Discussion on technical problems of multi cylinder intake manifold and carburetor

[sition]

Multi cylinder engines, such as in-line four cylinder engines, usually use carburetors for fuel supply. They are sensitive to the balanced length of the intake manifold. Have you ever had such an experience? The mixing ratio of a single cylinder is too rich, resulting in the cylinder not working and affecting the power。

 

[sition]

For example, for an in-line 6-cylinder engine, whether two cylinders near the carburetor will suffocate due to excessive fuel intake. I saw this video of the perfect in-line 6-cylinder engine, and it seems that this problem did not occur

 

[sition]

For carburetors, we often lazily use the rc carburetor as the fuel supply. At low speeds, the carburetor throttle is biased to one side, which will also lead to excessive gasoline on the other side, or this can be achieved through fine tuning, but it is always imperfect

 

[bluejets]

4 cylinder inline I built never had a problem and it was built with manifold same as the 6 above.
Many instances too much worry goes in to nothing.
Most are satisfied if the thing runs reliably, which is what my 4 cylinder does.

 

[peterl95124]

when I ran my inline 4 without the exhaust manifold so I could feel the exhaust from each cylinder separately it was obvious that the inner two cylinders nearest the carb were doing most of the work as the exhaust from the outer two cylinders were remarkably cooler. this was a non-standard inline 4 configuration where the valve order was in, ex, in, ex, in, ex, in, ex, instead of the more traditional ex, in, in, ex, ex, in, in, ex for which it is easier to make all the intake flow lengths the same to help ensure equal running in all four cylinders. I'll repeat this test on my De Havilland Cirrus when I get around to finishing it and report back again, as it uses the traditional order.

 

[Steamchick]

Hi guys. Just a few odd comments - based on experience with single carb car engines in the Manufacturer's design office.
One of the things that affects petrol (and other "wet" fuels) and air mixtures is the simple dynamics is a column of gas withe "wert" droplets in it. The gas goes around corners, but the droplets don't.., (Newton's laws of motion - Laws we can't avoid!) so the "wet" goes straight on and doesn't all turn the corner with the gas (air and vapour). Carburettors give a mixture of air plus fuel vapour, Aerosols (very fine droplets) and "wet droplets".
So consider a single carb feeding a single cylinder. When the valve opens, the air column accelerates, and as the accelerating air increases velocity in the carb it starts to draw up droplets of fuel. - Related to the square of the speed of air in the throat of the carb.... Initially the speed of air in the carb is too low to shred the droplets into smaller "aerosols" and heat has to flow into the aerosols to turn them from liquid to vapour.
(Hmmm.... getting a bit complicated, so I'll carefully describe the next steps):
When the carb is at "full gas flow" for the wide-open engine valve and mid-stroke of the piston, whatever the throttle valve is set at, there is a good amount of aerosol (and a bit of vapour) and not so much as "wet" droplets.
Then as the inlet valve closes, the column of gas slows, so we get more droplets against less aerosol fuel.
The combustion chamber heats the mix of air, aerosol fuel, vaporising some of the aerosol, and heating the droplets.
The spark ignites a fuel-air mix of vapour and oxygen in a volume of nitrogen.... and when correct the flame heats the aerosol further to assist igniting all the mixture. But (especially at higher engine speeds) the droplets don't all burn, just the "outer-part" so there is some residual hydro-carbon in the exhaust. - We have all seen the soot on older cars, even when we think the mixture is correct. What the plugs show is the mixture that has burned - at least as far as Carbon monoxide, not the results of all the fuel burning completely. Modern fuel injected engines with O2 sensors correct this "impossible process" from carburettors.
Now consider a manifold with centre carb and 2 legs going left and 2 to the right to feed 4 cylinders. - We will discuss one half. Say cylinders 1 and 2. These fire at some spacing, not necessarily 360 degrees apart on the crank - so maybe inlet of one, - then 180degrees crank - to inlet of next... then 240 degrees to inlet of the first? Oops! Sounds like we cannot expect the columns of air to have the same starting conditions when the inlet valve opens. - This imbalances the air per cylinder and thus the mixture.
But what about the fuel as it travels along a z-shaped passage to the valve? Going to the furthest cylinder, the droplets try and go straight on at the end of the manifold, instead of following the turn made by the air and aerosols into the cylinder. These droplets hit the inside of the end if the manifold, and eventually get sucked into the engine as larger, but hotter, droplets. That does NOT help good combustion of that bit of fuel. So we need a richer mixture to compensate for that "loss" of combustible fuel. (It probably burns in the exhaust pipe, not the cylinder!). But consider the column of air and aerosol to the nearest cylinder to the carb? The air and aerosol turning from the manifold to the inlet tract to valve and cylinder allows a lot of the fuel droplets to fly past the inlet tract and hang around in the relatively still air at the next cylinder inlet tract... (furthest from the carb). So Middle cylinders are thus starved of some of their fuel, which goes towards the outer cylinders...
Hence, we have a lot of factors that affect the individual mixtures in cylinders:
# Effect of a "stopped" column of gas accelerating to sucking adequate aerosol-fuel for the engine to run but getting droplets of "excess" fuel when the gas column is slow at the start.
# Imbalance of position of each cylinder "sucking" in degrees around the crankshaft = different "dwell times" between valve openings.
# Imbalances of position of cylinder along the length of manifold.
And in truth it is more complicated than that!
I worked alongside an engineer who's job was to make the inlet manifold castings with internal vanes to direct the fuel and air mix around the corners, to try and correct the "droplet" problems. - These problems "disappeared" when the manifold only had to carry air, because the fuel was injected at the inlet tract close to the inlet valve.

