingly the length of my normal intake works quite well on a stock cam engine at lower rpm also. ua-cam.com/video/kESMOV0sRkE/v-deo.html
@@Garage4age Ever thought about making and selling some of those 'normal' runners? Or would they only work with the BMW ITBs or whatever you are using?
R26b rotary pictured in thumbnail has very diff characteristics. Not saying these experiments aren’t worth while, just saying they may be more beneficial on a rotary
Great work as always! I wonder if you could provide the length from the intake valve head to the junction where you put the intake runners? If I know the total length from bellmouth to intake valve head I can provide you with some wave acoustics calculations. Thanks, Martin
For shits and giggles you could try and curve the intake ontop of itself using 45s or 90 to make it longer and see if there is any bottom end to be found at all from a longer intake.
that pull from 7:06 to 7:16 gives me a fuckin goosebump! Bro that sound! Imagine if it cuts the ignition and make that RAN DAN DAN DAN DAN DAN (cutting sound like a motorcycle)! THAT WILL BE A PLEASURE, INSANE AF!!!!!!!!!!!!!!!!
You mean the stock 4AGE intake on the MKI MR2? T-VIS. Those work by keeping the velocity up. One runner at low rpm, both at 4k ish IIRC. Variable runner length works, just maybe not in this situation, where everything is already pretty short. I think... I would use long trumpets, connect 2 and 2 in the middle with a tube each, buttefly valve in the 2 pipes for the first stage, second stage all four trumpets connected across, near the base of the trumpet, with 3 butterflys to separate them. One spring with a rod connecting them, one valve per stage to operate them. At idle they all draw air full length, mid rpm 2 pistons share with each other and effectively shorten, high rpm all 4 cylinders can draw from 3 directions effectively shortening the length again. The butterfly valves don't need to form a perfect seal, they just slow the sharing. It works. I've got a similar system from Mazda on a V6. Adapted it as well as I could describe it to work for 4 cylinders. A dyno would not show an increase in power, however it keeps tq up for a wider range. You definitely feel a loss if it stops working. The engine will feel flat. You won't make more power but it feels like it always has some left. For a controller I'd use an Arduino, they're 10 bucks, easy to program, open source. Some guy made a tach for his car with one... linked the code to the UA-cam video. The code could be copied and adapted to program a variable intake or anything rpm related with a clean tach signal. Another thing lacking in this dynoed setup is resonance. Since there is no intake box or boxes connecting the cylinders here, there is a loss of reflected air when the cylinders valves slam shut. Air builds up on the back of a suddenly closed valve, the air stacks and then springs back the other direction in the runner, which could be used to help charge another cylinder that is open. Not having a connection it just goes back at the incoming air in each trumpet and is wasted energy. It usually only does this well at a certain rpm, but can also have effect at double that rpm, half that rpm, resonant frequencies. Mazda KL v6 if you're interested, there were 4 different intakes each one tuned a little different, different rpms for the stages.
make the intakes much bigger then taper down to the size of the TBs. have them hook upwards for clearance from the fender and you can alternate them like a VR style cylinder spacing to make the opening of the trumpets huge. This should really help with intake velocity
Have you looked at the yamaha r1 (2007) variable stacks? You get a choice of two runner lengths eg your 65 plus a 100 extension... Not a huge packaging loss either
There's a study for an intake runner, for one cylinder, with that two stroke exhaust type double inverted come. You should take a look. Also, the absolute perfect runner needs to modify length and section, just like those Chinese traps!
how about making a turbine shaped variable length intake manifold? basically like an LS manifold, with the core rotating inside, opening the runners to a longer path when needed. would be quite bulky to print, but if you can fit a motor at the back and make the turbine twist downwards or upwards, there's 200 millimeters to fit 40mm runners and the core.. considering wall thickness and a taper, the core should be near 80mm ID... that would give you runners that would be be 30-300mm long. (170ish mm runner twist diameter*3.14, minus the taper at the end) and to help velocity and flow, the body opening would also be on a taper, starting at 60% width of the runner and opening to a full width at 180 till 270°
That's more or less what I thought, length makes a difference but not anywhere near as much as you think unless they are stupidly long like some OEM cars with runners 300-400mm+ which really suck at any decent rpm. But the min diameter is very important. My engines (3800 series 1) have a tapered trumpet one piece cast common plenum intake with a pretty short runner of something around the 150-170mm mark iirc. But due to their very decent taper, they get quite large quickly after the head flange. This explains why they make much better power and power curve NA than the later 3800 II engines despite the limited and smaller valves, as they have a very tiny mostly un-tapered runner, even though their length is more ideal for most combos. I measured the runner of the series II and it's like 30mm diameter for the most part.
engine harmonics disturb flow allot more than one might thing. it's why you find OEM air boxes isolated from the engine usually fastened to the body with a soft rubber marrying it to the throttle body. you should have those runners bolted to the body with soft rubber joining the carbs. harmonics can reduce runner surface area by 40% the whole back pressure argument is also related to vibration disturbances of the exhaust flow
I’ve watched your videos for a while now. Very cool experiments with lots of intriguing ideas. I had a look at your original post and I’m wondering what you used to move the stacks up and down? I think this would be great on my dragster.
