Cam duration for slow running four cycle model engine

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Brian Rupnow

Design Engineer
Project of the Month Winner
Joined
May 23, 2008
Messages
14,915
Reaction score
8,219
Location
Barrie, Ontario, Canada
I have technical books that are a great source of general technical information about cam design for slow running (1000 to 1500rpm) four cycle engines. I know the criteria used for valve lift, cam flank radius, nose radius, base radius, and almost everything else related to cam design. The one thing that I am vague about is cam duration. I make both intake and exhaust cams identical on the engines which I build. Most of my cams are made to more or less match the profile of cams designed and built by Malcolm Stride of the U.K., which were used on his Lynx and Bobcat engines. These cams had a duration of 115 degrees as per his drawings. Since there is a 1:2 ratio between the crankshaft and camshaft, this means that the cam affected the engine through 230 degrees of the crankshaft movement. This meant that the intake valve would begin to open 20 degrees before the piston reached top dead center, remain open thru 180 degrees of crankshaft rotation, and close 60 degrees after the piston had reached bottom dead center. But if you add those numbers up, that is a total of 260 degrees, so his cam should have a duration of 130 degrees, not 115 as per his cam drawing in one of his published plan sets. Maybe I'm reading his drawing wrong.---Or maybe his timing diagram is wrong--Or maybe he's taking valve tappet clearance into consideration. I have also read elsewhere that cam duration for a slow running engine should be 120 degrees, which translates to 240 degrees of affected crankshaft rotation, which lets the valve begin to open 20 degrees before top dead center, remain open for 180 degrees of crankshaft rotation, and close forty degrees after bottom dead center. George Britnel, if you read this, what duration are the cams you use or recommend for slow running four cycle engines?---Brian
 
Don't know the technical answer but my Silver bullet runs pretty well and slow with no overlap and the intake opens at TDC, If it opens too early you have air rushing out the carb before it reaches TDC, ok at high speed but seem like a bad idea at low speed. In fact I've seen some of my engines with an early intake opening blowing a fuel mix out the carb! Don't believe overlap is your friend at low RPMs either.

Maybe make a cam with the lobes easy to set, set screw, loctite?, and try various settings.
 
I've found a clue!!! On page 24 of "Miniature Internal Combustion Engines" by Malcom Stride, he shows two valve timing diagrams, the one on the right is for a "High speed Sealion Engine", the one on the left is for a "Slow revving Wyvern engine", both by Edgar T Westbury.
If you look at the intake valve on the left, it begins to open 5 degrees before top dead center and closes fully 45 degrees after bottom dead center. Thats a total of 230 degrees, which means a duration of 115 degrees on the cam. The exhaust begins to open 55 degrees before bottom dead center and is fully closed at 15 degrees after dead center, a total of 250 degrees, which translates to a cam duration of 125 degrees. So---A happy medium seems to be a duration of 120 degrees on a cam that will work on both intake and exhaust valves. I'm going to do a solid model overlay of a 130 degree duration cam and a 120 degree duration cam, and see how much difference there actually is.
lQcIWP.jpg
 
So here we have two cams, overlayed. One cam is 120 degrees duration while the other is 130 degrees duration. The difference is insignificant, as I suspected.
XdW3gT.jpg
 
Brian, I think you may have LC Mason's book too. There is a chapter (5 pages) on valve timing and it explains a lot, including why a slow running engine will generally have cams with a smaller duration than a high speed one. Curiously, both Mason and Stride give timing diagrams from almost the same set of Westbury designs. Most, if not all, Westbury engines have more duration on the exhaust than the inlet. Your question also gave me an excuse for an extended flick through Modern Petrol Engines, Judge, 1955. It is a very intersting book, but no help here. He does tell us that the Jaguar XK engine (quite a bit undersquare) has symmetrical timing 15° either side of TDC and 57° either side of BDC.

Your diagram appears to shows a difference in duration made by altering the flank radius. All you are doing there in going from 120° to 130° is slightly reducing the acceleration - a useful thing to try if you are getting valve bounce! If you try making the difference by using identical flank radii but altering the nose radius, you should see a distinctly 'fatter' cam.
 
