Coil Pack Ignition and Custom Intake Manifold on a Nailhead

[Beemon]

I had an epiphany while visiting family this weekend. Modern cars use a crank position sensor and a cam position sensor to fire the coils. Not much has changed since 1956, the crank position sensor is the timing mark and the cam position sensor is the breaker points. Therefor, I've devised a rudimentary system for a coil pack system using archaic points based ignition. Applying the same basic principals of the ignition system of yesteryear, I devised this schematic below.

 

My initial drawing has the breaker plate energized with the voltage on the resistor side, switching voltage to the coil. Now that I'm thinking about it, I could have each coil wired hot to the ballast resistor on the resistance side and then trigger the ground with the points, just like how it original was. Instead of spark plug wires, they would be lead wires to a terminal on the coil pack, where the coil pack is directly attached to the spark plug. Another method I thought of devising would be to mount the coils like an LS engine on the valve covers and then run the plug wires through the plug cover like original, but I feel the coils on the valve covers would look ugly. Plus, having the coil directly on the plug would reduce resistance in the high voltage circuit, reducing ignition losses.

 

The main goal is to eliminate the points, cap and rotor permanently as they are becoming increasingly rare and costly - thus, this type of conversion not only updates the car, but makes components and modern. As it stands, the current setup still relies on the cap and rotor. The Pertronix hall effect kit, while it has never failed me unlike the Ignitor II kit (it was removed because I preferred the points ignition and felt the hall effect kit was causing issues - turns out my distributor was re-curved when re-bushed without my knowing) would be a perfect contender for a cam position sensor. 12V to the module and the negative wire goes to the cap center terminal. Coupled with a 7.5" mag wheel from the MSD Flying Magnet series would add an electronic crank position sensor, in which a junkyard ECU for a 5.3L LS engine could be substituted. Then the coil packs could be routed properly (perhaps wire loom can be salvaged?), connected to the Pertronix unit and the MSD unit. Advance weights and vacuum advance could also be eliminated, and a 3D printed and polished ABS cap could be put over the top of the distributor for a clean look.

 

An alternator would be required for a stable voltage, eliminating voltage regulator and wiring directly to the battery. These additions, plus the custom ABS intake manifold I've been scheming, would give the modern look and reliability of a modern car. It would also be original and an interesting talking point, with only buyer's remorse to give you any heart ache.

 

Anyways, thoughts?

 

Edited November 6, 2017 by Beemon (see edit history)

[old-tank]

Could probably work as a learning exercise.  Maybe you could become the next Rube Goldberg!  

[NC-car-guy]

You do know you can get an electronic one-wire distributor for a 322 right?

[Ben Bruce aka First Born]

1 hour ago, old-tank said:

Could probably work as a learning exercise.  Maybe you could become the next Rube Goldberg!  

 

 Be nice, Willie.

 

  Ben

[old-tank]

3 hours ago, Ben Bruce aka First Born said:

 

 Be nice, Willie.

 

  Ben

...never been accused of that...

[Beemon]

5 hours ago, wndsofchng06 said:

You do know you can get an electronic one-wire distributor for a 322 right?

 The goal is to use OEM components in the event something like MSD discontinues support (as they have done in the past) or goes out of business. Coil on plug also eliminates resistance from plug wires. Also, to put into perspective, it reduces coil load at the plug where the coil only fires 6 times per second at 2800 RPM and is allowed to reach an ideal charge versus firing 47 times per second at 30 degrees of dwell to hope to charge the coil enough to burn off the last cycle's cylinder. At higher RPM where duty cycle is critical, and charge saturation is weak due to decrease in charge time, depending on the coil's ability to build up a strong enough field in the secondary windings, your engine may not fire a complete burn in the combustion chamber. We don't notice this because the burn is an average of total combustion per cycle per minute and the rotating assembly is turning in a way where inertia can keep things I  motion. This is where electronic ignition systems come in handy, where the voltage recovery can be increased due to variable dwell settings, but you still run into issues with coil saturation. MSD's fix was to fire the coil multiple times per cycle to complete the combustion burn, but it is ineffective as it increases coil output and further increases coil saturation to 141 fires per second, in an attempt to completely burn the fuel (this is alleviated with their ignition box, but there are still fall offs at increased RPM). Thus, coil pack is the best way to go for performance and economy, where (especially on a car with a max RPM of 4400) the combustion chamber has a much higher success rate of completely burning the fuel during the power stroke. 