was just a windscreen wiper motor, with switch at each end telling it to stop, controlled by the ecu. the 16year ago me built it , it was too slow and unreliable. I was always going to build a better setup. for a kinda end game thing, once went as far as could with everything else. but yeah, got put in the not worth doing basket
I bet if you did a 400-500 mm length at low rpm, you'd pick up some power down low. Obviously you got a cam made for the top end so a smaller cam would really Benefit from long runners but I think it could still help. Dodge made some crazy intakes back in the 70s and one of them was the cross ram. The driver side head was pulling air from the passenger side and the passenger side head was pulling air from the drivers side. Crazy power for the street but they were pretty much dead after 3500 rpm.
I’m late to the party but; I’ve watched quite a few of your videos, and it seems your combination is really well optimized at this point / not much low hanging fruit to be had without internal changes…..I’m not sure if you’ve considered this; but a 3-D printed plenum (to fit over your current ITB setup) could possibly result in some mid-range gains……the attenuation of the 2nd / 3rd order standing waves that hurt power at non-ideal rpm-ranges can be weakened / smoothed out by interference with the waves of other cylinders interacting on overlap…….the shape / volume that is most effective may /may not fit with straight runners, BUT it’s another rabbit hole to be jumped down…….I mention this because you’ve made more power with the hood closed vs. open; signifying there may be merit in having closer surfaces to reflect intake pulses off of.
Apparently, there is a pedal to the floor and we went to measure the power. The tests are extremely interesting, but not complete. Here, either the second row of throttles or pressing the pedal will gradually help to remove the dips in the initial stages, which is indirectly confirmed by a more even characteristic with narrow pipes. В любом случае - крайне интересная информация
Super long runner for a torque monster! also can you dyno part throttle pulls to simulate drive ability? some of these setups might be good a full throttle but a dog to drive. There are also fuel chillers, oil weight. Just some ideas. Great testing as always.
I always get something out of the test results for my jetski .I keep going the wrong way hoping to see a different result topped as well .I have worked out restriction in the exhaust doesn’t like a tight lsa on the cam timing.Power has doubled on the new set to a point i actually have get out of it on a windy day otherwise I’m going to be in trouble.
Would having both short and long runners in different cylinders improve anything? Honda did it in F1 in the 60s (i think it was the RA271, it had short runners on 4 cylinders, mid runners on another 4, and long runners in the last 4)
i also wonder about this, as some run a mix of long and short stack on dcoe carbs. where every 2nd cyl get a short and long one on I4 engines and i wonder if it's a way to compromise
Yea most motorcycles have this feature, basically flattens the curve a bit and possibly saves a few mm of space in the airbox, would be interesting to see here
@@brendanlist4811 A lot of bikes use a cross plane crank, not flat plane and they are also trying to smooth power as they have far less traction. For a car that has far more traction than power, it is not an issue especially when you have so much torsion damping along the driveline. Chains don't allow for torsion damping as an axle does.
As above answers, it can be used to smooth out the power curve. as you can see in these tests the average power would be touch lower, if you had to pair up with the less the ideal runner to smooth the curve. I have a video testing 2 different lengths on my channel page
Very interesting. What longer lengths actually make a difference? Would a u shape like a trombone tube work any better for allowing longer length runners? I mean, Look at how 2 cycle exhaust pipes are tuned with and shaped for different applications.
can you maybe test these with a pipercrosd filter? would be really intrested how much power they take from your setup. awesome video. always love when you experiment on crazy concepts
I'm guessing if you could fine tune the torque release of the dyno (not sure how) but you would see a smoother and higher output. Are you able to do a roll down run (brake dyno)? Start at Redline and slowly increase the torque applied by the dyno. The peaks would be your actual torque/power and the drops would be the dyno letting off to allow the rpm to increase.
This Engine makes the same power over a pretty big afr window, (is a vid on my page showing it) so dont usually bother correcting afr unless there is a big change in power /afr. I usually have it set in mid 12's, so most changes will stay within 13 to 12 afr, if makes more or less power. which is well within the "window" If goes out of that window i usually do some corrections. Here it is : ua-cam.com/video/yjvPmG123nI/v-deo.html
I also thought variable length would make more difference than that but you've found the answer without realising it. Stick with the length of the normal setup and make it variable diameter ?
im building a 1986 aw11. just for street/autocross atm but it would be fun to race someday, do you have a parts list or something for the good4a or a budget 4a build?
Boss video. What is the average port diameter, and throttle ID? Also what is the large runner ID? Trying to determine the restrictive point when driving the 35mm ID runner (obviously it is but i don't know the rest of the ID's of the induction system)
only compared to a smallport head that already had a little work done to it. the original turd engine was smallport if go back and watch them. running the same inlet manifolds and both heads stock would be very little in it.