Last edited:
BAEM club member Bob Hettinger put together a compendium of cam angles for a variety of model engine designs plus a full size Chevy V8 road and Chevy V8 race, and I added the full size Rolls-Royce Merlin early and late cams to it. The upshot is that exhaust duration is clearly centered around 130-deg (with Chevy race 150, and late Merlin 140 exhaust duration), and intake duration only slightly less clearly centered around 120~130-deg (with Chevy race 150, and late Merlin 145 intake duration).

So after much anguish I've concluded that cam angles for model engines don't matter that much, and settled on identical 130-deg with "three arc" design (no straight sections, for smoother transitions) for both intake and exhaust, and 120-deg split, symmetrically centered on TDC (after taking tappet clearances into account these are actually 120-deg cam profiles, and there's no overlap, which is important for easy starting, and in the case of the Merlin prevents backfires, which on account of its carburetor being ahead of its voluminous supercharger and intake manifolds would be a disaster, and runs great with this simplified cam, and according to independent spectators is as loud as Paul Denham's supercharged model V8 (but no idea what cam profile he's using).

if you want both high power and high RPMs then you might need the larger duration angles (eg Chevy Race or late Merlin), but if you only want high RPMs then it probably doesn't matter and you should stick with easy starting smaller angles with little to no overlap, regardless of whether you want a low RPM engine or a high rev'ing engine. Its just a model, we're not going for the quarter mile, or 0-to-60, or ..., nor do we care about EPA emissions and milage, etc, etc, etc that cams can be optimized for, we just want easy starting (and loud sound, for some reason :) !!!)
 
Last edited:
A cam which theoretically called for no valve overlap would have a duration of 90 degrees at the cam, 180 degrees at the crankshaft. I've made a pair of cams with 130 degrees duration, which gives 260 degrees at the crankshaft, so the intake valve begin to open at 20 degrees before top dead center and remains open until 60 degrees after bottom dead center. The exhaust begins to open 40 degrees before bottom dead center and remains open until 40 degrees past top dead center.----Is this too extreme for a low speed four cycle engine?--I honestly don't know, but I think I have too much duration. I am using the three arc design. It's driving me crazy that I can't come up with firm numbers for what I am trying to do.
 
Low-speed open-crank Westbury designs are the Wyvern and Centaur.
http://www.hemingwaykits.com/acatalog/The_Wyvern.html

Wyvern: inlet opens 5 before TDC, closes 45 after BDC, exhaust opens 55 before BDC, closes 15 after TDC, so 115 and 125 degree cams.

Centaur: inlet opens 15 before TDC, closes 45 after BDC, exhaust opens 50 before BDC, closes 20 after TDC, so 120 and 125 degree cams.

So I would agree that 130 degree cams are probably not ideal for a low speed engine. Even the Sealion, an overhead camshaft job with a 140° exhaust cam, does not close the exhaust as late as 40° after TDC (it's 30°). I think conventional wisdom would argue that narrower cams and less overlap should give better low speed running, including better compression, less unburned fuel in the exhaust, more torque, and easier starting. Just how much difference should be expected, I don't know.

Both the Wyvern and Cetaur have roller followers rather than the flat tappets you usually use. This means they can have cams with flat flanks, and possibly some dwell at fully open. I don't have drawings for these engines so I don't know the cam design details. For a given engine rpm, the flat flank and roller follower combination has higher acceleration than a harmonic (three arc) cam and flat tappet, which is why the former are more often used on low speed engines.

Westbury's instructions on cam manufacture tell you to put the clearance in by relieving the heel of the cam, so that the tappet clearance is already taken up at the start of the flank - 120° means 120°. Putting clearance in by an allowance at the bottom of the flank is a poor way to do it, putting (in the model case, very small) impact loads on the valve train, leading to unnecessary noise.
 
Last edited:
Well Brian I think the problem might be that cams are not designed for "low rpm". They are designed to give high end or low end performance.

Comp cams had at one time, a catalog online with everything they made at the time. I'll send it to you if you can't find it.

