Better Insight into Cam Timing

[Swifty]

Just to satisfy my own curiosity, I drew up your cam profile and then allowed for .005" valve clearance, the main difference that it makes is in the duration from when the valve starts to open to when it closes, yours has 120deg, while allowing for the clearance it only gives 90.6deg duration.

As it's been stated, it's only for model engines and will work fine. I assume that full size engine makers take into consideration the valve clearance when designing cams.

Paul.

 

[Lakc]

The amount (degrees) it changes depends on your flank radius. If your lift begins at 10 BTDC but it takes another 20 crankshaft degrees before the lobe moves to make up for the lash your opening is now at 10 ATDC.
Now if you undercut your base circle by the amount of lash everything goes back as you designed it.
Next is factoring in how different cam follower diameters change the duration.

 

[Brian Rupnow]

L.C. Mason states that " Incidentally,a point on cam shapes for petrol engines is that it is normal to undercut the base circle diameter behind the nose below the nominal size by the amount of the valve clearance. In the model engineers workshop it is somewhat easier to machine the shape of the cam so that the opening and closing points correspond exactly to the timing diagram. If the base circle around the "no lift" part is then machined below size by the amount of the clearance, then the clearance will be taken up by the time the shaped cam flanks come into operation, and the valve will then lift and close at a more nearly correct time." ---This is not really clear to me. From my perspective, it's hard enough to machine the exact shape of the cam as dictated by the drawing I posted, without having to consider this additional bit of information. If the valve lash is a known factor BEFORE the cam layout is made, (which it should be) then does one just subtract .010" from the nominal base circle before the cam layout is made, or do you make the entire cam first, then go through a secondary machining sequence to put in the .005"undercut" around the base circle?--This seems like it would be terribly impractical to me.

 

[gbritnell]

Hi Brian,
Here's a layout of the cam lobe for my 4 cylinder engine. The drawing shows the rotation angle with -0- clearance and the rotation angle with .005 clearance.
Not enough to affect the operation of the engine but if one was building to certain opening and closing events then the .005 would make a difference.
gbritnell

View attachment CAM CLEARANCE DRWG ROT.pdf

 

[petertha]

... I drew up your cam profile and then allowed for .005" valve clearance, the main difference that it makes is in the duration from when the valve starts to open to when it closes, yours has 120deg, while allowing for the clearance it only gives 90.6deg duration.
Paul.

That's pretty significant timing difference actually. I guess it probably makes sense though, a seemingly smallish 0.005" gap relative to a 0.070" lift is clearance 7%. I recognize that's not what translates into angular change between theoretical open/close. Just saying the same 0.005" feeler gauge gap on a 'big' engine with 0.5" lift is only 1%.

Swifty, did you do this in a Cad program by chance? I was going to suggest to Brian, he probably has the tool (Solidworks) to simulate this with layout sketch blocks. I've only dabbled with this, but I think all the features are there to set up the geometry & mates & measure the difference like what you appeared to do. If you mean pencil & paper, then I bow my head to you please provide details! Upper pic = valve assembly. Lower pic = SW tutorial on sketch block linkage motion.

 

[Swifty]

Petertha, yes I drew it up in CAD, it was straight forward to draw Brian's cam and then draw the .005" clearance around the base circle. I could then see when the cam actually started moving the valve and from there work out the angles.

Paul.

 

[Jasonb]

Brian, have a look at the last paragraph here, you may find it easier to just draw it up and then put adjusted angles into the camcalc so that with the valve clearance you get the durations that you want. Far easier than undercutting.

http://modelenginenews.org/design/CamTable.php#bcr

 

[mu38&Bg#]

.005" of lash is quite large at this scale, especially if you have small lift. Production model engines recommend, .04-.1mm(.0015-.004") and some users set it as tight as possible, while still allowing reliable starting, to get a little more power. This requires high accuracy of the cam profile. The more accurate the cam is, the less lash you can run. There is also a slight difference between lash and base circle clearance when a rocker is involved.