Would you do a video of different throttle sizes going into the 4a head? (20v 4age itb’s are becoming unobtanium status when motorcycle throttles are still plentiful)
@@mikemikeyee i was running blacktop throttles with 2mm oversized 47mm plates up to 150kw. changing to the bmw 52's was no gain in power at that level. i haven't tried anything smaller at that power level. but can say a 40mm runner in front of them chokes the engine pretty bad. so silvertop 43's would be a no go considering the area the plate takes up. blacktop 45's probably pretty borderline at that power level. but also depends where they are placed, further away from the head need to go bigger. there are some pretty cool 47mm bike e throttle throttles, i forget what the come off now. but they would be the go. Also thanks for the thanks
The tests with orange pipes were from before the engine had vvti. But yeah the vvti lets you kinda effectively shuffle the length bit. since you are changing the valve closing time. vs the ideal length, can change the length a little bit and then retune the vvti to some of the power back. but the ideal runner still wins. intake valves are 32mm
What happens when you straighten out the airflow going in vs. creating turbulence or purposely spinning the air into a vortex on the inlets. Forgive me if there is already a video covering that, but I am super curious what vortices or turbulence in key areas can do to help or hurt power delivery. For example, a small hump on each side of the inlet to bounce air up and around the throttle blade. Things like that.
I actually have basic video just testing with bonnet (yeah we call it that here) open /shut. as people were saying to run the tests with it shut. I run it open as its easier to keep all the temps the same, cut down the variables and keep the results more consistent, on the part im testing. it actually made a touch more power🤷♂ ua-cam.com/video/I70KtxG4hWM/v-deo.html Is also a video on my channel somewhere i hooked a dyno fan up to the intake..
Do you mean the same shape but just a different material? I’d guess the (slightly) greater heat transfer from the head into the incoming air would make a less dense charge, but not enough to even be measurable
Great channel mate. 😃 BMW used a variable length manifold (Disa/Diva) that had long and short runners sharing the same plenum. Air flow would be directed by a valve operated by a vacuum actuator.
I'm actually debating if I should keep the DISA intact or not, in my E46. To date, I haven't found any dyno tests that actually show how it affects performance. Just a bunch of speculation.
@@michaelblacktree Dyno results show that at low RPM it helps power, once its shut its no difference. If its a road car keep it, if its a race car get rid of it.
Nice, but I was really hoping to see a sliding barrel (change tube length) or a moving extension (v-stack pushed into v-stack). The following Ferrari has an intake with trumpets that slide downward with RPM to alter runner length. ua-cam.com/video/MRZW31F7McY/v-deo.html
I like how your crank vent nut comes more loose with each run. 😂
It's fine, re-tightens itself on the rundown!
yooo I rushed to the comment section because I saw that an fitting doing the danger dance
The best fact/science based dyno videos on the web. Please keep them coming!
OK, so in the day I raced Toyota Atlantic. I've got a lot of time playing with intake trumpets on a 4A-GE. You definitely want the trumpet on the end, but I understand why you tested like you did. We would run as short as 50mm trumpets on ovals when we were only concerned about 9200-10000 rev range. We would run from 80 to 150mm trumpets depending on if we were looking for more high end or more mid range. In the last few years, the best road course / street course setup was with cylinders 1-4 about 30mm shorter than cylinders 2-3. A high rev track might be 80/110 while a street circuit might be 130/100. It seemed to help mid-range without hurting the top end. We would also run a tri-y header on everything except the ovals, which was 4-into-1. The airbox sizing is very important. You're missing something if you haven't gone down that bunny-hole.
Interesting, thanks. Numbers sound similar if consider the difference in the itb section length and I'm targeting a lower rpm. Goal for this engine was always to have good top end while having most power possible through the middle. It did have a more aggressive cam combo at one staged that made it more similar to Atlantic power curve (the one i had on my dyno at least) but now runs a smaller cam combo and vvti which is a game changer for a street car
That spinning AN line nut was driving me crazy 😂
Another good movie and a lot of work. Best begards. In motorcycles, the higher the runner rpm, the actuator shortens it. In motorcycle engines, if the producers wanted high torque in low parts, they used a narrow and long intake to accelerate the air flow and cylinder filling, but if these were high-performance engines where the lower parts were not as important as the high rpm parts, a short intake was used. To increase the torque at the bottom of these engines, at first a throttle was used in the exhaust system, and later a variable intake system was added. In the cases I have encountered, double fuel injection is used with variable intake. One injector in the intake and the other above the runner. Finally, one important thing that is used in all these engines and affects the operation of the engine is the airbox.
Well done for conclusively dismissing the calls for variable intake length (and yes I was one of the ones curious about it)! Its almost like the designer of the G4A knows what he / she was doing.
I'm a he, last time i checked haha. Thanks
@@Garage4age Well that got a bite!
The bike variable inlet where the longer trumpets are lifted and have shorter trumpets underneath seems easiest to implement in this case. And with less disturbance for max power.
I think Suzuki do this on the gixxer, you could lift 100mm of runner to turn the 165mm into 65mm runners at the rpm cross over point
That makes sense when the bike is used in the 7000 to 14000 rpm range. This engine only needs to cover 5000 to 9000 approximately.
