Are there any small, OHV (ideally OHC as well, but not necessarily) four-stroke engines out there that have an engine head you can remove? The smallest I've found is the 50cc QMB139 engine used on imported scooters. http://www.partsforscooters.com/cyli...9&category=465
I'd like to find something in the range of 25cc.
Honda, Subaru/Robin, Stihl etc all have small 25cc engines, but the head and block are cast as one piece.
I want a removable head because I need to adapt this head to another engine. Obviously an OHC engine would work the best because I just need to link the overhead camshaft to the crankshaft with a timing belt.

Another possibility is using the cam and valves from the Honda GX and fabricate my own head. I imagine this would be pretty difficult though with machining the valve seats/guides and the complex geometry of the intake and exhaust manifolds. I imagine a 5 axis CNC mill is needed. I have a 4 axis cnc mill. Another method that people used to use to make experimental engine heads with complex gemoetry before the advent of CNC was to machine a bunch of thin plates at incremental cross sections and stack them together, like how composite resin molds are made. Does anyone have any experience with making engine heads?

Or another way is to cut off the head from the GX and weld a plate to it, machine it, and bolt it onto the engine I have.

If any of you are wondering why I have to adapt a head to an existing engine, it's for a competition and it's part of the rules, so there's nothing I can do about it!

I'd like to hear some of the experts' advise on this forum!

Thanks!

Engine were made long before 4 and 5 axis machines existed. It just involves fixturing and time to machine or intricate patterns for casting. Can you share any more information about what you are working on?

Thanks for the reply,
The rules of the competition states that everyone has to use the same engine (148cc B&S L-head), but you can modify the hell out of it. The goal is to improve fuel efficiency.
The most significant sources of improvements will come from:
1. increasing the compression ratio as much as I can to increase the thermal efficiency
2. leaning the fuel air ratio to about 0.7

To accomplish this I need to:
1. Use fuel injection to provide better mixture quality to prevent knocking at high CR
2. Use overhead valves to decrease combustion chamber volume to reduce combustion time to prevent knocking at high CR
3. Use squish chamber design and maybe intake valve swirl design to increase turbulence in the chamber to improve mixture quality and increase flame propagation speed to prevent knocking
4. Use electronic ignition to facilitate tweaking of spark timing to suit the new engine characteristics

Some rules:
Iso-octane fuel
I must use the stock engine to retain the piston
I will need to reduce to engine size to produce about 1HP

It's true that machining a new head could be done, but my question was how easy is it?
I mean, I could design something fairly easily, and machine it without even using CNC, but will it work well? Do I have to spend half a year to tweak it so that it will work? I have no practical experience with engines so I have no clue. Is it better to just take something that's already designed and adapt it? The engines on the market are not designed absolutely for fuel efficiency, so they would not work as well as something that I designed only for fuel efficiency. But it would work for sure. Is it worth it to design and build something and spend a lot of time to tweak it?

post up a picture of the engine you are talking about and then we may have some ideasI do but need to see what it looks like

Is this for the Shell EcoMarathon? Which school are you with? Have they not competed before? I realize there is little to be found online about specific engine modifications for the Eco Marathon. The schools seem to keep the secrets pretty well.

I don't think it's necessary to have pictures of the stock engine, since it is the head that I'm trying to modify/fabricate. The head from the stock engine is going right into the trash.
But if you still want to see it, here's a picture of the engine with the head off from Berkley's site:
http://www.ckmcgraw.com/afmcgraw/smv.htm
They used a 50cc Honda, which is still too powerful to power the vehicle.

It's for the SAE Supermileage. It's similar to the shell, but the main difference being that everyone starts with, and has to use part of the same engine model, and it's only for college kids.
For comparison, the best mileage from SAE is just over 3000MPG. The best mileage from Shell is over 10,000MPG.
It looks like Berkley did a lot of modifications: EFI, overhead valves, pressurized lubrication, etc. But they still only got around 2000MPG I think. I wonder what their compression ratio is. The compression ratio for the top Shell teams are around 16.5, which is the most efficient. Any higher and piston ring friction from the high pressure takes over and you get a reduction in efficiency.