So "well done" anyone who manages to make a multi-cylinder engine with unequal inlets through the manifold run smoothly.
I hope I have not made this either too simple or too complicated for you? It really is more complicated than I describe when tuning production engines for the optimum performance and to meet emission laws.
K2

 

[sition]

4 cylinder inline I built never had a problem and it was built with manifold same as the 6 above.
Many instances too much worry goes in to nothing.
Most are satisfied if the thing runs reliably, which is what my 4 cylinder does.

Hello, Blue Jet! Thank you for your reply. I saw the video and your engine is running well. However, I heard the sound of the engine running. One of its cylinders seems useless. Of course, I just judge from the sound, maybe it's my illusion

 

[sition]

when I ran my inline 4 without the exhaust manifold so I could feel the exhaust from each cylinder separately it was obvious that the inner two cylinders nearest the carb were doing most of the work as the exhaust from the outer two cylinders were remarkably cooler. this was a non-standard inline 4 configuration where the valve order was in, ex, in, ex, in, ex, in, ex, instead of the more traditional ex, in, in, ex, ex, in, in, ex for which it is easier to make all the intake flow lengths the same to help ensure equal running in all four cylinders. I'll repeat this test on my De Havilland Cirrus when I get around to finishing it and report back again, as it uses the traditional order.

Hello, Peter! Thank you for your reply. I also think that the setting of only two intake pipes in the in-line four cylinders will be very stable. But if it is L6 or V8, it may also be different

 

[sition]

