As the thread title says, it is a question asking if Ti VCT is continuously variable or if it is only stepped.

The reason why I ask is because continuously variable VCT will give you more area under the torque curve. And more area under the torque curve means better acceleration. This is due to the better VE the air flows for the RPMs would give you.

The difference between continuously variable and stepped is this:

Continuously variable allows you to phase the cam lobe forward (advance) or backward (retard) and take a position at ANY angle between its range of motion. Even minute-of-angle positions can be taken if the system were fine enough.

Stepped means that the cam lobe can only be set at specific angles between its range of motion. The simplest step cam VCT system is the primitive 3 position system: either full cam lobe advance or max cam lobe retard and of curse, its default position between max advance and max retard.
Some of the more better stepped VCT systems have many possible cam positions between max advance and max retard positions.

For stepped VCT systems, depending on the number of possible cam positions between max retard & max advance positions, the amount of area gained under the torque curve varies. For a VCT with only 2 or 3 possible cam positions, you do not gain as much torque area as you would if the VCT had many more possible positions.

Since continuously variable VCTs have effectively an infinite amount of cam positions between max advance and max retard cam positions, the torque area gains under the torque curve is much better and also the torque curve could be much more smoother.

 

It is a continuous system, within accuracy limitations of the oil control solenoids. It is reputedly capable of complete end-to-end adjustment within one revolution at 7000rpm, though it would seem to me that accuracy would suffer at higher RPMs as the system has less time on each compression/decompression cycle to make its adjustments.

I don't know for certain but I would have to imagine its accuracy is measured in degrees, possibly +/-2 or 3 degrees at high RPM. At low RPM I can't imagine why sub-MOA accuracy isn't possible, but what practical application could it have? You'd be talking about tenths or hundredths of a HP at that point.

Just by looking at pictures of the system, I would hazard a guess and say the system offers +/- 40 degree adjust-ability, with 80 degrees in each lobe. Somebody please correct me if I'm wrong.

 

After surfing the net for about an hour, the only reference to Ford's Ti VCT being continuously variable is in Wikipedia.

Also, after thinking further about how fast the Ti VCT can react, it is stepped-variable by this limitation (even though it is not mechanically-stepped.
The Ti VCT can act at its very fastest 0.20 seconds.

Assuming that Ti VCT is NOT stepped mechanistically, what you are saying about accuracy at higher RPMs is very correct, and the desired cam position would be off by more degrees at higher RPMs.

For those of you all who do not get the concept of what we are talking about, at low RPM, around idle through grandma-like acceleration (750-2000RPM), the 0.20 seconds to phase the cam lobes into proper position is very precise.

But as you go higher spirited driving RPMs, say 3000-5000 RPM, the cam phase position becomes less precise because in that same 0.20 second time, the cam turns much faster and the solenoid cannot lock the VCT phaser fast enough to catch that desired position.

At racing RPMS, 5000-7000 RPM, that cam turns much more faster for the VCT solenoid to react, so the cam would be locked into position by more degrees later from its desired position.

What I do not know is if the computer has tables with curves for RPM bandwidths to anticipate the 0.20 action time to lock the cam. If it does, and the computer knows, for example, that the solenoids must act 40 degrees of crank rotation to lock the cam to achieve the desired lock position for the given RPM, the computer would of course order the solenoid to act 40 degrees before that position because it knows that in 0.20 seconds, 40 degrees of cam angle would have passed.
And to complicate things further, if the engine RPMs are accelerating or decelerating, the ECU must have algorithms to calculate and act for those cases too.

 

According to this write-up, the Coyote's Ti VCT can phase the cams up to 50 degrees of crank rotation.
http://www.mustang50magazine.com/tec...rformance.html

I assume they mean 50 total crank degrees, or +25 max advance and -25 max retard in relation to crank.
And I of course assume that all Modular engines use the American automotive industry standard of 2:1 crank to camshaft rotation ratio. This would mean (if this assumption is correct) that the cams can be varied by a total range of 25 degrees in relation to its own shaft centerline axis.

 

IIRC there are no fewer than 17 separate tables for Ti-VCT actuation in the ECU.

