[paulsv]

at this point the only real concern I have on this build is the carb. based on the size of the throat on the carb It makes me wonder if its going to be big enough. the only reason I say that is. I have 50cc plane engines that I can almost put my pinky thru the throat and this engine is sitting at 92cc. But I also know that a blower can make a carb double its normal flow capacity too. so time will tell they say.

Remember, too, that an 8 cylinder engine doesn't need a carb that is 8 times the size of a single cylinder engine. Each cylinder draws fuel sequentially, so it is more tied to the volume of a single cylinder than to the total volume of all the cylinders. So a 50 cc single cylinder engine may need a much bigger carb than a 100 cc V-8.

[cforcht]

ya know. now that I think about it the 50cc engines I refer to are 2 stroke also. which has double the intake strokes as a 4 stroke. so it would naturally seem the carb might need to be an bit larger to handle that too vs a 4 stroke engine.

doesnt really matter. I was just stating for the record my only concern in the whole set of plans was the size of the carb. I am making it to the plans. its always easier to make it bigger rather than smaller if need be.

[cforcht]

Dyno,
Im sitting here reading your post on the curves and valves opening and closing etc etc. Im shaking my head laughing thinking to myself it sure seems like he knows what hes talking about because I have seen mention of all those key words elsewhere. I have a pretty good idea what most of them mean but what their importance is and how to figure them is beyond me. looks like you're doing a fine job. otherwise I would still be in search of a solution to my cam problem.

[dynosor]

looks like you're doing a fine job.

Thanks, I am having fun.

Craig, the short answer to your question of appropriate cam lobe shape when going from a flat to a roller follower is this:

Increase the nose radius of your original cam lobes by 25% to 0.235", and replace the radial flanks with flat ones. The base circle and lift stay unchanged.

This results in very nearly the same displacement curve as the original cam using a flat follower, but with a gentler initial valve opening and final closing when using your roller follower – see chart:

I first tried a 10% increase in nose radius, but this was not enough. The first SW model below shows the 0.235" nose radius cam lobe and the second, the 0.207" nose radius cam lobe. The lobe image by itself is your original profile with curved flanks.

All that said, I am sure your original cams will work pretty well with the roller followers too.

.

[cforcht]

Dynosor, what is your given name? I like to be able to address people properly especially those who lend a hand. I do appreciate the help. I will be re-making the cams to your specs. I realize I don't HAVE to do it, but there are some things I need to have a certain way or I'm just not happy with it. I also see you're up late too. I just got done making a couple distributor caps. I made them from plexiglass (acrylic) mainly because I have a crap load of it 2" thick. leftovers from a job (demo blocks) I used to run frequently but has since died because the product finally got off the ground and they sold it to a larger company who doesnt need the demo blocks to promote the product. I would prefer black delrin but all i have on hand big enough is white. and a distributor cap in white just seems wrong.

[dynosor]

Dynosor, what is your given name? I like to be able to address people properly especially those who lend a hand.

Hi Craig

I appreciate the appreciation. Your handle is your name and it serves to draw potential customers to your business. I am active on a number of forums and have elected not to use my real name. Since you are not "incognito" I have PMed you the same level of info.

Cheers

[cforcht]

OK I can understand that. Although using cforcht as my "handle" was not intended for any business related reasons. I have used it that way ever since i can remember. only in the last few years have I officially gone into business. and that revolved around getting laid off as the production supervisor when the manufacturing industry started circling the drain a few years ago. I wasn't the only one to be led to the door. pretty much everyone in the office got it. including the plant manager.

once again, thanks again for the assistance. I am already working on the "new" cam design. gotta order some material though.

[dynosor]

I am already working on the "new" cam design.

I figured I would do a sanity check by comparing the dynamics of your original cam with my proposal for the roller follower version. The attached charts verify that the roller follower version should have essentially the same performance as the flat follower version, but with a more gentle action opening and closing the valves:

Notice that the acceleration and jerk are lower on the roller follower version - jerk is how abruptly acceleration is applied. The units shown are for displacement, with the same scale on both charts. The measure for velocity, acceleration and jerk are proportional, so no absolute units are given.