So back to the issue, how difficult is it too fabricate an engine head? I imagine it would be much simpler if I used the cam and valve assembly from another engine.

Quote:
“So back to the issue, how difficult is it too fabricate an engine head? “

Interesting project.

Billet or casting?

Are you asking for design assistance, or manufacturing methods?

Do you have a preliminary engineering drawing that you could post?

I’m sure the home, and professionals would chime in with fabrication details, if they could look at a rough sketch or two, in order to better understand what you have in your minds eye.

R=16.5:1 with iso-octane and such a small bore should work OK. I've heard that the engine is only run on the course at WOT for as much time as it takes to accelerate from the minimum speed to the maximum speed, then you shut down and coast until you reach minimum speed, repeat until the laps are up.

I wish one of these mileage competitions was open to the general public. I would give it a shot.

Access to a five axis machine would allow intricate porting without getting into castings. However, casting could accomplish similar results with some clean up. It depends on what you have in mind for port design and what equipment and processes you have available and are familiar with.

I think you are overlooking some very basic engine design principles by using a cam and valves from another engine, unless you have already evaluated the parts to meet your needs for your engine design.

Yes a good technique is to run them at WOT only to reduce throttling pumping losses.
There is no minimum speed or maximum speed. Only a minimum average. But obviously it's better to make the ampltidue of your speed oscilations small, because resistance increases exponentially with speed.
The shell competition is open to the public.

The given cams and valves I think would work if the valves were large enough, no?
To find out if they were large enough, I would need to calculate the mach index, which takes into consideration how well the valve is designed (flow coefficient), the mean piston speed, the diameter of the valve, and the diameter of the bore.
By the way, does anyone have a GX25 and have measured the valve diameter?
I would need to play with valve timing. But I was under the impressoin that all of the OHC engines have an adjustment feature on the rockers?

To design and fabricate new valves, a head, rockers, cams, camshaft, etc. etc. and tweaking it would take well in excess of 1000 man hours. The competition is in the summer, and I'm still a full time student. Remember I still have to design and build a vehicle! The marginal benefits of fabricating my own vs. adaptin an off-the-shelf engine wouldn't be worth it.
Are there any off-the-shelf engines around 25cc where the head is removable?

I don't have any proper engineering drawings right now, but it's the same basic design as all engines:
Inlet and Exhaust ports, valve seat and guide, holes for the spark plug(s), and a pent-roof chamber

Quote:
“The marginal benefits of fabricating my own vs. adapting an off-the-shelf engine wouldn't be worth it.�

You seem to know what you want.So take your slide rule down to the local implement shop(s) and ask nicely to measure up some examples.

You will have to compromise it you want to start with off the shelf parts to save time.
You should be able to slice the head off the most valid candidate then fab an adapter plate to the Briggs block.

How would I attach the adapter plate to the head?
Welding? Would that not distort it a whole lot?
I could pre-heat the part to 600 deg. What that be ok?

Brazing would probably create less distortions, since I can apply heat to the entire block in an oven and melt the braze instead of locally applying intense heat. I calculated the force to be around 1500 lbf.

I am good at answering my own questions

The welding process you will want is called TIG.

Your instructors should have local sources to have it done,
as it is a skilled welding process.

You will have to have it fixtured before turning it over to the welder.

yes I've done TIG before. The problem I'm worried about is distortions and warping due to localized heating.
If I used brazing, the entire part could be heated, since the melting temperature of the filler is lower than that of aluminum, reducing temperature gradients and thus warping and distortions.

Do you have any experience with brazing?

Quote:
"Do you have any experience with brazing?"

Please explain your “brazing� process as it applies to cast aluminum.