 

17 separate tables? Wow, that is impressive! I hope some of these tables can be interrelated to each other in an intersecting manner where one table for example would span 5 others, for example.

The concept I am speaking of is dealing with matrices, where each table is a matrix. By spanning, I mean that one table directly ties or interrelates multiple tables together. Figuring out this sort of thing would undoubtedly require knowledge in vector calculus.

 

This is why there are very few home tuners for this ECU. Tuning a gen 1/2 SHO (89-95) is hard enough, and there are only about 40 tables and maybe 5 or 6 matrices to deal with grand total. From what I have heard, there are hundreds of tables and dozens of matrices for the Copperhead.

 

the accuracy of the ti-vct isn't as much of an issue as having to go from a low map column which may be at -10 degrees, to a higher map column where the desired position may be 25 advanced at that given rpm range. of course a good tuner makes tunes so the position hunting doesn't happen. the systems i have any experience with, hondata for hondas, vipec and kms for bmw, and aem on a subaru, the datalogging shows the cam sensors doing an excellent job in getting to the desired cam angle, hunting really isn't a problem with them. as we all know, honda likes to spin the **** out of motors, even around 8,500 on a stock motor, and they always do well. as long as the gear and cam sensors are functioning, they do well. this is based on the viewing of the datalogs after doing tuning on those systems btw.

so i can't say i've tuned one of these, but i don't see ford doing it wrong having other systems to look at before introducing the vct in 2005 (and now ti-vct.) i would have a little more research under my belt on ford's system, but it hasn't enjoyed the same popularity with the ford and gm aftermarket like others such as bmw and honda. plus, i didn't feel particularly impressed with the 2005 gt when i went to test drive. now that i've been impressed by the 2011 5.0 and decided to get one, i'll be trying to get a little more experience with it. the tuning of it isn't tough, but it is more time consuming since you're doing more than just a/f and ignition for the bulk of it.

with the standalones you're typically going to see columns arranged by manifold pressure, and intersecting rows with rpm, and you set a cam angle for each intersecting cell, then for each came angle you'll do your fuel and ignition. if the cam angles to be selected are incremented in 10's but the cam angle is 15 (0 advance) you would adjust fuel and ignition for 10 and 20 and it would interpolate. you're typically going to have the same or similar cam advance at the higher rev range in the low throttle areas as the full throttle areas, and incremented closely together going up in the rev range, so you can make the time of the change a non-issue.

here's a screen shot of the cam advance table on one of the standalones... it will give an idea of how you can set the advance up so the timing isn't an issue. this is for 50 degrees of advance, 0 = -25 degrees (retard), 25 = 0 advance, 50 = 25 degrees (advance)

actually, i don't have picture/attachment priveleges... but here's the url for the picture
http://i184.photobucket.com/albums/x...g?t=1285655677

 

Wes, what are you using to do direct edits like that?

I have been waiting for hp tuners to release support for the 5.0 but it doesn't look like it's going to happen anytime soon, and I want real control over the tune not just a canned tune from sct.

 

well, that particular one won't help us, that's k-pro on a honda that i did (showing an example of how it works) but, aem ems, and megasquirt seem like two promising units for us. i don't have experience with the megasquirt from http://www.diyautotune.com/ so i can't tell you all the features (and i think you may have to buy before you can download their software for your laptop) but it's supposed to be really good from what i hear. i've also been in touch with aem about their ems, still awaiting info, but it's ridiculous how much you can control with these ems.

the aem is very thorough, you can control injector pulsewidth, change to open loop to get better gas mileage on a part throttle tune, change cam advance to help with exhaust drone at part throttle, force the radiator fan on at whatever temperature, set up a launch limiter, pretty much anything the computer can control, you can manipulate with these. it's just a matter of compatibility for now, which is what i'm trying to find out from these guys. i'll update on my findings.

the aem is a known quantity for the fox body, and sn95, but the majority of mustang owners like to think inside the box, and associate stand alones with imports (toyota and honda mostly) and accuse people of being ricers when they go stand alone. it's a shame really considering you can set up your own temp and gear comp tables, and run the parameters that work in your location rather than something generic.