I've got a 400 small-block, GM 454 big block TBI, flow-matched 65-lb injectors, Mallory external pressure regulator at 22 psi, on a Holley ProJection TBI manifold (rated "Idle - 6000 RPM"). 4-wire O2 sensor.

My total build details are in my sig.

The problem is, I have a "flat spot" between 1700 and 2000 rpm. Right in the highway cruise range (65 mph - 75 mph is 1750 - 1950 rpm)! It occurs under any load, any throttle position. It can cause the car to lose speed with the cruise set, which tries to compensate, then it gets rough and might even give a backfire before it starts to pull. It's a little worse at cruise once the ECM enters highway mode (5 degrees additional advance, O2 sensor reading about 0.04 volts).

I had this same problem when this same system was on my old 305 build, only worse because of its radical cam.

The ECM is the stock '88 1228063.
The bin is stock AKRM, with EGR test turned off and BPW set for the displacement, injector size, and fuel pressure. (It calculated out to 113.)
No codes.
BLM's range 116-132.
O2 sensor readings fluctuate normally.
MAP operates full range. MAP at steady cruise in this speed range, on level road, reads 36-40.
TPS is working correctly, across its full range.

Since it's worse when the ECM enters highway mode, it appears to me that it's a lean condition. But the O2 sensor doesn't say so.

The only other little problem is, I'm sure it's going lean at WOT. But I haven't done any tuning on the PE tables yet, so I'm not concerned about that.

I plan to add the EGR back on after a few tanks of fuel. That might help it in the cruise range. But it certainly won't have any affect on the flat spot under acceleration.

I'm going to call Holley's support to see if it might be something with the manifold. But I doubt it.

I thought about a spacer under the TBI, but there' barely room under the hood for the stock air cleaner now.

Any ideas?

 

I am going to tell you what I found out and what I told Teelton when his Vortec headed 4.3 was doing the same thing. It was lame under load at part throttle, but would rip when he hit WOT. I would move the TPS% for PE to a lower setting and try again. Lets say PE at 40% TPS.

Also check your initial timing at the distributor to make sure it matches what you have in your .BIN.

I have always had trouble with stock GM like advance curves with even a mild cam.

 

Hey, thanks. I had briefly considered that, but somehow forgot it. I found on the old 305 that setting my PE/TPS as low as 25% worked wonders.

I started back with a stock AMYB because the cam is so mild. So far, except for some small irritations, it's working great. It'll "light-em up at will" and is giving me 20 mpg on my commute, without the EGR installed yet.

As I tune, I'll be sure to work the PE/TPS down in steps until I find the sweet point.

Also, I noticed that the "Highway Mode Threshold' is set at something like 98.9 kPA MAP! No wonder it won't come out of highway mode until I really push it down, or I drop the throttle fully closed! I'm gonna move that down to around 50, then adjust up or down until I like how it acts. Any opinions on that?

 

Thought I'd update you.
PE vs TPS threshold is now 25% at all rpms except 1200 - 2000, where it's 20%. This makes a huge difference in the overall throttle response of the motor, but only helped the flat spot a little.

By the way, this makes a lot of sense. If you think about a 4-bbl carb, PE comes in when the secondaries start to open. With a carb, you have to think of throttle percentage as percentage of air flow rather than plate position. So in a square-bore carb, PE starts at 25% throttle. In a spread-bore (which has better low-end response), PE starts at 20% throttle, or as low as 15%, depending on the difference in the throttle plate sizes.

Another big help from I got from you ()-- I looked at the master spark advance table. Man, what a mess! I know how and why GM did it that way. But sheesh!

Anyway, I got a wild idea. I went to the MSD ignition web site and pulled down the timing curve stats for their street performance distributor. I chose which set of springs and bushings would fit my build. I also figured on connecting the vacuum advance (10* total) to manifold vacuum, the best setup for performance engines. Then I totally emulated that distributor in my main spark table. Here it is....

> 0* base timing.

> 0-40 MAP (full "vacuum" advance):
>> 16* at 400-800 rpm (covers idle)
>> 34* at 3200

> 70-100 MAP (zero "vacuum" advance):
>> 6* at 400-800 rpm (covers idle)
>> 44* at 3200

> 55-65 MAP: even 2-degree steps with MAP change

> Advance rate is nearly staight-line from 800 to 3200 rpm. Rate is slightly faster up to 2200, where it slows a bit. (There's a very slight "peak" in the curve at 2200, where the advance rate changes just a little).

> 2 degrees per 1000 slope above 3200.

In the TunerCAT graph view, this is like nothing I've ever seen, but it's definitely clean and smooth. I've never heard of anybody taking this approach, but I figured it had to be better than what was there.

The result? Wow!
> Big jump in power and throttle response.
> Easier cold start.
> Much smoother cold fast idle and decay.
> MUCH improved warm idle (very stable).
> Major reduction in that flat spot (but not gone completely).
> Engine "feels" much smoother at cruise from 40 to 80+ mph.
> Improvement in fueling at WOT, but I think I'm gonna have to get bigger injectors if I want to wind it up tight very often.

So I think I'll leave the spark table as it is now.

 

I started out with the old Corvette L82 advance curve when I had my L82 cam in my 312.

 

Here is what I am running in my current 350, ported 081 heads, ZZ4 cam.



I have 5* of PE timing.

 

Well, I found the problem, and discovered something new about tuning. (My ECM is a 1228063. Very similar to the '7747, but it uses an Intake Air Temp sensor and has some table differences.) Apparently, the ECM does more than just add the values in the #2 Fuel table to the #1 table values.

BLM's in the range 1600-2000 rpm and 30-50 MAP were running in the 160's. I tried rasing VE settings in the #1 table, with zero affect. I ended up with a ridiculous high peak in the middle of my #1 table graph, adding as much as 40 to the VE values in that range. Same high BLM's. Same flat spot.

In looking at the graph on the #2 table, I noticed it had a "dip" in that RPM range. So I just took the dip out, giving the graph a smooth transition between the ranges above and below that range. To take the dip out, I only had to increase the values there by 8.

Then I put the #1 table back as I had it before. With the new values in #2 table, BLM's dropped to 130 and the flat spot disappeared.

If the #2 table values are simply added to the #1 values, then I should have been able to make the corrections in the #1 table. But that didn't work. Only when the correction was made in #2 table did the problem go away.

I'm also keeping the spark table that I put in trying to fix this (emulating a Mallory Street Performance distributor). It's really sweet.