HOWTO: Tuning cams

[Carl]

Hi everyone 

I have fitted a set of stage 3 crow cams to my fg and are unsure where I should start and what preamaters I should be looking at and is there a process that I need to follow. Any help would be appreciated.

[Roland@pcmtec]

You've just signed yourself up for a fun process!

Firstly your injectors must be dialed in perfectly BEFORE installing the camshafts. Eg 1-2% trims. If you haven't you will have a chicken and egg problem. You either need stock injectors and camshafts, or dialed in injectors and camshafts, otherwise you will not be able to tune the vehicle. If your injectors are not dialed in, the easiest way to tune the camshafts will be to put stock injectors back, once you are happy then install the injectors again. There is no way around this, as there is no method to determine if your speed density is incorrect, or your injector slopes are incorrect.

Next step, make sure all your speed density settings are standard.

Process if you are retaining VCT:

I recommend you make sure your camshafts are ground with zero overlap, or you install an offset pin to get 0 overlap. Overlap should only occur when commanded, it shouldn't be present at all times. This is how the factory PCM works and is tuned. If you install camshafts with overlap (almost everyone does) you are going to have a fun job trying to fudge all the speed density maps to make it work. As a result most people end up with a terrible result and stop driving their cars as they become a pig going rich/lean, bad brakes due to low vacuum and stalling all the time. 

First, do a dyno run and check lambda across the rev range vs your commanded lambda so you have a baseline. Also log both cam angles and camshaft error. If you see large error (eg camshaft gets stuck at 50 degrees) then your VCT cannot control the camshaft and has failed. This is very common with camshafts with high ramp rates and heavier valve springs. If this happens your camshaft will be stuck in a huge overlap position. This results in most of the aircharge going out the exhaust and the car being hard to start.

Now something to be aware of, with large amounts of overlap you will have a large amount of air and fuel going straight out the exhaust. This results in your wideband reading a false lean condition (as it measures oxygen, not fuel). So you will have your wideband saying your vehicle is lean and at the same time fuel spitting out the exhaust.

If your VCT gear has failed you have two options. Speak to one of the workshops who sells modified VCT gears that won't fail in this manner (Joe at hoontune makes some). Or you can lock your camshafts in place and remove the VCT gears. This is not recommended as your power and fuel economy will suffer, however it is the easiest method to tune.

Now assuming your VCT has not failed and is still functional you would move on to the speed density tuning guide and follow the steps found there:

If you have installed the camshafts with zero overlap you will find you don't need as drastic changes above and can start commanding more overlap across the rev range and noting power. If you increase the overlap you will need to follow PART4 of the speed density guide linked above. Remember that at large overlap positions you will be getting a false lean condition, so you need to use some educated guesses here.

Here is one of the overlap modifiers you can modify.

 

Process if you have locked the VCT Gears

If you have locked the VCT gears you can simply use our custom operating system and enable speed density vs map pressure via the guide below:

 

 

This allows you to do typical VE style tuning where you simply fudge the values at different map pressures to get the result you desire. This is much easier to tune, however you will lose fuel economy and power due to the lack of VCT.

As you can see this is not a trivial process. It is recommended that if you are a workshop that you pick 1 camshaft, spend the 2 weeks tuning it perfectly, then only sell that camshaft in the future and use the base speed density tune you have created. If you have random camshafts turning up every week you simply do not have the time to tune it correctly for the money they are paying. Most tuners report it takes well over a week to properly tune the speed density (if you retain VCT) for a specific camshaft.

Ford spends half a year with 3 OEM calibrators and a dyno cell to create the stock tunes. Trying to replicate this in a backyard to the same level of quality is simple not possible so you will always end up with some compromises here.

edit: This is relevant as well if you are a customer looking at having this done, it might help set your expectations.

 

[Carl]

Does the offset pin go behind the phaser to achieve the 0 overlap and if so where would I source them from. Thanks 

[Roland@pcmtec]

Talk to crow cams, I'm sure they can help.

edit: If they are already installed it isn't worth changing, as depending on the cam design it may not be easily achieved as the offset goes on the camshaft itself. This is more a guide if you were getting a camshaft ground it from scratch and wanted the best results.

[JMSMotorsport]

First time playing with a cammed Barra. It’s NA FG ecu and in a speedway car so street manners isn’t an issue. But trying to nail down an issue with it going lean with it loaded on the dyno at say 1200rpm and then laying into it. Trying to replicate the rolling start they do.

Am I best to use the Tuning Correction table rather then fudging the SLOPE. As I find just light throttle cruising at around 30deg VCT on the slop I have it around lambda 1 but if I had some throttle holding steady load it goes super lean and starts to pop and carry on while still being in the 30deg are on the SLOPE?
 

I’ve also had zero gains from either retarding or advancing the VCT slightly, Am I best to just Zero out the table’s both Overlap and Intake VCT at load 1 and then advance or retard each RPM point individually until return are null then move into the next one? 

[Roland@pcmtec]

First things first. 

Datalog the camshaft error to make sure you don't have a mechanical issue. It should never be more than a few deg. 

Next check your fuel pressure is steady. 

If both of those are OK I would zero all your camshaft tables (at least make them flat) and do a scatter plot/histogram of LTFT or wideband vs map pressure. Avoid transients as much as you can and use steady state until you get at least 200 samples per cell. 

If you find you have a large error at low map values only then you should start by tweaking the speed density offset tables. If the error is linear across all map values then you either have injector scaling issues or you need to adjust the speed density slope tables. 

Once you've identified where the problem is start by making a histogram against the actual table of interest using the trace function in the data logger. 

Rinse and repeat in steps of 10 deg intake cam angles. Once this is ok start playing with the overlap tables. 

Standard they command 30 deg overlap when coming into boost. How are your cams degreed? If they already have say 20 deg overlap then this would almost certainly cause a misfire. You may want to remove all commanded overlap all the time or even use negative values. 

Once you've dialed in all intake cam rows it's time to do some dyno runs and figure out the best angles per rpm for max torque. 

Expect to spend several days to get it correct. This is time consuming to do correctly. 

Edit: tuning correction is an alpha N correction table based off accelerator pedal position. It is not real load and not the correct way to tune cams. You may get an acceptable result if your cams are locked but it is lazy and not the right approach. 

On the v8s or na vehicles with locked cams you should use the custom os speed density patch to turn the speed density table into a traditional ve table. 

[JMSMotorsport]

Roland, cheers for the reply. It’s NA so no boost to worry about and stock injectors so a win there. Managed to fudge it enough to get it on the track running safely. Ended up zeroing out both overlap and advance tables and then played around with them from there. Will log the cam error next time to be sure all is well. Not sure about the cams but the way it was behaving I feel it most definitely had overlap ground into them as I used some negative numbers to help the popping. It needs a lot more time to get it dialled in but run out and does what it needs to hold a rolling start and then go flat in 2nd for the remainder.  

[rfh]

Hey Guys

When the inlet cam is reading zero degrees, is this the point when the valve begins to open when the engine is bang on TDC or is it lobe centerline?

Positive numbers = more advance?

Negative numbers = more retard?

Thanks Rob

[dat111]

1 hour ago, rfh said:

Hey Guys

When the inlet cam is reading zero degrees, is this the point when the valve begins to open when the engine is bang on TDC or is it lobe centerline?

Positive numbers = more advance?

Negative numbers = more retard?

Thanks Rob

other way around 

negative = advance 
postive = retracted