Look for cams that give the best low end torque like rv's, tow trucks, and busses. Also look at other factors like flywheel, rotating mass has an effect. Also remember that cam specs are called out at .050 lift. Duration is from .050 open to .050 from closed. Actual duration from zero to zero is more. Most model engine builders put 220 -280 from zero to zero. They are likely way under what is used in full size automobile performance engines. That's why most cam lobes look like they have a sharp point.

I'm not going to try to steer you in any one direction but if you consider what I have typed and download the catalog I think you can make a pretty good decision.
 
A cam which theoretically called for no valve overlap would have a duration of 90 degrees at the cam, 180 degrees at the crankshaft. I've made a pair of cams with 130 degrees duration, which gives 260 degrees at the crankshaft, so the intake valve begin to open at 20 degrees before top dead center and remains open until 60 degrees after bottom dead center. The exhaust begins to open 40 degrees before bottom dead center and remains open until 40 degrees past top dead center.----Is this too extreme for a low speed four cycle engine?--I honestly don't know, but I think I have too much duration. I am using the three arc design. It's driving me crazy that I can't come up with firm numbers for what I am trying to do.
Brian, take a look at your 120-deg and 130-deg cam profile drawing (showing that they're practically indistinguishable), now make three copies and place them over each other such that you have an equilateral triangle of cam lobes (looks kind of like a Wankel rotor), they are all 360/3 = 120 degrees apart, now remove one of them, the remaining two are still 120 degrees apart and have no overlap. This is the ideal cam if simplicity is your goal (intake and exhaust are the same and are symmetric around TDC), and my working model Merlin engine is proof that it works AOK (easy starting and fast if you want it) for a model engine.

I'd advise to stop converting to/from crankshaft degrees, just focus on camshaft degrees, all that conversion is confusing.

HTH, YMMV, etc...
 
IMO it's silly to be worrying about duration or lobe separation angle. Instead think about when you want each event to happen, in terms of crankshaft rotation. By far the most important parameter is the inlet valve closing point: closing the inlet valve sooner will move peak torque lower in the rev range, and improves static compression leading to easier starting. Moving it later moves peak torque higher in the rev range. One can then consider overlap. More overlap improves torque in the midrange and top-end, but reduces torque and impairs scavenging and compression at low speed: extreme cases cause intermittent misfires leading to the classic 'lumpy' idle of a hot-rod.

For a slow running engine you want inlet valve closing to happen shortly after BDC, and you want low overlap i.e. inlet valve opening and exhaust valve closing both happen very close to TDC. Exhaust valve opening can often be the same as the inlet valve closing except before TDC rather than after (this means inlet and exhaust cam profiles can be the same, simplifying manufacture). Therefore, a basic set of parameters for a slow engine might be as follows:

Inlet valve opening: 0 degrees (TDC)
Inlet valve closing: 20 degrees after BDC (200 degrees crankshaft rotation)
exhaust valve opening: 20 degrees before BDC
exhaust valve closing: 0 degrees (TDC).

This gives you 100 degrees of duration on the cam lobe.

To tune for mid-high RPM instead, you'd use more duration to gain overlap and close the inlet valve later in the cycle:

IVO: 12 degrees before TDC
IVC: 58 degrees after BDC
EVO: 58 degrees before BDC
EVC: 12 degrees after TDC.

This gives you 125 degrees of duration on the cam lobe, and happens to be the default cam timing used by Lotus for their free engine simulation software.
 
Ok, just to illustrate my comment better: I simulated a 25mm bore x 25mm stroke single cylinder engine with a compression ratio of 6:1, 12mm inlet valve fed by a 50mm long inlet runner and 10mm exhaust valve with 200mm long tailpipe in lotus engine sim, with the two cam profiles described in the previous post. The line marked with triangles is 125 degree cam, while the crosses are the 100 degree cam. The most useful curve here is the BMEP curve in green: note where each cam profile produces its peak BMEP, and how the 100 degree cam's BMEP is falling below the 125 degree cam at the end of the pull where the engine reaches 6000 rpm.