 

Its just an idea. I've never seen this done before except on SBC engines that share components with modern LS and LT platforms. 

[Frank DuVal]

Perhaps I am missing something, but you have the coil's primary leads going through the rotor and cap of the distributor? You know there is no contact between the rotor and terminals of the cap, right? The gap is close enough for 15kV to jump, not 12 V.

 

Modern coil packs are switched by the ECM, using information from the crank sensor (and cam sensor if used, Saturn does not).

 

On 10/30/2017 at 5:28 AM, Beemon said:

An alternator would be required for a stable voltage, eliminating voltage regulator and wiring directly to the battery.

???? Of course an alternator has a voltage regulator. It may be internal to the alternator housing, but it HAS to be there.

 

Since the electronic system actually does connect to the battery, a generator system should do just fine. Well, unless the load on the battery (winter, heater on full, headlights on, etc) is enough to drop battery voltage below 10 volts at stoplight idle.

On 10/30/2017 at 5:28 AM, Beemon said:

These additions, plus the custom ABS intake manifold I've been scheming

 

Ever work on a 3.8 L  3800 engine with the plastic intake manifold? Cast iron never had these issues....  And you have exhaust in the intake manifold too, IIRC. OK, looked at pictures of intake, maybe they are coolant passages.

[Beemon]

Thanks for the reply Frank! I've been doing quite a bit of research on this subject since posting, but haven't updated the thread since there seemed to be a lack of interest on the coil on plug ignition system. I've since revised my plan to make more sense. The first hurdle I'd have to overcome without using an ECM would be to switch 5 volts to each coil. This is pretty far fetched at this point, so an investment into an ECM would probably be ideal. Once you look past programming your variable dwell characteristics and match a basic timing curve to the stock curve in the book (or a later Nailhead, if applicable). Making an ECM work with the ignition system would require a crank trigger. This can be bought from some venders and can be mounted to the front timing cover just like the traditional timing marker. Basically, a sprocket of some kind is bolted to the front of the harmonic balancer and a hall effect pickup sensor is air gapped next to it on a bracket bolted to the timing cover. A blockoff plate would be made for the distributor to delete the vacuum advance and the advance plate would be locked out, with a hall effect sensor fixed to the breaker plate to signal cam timing. On a 3 or 4 pin coil, from research, it looks like two pins are grounded, one is switched 12V and the other is the 5V signal to the coil. On the 4 pin connectors, the 3rd ground pin is to signal the ECM that the coil has fired as the magnetic field collapses when the circuit opens. Coils off a more modern Buick or other GM is probably what I'd use as they're smaller and are oriented where I can put the plug co-linear to the valve cover with the mounting hole perpendicular to the valve cover, which would put the wires towards the back of the block in a clean fashion. The studs on the block for the plug covers would then be utilized with a mounting bracket for the coils. Without an ECM, there would have to be 8 cam triggers in the distributor (one for each coil) in order to work properly.

 

The trick I guess is finding an inexpensive ECM that can be easily programmed. There are kits out there (DIY kits, pre-assembled kits, etc) that would be the best course of action. I think writing the program yourself would be the easiest part of the whole conversion since the timing curve is already given to you in the shop manual, and input parameters are as easy in most cases as initial, max, and cruise timing.

 

The thing with alternators vs generators is that they use solid state voltage regulation versus points based voltage regulators. When the points open and close, it creates a collapse in the windings magnetic field and can give off large spikes. This is usually suppressed by a capacitor to reduce RFI, but the system is still open and prone to causing issues with electronic components that are designed to play with modern systems. Hence the stable voltage, but I see your point running off of the battery alone. Since the idea is to also include ported fuel injection, and the increase of reliance on other electronic parts, the charging system needs to be adequately sized. For a COP ignition conversion, the alternator is probably sufficient enough in it's own regard. I have never seen my battery drop past 12 with all accessories and lights on, but having an alternator to reduce those spikes would help in its own way. The comment about eliminating the voltage regulator comes from the generator voltage regulator being removed from the inner fender. For simplicity's sake, I would probably opt for a Powergen with the terminal tower to retain the gas pedal relay, as the terminal variants of the Powergen emit a 5V signal for a dummy light, which is perfect to open the circuit on the starter relay.