Hi guys. Just a few odd comments - based on experience with single carb car engines in the Manufacturer's design office.
One of the things that affects petrol (and other "wet" fuels) and air mixtures is the simple dynamics is a column of gas withe "wert" droplets in it. The gas goes around corners, but the droplets don't.., (Newton's laws of motion - Laws we can't avoid!) so the "wet" goes straight on and doesn't all turn the corner with the gas (air and vapour). Carburettors give a mixture of air plus fuel vapour, Aerosols (very fine droplets) and "wet droplets".
So consider a single carb feeding a single cylinder. When the valve opens, the air column accelerates, and as the accelerating air increases velocity in the carb it starts to draw up droplets of fuel. - Related to the square of the speed of air in the throat of the carb.... Initially the speed of air in the carb is too low to shred the droplets into smaller "aerosols" and heat has to flow into the aerosols to turn them from liquid to vapour.
(Hmmm.... getting a bit complicated, so I'll carefully describe the next steps):
When the carb is at "full gas flow" for the wide-open engine valve and mid-stroke of the piston, whatever the throttle valve is set at, there is a good amount of aerosol (and a bit of vapour) and not so much as "wet" droplets.
Then as the inlet valve closes, the column of gas slows, so we get more droplets against less aerosol fuel.
The combustion chamber heats the mix of air, aerosol fuel, vaporising some of the aerosol, and heating the droplets.
The spark ignites a fuel-air mix of vapour and oxygen in a volume of nitrogen.... and when correct the flame heats the aerosol further to assist igniting all the mixture. But (especially at higher engine speeds) the droplets don't all burn, just the "outer-part" so there is some residual hydro-carbon in the exhaust. - We have all seen the soot on older cars, even when we think the mixture is correct. What the plugs show is the mixture that has burned - at least as far as Carbon monoxide, not the results of all the fuel burning completely. Modern fuel injected engines with O2 sensors correct this "impossible process" from carburettors.
Now consider a manifold with centre carb and 2 legs going left and 2 to the right to feed 4 cylinders. - We will discuss one half. Say cylinders 1 and 2. These fire at some spacing, not necessarily 360 degrees apart on the crank - so maybe inlet of one, - then 180degrees crank - to inlet of next... then 240 degrees to inlet of the first? Oops! Sounds like we cannot expect the columns of air to have the same starting conditions when the inlet valve opens. - This imbalances the air per cylinder and thus the mixture.
But what about the fuel as it travels along a z-shaped passage to the valve? Going to the furthest cylinder, the droplets try and go straight on at the end of the manifold, instead of following the turn made by the air and aerosols into the cylinder. These droplets hit the inside of the end if the manifold, and eventually get sucked into the engine as larger, but hotter, droplets. That does NOT help good combustion of that bit of fuel. So we need a richer mixture to compensate for that "loss" of combustible fuel. (It probably burns in the exhaust pipe, not the cylinder!). But consider the column of air and aerosol to the nearest cylinder to the carb? The air and aerosol turning from the manifold to the inlet tract to valve and cylinder allows a lot of the fuel droplets to fly past the inlet tract and hang around in the relatively still air at the next cylinder inlet tract... (furthest from the carb). So Middle cylinders are thus starved of some of their fuel, which goes towards the outer cylinders...
Hence, we have a lot of factors that affect the individual mixtures in cylinders:
# Effect of a "stopped" column of gas accelerating to sucking adequate aerosol-fuel for the engine to run but getting droplets of "excess" fuel when the gas column is slow at the start.
# Imbalance of position of each cylinder "sucking" in degrees around the crankshaft = different "dwell times" between valve openings.
# Imbalances of position of cylinder along the length of manifold.
And in truth it is more complicated than that!
I worked alongside an engineer who's job was to make the inlet manifold castings with internal vanes to direct the fuel and air mix around the corners, to try and correct the "droplet" problems. - These problems "disappeared" when the manifold only had to carry air, because the fuel was injected at the inlet tract close to the inlet valve.

So "well done" anyone who manages to make a multi-cylinder engine with unequal inlets through the manifold run smoothly.
I hope I have not made this either too simple or too complicated for you? It really is more complicated than I describe when tuning production engines for the optimum performance and to meet emission laws.
K2

Hello, Steamchick. Thank you very much for your earnest reply. Your professional knowledge has greatly benefited me. This is indeed a very complex knowledge. Although they cannot be exactly equal, it is very good for a model engine as long as each cylinder can run. I watched a lot of videos. I think most of the time when a certain cylinder does not work is at low speed, or the carburetor is not atomized enough or the carburetor is too large。
You are right. Anyone can build an unbalanced multi cylinder engine, but not everyone can build an engine that is nearly balanced and does not need to spend too much time adjusting the intake or carburetor.

 

[Steamchick]

Hi Sition: I have never made nor balanced an inlet manifold. Most of my motorcycles have been twins with 1 carb per cylinder - MUCH easier to manage than my colleague who was trying to balance a UK made equivalent intake manifold (or 3) to the Japanese ones which we were "copying". Extensive engine calibration, followed by lots of additional sensors (per cylinder) and tweaks that had to find their way into the patterns in the foundry.
Similarly, a colleague at my local model Engineering Society made 6 manifolds for a straight 3-cylinder side-valve engine he made, before he was happy with smooth running. He does not make replicas of full-size engines, so is not constrained that way, but makes engines that are freelance... in extraordinary configurations, - because he can. (See some "odd" engines= photos attached - In don't have his multi-cylinder engines though!).
On a 4-cylinder straight engine his best (only successful?) manifold was made from brass tube, as a "Y" from the carb to a pair of "Ys" to the engine, so each inlet tract saw the same timing of suck and same tract length and "shape" as all the others. He was convinced you need geometric symmetry for the gas flow for a smoothly balanced engine, determined by the change of engine note (speed) when you remove each plug cap in turn. (That is also how I tune my multi-carb motorcycle engines). - e.g. if you remove 1 plug cap and nothing changes, then that cylinder isn't really working... but if 1 cylinder stops the engine when you remove the plug cap then that cylinder is doing a lot of the work. - Simple and effective for balancing carbs, but not so easy for manifolds!
K2

