Jump to content
Venom351R

Today I started the process of figuring out why the Cobra is not making any power.

Recommended Posts

Hoping for an easy solution for what is wrong and why its not making more power with the new heads and long tubes. Doing a compression check later this month. I'm starting small with stuff that I can check that does not cost money instead of throwing new parts at it from the start. I check the port alignment with the lower to the heads and this is what I came up with.

I used my magnetic extendable bolt pick up thingy to reach down there and see if it was a smooth transition or not. When I put it past the lower and pulled it back up I could feel it catching on the edge of the lower intake b/c it was smaller then the opening of the head ports. It was about the width of the base on one side that was catching. So how much mismatch is acceptable? Is the fact that I can pull this tool back up through and have it catch on the lower intake mean the ports are way mismatched? Few pics to help with what I'm explaining.

This is what I used to reach down and feel for the port alignment. When I pulled it back up to where the head port and lower intake port meet the edge was catching on the mismatched section.

F2E98CD2-5EC8-47B8-83B4-6EC47174B929-1468-0000010E0D2A5736_zpsb64b55f1.jpg

A6C0A6A4-9B5C-48D9-961B-96487D7CE798-1468-0000010E08356F90_zps4539a7c8.jpg

A7F778F3-8CC1-4B36-AFFD-9865242FAEF2-1468-0000010DFE4EF190_zps956b2df0.jpg

3B846C99-A717-45B6-BF68-E0E28DB9AC9C-1468-0000010DF8691F8D_zps19372a3f.jpg

Link to comment
Share on other sites

yeah its the RVT gasket maker as far as I can tell. The ports are off slightly. If you look at the tool that I used and the outer edge that is raisded up from the center section, the width of that section is about what the mismatch width is

Link to comment
Share on other sites

WTF trav Rtv on the intake...absolutely not. Pull that off, grab the proper felpro for your intake and make sure you trim it up. Reassemble with a bit of spray adhesive just to keep the gasket inplace. The only place you need rtv is on the china walls in each corner.

Link to comment
Share on other sites

Its not RVT that was used around the ports on the heads. It has a felpro gasket on it. I just went out again and checked and its just around the outside part of the lower intake, its not squished up between the ports, its very little material and almost scrapes off.

What I think that stuff is radiator stop leak. When I first bought the car down in FL the radiator sprung a leak the day before I was to drive back here. Since then the system has been flushed a few times but I still think there is a little bit of that stuff left. Looks like a little of it set up on the outer edge of the lower intake port and dried there. I checked some of the other ports and none of it is present.

The slight gap that is present between the lower intake and head ports is not enough to explain what the problem is. Its not exactly matched but not really that far off either. I think my issue still lies in either the timing, camhshaft or compression.

Edited by Venom351R
Link to comment
Share on other sites

Yes, then it was done again by the shop b/c the fucking shitty MR Gasket head gaskets did not seal like they should have. Knew I should have gone w/ Felpro despite what I was told. $35.00 part cost me 1K in shop fees

Link to comment
Share on other sites

Is your car currently running a programmed chip? If so, would you be willing to mail it to me so I can look at your tune?

If your cam timing is good, and your balancer is at a true zero... There is nothing short of your tune that will hold you back. I added about 5 degrees of total timing to my tune and picked up .3 at the track.

Help me help you.

Link to comment
Share on other sites

Your new Dyno tune won't drastically differ from what you have now... Unless your old tune was just a gigantic pile of crap. My offer still stands though. If you are interested, I would like to go over the map in your chip to see if perhaps I can find something off.

In the mean time, you should verify zero on your balancer. A few degrees will make a huge inpact.

Link to comment
Share on other sites

Verifying the correct timing is on the list as well as a leak down test. Send me the info on where to send you my chip I have no issue with you taking a look at it

Link to comment
Share on other sites

MK2- I''m going to PM you my cam specs and cylinder head specs and everything power related so you can go over everything as well. I'll do that friday as well I'm to tired now and prop will be after work tomorrow to.

Rich, the car has no punch or now power at all. There is no pin you in the seat feeling anymore. The STI pulls harder.

Link to comment
Share on other sites

MK2- I''m going to PM you my cam specs and cylinder head specs and everything power related so you can go over everything as well. I'll do that friday as well I'm to tired now and prop will be after work tomorrow to.

Not a problem. I'm just trying to kill time while I wait for Ed @ FTI to finish my heads...