 

Exhaust does run through the intake manifold, but it can be blocked off with a stainless steel plate. The intake manifold would be made with modern ABS printing processes, where it would be designed for stress concentrations around bolt holes and heat resistance at the cylinder head. It will probably be a 3-piece process where the top cover, runners and base are all separated for removal of build material. Final assembly would be dipped in curing resin and then baked in an oven to fill porous media from the printing process. Ideally I'd want to cast an aluminum intake manifold, but this is probably much more expensive than having an assembly printed. I don't have access to a vacuum forming machine, but that would probably be the best way to go. To ensure runner length, it would probably be tall like manifolds seen on modern V8 trucks - the one that comes to mind is a Toyota Tundra intake, or a high torque HEMI, where the mounting flanges are flat like a Nailhead. The longer than normal runners on the intake will give the Nailhead an immediate boost in power over the stock manifold, increasing street torque. This can be further refined with Helmholtz resonance theory. A 78mm throttle body would most likely be used, and the shaft on one end would be machined to actuate a ball bearing connected to a vacuum chamber for the gas pedal start feature.

 

This type of mod actually excites me quite a bit. A lot of people bolt on and go with aftermarket parts without thinking about or designing a whole system. This allows me to apply my schooling to something practical, and in the end I could even sell a conversion kit like this since there isn't a market for it. With all these TBI EFI kits coming out, I'm surprised no one has done it yet. Crank triggers are being sold, but they do nothing for you when all the aftermarket stuff is still focused on single coil distributor ignition. The best part is that it will be all bolt on and go, so no modifications to the cowl or wiring harness needed except to tape off unused wires.

 

By the way, I have looked into the TBI EFI kits. A lot of them have come down in price in recent years since they started coming out. However, there is an inherent issue with the Nailhead intake manifold - it makes an S curve from the intake runner to the cylinder head and this type of turn really disrupts flow and kills efficiency. To make matters worse, the intake manifold is small bolt pattern. I have teased the idea of porting one of my spare 4-barrel intake manifolds and using the base of one of my 4GCs for the throttle body. A TPS, MAP and IAC sensor would have to be installed somewhere... the MAP could be plummed into the the port for the power brakes and the IAC could be put up stream of the throttle plate where a 3D printed assembly would be made to mount an air cleaner. The TPS could be fashioned to the end of the throttle shaft on the opposite side with a mount affixed to the 3D printed "fuel bowl". You would still run into efficiency issues with the original intake manifold, but could be a quick start. If dropping one of the aftermarket TBI systems on the current intake with an adapter spacer, then you would run into choking conditions and incorrect MAP readings. That's one of the big take aways from being on the Formula SAE team, where we're dealing with a 20mm restriction at the intake manifold. MAP will be lower than what the computer simulates at the throat. I have heard of AFB intake manifolds existing for the 322, but they're probably experimental and extremely rare.

 

Anyways, that's my analysis so far. I'm still adding to my notes as I move forward. electronic ignition timing is kind of new to me, but it's not complicated. Intake manifolds on the other hand, have been on my mind since September.

[Ttotired]

Just a note on your voltage output for the light/start switch you said from the powergen, it will be equivalent to the output

voltage of the alternator, otherwise, the light (which is what its normally used for) will stay on at 1/2 brightness using your 5V value.

Some alternators do have a 1/2 volt tapping (early 70s fords over here and I assume in the US as well, used it to run the automatic choke) and on 

some diesels (in particular trucks like Kenworth, Mack, Peterbuilt ect) used it to run the tachometer, but the important thing to note here is that the 1/2 volt tapping is an ac voltage.

It can be used to run anything that isn't fussy about clean current (it alternates obviously) or as above, where it uses the alternating current as a speed signal (tachometer)

This output is not designed (normally) to run any real load from it, its really just a signal output and can normally cope with about a 1/4W load safely.

 

Back to the powergen, I think you will find that the studs are 1 main output and 2 warning light. As said above, warning light output will be equal to alternator output.