 

[gbritnell]

All of what Steamchick stated is correct but there are thousands of full sized engine and models that use the single carb, multiple runner feed setups. The only way around it is by direct port fuel injection or multiple carbs, not so easy for models. I have built numerous engines with more than one cylinder, twin, triple, four, 5 radial, 6 and several V-8's. There's not much you can do design wise when you're making a replica to improve flow characteristics through tiny ports and runners. The best you can do is fine tune fuel mixtures and ignition timing to get the engines to run well.

 

[HMEL]

Multi cylinder engines, such as in-line four cylinder engines, usually use carburetors for fuel supply. They are sensitive to the balanced length of the intake manifold. Have you ever had such an experience? The mixing ratio of a single cylinder is too rich, resulting in the cylinder not working and affecting the power。

Yes, but only with larger engines. Carburetor design is tricky because you want the fuel to atomize into small droplets and be transported evenly to each cylinder. This is dependent on the manifold sizing or the runners to each cylinder. The values for velocity can be calculated along with fuel air ratios but for model engines I think this would be more difficult. I have an ultralight aircraft and carbuerators are more than an academic interest. Also when I was a lot younger my father often did such work for stock car racing engines. Trial and error coupled with some good engineering work will often yield results. Inspections of the engine will also tell you what is going on. Fuel jets into the carb venturi are also a key player. The manifold design of an engine can definitely effect performance and that part is a balancing act of pulsating pressures and flow. I used the Helmholtz equations along with some classical flow equations when I messed with it. Now what I have said is pretty general and is based on my experience. For a particular engine small or large you have to start with its displacement, determine max air flow, then the fuel it requires to develop the horse power you want. And then you can select carburetor and jet sizes. Each runner has to be the same hydraulic length not a function of length. It depends upon the number of elbows etc or turns the air makes. And then a good mechanic will test that and give you a thumbs up or a thumbs down! And finally carburation is becoming a lost art with fuel injection at least that is my observation.

 

[Zeb]

Hey gang.
I've had to send a few jugs out because of aggressive leaning on carbureted piston aircraft. Most of what we do in this forum is at low power settings, so not sure if it matters. If you want to dig in to the rabbit hole of fuel atomization for aircooled engines, google-fu and read about "lean of peak". There's a good point to be made as to why fuel injection is so much better...and less interesting for our hobby. ha

Tinfoil time...
For carb engines with multiple cylinders, you can lase your cylinders at the exhaust (or hookup probes) and enrich your mixture until the hottest cylinder starts to cool. I never went to school on engine design, but some ideas come to mind...

With an engine where all cylinders perform identically if swapped and carb is atomizing rich of stoichiometric:

• Smooth intakes that minimize turbulence would be most beneficial. Sharp walls/edges and low pressure vacuum on the backside of the bend might precipitate the existing fuel atomization. What Steamchick said was interesting too, where particles of enough mass inertially separated at a bifurcation. Couldn't resist haha.
• Length of intake path might be less of a factor, unless a sharp temperature gradient were applied across it. Some engines normalize the temp by passing the tubes through a heated oil pan. Maybe it's just to heat a cold mixture, I can't remember. Heating lowers the air density and enrichens the mixture.
• Intake pipe diameter would have a noticeable effect. Particularly if you started to increase friction against flow.
• Carb location is sometimes optimized for shortest path to all cylinders.
• Along the lines of what Steamchick mentioned, I've noticed some aircraft intake designs use an extension at the end of the intake tube...as if the tube was going to continue to another cylinder, but it was chopped off and capped. Always wondered why. Cylinders up front run a lot cooler than the rest, so not sure if that factors as well.

Pilots love to stare at their exhaust gas temps and freak out about 20 degree cylinder temp variances. They're just going to run that way no matter what. A good option is to tear out the snazzy engine monitoring and keep the oil pressure gauge....or in our case, oil cups filled up the glass.