The difference in runner length is probably too small to make that work, and the shortest length isn't that short.
A different way is to use resonances. Many car intake manifolds employ this approach. One of the most complex is that which was fitted to the Ferrari 458, itself an evolution and complication of the earlier engines.
But the style they adopted for their V12s might give some inspiration, and is more appropriate here. The idea is that a valve opens up in the manifold to an extra chamber at one end, increasing the volume of the plenum. This changes its resonance(s) and hence cylinder filling in the same way runner length can.
I don't know if this engine normally uses a plenum over the bellmouths, but it's generally good practice even with ITBs. "Unfortunately" it's another thing to tweak and tune. We've sort of seen this effect with the K24 EP already.
Yes this design works well, proven, reliable. But i think a sliding single trumpet is just best. it is the no compromise solution. As far as area under the horsepower curve, we are talking a few % at most either way. So having a good cost effective setup like the double trumpets makes plenty of sense.
Miata NB1 intake also does this (not NB2, which instead has a VVI)
we could test it but the goal is to reach the second harmonic… if we use a bike such as the yamaha r1 for an example and i put in some vague non accurate numbers because i dont know what yamaha has (and because yamaha knows air - beams engines toyota engines ford sho v6 engines etc)
2007 yamaha r1 - has variable itb. 13,700rpm rev limit… usable power i dont know, google a dyno graph.
second harmonic wave is the goal. divide that by the rpm you want your second harmonic to come on song… and thats your required total runner length in inches from the beginning of the opening to the end of the nearest side valve face (ill use valve face because i cbf figuring out if they use the outer side or average it or another part of the end of the valve area)
so… 132,000 - the second harmonic, divided by… lets say the rpm of the yamaha - 13,700
9.6 inches.
the yamaha itb looks like its around 6 inches total, with the rest of the length in the port to the valve… probably (this is all just to get things in perspective, nothings accurate)
so you target that, then go longer or shorter and test and test and test… until you get what you want.
this is where the question lies with the 4age and any other car… can u make a variable itb 13.2 inches long, or longer?
because the length gets a lot longer if you want your varying itb to generate second harmonic torque from 5000rpm… 26.4 inches…
will it fit? will it work? or will it ake uhhh… 4ages for anything to happen?
if you go shorter, for the 3rd harmonic… can you make your long early torque generating variable itb shrink down to the right length at high rpm?
will they hit anything if you pop the long runners off like how it is in the yamaha?
the r26b quad rotor has all the space… when i wanted to try that in a 2jz toyota chaser.. i realised i had no space.
but then theres real world… weve seen dyno tests already with ultra long runners tested here in another video, cut down bit by bit… and the results didn’t come anywhere near the power of the g4a itb setup when trying to aim for the second harmonic, even at 10 inches.
(and they were over 1 metre long to begin with - in an attempt to catch the first harmonic wave - the car didn’t want to rev out in some cases)
so what gives? runner diameter (and therefore, itb size) another factor perhaps? do cam specs need to match? does it need vtec and dual vvti? does the header need to be just as long to work with the runner length???
Wow! Thank you for this tests ❤ to some, like me who loves doing tests and dyno this is one of the best I’ve seen and most of the good ones I see are from your channel!
props to you and stay awesome!
I watched all of it and was feeling sorry for that little motor by the end of it! A lot of useful info there, I have a na 1.6 mx5 motor on throttle bodies and always wondered what changing lengths would do, thanks for putting this up.
Quite interesting to see that all youre test with various trumpet designs didn´t catch up with youre original setup.
What i have found was a difference when you design the whole inlet runner on a taper....quite better in top end power with little lost on low and mid range.
Was a qonsequent setup change from straight to a 7° taper
And for youre last test if the strut tower will make something different....
Once i hold my hand in front of the trumpets on the dyno, even with a quite short distance of 20mm you couldn´t feel THAT much force on it...wich concludes that most of the air will be sucked in from the side (why a correct bellmouth is so important)
I always enjoy watching youre videos as i have seen similarity to things i have also done before.....and to often throw them in the "not worth bin"
Awesome work as always a Boet, I only prefer the 165mm runners, but also very interesting to see what the 65mm did on the graph, but hey the blue guys were cool 🤓🙌🏼
Thanks for always being our 4age mythbuster
my 45 stock double have 40mm and a bit and they have some kind of seperator in front of the venturi.. thats why it has top end and my car cant feed it due to small port stock head... my 38 4/2/1 to 45 exhaust will stay.. but no room for longer trumpets.maybe some semi diagonial like those alfa romeo junior ones will help . if you ever find a 45 carbed please try some with curve on trumpets because engine is on ke70 and its not diagonial but full horisontal so even less space at inlet side and even more at headers but made me weld handmade stainles headers due to no ready manifold is made to pass the right steering rack system and medium sandwitch with oil cooler ... was a headache but it worths work on 4a ///
Same for runner lengths with dual plane intake manifolds.. or the plastic intakes like youd find on an LS. Longer runners gives you more top end shorter ones low end. Its however you want to tilt the curve. the last generation of 5.7L Hemis had a valve system to switch between long and short, so it was the best of both worlds.