My Tig experience is only moderate. There are guys on here who obviously have done this for a living.
One of those guys may happen upon this thread and quash your warping fears in the hands of a journeyman TIG welder, using a nice ridged fixture to hold things in alignment.

OK if your going to down size the cylinder and need a 25cc cylinder there was an engine made called a Conley or Abitar.
Removeable head, 4 valves, DOHC.
It will $$ they are simi collectable
Enya and OS both make OHV, 2 valve heads. 20-30CC

Not sure...but I think Webra made a 1.20 4 cycle with a removable head. It had rotary valves in head. Mabe MRC sold them?? Capt,n

Thanks cap'n and TNK for the reply. Some of those you suggested were glow engines. But I imagine it's pretty easy to convert, as long as there is room on the head to put a spark on (it's larger than a glow-plug I imagine).

As for brazing aluminum. I've never done brazing before. But I read somewhere about brazing cracked engine blocks. The melting point of die-cast alloy 380 is 1200-1400 deg F, which is above the melting point of most fillers. Again, I don't know much about the process at this point, but I imagine I would clamp the two pieces together and stick them in an oven. Then after they have reached the desired temperature, take them out, lay the filler on the area I want to be welded, and stick it back in the oven. It's possible to obtain braze welds that are stronger than the parent metal.

CH Electronics has a 1/4x32 spark plug, same size as glow plug.

The italian OPS made a beautiful 20cc OHC belt driven ,removable head, very elegant engine some 20 years ago .if you find it, get it.

The engine design IS important.
But the WEIGHT, FRICTION, STREAMLINING & RELIABILITY could be more important.
Try to COPY the winner with the highest MPG. Improve from their success.

Since you are only operating at idle & WOT, a carburator will do almost as well as a $ 3,000 electronic F I & ignition setup.

PLUS.

It is a Hell of a lot more reliable than a prototyped computer program. Most "Did Not Finish", are due to the computers & wiring, failing in the final runs. A carb only can have at MOST is 4 jets for your use. Idle fuel & air jets.......WOT fuel & air jets.
Know the temperatures and humidity at the place & time of year of the contest. Set the carb accordingly. Carry 2 different setups of jets for a worst case ever run condition. You can absolutly change jets faster and more reliably than a programmer can. Just have the 2 different complete carbs to cover conditions. Most reliable way.
Change the carb to the condition.

Planning for reliabilty always beats R & D.

Dumb luck overrides either one.

# 1 priorty is to make dozens of runs with absolutly NO problems.

Cyclops:
Thanks for the adivce, it seems like you have experience with this kind of competition?

Engine is the most important factor because it has a high output compared to the resistances. For example, using a 1HP engine produces about 100N at the wheel. The sum of the total other forces, on a flat road, is about 10N. So a 10% increase in efficiency of the engine has 10x the effect as a 10% increase of the other factors. Plus, there is much more potential for efficiency gain in the engine. Weight is the second most important factor,but there's not a lot of room for improvement. Everybody has 3 wheels, an engine, and and a driver. If you spend a year developing the frame and use a super highly optimized composite sandwich monocoque design, you will save max 10% weight a well designed aluminum frame design, which is still nothing compared to the improvements you could've done on the engine.

A carburator works fine at WOT under steadystate under low compression ratios
Remember that in the competition the vehicle is started and stopped frequently, running for max 10 seconds at a time.
A carburator needs a much richer mixture than EFI when it is cold and starting, since the AF ratio outputted by the carb depends strongly on the RPMs.

There are EFI kits you can buy specifically for small displacement engines. Infact, since I'm running at WOT all the time, the fuel injection doesn't even need a metering device to change the amount of fuel when the engine is throttled. Only a temperature sensor is needed to compensate for the change in density of the fuel. This greatly simplifies things.