 

[oldCB]

Well I'm gonna start putting my 1/2 VW back together for my "SuperBike" a modified Airbike that I NEED to fly as life has kinda got in the way, meaning other jobs and demands prevented my lavishing love on her
I used 2 mikunis, and had good results...she only burned 1.3 GPH using auto fuel so cheap flying it is!
Ya I had EGT's in each exh, but plan a tuned 2 into 1 exh system with a small "Supertrap" type muffler. The plan is to extract more torque since my prop's max tip speed is 3200 RPM's.
Bragging a bit MoJo was the 1st 1/2 vw powered Airbike in the universe, I spoke with Wayne Ison whilst figuring out the engine mount, he was skeptical at first but was pleasantly surprised when he saw my C/G was just about optimal
I'm a Mechanical Fascist...........There was no way I was gonna FLY behind a 2 smoke!! In fact the only thing their good fer is cutting trees.
I'm replacing the Scat cylinder heads (ports way too big for my application) plus they are HEAVY suckers, the OEM heads are much lighter and should make more power in my application.

 

[Zeb]

The airbike looks awesome. I hadn't seen them before. You could always add a slug of lead in the tail to dial in the CG.

Must resist getting one.

 

[oldCB]

I had a small batt near the tail for a strobe I had mounted on the Vert Stab.............DO NOT RESIST, OBEY yer desires.


It's an absolute BLAST to flyy!!

 

[oldCB]

That pic is when I had a Military Standard 4 cylinder on her.........they are great engines and VERY smooth but it exceeded the limit so I'm putting the 1/2 VW back on. Besides it climbed out better w/the VW!

 

[HMEL]

Hey gang.
I've had to send a few jugs out because of aggressive leaning on carbureted piston aircraft. Most of what we do in this forum is at low power settings, so not sure if it matters. If you want to dig in to the rabbit hole of fuel atomization for aircooled engines, google-fu and read about "lean of peak". There's a good point to be made as to why fuel injection is so much better...and less interesting for our hobby. ha

Tinfoil time...
For carb engines with multiple cylinders, you can lase your cylinders at the exhaust (or hookup probes) and enrich your mixture until the hottest cylinder starts to cool. I never went to school on engine design, but some ideas come to mind...

With an engine where all cylinders perform identically if swapped and carb is atomizing rich of stoichiometric:

• Smooth intakes that minimize turbulence would be most beneficial. Sharp walls/edges and low pressure vacuum on the backside of the bend might precipitate the existing fuel atomization. What Steamchick said was interesting too, where particles of enough mass inertially separated at a bifurcation. Couldn't resist haha.
• Length of intake path might be less of a factor, unless a sharp temperature gradient were applied across it. Some engines normalize the temp by passing the tubes through a heated oil pan. Maybe it's just to heat a cold mixture, I can't remember. Heating lowers the air density and enrichens the mixture.
• Intake pipe diameter would have a noticeable effect. Particularly if you started to increase friction against flow.
• Carb location is sometimes optimized for shortest path to all cylinders.
• Along the lines of what Steamchick mentioned, I've noticed some aircraft intake designs use an extension at the end of the intake tube...as if the tube was going to continue to another cylinder, but it was chopped off and capped. Always wondered why. Cylinders up front run a lot cooler than the rest, so not sure if that factors as well.

Pilots love to stare at their exhaust gas temps and freak out about 20 degree cylinder temp variances. They're just going to run that way no matter what. A good option is to tear out the snazzy engine monitoring and keep the oil pressure gauge....or in our case, oil cups filled up the glass.

I have an T-Bird. It runs a 277 2 cycle Bombardier 27hp engine at 6300 rpm with gear reducer. I have a cylinder head temp and an exhaust temp. Exhaust gas thermo couple has to be within 6 inches of the port for accurate measurement. It is a simple gage to monitor engine health. There is a term called the ultralight salute. Its when the engine freezes up. Funny term for an engine too hot. Some guys like to adjust mixture while flying. Oil is mixed with gasoline or you can get an injector for that. This engine runs great but is no longer made. Parts are still made. The engine is tuned on the ground. Exhaust temps are like a go no go gauge. They can give you fair warning when to get your butt back on the ground. The carburetor has a sliding venturi. A lot of people think that the 2 cycle is similar to a 4 cycle in care and operation. It is not. I think the exhaust temp is a simple low cost safety device for any ultralight flyer.

 

[oldCB]

Like I was saying........2smokes are great fer cutting trees.