I love your videos and it's your fault i'm putting ITB's on my car. I learned alot from your videos and soon I'll be able to research a good trumpet design for my application! Thank you so much for the time and effort put on all your videos, I've been here for years now lol Hope someday I can grab some merch too! Much love from brazil!
Nice to see a fellow brazuca round these parts! I'm trying to learn how to build a proper ITB system too, and this guy sure provides a lot of useful content
I think most people are forgetting that in bikes with very wide rev ranges and sensitivity to power, smoothing of the torque with different runner lengths(Variable or different length stacks) you may have some advantage in driveability. I think the reason it worked so well on the 787B was more due to the rotary engine which behaves much like a 2 stroke in that intake and exhaust design varies the power and torque far more. Would be awesome to do a test on a rotary to see if there are benefits, sadly I dont have one anymore or id offer it up for testing
Yeah they are both a different animal. can see it working better with more rpm and yeah rotary. I'd like to do some rotary stuff, but the rotary bills put me out of that game.
Yeah the fuel bill alone made me gap, even with a crap exhaust setup 160WHP at 6900RPM is pretty impressive for a 4 port RX8 motor which are fairly cheap@@Garage4age
@@Garage4age If I ever stumble upon large sums of money, I would 100% fund your rotary sidequests. 😂
Always great and interesting. Variable Intakes are fascinating
Love the "Made in NZ" on Cilinder 4's orange pipe.
the power ducts!
I'd love to see one of those old-school ccv scavenge fittings on your exhaust. If you did it properly right at your header collector when it's the narrowest it could pull very nice vacuum into your valve cover. Awesome video
electric water pump conversion with pwm control like the Davis Craig pumps, gut the mechanical pump internals so you can still run it as a pulley if needed. should pick up a few ponies in the top end.
my 2008 Yamaha R1 has IBTs VTIR and its just amazing the difference it makes compared to my 2004 R1. They almost have the same peak HP but all the way to peak the 2008 is making more power, especially down low. It feels like a whole different beast of a bike.
love this channel. My 1G has equal length headers, variable intake runner length, and vvti. When it hits peak power of 160hp oh boy 🤣 ur nuts start quaking buddy
Very nice now I don't have to worry about trying to design some crazy telescopic system in my tight engine bay. Would be very interesting to see different length runners being used at the same time like @dantevito1193 said
Excellent work keep going!
I like the idea of the variable intakes. Harmonics battle I guess that's why toyota used yamaha! Temperature is what affects the speed of sound in air. Harmonics are really crazy how much they affect everything. Maybe all those hippies are not crazy! haha I know there is something about harmonics and motion where you can get the same power back of the motion, by playing the right side of harmonic tones to it. Wouldn't that be cool if we could make that work on the mighty 4age haha
Another good comparison... Nice!
Even though 9k is a pretty crazy redline for a car, i don't think it's high enough to get a noticeable difference. Bikes rev much higher. R6s redline close to 18k and come with variable stacks from the factory, but the trumpet shape leaves some power on the table that can be gained via replacement stacks that retain the variable length system. Some aftermarket stacks claim from 7-14rwhp, on a 95whp engine that's pretty nuts.
Thanks!
Always interesting to see your vids!
This is where intake gimmics go to die, Love it!
Long runners are quite limited on lenght there, usualy they go over the top on short places like that. You could go extra long and short and have flap openin for short side at 5000rpm or like it is in many cars but maybe in this engine thats lives in high rps that is unnessacry :D
Different cams may produce different results. I can see cams with high intake/exhaust overlap having much more of an impact with intake length.
Lots of overlap is kinda the point of itb's
This engine is running pretty big overlap. Interestingly the length of my normal intake works quite well on a stock cam engine at lower rpm also. ua-cam.com/video/kESMOV0sRkE/v-deo.html
@@Garage4age Ever thought about making and selling some of those 'normal' runners? Or would they only work with the BMW ITBs or whatever you are using?
Awesome job as always.
R26b rotary pictured in thumbnail has very diff characteristics. Not saying these experiments aren’t worth while, just saying they may be more beneficial on a rotary
Great work as always! I wonder if you could provide the length from the intake valve head to the junction where you put the intake runners? If I know the total length from bellmouth to intake valve head I can provide you with some wave acoustics calculations. Thanks, Martin
For shits and giggles you could try and curve the intake ontop of itself using 45s or 90 to make it longer and see if there is any bottom end to be found at all from a longer intake.
anyone else just watching the running nut on the rocker cover turn
my Ford Focus ST170 has a dual runner length inlet manifold from factory.
it is a 2.0L 16V engine which makes peak power at 7000rpm.
that pull from 7:06 to 7:16 gives me a fuckin goosebump! Bro that sound!
Imagine if it cuts the ignition and make that RAN DAN DAN DAN DAN DAN (cutting sound like a motorcycle)! THAT WILL BE A PLEASURE, INSANE AF!!!!!!!!!!!!!!!!