Obviously this is only a computer simulation with a lot of arbitrary variables, so don't take this as an accurate prediction of the performance of such an engine. It just gives you an idea of the differences between cams, all other factors being equal.
1663554506606.png
 
The actual angle the cam profile produces will not equal the valve open time as there is some lost rotation as any gaps (lash) is taken up in the rods, rockers, etc so you will get less duration at the valve than the cam theoretically gives. hence the difference in Malcoms cams and his diagrams. BTW his are not slow speed engines
 
The actual angle the cam profile produces will not equal the valve open time as there is some lost rotation as any gaps (lash) is taken up in the rods, rockers, etc so you will get less duration at the valve than the cam theoretically gives. hence the difference in Malcoms cams and his diagrams. BTW his are not slow speed engines
This appears to be considered the normal arrangement round here, but it is suboptimal.

Ideally, the clearance (lash) should be provided round the heel of the cam by making its radius a few thousandths less than the nominal base circle diameter. This clearance should be taken up by ramps at either end of the heel, so that it is all gone by the time the cam comes round to the start of the flank. This allows smooth acceleration of the valve train from zero velocity. If you don't do this you loose the early part of the acceleration curve and the clearance is taken up with a bump.

Here is another diagram showing a grossly exaggerated clearance. In practice, with a sensible amount of clearance, the ramps are much shorter.

On my camshaft grinder the cam template runs at 10 rpm. If you watch it carefully you can see a slight movement as it rides up the ramp, an instantaneous pause, and then the lift up the flank.

Before someone says none of this matters in practice, I did start with the word "ideally".
 

Attachments

  • Cam Illustration 2.pdf
    86.3 KB · Views: 18
Ok, just to illustrate my comment better: I simulated a 25mm bore x 25mm stroke single cylinder engine with a compression ratio of 6:1, 12mm inlet valve fed by a 50mm long inlet runner and 10mm exhaust valve with 200mm long tailpipe in lotus engine sim, with the two cam profiles described in the previous post. The line marked with triangles is 125 degree cam, while the crosses are the 100 degree cam. The most useful curve here is the BMEP curve in green: note where each cam profile produces its peak BMEP, and how the 100 degree cam's BMEP is falling below the 125 degree cam at the end of the pull where the engine reaches 6000 rpm.

Obviously this is only a computer simulation with a lot of arbitrary variables, so don't take this as an accurate prediction of the performance of such an engine. It just gives you an idea of the differences between cams, all other factors being equal.
View attachment 140069

can you add my cam to the mix, 120-deg duration, 120-deg separation, symmetrical around TDC.
 
I am just learning... But I am aware of the various profiles (radii) of cam follower (including flat) that are used in various engines, so I think (?) that there is something critical about how the cam and follower interact to give the lift timing. I.E. You cannot study the cam in isolation but must consider how it interacts upon the cam follower. I guess this is easier done with CAD - even including the introduction of cam-follower clearance of a thou or 3...?
If you plot lift versus cam rotation against 1-degree rotational movements of the cam, I think you will get a much clearer picture of valve lift and timing than can be explained in words?
I am not CAD competent, so may be completely wrong here...

There are further complications in engine design, such as the gas dynamics. Considering model engines, we generally try for 1 or other of 2 options:
Performance based - e.g. to fly an aircraft, pull children around on a train, etc. - or "looking like the original" = slow running - I.E. not powering anything, just balancing fuel and power generated with the friction and pumping losses of the engine. (Idling).
Hence, valve timing for a slow running engine is unlikely to be in any textbook that I can imagine... as it is entirely un-representative of the "real world" of power generation (replacing man- and horse-power) that created the engines in the first place.
I should consider intake valve opening at 10deg BTDC ~ TDC, closing not later than 20deg After BDC; and exhaust valve opening 10deg BBDC, and closing at ~TDC...?? The gases will travel much faster than the piston, when below 1000rpm. I.E. Overlap should be minimised for a slow running engine, with minimum gas flow.
But I am an amateur, so may be wildly off the mark?
K2
 
can you add my cam to the mix, 120-deg duration, 120-deg separation, symmetrical around TDC.
1663664228713.png

Yours is the 'up' triangles while Lotus's default is the 'down' triangles. The 'super slow' 100 degree cam remains the crosses. Based on this, your cam eliminates the downsides of too much overlap seen on the stock cam, but still isn't a good match to this engine's intake and speed range when compared to the 100 degree cam.