 

 

[Beemon]

Mick, correct me if I'm wrong here, but I thought the light circuit was wired into the stator field and not the alternator output circuit? If it happens to be alternating current, then a diode can be put in line to shackle the AC voltage and force it into DC voltage. On the Powergen, if I recall correctly, it can be wired like a normal 12SI by popping off a concealment plate - or as a 1 wire alternator from the main terminal. The optional two post Powergens come with 1 AC terminal and 1 light terminal on top of the case, as well as the two pinned connector that is usually concealed. I can't find information on it now, and the Powermaster website is quite terrible, but from memory I am pretty sure the light circuit is wired into the stator windings and passes through the rectifiers, giving out DC voltage. The original starter relay is ground through the field windings of the generator - it should be the same for the Powermaster, except its grounded through the stator windings. Ideally any voltage to open the circuit for the starter relay is ok, it's just to break the ground. If the load safety is 1/4W, a simple resistor could be put in series to decrease voltage to 5V at output.

[Bloo]

On just about any modern alternator, the idiot light terminal is the same voltage as the output.

 

The alternator is just a 3 phase AC generator, with three diodes to ground, and 3 diodes to the battery. The diodes only let the current flow one way. The result is pulses of DC to charge the battery.

 

 

Here is where it gets interesting, there is a third set of diodes, generally smaller, that are connected just like the positive diodes, and behave the same as the positive diodes, but are not connected to the output terminal. Since they are not connected to the output terminal (which is connected to the battery), they are only hot while the alternator is turning (charging). This set of diodes is used to power the field, and also the regulator, if the regulator is electronic. In this way the field and regulator do not need to be switched. You could just switch them with the ignition switch (like Chrysler), and then you wouldn't need the third set of diodes, but that generally isn't done anymore. This is shown in the graphic below. Ignore the "IG" and the "S" terminal.

 

The output terminal of the third set of diodes, which you are using to power the field (and maybe the regulator), is identical in voltage to the output terminal when charging.

 

If you connect a light between this terminal and the battery (switch it with the ignition switch), the light will glow when the alternator is not turning (or not charging) because there is system voltage (about 12 volts) from the ignition switch on one side of the bulb, and 0 volts on the other. When the car begins to charge, both terminals are at system voltage, ans since there is no difference, the light goes out.

 

The light has another purpose. It bleeds a little current into the field and regulator to get things started working. The GM "one wire" alternator is sort of a happy accident. It really should have a light, but starts up without it due to residual magnetism in the field. Other systems may or may not work if the light is burned out. Many work unpredictably from trip to trip, sometimes they will start charging without the light present, sometimes not.

 

You could power an electric choke from the third set of diodes. GM did this as I recall. You need some voltage that is only there when the car is running, or someone will leave the key on, the choke will open, and the car wont start because it is still cold.

 

Some older systems used the "Y" connection in the stator to run the light. It is hot only when charging, but is a little less than system voltage. Due to the difference in voltage, some resistance was needed to keep the light out. Sometimes they glowed anyway. These sometimes powered electric chokes as well. Maybe this is what Ttotired refers to. You don't see it anymore. Also it had to be a "y" connected stator, and some stators are "delta" connected.

Edited November 4, 2017 by Bloo (see edit history)

[Beemon]

That was one of the diagrams I had looked at and seems to be how the Powergen alternators are wired for their lamp circuit. As long as it's grounded until charging, it should suit the starter relay just fine. Since the the starter relay is wired so the ignition switch turned on causes the armature to energize, it's essentially a "bulb" where instead of a bulb turning on with the key on, the starter relay closes the starter circuit. When the charge is generated by the alternator circuit, it opens the armature circuit on the starter relay and the "bulb" goes out. The 12SI does not have a lamp circuit from my other meddling with alternators and is excited through the ignition on voltage, where that system is always hot with the key on. The trick would then to change from a 1 wire alternator to a keyed exciter circuit with the isolated bulb circuit so the armature doesn't open the circuit until after charging. I'm not sure what the kick on RPM of the Powergen is, but I do know 1 wire alternators have come a much longer way than previous years with  stronger magnetic fields when off.