Build a intake manifold with dual runners. Have a valve that switches from the long runners to short runners at higher rpm.
You mean the stock 4AGE intake on the MKI MR2? T-VIS. Those work by keeping the velocity up. One runner at low rpm, both at 4k ish IIRC. Variable runner length works, just maybe not in this situation, where everything is already pretty short. I think... I would use long trumpets, connect 2 and 2 in the middle with a tube each, buttefly valve in the 2 pipes for the first stage, second stage all four trumpets connected across, near the base of the trumpet, with 3 butterflys to separate them. One spring with a rod connecting them, one valve per stage to operate them. At idle they all draw air full length, mid rpm 2 pistons share with each other and effectively shorten, high rpm all 4 cylinders can draw from 3 directions effectively shortening the length again. The butterfly valves don't need to form a perfect seal, they just slow the sharing. It works. I've got a similar system from Mazda on a V6. Adapted it as well as I could describe it to work for 4 cylinders. A dyno would not show an increase in power, however it keeps tq up for a wider range. You definitely feel a loss if it stops working. The engine will feel flat. You won't make more power but it feels like it always has some left. For a controller I'd use an Arduino, they're 10 bucks, easy to program, open source. Some guy made a tach for his car with one... linked the code to the UA-cam video. The code could be copied and adapted to program a variable intake or anything rpm related with a clean tach signal.
Another thing lacking in this dynoed setup is resonance. Since there is no intake box or boxes connecting the cylinders here, there is a loss of reflected air when the cylinders valves slam shut. Air builds up on the back of a suddenly closed valve, the air stacks and then springs back the other direction in the runner, which could be used to help charge another cylinder that is open. Not having a connection it just goes back at the incoming air in each trumpet and is wasted energy. It usually only does this well at a certain rpm, but can also have effect at double that rpm, half that rpm, resonant frequencies.
Mazda KL v6 if you're interested, there were 4 different intakes each one tuned a little different, different rpms for the stages.
love the car. Long ones sound so nice as we all know :)
make the intakes much bigger then taper down to the size of the TBs. have them hook upwards for clearance from the fender and you can alternate them like a VR style cylinder spacing to make the opening of the trumpets huge. This should really help with intake velocity
Have you looked at the yamaha r1 (2007) variable stacks? You get a choice of two runner lengths eg your 65 plus a 100 extension...
Not a huge packaging loss either
There's a study for an intake runner, for one cylinder, with that two stroke exhaust type double inverted come. You should take a look.
Also, the absolute perfect runner needs to modify length and section, just like those Chinese traps!
Great Scott!
super clean starlet. good job.
how about making a turbine shaped variable length intake manifold? basically like an LS manifold, with the core rotating inside, opening the runners to a longer path when needed. would be quite bulky to print, but if you can fit a motor at the back and make the turbine twist downwards or upwards, there's 200 millimeters to fit 40mm runners and the core.. considering wall thickness and a taper, the core should be near 80mm ID... that would give you runners that would be be 30-300mm long. (170ish mm runner twist diameter*3.14, minus the taper at the end) and to help velocity and flow, the body opening would also be on a taper, starting at 60% width of the runner and opening to a full width at 180 till 270°
At that point just intall a turbo and ram cold air intake.
That's more or less what I thought, length makes a difference but not anywhere near as much as you think unless they are stupidly long like some OEM cars with runners 300-400mm+ which really suck at any decent rpm. But the min diameter is very important.
My engines (3800 series 1) have a tapered trumpet one piece cast common plenum intake with a pretty short runner of something around the 150-170mm mark iirc. But due to their very decent taper, they get quite large quickly after the head flange.
This explains why they make much better power and power curve NA than the later 3800 II engines despite the limited and smaller valves, as they have a very tiny mostly un-tapered runner, even though their length is more ideal for most combos. I measured the runner of the series II and it's like 30mm diameter for the most part.
Hero of the people
engine harmonics disturb flow allot more than one might thing. it's why you find OEM air boxes isolated from the engine usually fastened to the body with a soft rubber marrying it to the throttle body.
you should have those runners bolted to the body with soft rubber joining the carbs.
harmonics can reduce runner surface area by 40%
the whole back pressure argument is also related to vibration disturbances of the exhaust flow
I’ve watched your videos for a while now. Very cool experiments with lots of intriguing ideas.
I had a look at your original post and I’m wondering what you used to move the stacks up and down?
I think this would be great on my dragster.
was just a windscreen wiper motor, with switch at each end telling it to stop, controlled by the ecu. the 16year ago me built it , it was too slow and unreliable. I was always going to build a better setup. for a kinda end game thing, once went as far as could with everything else. but yeah, got put in the not worth doing basket
I bet if you did a 400-500 mm length at low rpm, you'd pick up some power down low. Obviously you got a cam made for the top end so a smaller cam would really Benefit from long runners but I think it could still help. Dodge made some crazy intakes back in the 70s and one of them was the cross ram. The driver side head was pulling air from the passenger side and the passenger side head was pulling air from the drivers side. Crazy power for the street but they were pretty much dead after 3500 rpm.
