My Flat Tappet Cam Wiped Out So I Decided To Go Roller

[L31MaxExpress]

My son came over yesterday and brought some of his tools including his fluid sucker and a vacuum filler. He sucked down the radiator to get as much as he could out of it. Then he pulled the CTS out of the head, I cut it down so it is as short as the original Dakota Digital sensor I had in there. The idea was to get it more into the center of the water jacket.

Next was to pull the T-Stat. It's kinda a PITA because one of the bolts hold the brace bar for the belt tensioner plus there's 5 wires grounded on those bolts too. Anyway, I had drilled an 1/8" hole in it before, now I added 2 more 5/32" holes to it.

You must be registered for see images attach

This is cool, he stuck the suction hose from the vacuum filler into his fluid sucker tank, built up a vacuum in the cooling system of my truck, turned off the vacuum, and watched the gauge, like you would do in an A/C system, to check for leaks - none. He open the suction hose valve and within a couple minutes all of the coolant was back in the radiator. This also eliminates the chance of getting any air pockets stuck in the system.

Here's when he started to build up a vacuum

You must be registered for see images attach

This is when he was filling the radiator.

You must be registered for see images attach

Then I started it up, watching my temp gauge on the dash, and the 'puter's the CST in the head stayed about 20° higher than the ECM's while warming up, it did take a little longer for the T-Stat to open but, who cares here in AZ. The dash gauge showed a little over 200° when the 180° T-Stat opened at ~190° on the display. Then they both cooled down to the point where the dash showed ~10° warmer than the ECM. That was how it acted all the time before this engine so, I'm happy

Now, time for her maiden voyage to go get some 91 octane down her gut. I took it easy, barely using 1/4 throttle most of the time
A couple times I hit it with 1/2 throttle - nice get up and go with no Knock Counts. Then I just couldn't help it. From a stop sign I floored it (well almost). and burned rubber across the intersection before I let up - nice! Then, we turned on to highway 85 heading to the QT station. I hit it in 2nd gear up to 5700 RPM before I let off, it still wanted to go more, nice smooth acceleration with no KCs

I still have it set up for VE Learn and, now that I have some 91 in it, I'll do some driving around to tune. It still surges when stopped, in gear so, I have work to do on that. Good day!

I looked at the .bin you sent me and the data log. I sent you one I messed around with and took some ball park dead reckoning on it to smooth it out. With the low rpm advance and launch mode advance looked to have had way too much timing at low rpm and high load. I halved the launch mode spark and pulled down the low rpm high load timing to be more like a traditional distributor. Your VE learning is not catching all the VE table cells. They definitely need to be smoothed after every change by interpolating between the values it has learned and the values it has not touched. The low-speed VE table alone could be contributing to the stalling.

 

[PlayingWithTBI]

With the low rpm advance and launch mode advance looked to have had way too much timing at low rpm and high load. I halved the launch mode spark and pulled down the low rpm high load timing to be more like a traditional distributor. Your VE learning is not catching all the VE table cells. They definitely need to be smoothed after every change by interpolating between the values it has learned and the values it has not touched. The low-speed VE table alone could be contributing to the stalling.

Thanks! I'll look at it tomorrow.

 

[PlayingWithTBI]

I sent you one I messed around with and took some ball park dead reckoning on it to smooth it out. With the low rpm advance and launch mode advance looked to have had way too much timing at low rpm and high load. I halved the launch mode spark and pulled down the low rpm high load timing to be more like a traditional distributor.

I flashed your .bin after I enabled BLM, HWY, and CL. Took it out for a drive, to get the truck washed, and back to the house. Most of it was back roads at 55 - 65 MPH. I noticed, at idle, it ran way too rich, ~11.8 AFR with 13.8 cAFR. At higher MAP and RPMs (3600 - 4200) I got a lot of Knock Counts which, in this snap shot, the ECM pulled 5°. The fueling looked pretty good with cAFR 12.6.

Notice the IAC is opened to 255 counts as you suggested a while back on my other cam, thanks for that too!

You must be registered for see images attach

Based on this run, I lowered SA at those higher ranges. We'll see what happens.

You must be registered for see images attach

Then I flattened out my VE table around idle to get closer to the 13.8 AFR.

You must be registered for see images attach

Thanks again for everyone's help! Keep those letters and comments coming

 

[L31MaxExpress]

I flashed your .bin after I enabled BLM, HWY, and CL. Took it out for a drive, to get the truck washed, and back to the house. Most of it was back roads at 55 - 65 MPH. I noticed, at idle, it ran way too rich, ~11.8 AFR with 13.8 cAFR. At higher MAP and RPMs (3600 - 4200) I got a lot of Knock Counts which, in this snap shot, the ECM pulled 5°. The fueling looked pretty good with cAFR 12.6.

Notice the IAC is opened to 255 counts as you suggested a while back on my other cam, thanks for that too!

You must be registered for see images attach

Based on this run, I lowered SA at those higher ranges. We'll see what happens.

You must be registered for see images attach

Then I flattened out my VE table around idle to get closer to the 13.8 AFR.

You must be registered for see images attach

Thanks again for everyone's help! Keep those letters and comments coming

My take, that is far from a natural VE table change.
Definitely would have multiplied that area by a percentage and smoothed it into the areas where it is closer. A big flat area like that will cause running issues.

 

[PlayingWithTBI]

My take, that is far from a natural VE table change.
Definitely would have multiplied that area by a percentage and smoothed it into the areas where it is closer. A big flat area like that will cause running issues.