IMO, this only matters if you strive for maximum power output. Valve timing and valve time-area only matter at high power output.

Greg

 

[Brian Rupnow]

The base circle diameter is a somewhat arbitrary figure, based on roughly 6 times the valve opening, which again is a somewhat arbitrary figure. The flank radii are based on the "arbitrary" base circle diameter.The only real "absolutes" that I put into the drawing of the cam is the angle and the "cam blank diameter" which is based on the "base circle diameter plus the desired cam lift." The only "absolute" figure I can mess about with is the angle of the cam. If Swifty's calculations are correct, and there is a real difference of 30 degrees (nominal difference between my desired 120 degrees and Swifty's arrived at figure of 91 degrees), then wouldn't it seem somewhat logical that if I wanted a REAL cam dwell of 120 degrees, I would just design the cam angle initially for 150 degrees? Then when the 30 degrees was "gobbled up" as a result of the valve lash, the cam would open and close the valve at the 120 degrees that I originally desired?

 

[Swifty]

Brian, I think that you are on the right track, I won't be able to check on my CAD for a while, it's early morning here and my wife and I are going away on an overnight trip for our wedding anniversary.

Paul.

 

[Charles Lamont]

then wouldn't it seem somewhat logical that if I wanted a REAL cam dwell of 120 degrees, I would just design the cam angle initially for 150 degrees? Then when the 30 degrees was "gobbled up" as a result of the valve lash, the cam would open and close the valve at the 120 degrees that I originally desired?

This really is not the way to do it. The smooth acceleration of the valve depends on the flank arc being tangential to the base circle at the actual opening point. The geometry of the arrangement also means that, at opening, the cam flank should be in contact with the centre line of the tappet. Anything else and you are opening the valve with a impact, which will create noise and wear. This is why the clearance is best provided by reliving the base circle round the heel of the cam, with a transition to take up the clearance before opening.

 

[dsage]

Wow !!
This discussion really took off since I posted my suggestion of allowing for the valve lash. But as usual everyone here is very knowledgeable on the topic and have given good suggestions to get around the issue. I agree that skimming a bit off the base circle at the bottom is probably the best way to solve the problem. (Something I'll consider asking John to add to the program) Right now when I grind them it's easy enough to add a bit to the duration to allow for the loss of duration when the lash is applied. It's just numbers in the Gcam program (doing so introduces the problems eluded to by Charles). Even so it has been a trial and error effort to get the cam to perform exactly the way I expect it.

In actual fact with such small lash values I'm find eccentricity of the lobe on the shaft is more of a problem. Wherein the lobe might be offset a couple of thou off the center of shaft rotation and the valve might open in the right place but not close where expected. Or sometimes even more lash has to be added to get it to close at all.

The only reason we developed the Gcam program was because it didn't make much sense to me to use cam calc to rough out a cam and then be so crude as to FILE it to shape when I had a CNC mill capable of grinding to the exact desired profile. Especially after I noticed a few thou here and there made such a large difference in the timing. Regardless of how you make the lobes, now that you are studying the problem you'll be able to do a better job at making them however you choose to do it.
In the end it makes little or no difference. The engine is going to run just fine. It's more me trying to be a perfectionist. Having said that if I have the ability to grind a (near) perfect cam lobe then why not try and make it so.

Sage

 

[Brian Rupnow]

dsage---I buy that line totally!!!---"In the end it makes little or no difference!!!" I have no CNC equipment, and find it a challenge to make a cam---period!!! As much as I love building these small engines, I love designing them as much or more. I think I will stick to the parameters I gave in my drawing, and forget about the undercut for valve lash. Not because the theory is flawed, (It isn't) but because there is quite enough work for me to do just designing and building an engine that runs. If someone builds an engine I design and wants to go that extra step when making a cam, then more power to them. I don't think I am going to worry about it.---Brian Rupnow