Some of the ripple in the charts is because I limited the resolution of displacement to 0.001" increments and the rotation to 5.625 degrees (64 divisions per rev) to get the job done sooner. The real deal should be a little smoother, assuming you can hold 0.001".

.

[cforcht]

LOL holding .001 is a daily adventure for me. its when you get down to the .0005 or less where it gets exciting.

I understand the lifter/valve acceleration from a mechanical aspect if not from an engineering aspect. I know that because of how the lifter contacts the lobe as it nears the apex of the lobe it tends to "snap" the valve open because of the relationship of the width of the face of the lifter vs the lobe profile. most of this is due to the width itself. because at point of valve opening the lifter is contacting further from center line of the lifter than a roller lifter would. hence the snapping of the valve/lifter when it "pops" up on the lobe. roller cams give you a smoother opening and closing as you stated and offer less friction too.

[stevehuckss396]

Have you decided which ignition system you are going to use?

[cforcht]

Steve,
I didn't know there were any choices. I was looking at the WYSIWYG one you mentioned you were using. although I am unclear as to which "kit" or combo you are using. Stuff like this I feel its best to go with what works instead of trying to re-invent the wheel. based on your ignition test video its pretty clear what you've chosen should work very well. more details on what you're using would be greatly appreciated.

[stevehuckss396]

Steve,
I didn't know there were any choices. I was looking at the WYSIWYG one you mentioned you were using. although I am unclear as to which "kit" or combo you are using. Stuff like this I feel its best to go with what works instead of trying to re-invent the wheel. based on your ignition test video its pretty clear what you've chosen should work very well. more details on what you're using would be greatly appreciated.

That is a good choice. I use the board and hall sensor. I used the 5KV wires because they are only .100 in diameter. Roy also has 10, 15, and 26? KV wires. With the increase in size you might want to step up to the 10 or 15 KV wiring. All you would need to do is increase the holes on the distributor for the increase in wire insulation size.

I guess the big question is, what size plugs did you step up to. 1/4-32's would work good and plug boots are available via a vacuum fitting. Dorman makes a fitting that can be cut down to look like a boot and fits the 1/4-32 plug very well. You can get them at the local auto parts store.

I like the 47410 for the 90 degree and the 74706 for the straight. Make sure they fit your wire.

[cforcht]

I drilled the holes in the cap undersize simply because I wasnt sure of wire size yet. easier to make them bigger than smaller. I was thinking of going with the 10kv wires. as for plug size I went with something I had tooling for. 5mmx.08 I know its metric but I have the tooling to make it. tap threading insert etc. I can always jump up to 1/4-32 if desired. seeing how the boots are available for that size it would make sense to do that. one less thing I would have to make. I see they have spark plugs in that size as well. although I do think I would prefer to make them instead of buy them. just for the sake of saying I made the spark plugs too. besides I already have a butt load of corian on hand for this purpose.
Is this the unit you are using. copy and pasted from cncengines.com site.

BASIC SINGLE SPARK CDI
IGNITION SYSTEM
INCLUDES:

* WYSIWYG BOARD
* SPARK COIL-attached to power board
* 1 FOOT SPARK PLUG WIRE-3/32 DIA.
* 1 FOOT GROUND WIRE

Add a hall sensor and magnet or 2 wire connector
for a complete CDI ignition system.
$55.00

[cforcht]

almost forgot. I have several hall sensors already on hand. I use them for rc helicopter engine governor sensors.

[stevehuckss396]

That is the board.

There are two schools of though here.

One: Have a magnet trigger the sensor by fixing the sensor and mounting the magnets on a disk either 8 in the distributor or 4 on the crankshaft (crank trigger). I have had good luck with this setup.

Two: Some feel that you should never have a flying magnet. There should be a single magnet and sensor. They use a "cup" with 8 slots cut into it and that rotates in the distributor between the sensor and magnet. The sensor only senses the magnet when a slot passes in front of the sensor/magnet pair. I have used this method also but the width of the slot needs to be perfect to get the correct dwell.


I have had good luck with the flying magnet setup and magnets on ebay are a few bucks a hundred so big deal. The second setup is what is closer to a commercial ignition.

I may not have to tell you this but the sensor only reads one pole of the magnet. Check all the magnets on a live ignition before you press them in the disk to be sure they trigger the ignition.