Great stuff, as usual. You make me sad about how USDM Starlets are impossible to find.
I’m late to the party but; I’ve watched quite a few of your videos, and it seems your combination is really well optimized at this point / not much low hanging fruit to be had without internal changes…..I’m not sure if you’ve considered this; but a 3-D printed plenum (to fit over your current ITB setup) could possibly result in some mid-range gains……the attenuation of the 2nd / 3rd order standing waves that hurt power at non-ideal rpm-ranges can be weakened / smoothed out by interference with the waves of other cylinders interacting on overlap…….the shape / volume that is most effective may /may not fit with straight runners, BUT it’s another rabbit hole to be jumped down…….I mention this because you’ve made more power with the hood closed vs. open; signifying there may be merit in having closer surfaces to reflect intake pulses off of.
I think you should try placing a 45-degree angle on cylinders 1 and 3, and the tubes for cylinders 2 and 4 straight and shorter.
Try a double throttle, to compensate the inlet back pressure on 4k and 4,7k rpm power drop.
Apparently, there is a pedal to the floor and we went to measure the power. The tests are extremely interesting, but not complete. Here, either the second row of throttles or pressing the pedal will gradually help to remove the dips in the initial stages, which is indirectly confirmed by a more even characteristic with narrow pipes. В любом случае - крайне интересная информация
Thanks for good content
Nice work as always mate!
7age are juuuuuuuunnnnnk
Maybe on rotary engines like 787b on preview pic resonance effects may vary from 4stroke piston engines and more similar like 2stroke….
yeah definitely a different animal.
alot of grate info here had a hard time watching keep up the hard work
Awsome topic. Thanks for sharing 👍
nice approach, man
1:51 the fitting moving lol
it's not crazy when I know yamaha have been doing this since 2009 but not in a linear way more on and off around 10k on the R1
Super long runner for a torque monster! also can you dyno part throttle pulls to simulate drive ability? some of these setups might be good a full throttle but a dog to drive. There are also fuel chillers, oil weight. Just some ideas. Great testing as always.
You dont show AFR on these charts. is your target AFR always the same?
I would be very interested to see a test with hood closed therefore potentially changing air temp, and one with cold air box connected 😊
Is a vid on my page testing with open/shut. closed actually makes more power.
I always get something out of the test results for my jetski .I keep going the wrong way hoping to see a different result topped as well .I have worked out restriction in the exhaust doesn’t like a tight lsa on the cam timing.Power has doubled on the new set to a point i actually have get out of it on a windy day otherwise I’m going to be in trouble.
Perhaps a collector plenum/hat ,like some bmws engines, could make a difference.
Would having both short and long runners in different cylinders improve anything? Honda did it in F1 in the 60s (i think it was the RA271, it had short runners on 4 cylinders, mid runners on another 4, and long runners in the last 4)
i also wonder about this, as some run a mix of long and short stack on dcoe carbs. where every 2nd cyl get a short and long one on I4 engines and i wonder if it's a way to compromise
Yes actually. It smooths out the peakiness of the torque curve because the cylinders resonate at different frequencies.
Yea most motorcycles have this feature, basically flattens the curve a bit and possibly saves a few mm of space in the airbox, would be interesting to see here
@@brendanlist4811 A lot of bikes use a cross plane crank, not flat plane and they are also trying to smooth power as they have far less traction. For a car that has far more traction than power, it is not an issue especially when you have so much torsion damping along the driveline. Chains don't allow for torsion damping as an axle does.
As above answers, it can be used to smooth out the power curve. as you can see in these tests the average power would be touch lower, if you had to pair up with the less the ideal runner to smooth the curve. I have a video testing 2 different lengths on my channel page
Very interesting. What longer lengths actually make a difference? Would a u shape like a trombone tube work any better for allowing longer length runners? I mean, Look at how 2 cycle exhaust pipes are tuned with and shaped for different applications.
can you maybe test these with a pipercrosd filter? would be really intrested how much power they take from your setup. awesome video. always love when you experiment on crazy concepts
He did a couple videos back, keep watching them and you'll find it!
Reminds me of VICS on the 2nd gen miata intakes.
Ďakujeme.
I'm guessing if you could fine tune the torque release of the dyno (not sure how) but you would see a smoother and higher output. Are you able to do a roll down run (brake dyno)? Start at Redline and slowly increase the torque applied by the dyno. The peaks would be your actual torque/power and the drops would be the dyno letting off to allow the rpm to increase.
Already Tried to Airbox them like an e30 M3 or 190 evo in the DTM
Make ideal shaped trumpets, but make them out of rubber so they can be stretched
I think longer is where you be finding resonance at streetable rpms
Do you correct air fuel ratio, or there is no need following the intakes you tested here ?
Also,what is the air fuel ratio in those tests ?
This Engine makes the same power over a pretty big afr window, (is a vid on my page showing it) so dont usually bother correcting afr unless there is a big change in power /afr.