I agree with studying winners, but not blindly. I have done my independent research and design without looking at their designs, and have confirmed that it is similar to the winning teams. I also have read many reports by various teams who have competed at the competition to gain experience through what they've learned, since this is my first time at the competition. I think the teams at the European Shell Eco-marathons do a much better job. Their record is over 10,000MPG vs. the 3000MPG of the SAE competition in the US. Although a lot of their top teams is

I've also done other competitions (AUVSI AUV), and know the importance of reliability, time management, and setting realistic goals. I have also one of the highest averages in my program and have more than 3 years of machine shop experience.

Believe me, I've thought this all through.
I appreciate your help though.

BINGO...that is the engine I was talking about. I think I may know where there is one new in the box. The guy wanted serious $$$ for it. I see him from time to time. Best Regards Capt,n

Do you have access to a intake vacuum to rpm graph for the cylinder configuration you are going to use? I am applying a bit of old logic to efficiency. But maximum torque was produced at about 2,000 to 2,700 rpm in older American V 8 engines. This also usually corresponded to best economy in steady state loads ( speeds ). Any higher RPM's was a trade off of peak torque ( most efficiency in consumption ) for speed.

I am curious how the logic of WOT, is better than a acceleration to peak torque and than a coast down. Unless a intake & exhaust manifold is tuned to spread & and shift the harmonic resonance points to a higher point.

Please do not tell me a hemispherical chamber with a maximum area of intake & exhaust valves is no longer one of the best shapes for efficiency.

Lowest BSFC is usually obtained at peak torque output. This is true. However this is usually measured at WOT. At part throttle you introduce pumping losses. You still have a torque curve at part throttle but it usually shifts a bit to the left. So your peak torque RPM is usually lower. BSFC and fuel economy in a car are not the same thing. In a car the slower you can turn the engine and still have a comfortable acceleration available, the better the fuel economy is going to be. Dodge Vipers cruise at highway speeds at 1200RPM or something. My old BMW 328 does it at 3500RPM and still gets 30+MPG from a 2.8L.

The logic is to use WOT in the range of the engine's best BSFC. This might be 500-1000RPM +/- of best BSFC. Gearing and everything else comes into play.

Hemi's are no longer the best. They haven't been for a long time. 4 valves pent roof chambers are ideal and most common today.

Cyclops:
I'm not exactly sure what you mean, but the engine is accelerated at WOT to peak torque and then turned off.
If you hold RPMs constant but adjust the throttle by adjusting the load, then WOT gives best efficiency due to pumping losses when throttling
If you leave the engine at WOT and accelerate it against a static load, then the best efficiency occurs somewhere in the region of peak torque at around the engine's middle operating speed. This is because there is less time to lose heat due to faster combustion cycle, but more cycles. The amount of heat loss is actually the same, but you're also producing more power at higher speeds, so there's less percentage heat loss. Higher friction at higher RPMs reduces efficiency.

Gkamysz:
This is true bsfc and fuel economy in a car is not the same. Although accelerating as slow as possible isn't always the best strategy. I've heard of people (in the real world) accelerating very fast to take advantage of the high BSFC at WOT, and then coasting on idle.

Generally the best combustion chambers have the best anti-knocking properties to allow higher compression ratios and lowest heat loss. Chances of knockning decreases with short combustion time.
The best chambers are usually shaped like saturn. This is because the center has a high volume to surface area ratio, and it takes about equal time for the flame to reach all areas of the chamber (in a flat head, the flame would take much longer to reach the sides of the chamber)
The "ring" squish area induces turbulence to increase flame propagation speed, better mixture quality, and reduce mix temperature at the end of the flame travel to prevent knocking due to high surface area to volume ratio. Too much squish decreases end temperatures too much, and leads to unburnt mix, which obviously decreases efficiency. If I had designed an engine, I would basically have to experiment with this squish area, which is why I was afraid of doing it due to lack of time.

HEMIs are two or three-valve pent roof chambers right? 4 valves are ideal because they have more valve area and give better turbulence due to tumbling.