Also I read somewhere that .040 is the optimum distance for the magnets to be spaced from the sensor. Any closer wont make it any better. That measurement should be maintained no matter the scale.

[dynosor]

This is a discussion on valve overlap for model engines.

Valve overlap represents the angular distance while both the intake and exhaust valves are at least partially open. Overlap is driven by a combination of intake and exhaust cam duration and the lobe (or cam) separation angle (LSA). For a given intake and exhaust valve duration, smaller LSA cams have more valve overlap and larger LSA cams less overlap.


The specific question is, should the V8 model with the blower have its cams made with less valve overlap (larger LSA) than the un-blown engine?

Optimizing valve overlap isn’t very important for a super charged engine if one is building a display models rather than to power an RC car or boat. One has to be pragmatic and recognize that the first obstacles you need to overcome are engine friction and obtaining good ring sealing. Without solving these, the engine won’t make enough torque at low speed to run, let alone idle - it would just stall when the starter is cut.

If you are serious about building an engine to power a vehicle it makes sense to apply the best known info and understanding appropriate to the given application, and this includes valve overlap for blown engines.

Racing engines use long duration cams with lots of valve overlap (LSA: 106 to 109 degrees) to so that high velocity exhaust gas can help draw in a larger volume of fresh intake at high revs – this tends to reduce torque at low revs and makes low speed idle unstable if not impossible.

If you assume the intake is pressurized with a blower you don’t need so much valve overlap to help cylinder filling. Large overlap then risks having the super charger blow usable mixture right out the exhaust. As such, larger lobe separation angles are appropriate (LSA: 110 to 118 degrees). Two questions that will help in deciding on appropriate LSA are: Is the blower going to supply air at significant pressure and are you planning on running much faster than 6000 RPM?

In general, using positive pressure at the inlet of a pipe to create flow is always more effective than by creating a negative pressure at the pipe exit. The simple reason for this is that we can raise inlet pressure to several atmospheres (increments of 14.7 PSA), but max vacuum is one atmosphere (unless you live in a pressure cooker).

The engine’s operating speed range needs to be decided early because this sets the frequency and timing of gas flow events – gas at a given pressure takes a finite amount of time to travel down a runner of a given resonant length - you want positive and negative pressure pulses to coincide with valve opening and closing event.

While commercial model engines tend to run at 10,000+ RPM, aiming for your model to run at super high speed is likely to have it shed parts. If that is your goal, then you should work up to it in several design & test iterations. Smaller bore and stroke engines can and need to run faster to deliver useful power. As such, they should use appropriate valve timing that may be somewhat different from a full size V8s.

If you stick to the speed ranges of full size V8s you can probably borrow a lot of the valve timing from them and get good torque and power across a similar speed range.

Valve lift, duration, overlap and phasing to the crankshaft are not the only factors: Runner and port area and shape are another huge part of the equation that needs to be designed to work well with the chosen valve timing. Then there is combustion chamber volume and shape, valve size and placement…

Ultimately, if you are greedy for power by using large carburetor throat, runner and port diameters and aggressive valve lift, duration, overlap and phasing, the engine may not run at all because of low gas velocity. Erring on the “chocked” side with smaller, less aggressive or at least moderate flow areas and valve timing will make it easier to produce a running engine that behaves well. That is, except at the top end of the rev range where it would feel like the throttle wasn’t all the way open – that can be fixed later.

[dwood]

(should the V8 model with the blower have its cams made with less valve overlap )

YES!!!!!

[cforcht]

gotta say thats a pretty detailed discussion on valve overlap. so in a nut shell I would say you're telling me don't worry too much about it. this is a model engine after all. not a racing engine. not so much as its a model engine. but all the other factors that would need to be addressed to make the valve overlap even make a difference. things that are much harder to control on the model like port shape, runner length etc etc. the goal here is to get a running engine. not a mini nascar replacement engine.

[cforcht]

Steve. Im sticking to the plan with the flying magnets. I see no reason not to. and yes I am aware the sensor only reads one pole. its the same in the RC helis.

[dynosor]

in a nut shell I would say you're telling me don't worry too much about it.

Yes, but it would be potentially better and certainly no worse if you decreased overlap by 2 to 4 degrees for the blower...