I usually have it set in mid 12's, so most changes will stay within 13 to 12 afr, if makes more or less power. which is well within the "window" If goes out of that window i usually do some corrections.
Here it is : ua-cam.com/video/yjvPmG123nI/v-deo.html
@@Garage4age Thanks.
Now it's time for spark plugs test.
Awesome, I love this testing. Have you tried a plenum?
try some singh grooves, perfect for the low range gains
I also thought variable length would make more difference than that but you've found the answer without realising it. Stick with the length of the normal setup and make it variable diameter ?
im building a 1986 aw11. just for street/autocross atm but it would be fun to race someday, do you have a parts list or something for the good4a or a budget 4a build?
these 2 vids should give you some ideas ua-cam.com/video/Ik7tK_zmZCM/v-deo.html ua-cam.com/video/CM_xHLRQCdg/v-deo.html
Boss video.
What is the average port diameter, and throttle ID?
Also what is the large runner ID?
Trying to determine the restrictive point when driving the 35mm ID runner (obviously it is but i don't know the rest of the ID's of the induction system)
have you ever directly compared a stock smallport head to a stock bigport head? I've only seen early BP head compared to the late BP head.
only compared to a smallport head that already had a little work done to it. the original turd engine was smallport if go back and watch them. running the same inlet manifolds and both heads stock would be very little in it.
Would you do a video of different throttle sizes going into the 4a head? (20v 4age itb’s are becoming unobtanium status when motorcycle throttles are still plentiful)
(Or how important are throttle sizes to a 4a head are)
@@mikemikeyee i was running blacktop throttles with 2mm oversized 47mm plates up to 150kw. changing to the bmw 52's was no gain in power at that level. i haven't tried anything smaller at that power level. but can say a 40mm runner in front of them chokes the engine pretty bad. so silvertop 43's would be a no go considering the area the plate takes up. blacktop 45's probably pretty borderline at that power level. but also depends where they are placed, further away from the head need to go bigger. there are some pretty cool 47mm bike e throttle throttles, i forget what the come off now. but they would be the go. Also thanks for the thanks
could you make the incoming air move in a twist like a tornado does with a special shape to the velocity stack. would that help performance.
Need to bring the hilux back and test intake designs with boost.
You also have variable oil sleeve nuts 🤣
probably canceling out the power changes 😂
You really need an engine dyno for this testing. You're a smart bloke, could you convert one of the sensors to an engine dyno.
Very comprehensive. Is the runner length effect not significant thanks to optimised VVT?
What size intake valves?
The tests with orange pipes were from before the engine had vvti. But yeah the vvti lets you kinda effectively shuffle the length bit. since you are changing the valve closing time. vs the ideal length, can change the length a little bit and then retune the vvti to some of the power back. but the ideal runner still wins. intake valves are 32mm
What happens when you straighten out the airflow going in vs. creating turbulence or purposely spinning the air into a vortex on the inlets. Forgive me if there is already a video covering that, but I am super curious what vortices or turbulence in key areas can do to help or hurt power delivery. For example, a small hump on each side of the inlet to bounce air up and around the throttle blade. Things like that.
do some tests with the hood closed, breathing hot air from the radiator 👍 make some machine pushing cold air into the tbi 👍
I actually have basic video just testing with bonnet (yeah we call it that here) open /shut. as people were saying to run the tests with it shut. I run it open as its easier to keep all the temps the same, cut down the variables and keep the results more consistent, on the part im testing. it actually made a touch more power🤷♂ ua-cam.com/video/I70KtxG4hWM/v-deo.html Is also a video on my channel somewhere i hooked a dyno fan up to the intake..
Maybe can try the 4age 20v style with the box see what is the result dyno
I'm curious if a carbon intake manifold makes a difference.. such as the e30 m3 dtm
Do you mean the same shape but just a different material? I’d guess the (slightly) greater heat transfer from the head into the incoming air would make a less dense charge, but not enough to even be measurable
Best fucking youtube chanel ever! Hello from France
Great channel mate. 😃
BMW used a variable length manifold (Disa/Diva) that had long and short runners sharing the same plenum. Air flow would be directed by a valve operated by a vacuum actuator.
Same design in the Toyota 3SGE
@@sigmamale4147 Mercedes m112 and 113 too. On the naturally aspirated versions of those engines.
I'm actually debating if I should keep the DISA intact or not, in my E46. To date, I haven't found any dyno tests that actually show how it affects performance. Just a bunch of speculation.
@@michaelblacktree Dyno results show that at low RPM it helps power, once its shut its no difference. If its a road car keep it, if its a race car get rid of it.
@@fellowship_of_the_goat - OK Captain Obvious. We know it helps low end torque. The question is, how much?
Not much info on the topic of this. Thank you!
Nice, but I was really hoping to see a sliding barrel (change tube length) or a moving extension (v-stack pushed into v-stack). The following Ferrari has an intake with trumpets that slide downward with RPM to alter runner length.
ua-cam.com/video/MRZW31F7McY/v-deo.html
been wondering... u know 2 stroke expansion chambers...
could they like work as intakes and create some black magic pressure waves :D
Hell yeah