[Beemon]

So I know this is kind of off topic, but the subject came up about intake manifolding... apologies for typos, its 3:48 as of this sentence and writing is after a Friday night binge.... well, social gathering. Here is a pretty rough draft of design, where total runner length is approx. 28" in this sketch. Using Helmholtz resonance tuning, the equation f = (Cross Sectional Area of Runner (dictated by cylinder head opening) / Engine Displacement in Liters) * ((((60 * Speed of Sound)/Pi)^2 * (Compression Ratio / (Compression Ratio - 1))) / Target RPM^2), which equates to (2.5/5.28)*((((60*1125)/3.14)^2*(9.5/8.5))/10240000) = ~24"

• Runner Cross Sectional Area = ~2.5in^2
• Engine Displacement = ~5.28L
• Speed of Sound = ~1125ft/s
• Compression Ratio = 9.5
• Target RPM = 3200RPM (Peak torque 341ft lbs is measured here)

I estimate my drawing to be about 28" long. Looking back now, total length also needs to account for cylinder head depth to top of valve, so intake runner only needs to be probably 21" long, or something. Might want it a bit longer or shorter than desired length for turbulence, etc... can be optimized in program to create near laminar flow. Smaller than expected. Equation also neglects cam timing, as it cancels out in derivation. It's too late to throw a mock up into the software. Also for scale, red line at top of drawing is stock air cleaner height; lots of head room! I think it's time for bed.. enjoy!

[60FlatTop]

If you check the existing system you will probably find that the actual spark voltage is 4 or 5 KV. If you have a lean cylinder it might get above 5,000V but I haven't seen that often. Any old coil car do 10,000 to 15,000 V. Ohms law has a lot of room on old cars.

 

A common conversion you can look into is the distributorless Jaguar V12 conversion. Backyard jobs use two triggered 6 cylinder GM units. Jaguars draw cooling air through the distributor and it can cause moisture to rust the internal components. Rather than maintain the distributor some remove it. Just a work around for maintenance, no real world gains.

 

If you did it all yourself the minimum is going to be at least $350. Go out and buy five sets of Echlin points and condenser. Invest the remaining $150 in a girlfriend.

 

In the meantime read this book : http://www.jag-lovers.org/xj-s/book/XJS_help.pdf

 It has tons of fascinating stuff in it, even if you don't own the marque. I have read most of it at least once, lots twice or more. Maybe I need a girlfriend, too!

Bernie

[NTX5467]

Seems like the distributorless ignition "aftermarket" system has been around for a while.  Electromotive, I believe, has it to go along with their EFI system.  Probably 10 years old, now?

 

If you wanted to look really impressive, you could use 8 individual coils (ala Chevy 3.4L V-6) and use the "waste spark" orientation so that one coil fires two plugs, but 180 degrees apart in the firing order.  Might need to gang-together two 4 cylinder coil sets, though, to use the "module" to trigger the coils' firing.

 

With the newer EFI TBI units, all you're getting is a finer fuel spray into the intake manifold.  Possibly finer fuel control for marginally better fuel economy and exhaust emissions?  But little else.  You can probably attach some "wings" onto the existing carb pad and extend the pad to match the "Holley" pattern of the EFI unit.  Then machine the pad to match "the holes", as you desire.  Or just remove the carb pad all together, then weld-in a carb spacer of the pattern you desire.

 

IF you really want to get "modern", you can plumb in an engine oil cooler under the new intake manifold, as some GM engines do.

 

As for MSD "going away", that company is now in its third or fourth ownership from the original entity.  Now owned by the Holley Carb people.  Many of the newer EFI TBI self-learning systems also have capabilities of "timing control", in ONE integrated unit.  Perhaps that could be used to trigger the individual coils in the proposed coil pack?

 

How about some flow bench numbers on the Nailhead cylinder heads?  Using mathematical model simulation to determine at what rpm the intake ports start to choke the power output of the engine?  How those ports are different from the later 401-425 heads?  Then do the same for the exhaust ports.  Then position that against cam specs for lift, valve event duration, and overlap.  With that information, a new cam profile might be devised, for good measure.  

 

While you're at it, investigate getting Ross to build you some light-weight pistons for the 322?  taking rotational weight out of the motor is the same as adding power to the flywheel, typically.  Of course, it would also require lightening of the crank counterweights (knife-edging?) that might further reduce crankcase windage.  More power to the flywheel with less fuel.  Then looking into adapting one of the variable-speed electric water pumps rather than a belt-driven item.