Link to comment
Share on other sites

I don't think that little of a port mismatch will make that much of a difference at all. Do the compression test and see if something is way off. I don't think there is that much power hiding in the tune either. I'd run the combo past anther engine builder to see if the combo is off.

Kurt

Link to comment
Share on other sites

Here are the cam / head and any specs regarding to power about the engine. One engine builder already said this combo should be good for 450 at the wheels. Other opinions are always good. One other thing that was discussed before on another site was that it was not enough intake or the ports were not aligned and the lower needed work. Well I already confirmed that the port alignment is fine and I figured the Victor EFI was enough intake but I'll see what others have to say.

They cylinder heads ( Ed did get them down to 64cc Combustion chambers, other then that they are as described )

http://www.flowtechinduction.com/tfs-sbf-high-port-cnc-230-hr/

Camshaft specs

605 / 565 228 / 232 113 Lobe separation

Edelbrock Victor EFI ( un ported or touched )

PMAS 95MM Mass Air Meter

#36LB injectors

Accufab 1 3/4 Long Tube headers

Custom 3" H Pipe with cats

Magnaflow 2.5" catback

This is my old dyno sheet with AFR 185's with 72cc Combustion chambers, Trick Flow TFSR intake and shorty headers with a Bassani X pipe with cats. This is what the current chip is tuned for.

petestune.jpg

New Dyno sheet with the new combo

NewDyno.jpg

Cam / Cylinder head sheets

camsheet.jpg

Cylinderheads.jpg

Just one other thing I wanted to add. When I measured for push rod length to go with the new heads this is what I came out with as far as the witness mark, I was told it was pretty much dead nuts on.

003-60.jpg

001-51.jpg

and the result of the PVC w/ the clay.

008-18.jpg

Edited by Venom351R
Link to comment
Share on other sites

that is one of the thinnest witness marks I've seen yet. it just sucks that it's slightly off center. that cam info is great, I'll throw it into my computer tonight to get the timing events for the tune. if you mail your trip out today I should see it by the beginning of next week.

the radial clearance on that clay job seems a little tight. did you happen to measure that distance?

Edited by The Mk2
Link to comment
Share on other sites

I read a lot about the mark being better to be on center and thicker or off center and to the intake or exhaust side but thinner. From everything I read a thinner mark a little off center within reason is better then a thicker mark more centered. I sent that same PVC pic to Ed and he had nothing bad to say about it so I went with it. I did not measure the distance I figured everything looked ok. I did send the chip out today, its coming in a much bigger package then the chip its self. Had to put plenty of packing around it.

Link to comment
Share on other sites

General rule of thumb for PTV is 0.08 on the intake and 0.1 on the exhaust. Then 0.2" radial clearance. But its not going to cost you a billion HP.

As for the pushrod length witness mark, the thinner the better. The advantage to keeping it centered is that you have less lateral wear on the guides in the head.

Link to comment
Share on other sites

Sent my chip out to "The MK2" to have him take a look at my tune just to make sure there are no issues going on with it. Turns out its completely fucked. Still don't know all the details but I had a feeling that tuner was a douche moron. No wonder my friends 03 Cobra melted a piston.

Link to comment
Share on other sites

Yeah it appears to be at least its something other then a big void of having no idea of what is wrong. I'll let keith better explain what is going on with it. Hes given me a basic run down so far. Should have known better then to let that place tune my car in the first place.

Link to comment
Share on other sites

Well, I'm not too sure on how to approach making this post. If some of you get lost, i'm sorry.

Looking through Travis' tune, it was immediately made clear to me that his tuner was afraid to make power. There are some very obvious things that were clearly overlooked, as well as some things where he just plain dropped the ball.

Hurdle #1. Why I hate "calibrated" mafs.

The below attached image is a collection from his tunes basic fueling settings.

venom1b.jpg

In the main image, you will see the MAF transfer function for his maf. If you notice where I circled in red, this is where the injector calculations switch from using the "high slope" value to the "low slope" value. It is a very common first step in tuning the CBAZA ECM, specifically the J4J1 as both the high and low slopes are set the same (stupid engineers). The green line looks to be pulling away from the red line. The green is his programmed transfer function, and the red line is the stock curve. This is referred to as "maf curve peeling". Very easy to fix if you can datalog and have a wideband.

As a secondary image, you will see a box with PID in the top left. These are some basic injector scaling settings. This is where you will see that his Injector High Slope and Injector Low Slope are programmed the same. In reality, the injector low slope should be closer to 28-30.