A side note to this: Lotus's software automatically put a very restrictive 3mm throttle body on this engine, which of course limits top end power (probably because I told the program I wanted peak power at 6000, and it sized the throttle accordingly). I tried changing it to 8mm, and gained a lot of power at the top end even with the 100 degree cam, though peak torque was still at 2000 rpm. I think you'd have to run a more efficient intake and rev to a lot more than 6000 rpm on this hypothetical engine to really take advantage of aggressive cam profiles.
 
Ok, here goes my explanation. It's going to jump around but in the end I hope it all ties together. Let's keep in mind were talking about model engines.
Slow running engine? The greatest factor I have found over the years of building engines is the centrifugal weight of the flywheel. Not just the total weight but the weight at the rim. Why because it has to overcome the compression pressure at slow speed. To reduce the size and weight of the flywheel we can reduce the compression ratio of the engine. A Ford model T has a compression ratio of 4.5:1. I would say that most of our model engines are in the neighborhood of 6-9:1. Ok now we have that issue sorted out let's move to Brian's original questions about cam timing. Let's start with some explanations.
This would normally involve a little bit of trigonometry but we'll simplify things. When and crankshaft is at the top and bottom of it's stroke the circular movement in degrees doesn't produce very much piston movement. On an engine with a 1.00 inch stroke rotation of 15 degrees creates a piston movement of about .017 and rotation of 30 degrees only moves it about .067. So we open the exhaust valve 40 degrees BBDC. Hopefully all the combustion has taken place by this time. So the valve opens the exhaust gasses start to evacuate and then the piston starts upward moving the rest of the spent charge out of the cylinder. Now we're approaching TDC. When should we close the exhaust valve? Even an engine that is spinning at 600 rpm the exhaust gases have some momentum out of the port and exhaust pipe so with my previous explanation of piston travel we could keep it open 15 degrees past TDC. Or we could close it at TDC. Now the intake has to open somewhere in the equation. With the exhaust gasses flowing out of the engine and the piston not moving very far in the bore we should be able to open the intake valve a little early with no detrimental affect. Or not! Any movement of the exhaust gasses should start to pull the fresh fuel charge into the cylinder. On a side note this is how 2 cycle engines operate and therefore aren't the cleanest in terms of burned gasses. The piston in it's downward travel first uncovers the exhaust port and the gasses start to flow out. As the piston continues it's downward travel the intake or transfer ports are uncovered. There is pressure built up in the crankcase which forces the new charge into the cylinder. Ah but the exhaust port is still uncovered so some of the new charge is pushed out of the exhaust until the piston in it's upward travel covers the port. Now back to our 4 cycle engine. In my opinion the intake valve could open at TCD or up to 10 degrees BTDC with no problems at all. You have to understand that when you say a valve is opening at X degrees that only means it's starting to open. It's not like at X degrees it just pops open to full travel. Now the piston starts downward with the intake valve open. The fresh charge is flowing into the cylinder. Now the piston gets near to BDC and it's movement is minimal but the movement of the fuel charge has inertia so we can leave the intake open past BDC for better cylinder filling. How far do we leave it open? Generally we can make a camshaft symmetrical meaning if the exhaust opens 30 degrees BBDC then we can leave the intake open 30 degrees ABDC.
In my personal opinion going from valve opening and closing at TDC or BDC and going to 10 and 30 or 15 and 40 isn't going to make any difference in the operation of the engine. As I have said in other past replies I don't know anyone who has built an engine with a specific cam design and gets his engine running says to himself, ghee I thing I'll make another camshaft with completely different numbers to see what will happen with the performance of my engine. I have never talked to anyone who has done this.
gbritnell
 

Latest posts

Back
Top