 

Many GM alternators are internally-regulated.  The "small wire" in the plug-in usually appeared for "idiot lights" and/or electric choke heater circuits, but always t'd back into the "big wire" later in the harness.

 

Will one of the street-rod type steering columns go into the Buick?  Using electric power steering, of course.  Still use the stock steering wheel and such?  The wiring harness should be color-coded to match the existing GM harness, I suspect.

 

Enjoy!

NTX5467

 

Edited November 5, 2017 by NTX5467 (see edit history)

[Beemon]

Since September, I've been a member of our Formula SAE team and have been tasked with creating and designing the race car's intake manifold. I signed on for the intake manifold because prior I wanted to design one for the Nailhead. I figured this would be my best bet for a crash course, and after this one is said and done it's just a matter of plugging the 322 specs into the Excel spreadsheet. Our only restriction is that we must have a 20mm restrictor below the throttle plate for a naturally aspirated engine. We are going with ported fuel injection so it's straight air to the runners.

 

The plenum design is what is coined as the Lehmann design, originally drafted by a Eastern European gentleman that is known for working on some of the best top performance Audi engines. I stumbled upon his designs by accident, after doing intensive research on four cylinder intake manifolds (for clarification, everyone in the FSAE seems to use about 3 types of intake designs - a perpendicular flow plenum, a straight flow plenum and the a round plenum, where the four runners are next to each other instead of in a line). I chose to use and optimize this specific design because I've never seen it before, and because I don't think anything like it exists in the realm of FSAE. After doing a mock model and further optimization, I was impressed with the results. I've finally reached optimization of our model and it flows pretty well. After calculating mass flow rate into each cylinder based on cam timing, where duration, lift and intake timing comes into play, I finally set up a transient flow analysis.

 

This is the initial data that I used, calculated in an Excel spreadsheet. This was imported into the program as a function of time, where t = seconds and f(t) = mass flow rate. Here's the video:

 

Notice the swirling into each runner after the overlap period collapses. This swirl is generated by the air traveling above the bell mouths, circulating around the outside wall and sucking into the runner at a tangent, giving it a vortex effect as it goes down. This is extremely desirable as it increases the density of the air stream going into the cylinder. Optimization parameters included the height at which the air from the first plenum enters the second plenum, a change in diameter at the end of the first plenum to create an equalized velocity flow to the outside port, and the distance the first plenum is from the second plenum. There are four entrance holes that separate the first plenum from the second plenum, and these four slots have the combined cross sectional equal to the combined cross sectional areas of the four runners (this was to balance flow). I'm pretty proud of myself over this one, and the team seems to think we've got a good one here.

 

Also included in the analysis, but not directly affected, is a custom made throttle body, and air cleaner assembly. Prior years, they were using $50 K&N filters. After a few cross section calculations, I selected a paper filter that would not restrict flow at choke conditions and only costs $7 over the counter for a replacement (off of a Pontiac GTO Tri-power; readily available). The throttle body is a waste gate design prototype simply because I don't have a throttle body in hand to play with and I needed something to attach the air cleaner to the restrictor. The restrictor is a de Laval nozzle with a diverging and converging angle that is more radical than a traditional venturi nozzle used in most FSAE applications. As flow crosses the 20mm choke, it accelerates the flow out while reducing atmospheric pressure loss on the other side. The intake runners, air cleaner assembly and nozzle will be 3D printed with ABS plastic to achieve accuracy while the throttle body (if manufactured) and plenum will be CNC cut from stocks of aluminum (the plenum will be divided into multiple parts: nozzle bolt flange, taper made from cut 1/16 aluminum sheet, same with plenum 1 end and intake walls, roof and ceiling, and the throttle body will be cut from block aluminum).

 

As this project nears its end for this year, the 322 project will begin. Our CNCing for the club is a sponsorship program, so I will unfortunately not be able to make my plenum from aluminum, but using ABS for the whole thing was a goal from the start. I'm also really looking forward to seeing how this intake compares to a real application versus a computer simulation.