The next image down is the Cranking Pulsewidth vs ECT (engine coolant temp). I have both his programmed values and the stock values side by side. Trav, if you have problems with the car cranking too long, stumbling after cranking, surgind idle or even maybe the need to hit the gas a few times to get the car started... this is why.

Next I wanted to look at his actual commanded fuel tables.

This next image is a shot comparing the commanded open loop AFR that is programmed into Trav's chip.. Followed by the way it is tuned from the factory... followed by MY OL AFR map. You will notice that at anything under 50% load, the programmed map commands a stoich AFR. This is really good for emissions, sorta good for fuel consumption... but terrible for performance. Then you will notice that his map makes a large and very sharp jump from stoich to 13.04 at 2000rpm and up over 50% load. Its incomplete... Definitely needs to be refined.

When compared to my map, you will see that at idle, I have a very respectable 14.98 commanded. This is a trick I use to prevent idle surging, which is caused by a rich condition at idle. Anything 30% load and below gets bumped to a more power friendly 14:1 and smoothly transitions to 12.47 at 70% load and higher. Adding some fuel to this tune should help when we put some more timing in.

You might also notice that my RPM scaling is different. It might not mean much with this fueling table, but when we get to the spark maps, youll see why it's like that!

venom2i.jpg

Link to comment
Share on other sites

Moving on...

Lets take a look at the spark tables.

I can see that the tuner put in a decent amount of work here... but he actually did too much and hurt himself. One of the weirdest lessons I learned is that you can command such a high amount of timing that the computer will actually ignore the cell and utilize data from the "Sea Level Table". Anything that shows 55 is an ignored cell. Now, the biggest downfall to this map is that there just simply isn't enough timing. The Altitude table shown below is actually exactly as it was stock. This isn't too surprising as its not used much. The seal level table however is where most of the car's timing is pulled from. You can see that at WOT even at max RPM the car barely sees 28° of total timing. The Borderline table has slightly more timing at high RPM, but the biggest downfall to this tune is that the programmer has made sure that the car wont even use the borderline table. Some of the basic scalars are not set correctly and the car would basically never reference off that table. The Borderline table is defined as:

This is the spark to prevent the engine for knocking when running 14.64:1 A/F ratio and is typically setup for around 200F coolant and 75F inlet air temp. This table is typically determined using 87-octane fuel unless the car was designed for premium fuel then it should have been utilizes with 92-octane fuel. The x-axis of this table is engine RPM and the Y-axxis of this table is calculated engine load (volumetric efficiency).

Which basically means that it wont be utilized during Open Loop driving...

Now the Sea Level Table is defined as... :

FN2100 - The main spark table used to determine the base spark advance. If required, the ecu interpolates between this and the 'Spark Altitude Table' (aka FN2110) table to adjust for emissions at altitude. Use the 'Spark BP Bias' (aka FN712) parameter to control how the ecu interpolates between the two tables.

Looking at the Spark BP Bias table, the ECM will utilize the Altitude table only when calclated barometric pressure is below 26.5, and anything above there it will average the matching cells in a 50/50 ratio. In laymane terms, it adds the X/Y cells together and divides by two. At 4000 rpm and WOT, his car would command 25.5 degrees and at 5000 would be 27. Pure poo poo.

I am not too happy with these timing tables, so I'll be fiddling with them a little bit.

venom3p.jpg

Since we have ways to bypass the multi-table lookup for timing, we can actually set both the Borderline and Altitude tables completely to 55. This will make the ECM ignore those tables and work off one single (sea level) table. We can also change the BP bias to make the ecm only look at the sea level table... Much easier to track!

Link to comment
Share on other sites

Thanks Keith for everything thus far. I did not have time to read all of that b/c I have to leave for work but I will when I get home tonight, and Im sure a lot of it I wont understand but I'm glad you have found something with this set up. If the next logical step in all of this is for you to get some seat time in the car then let me know and I'll try my best to resolve the other issues with it soon and find a time we could meet up for a street night at NED.

Link to comment
Share on other sites

Can you explain further on what you mean by he programed the CID counter wrong? And it has at times showed the start up and surge description you talked about before. Just a rough guess but if I was leaving 5* of total timing at WOT on the table along with other tweaks that can be made how much power was I leaving on the table considering it pulled 349 before when the chip was in

Link to comment
Share on other sites

there is nothing obstructing the ports, granted I only had the upper off but i really dont want to pull the lower again. The gasket fit very well around the head ports.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


SN95 Source ©

The premier SN95 Community

×
×
  • Create New...