I'm getting close to switching to a roller cam in the big block!

[V_TastyTacos_V]

2) The reality is for a street engine, having installed height differences of even .030 isn't going to kill you.

What I'm wondering is how can a couple ten thousandths of an inch matter when the cam companies call out for the same valve spring for several different cams, with different durations, and different valve lifts. I also want to know if anyone uses the valve springs that come with new heads, or do people just chuck those? Like Trick Flow, for instance, sells several different part numbers of the same head, but one will have "Hydraulic roller" valve springs, and another will have "Hydraulic flat tappet" springs. WTF does that mean?

 

[HawkRod]

What I'm wondering is how can a couple ten thousandths of an inch matter when the cam companies call out for the same valve spring for several different cams, with different durations, and different valve lifts. I also want to know if anyone uses the valve springs that come with new heads, or do people just chuck those?

This could get detailed, but I'll keep this very high level: The ONLY JOB of a valve spring is to keep the valve following the cam profile. In very general terms, if it does that, then you are fine. As a cam profile and/ or RPM increase, then the spring must be stronger to keep the lifter from doing a "ski slope jump" off the peak of the cam. That's bad. There are all kinds of details to worry about at high RPM. For example, springs get some sideways harmonics, and designing the height so they are NEAR but not at, coil bind will reduce that. There are many more details that could be discussed too.
The bottom line is spring pressure is not like horsepower - more is not always better. But as long as the spring works to keep the valve following the cam profile, you are fine. Most of our street engines don't get stressed in RPM to a point where we need to be that crazy about the specs. For that reason, many springs will work for an average street engine. 100# "ish" closed and 300# "ish" open woks for a broad flat tappet range. If you have a roller cam, then spring pressures need to be upped a little bit. (If you are building a high RPM race engine, then ignore this post and sweat the details.)

Like Trick Flow, for instance, sells several different part numbers of the same head, but one will have "Hydraulic roller" valve springs, and another will have "Hydraulic flat tappet" springs. WTF does that mean?

A flat tappet cam has a lifter with a (nearly) flat bottom that rides directly on the cam. It spins and does stuff to avoid wear, but you have metal on metal and wear can be an issue. A roller cam has a lifter with a little wheel on it and it (somewhat) rolls on the cam, leading to much reduced wear.
Generally, all our old cars had flat tappet cams, and all new cars have roller cams. But they make retrofit kits that allow builders to put roller cams into old engines. In general, roller cam engines have stronger springs.

There is LOTS missing in the above details, but hopefully that covers some very high level basics.

 

[Kern Dog]

Something that has had me curious for years is this:
Looking at the profiles of the lobes of a roller cam versus a flat tappet....

If both cams have identical peak lift, the rollers always seem to have much more duration at both sides of peak lift.
I have a hard time understanding why I don't see this reflected in the published specs of the cam.
For example, first....here is a picture of the cam in my engine now, a Lunati solid lifter flat tappet:

Below left is the '528 solid flat tappet cam that went bad in June 2022. Note the shape of the lobes compared to the roller cam on the right. The published specs for the roller cam are similar for duration but higher for lift. The old Mopar Performance cams didn't come with a @ .050 number, later they caved in and told customers to multiply the gross duration numbers by .85 to get an @ .050 number. The '528 was listed as a 284-528 so 284 x .85 is 241. The roller cam has similar numbers at 238/244 degrees. I understand that duration numbers are taken at .050 of tappet lift but the amount of duration from there...to peak lift and back to .050 on the other side is a whole lot more yet I don't recall it being explained.

 

[qkcuda]

That is exactly why a roller cam of the same specs will outperform a flat tappet cam.

 

[HawkRod]

If both cams have identical peak lift, the rollers always seem to have much more duration at both sides of peak lift.

Not true - you are thinking about duration wrong.

Duration is measured from when the lobe begins to raise the lifter (typically at .050"). Both a flat tappet and roller cam might have the exact same duration (and exact same lift), but the roller cam has a STEEPER SLOPE. This means that there is more area under the curve, or in simple terms, it will open the valve faster and hold it "more" open for a longer time, allowing for more airflow than a flat tappet cam of the exact same lift and duration.