 

Anyways, hope you guys enjoy this bit of school project that has had me wrapped up for a bit.

[60FlatTop]

So the mass flow is zero every .2 seconds. What affect does the velocity pressure have on the Total Pressure of the runners? I am figuring you are using standard air at 38 C. Imagine a side stream (waste) flow that always maintained a Mass Flow greater than zero. It is only air with no entrained fuel since you have direct injection.

 

I would make three manifolds based on three mass air values, send them out to the test track with the instructions "Paint the best one gold and send it back."

 

Air has little tolerance for engineers. It is art in transition.

Bernie

 

 

[Beemon]

Bernie, you'll have to give me a bit to compile the data. We don't have the money to make a few, so what comes off my computer is what we get unfortunately. I just spent the better half of last night running a multi-variable simulation with 36 different iterations, all measuring total, static and dynamic pressures, as well as peak mean velocity and individual runner velocities and pressures. The study has finished, I just have to sift through the Excel file and pick the best one, and then run another study on convergent points. Because of the valve overlap there is some unwanted scavenging going on, which will impact resonant characteristics before the valve closes. You are correct on temps and air pressure. The program does a pretty good job at simulating wasted air, as it circulates around the plenum. Pressure is lower than ambient in the plenum, even more so as it enters the plenum, reducing static pressure. Based on the graphs thus far, flow seems to be consistent across the runners.

[Beemon]

So the program only analyzes static flow analysis. This means my simulation data is garbage. The transient flow study itself is accurate as the results mimic simulating a single cylinder, however the variable calculator does not take in these values and rather an average of the whole study instead. So now I need to run a variable analysis across four different studies and compare the results. I consulted with the professor of the CAD class about this issue, and while he was unaware, the issue was validated. As it stands right now, runner velocities seems to be around the 35ft/s mark, which I don't think is bad for a 36.6 cubic inch engine. It could probably be better, but without a flow bench and the ability to pump out prototypes, it's difficult to know how good it actually is.

 

Looks like I'm going to have to push for a flow bench to be made with a Hoover vacuum cleaner, scrap aluminum heater hose.

[60FlatTop]

8 hours ago, Beemon said:

This means my simulation data is garbage.

 

That's where the "make three and paint the good one gold comes from".

 

Be careful about deeper study of the Otto Cycle. You will wonder why they have used it so long.

 

Anyway, you are bringing the mass air flow to zero behind the intake valve and accelerating from zero too much, but that's the cycle. Think about a well rounded orifice (rounded upstream and downstream in the round runner, one each, at an optimal distance from the valve seat, I can minimize the effect of the inertia on the total induction system and add velocity to the inrush flow.

 

If you want to have fun, plot the enthalpy of the mass flow from the entrance of the air filter to the tip of the tailpipe. And see where you can adjust it.

Bernie

[Beemon]

3 hours ago, 60FlatTop said:

 accelerating from zero too much

 

A part of this project was for the CAD class project, too, and after I had figured out how to setup a transient flow analysis, I didn't have time to do a motion study on a camshaft. Ideally, I would want to model the cam and follower and then do a motion plot to see how quickly it ramps up to full lift. Instead, I just used half of a sine wave to "simulate" this effect. This is my next step, now that I'm not being stressed over finishing two projects and this.

 

Plotting enthalpy would be fun, but impractical given the skill level of the gentleman doing the exhaust system. I don't think it's even been started yet and we're looking to get it running by February. Group projects..

 

During study, I preferred the Brayton Cycle over the Otto Cycle - keeping the volume constant and changing the pressure just made more sense to me on paper. It's too bad the Chrysler turbine didn't go anywhere except to Jay Leno's garage.

[dei]

Ben,

I recognise and hear that young engineering student mind at work! (you have a bright future)

Most of the technical detail you all are discussing is above me but my son tried sharing with me some of his projects and studies, when I asked, as he went through his years at school. In fact he did the Formula SAE car project two years in a row and the team finished in the upper 30 percent out of 110 teams that year. Was told that it was the first time placing above 40% is something like 10 years. They even took it down to Texas for some fun and met some fascinating people with interesting ideas.

He just finished his Masters and has a job application in place. 

Graduating might seem a long way off to you Sir but enjoy the ride as you are building your future!