Yes, it's a Q&D mod but, I have it set to run OL at idle and, during the log, I saw I have as high as 70 KPa and 1100 RPM at 11+ AFR, while coasting with no throttle. Once I get the idle AFR and surge taken care of, I'll do some VE Learns and smooth out the table. I may even switch to CL at idle then.

Or, am I taking the scenic route here?

 

[L31MaxExpress]

Yes, it's a Q&D mod but, I have it set to run OL at idle and, during the log, I saw I have as high as 70 KPa and 1100 RPM at 11+ AFR, while coasting with no throttle. Once I get the idle AFR and surge taken care of, I'll do some VE Learns and smooth out the table. I may even switch to CL at idle then.

Or, am I taking the scenic route here?

I know the intent was to reduce the fuel delivery, but a large flat spot in your VE table is just going to induce surging rather than cure it. The MAP will just keep rising until the engine gets the fuel delivery it needs, which will make it have a large flat spot. You are better off looking at the VE table in the area you are having fueling error and multiplying across the VE whole table at the corresponding rpm range. If it is at 11.8:1 and commanded is 13.8:1, percentage difference formula shows about a 15% error. 11.8:1 is really not that rich, I often find big cam stuff typically likes 12.5-13:1 to idle smoothly. Yours is not quite to the point I would call a big cam so it should run smoothly in the 13.5:1 range if not a little leaner. Your stoichiometric value is probably 14.1:1 anyway given 95% of the nation only has E10 at the pump. The VE table should have a natural sort of curve to it. Lower MAP and Lower RPM should have lower numbers and as the MAP increases the numbers should increase. As the RPM increases up toward peak torque the VE table should increase. Once over torque peak the values should start to taper back off. What I would suggest instead of the large flat area, is to grab the VE table from 20-100 kpa, 400 rpm up to 1,200 rpm and multiply the whole chunk of the table in that range by 0.85. Then grab the VE table from say 1,200 rpm up to say 2,000 rpm and use the Interpolate Y function between that area. What that will do is reduce your whole VE table down at lower rpm by 15%, then smooth out the VE table as the RPM increases. You want the VE table to have somewhat of a VE table shape before you start trying to learn the VE table using the WBO2 feedback.

As for the spark knock, you still want a smooth timing map as well. I would suggest a global decrease of say 2-3* from the whole timing table and see how it responds, at least until you get the fuel dialed in. Keep the same basic timing curve for now but, pull some advance from the whole thing. Once you get the fuel correct you can come back and work with the timing some more. The advance curve I tried to mimic is similar to the one I ran in my HEI on my 383 when I fired it up and first ran it.

Surging is usually VE table induced although too much idle timing can also contribute to a surge. I put the idle timing value at 25* as that typically works pretty well. You can experiment with less timing if you feel the desire. Perhaps it will smooth out with less timing. One thing I forgot to look at in the data log is whether the MAP AE flag was being triggered at idle. The vacuum oscillations of some large cams can actually cause the MAP AE to be activated as the idle vacuum jumps around, which causes the air/fuel ratio to jump around. If the AE is being triggered and the ECM is dumping fuel at random times that will also create a surge, which will then cause the MAP to dance around, creating an even larger change in fueling, which then just keeps compounding until the thing is surging like a cammed up blown dragster.

Thanks for reading my tuning "Ted talk".

 

[Rustjunky67]

Theres been a lot of issues with flat tappet lifters lately. The only ones made in USA anymore are made by Hylift Johnson in Muskegon Mich. The only retail supplier of these lifters seems to be Isky. The rest of em its hit or miss getting good ones. They are junk made in china or india with incorrect hardnesses, or not ground to allow the lifter to rotate in the bore as the cam lobe pushes it. Additionally I have left the weak valve springs installed just for break in with a flat tappet cam along with using Shell Rotella oil on break in. Then switch to your heavier valve springs.

I would suggest thoroughly reading the information here:

540 RAT - Tech Facts, NOT Myths

540ratblog.wordpress.com

I'm in the process of replacing cam (flat tappet) and Hydraulic lifters..... he's got me almost 100% convinced, from both his unbiased independent testing and the testimonials of people who have followed his advice, that I don't think I'm gonna do a break in OR use any zinc additive EVER.....
I think it's well worth considering......IMHO

 

[L31MaxExpress]

For roughing out the VE tables, here is an example of what I do. Grab the area where it is rich (could also be lean). Then adjust by the % of change I want to see in that area. If it is rich multiply by less than 1 to reduce the VE#. If it is rich, multiply by more than 1 to increase the VE#. Remember how I said that 11.8 vs 13.8 was 15% difference. Multiply the area by 0.85 to reduce the fueling 15%.

You must be registered for see images attach

You must be registered for see images attach

Once that is done, will look like this, as you see I then select the Interpolate Y tool and use the function to smooth out the VE table into an area where the fueling is closer. You can do this in relation to MAP change as well. If it is lean at idle and rich at higher map, you can multiply the MAP in the area it is lean or rich and interpolate X which would be across the VE table.

You must be registered for see images attach

You must be registered for see images attach

 

[L31MaxExpress]

Once you interpolate the MAP, it will be much smoother. You do not want to see large dips or spikes in the VE table.

You must be registered for see images attach

You must be registered for see images attach

 

[Moparmat2000]

I would suggest thoroughly reading the information here:

540 RAT - Tech Facts, NOT Myths

540ratblog.wordpress.com

I'm in the process of replacing cam (flat tappet) and Hydraulic lifters..... he's got me almost 100% convinced, from both his unbiased independent testing and the testimonials of people who have followed his advice, that I don't think I'm gonna do a break in OR use any zinc additive EVER.....
I think it's well worth considering......IMHO

I like that blog. Think I'm gonna just use that Quaker state 5w30 full synthetic in my next build.