My grandpa has a sb 400 with a tpi setup from a 91 firebird in his '32 studebaker. The problem is the engine falls on its face at 3500 rpms. I have been doing some research and I read that the stock tpi manifold doesnt flow enough air for the 400. Is that accurate? We have an LT1 efi setup also, minus wiring and ecu that we could use, but we don't know what is involved in the swap. We are also open to using an aftermarket manifold if that would be better. The goal for the engine is to have a usable wide powerband we arent looking for something that makes tons of top end power, we just don't want it to fall off so fast. Any information would be greatly appreciated.

 

heres the dyno sheet

 

My guess is he:
A) Put a tiny cam in there.
B) Put small port heads on there.
C) Has a severe choke in the intake somewhere
D) Has a severe kink in the exhaust somewhere
E) Never set the timing correctly

A 400 with good parts will still be able to make more power and torque than that with a TPI on it. With really good parts, magazines and other people have shown that even though power levels off at 4-5000 RPM, it can still be pretty high.

It takes more than cubes to make power. It takes good combinations and good tuning.

 

There is something besides the intake that is wrong with the system.

It's a MAP system, so we will start with the obvious -The Prom-.

What kind of heads/cam/exhaust he has would be my second consideration. A MAF system would have fallen apart under those circumstances and bad Eprom work, but a MAP system doesn't even stand a chance if not done correctly for 400ci.

 

The wideband readings look acceptable, so I don't think the MAP prom tune is truly to blame here. A scantool may reveal some problem though.

Something is probably choking the intake side. Check to see if the throttle blades open all the way. But my bet is that he built one of those "hotrod pump gas driver specials" with 8.0:1 compression on bad heads, and a smaller than stock cam.

 

Stock heads, it has a cam i'm not sure on the specs, but it doesnt seem like anything too big, long tube headers and 2.25 header back exhaust. It has a custom chip. The guy who tuned it has tons of experience with tuning gm efi setups. His personal car has 440whp and getrs 25 mpg. I'm not saying he didnt mess up the tune, I'm just saying it wasnt done by some guy with a burner and no clue. By looking at the sheet and riding in the car it feels and looks like the ecu is pulling a ton of timing. Do these systems have a knock sensor?

 

Yes, they're supposed to use a knock sensor. The sensors don't work very well when used in different displacement blocks. Do a search on knock sensors and different blocks, and you'll see a ton of posts probably. You could disconnect the knock sensor (but listen carefully for knock) during a dyno run (or a romp down the street), or simply put on a scantool and watch knock counts and knock retard.

 

I know that on the import engines i usually deal with if you disconnect the knock sensor the ecu will usually pull about 6-10 degrees of timing so will the gm ecu do the same thing. I don't think an 8:1 motor would make that much torque so low in the rpm range. It had much more power before the tpi swap so I don't think it's anything to do with the engine or the exhaust system. Intake or timing problems seems more likely to me. Would the timing chain cause enough noise to make the knock sensor think there was detonation? And do these systems use a vehicle speed sensor?

 

A loud gear timing drive would, but if he just used a run of the mill timing chain I wouldn't see a problem there.

And I totally missed the a/f chart below... whoops

 

I've got a TPI 400, it dynoed 338rwhp, and 448 rwtq. Both power and torque peak were at 4000RPM. It has AS&M runners and intake. The power should have peaked much higher but I found some major engine problems which were preventing that. It should be in the 370-400rwhp range when its all fixed. So TPI can support that much power, just not stock TPI.

LOL, just looked at the dyno graph numbers. My stock TPI 305 did better than that, 230rwhp. Just with dual exhaust and headers. Stock chip too. Something definately wrong with that.

 

I agree in that the TPI is not your major problem. What kind of an intake system is on the car and the size of the air filter? You should be running at least a 2 1/2 inch exhaust assuming dual exhaust. Knock sensor could very well be timing out and cousing a big power loss.

 

The exhaust is 2.5" I know I said 2.25" before but that was wrong. I don't know what air filter it has I'll find out tomorrow. I also found out that the headers were glowing red after some hard acceleration. Since the a/f is ok then the only thing I can think of is timing. Thanks for all of the help guys any more suggestions would be awesome.

 

what are the specs of the CAM and CC of the heads? what about the pistons? are the dished or flat topped?

 

I dont know the specs of either but i do know the had a ton more power with the holley carb and edelbrock intake manifold. So I dont think it has anything to do with the engine. There is no audible detonation, we only put 92 octane in it nothing lower.

 

It could be getting "false knock" from some other engine noise, causing the computer to remove timing when it doesn't need to be removed, thus cooking your headers.
But first:
Check the base timing (est bybass disconnected), then connect it and free rev it just to check that if everything is working. Base timing should be 6 deg BTDC. With bypass connected, idle timing should be 18-22 degrees. Free revving should go up to about 40, depending on the calibration.
If all that checks out ok, then make sure the timing marks on the damper and block are correct. A piston stop would help determine that.
Setting timing on an efi car cannot be done "by ear." It must be done with a timing light.

 

How do I disconnect the bypass? Thanks for all of the information.

 

It's a Tan and Black wire in a stock harness. Found near the firewall on the passenger side. Stock harness has a single weatherpack connector for it (the connector is in the middle of the wire - not at the distributor or anything). You may want to add an accessible connector if your harness is butchered, or aftermarket, and/or tucked away neatly.

While disconnected, a check engine light will come on. It should go away when reconnected, and restarted. Set timing with it disconnected to 6 (stock).

All timing is relative to this, and only the knock sensor response will remove timing (it won't add more than the ECU's calibration, just remove timing when it hears knock - and it learns nothing about knock, just reacts if it's there).

 

I know the ecu retards timing when the knock sensor detects knock, thats why I think the engine doesnt make the power it shoud and why the headers glow. Retarded timing causes high egt's and kill's power. Thanks for the info on the bypass. I'll check it out this weekend.

 

Has anybody used tunerpro rt? Will it work on the tpi ecu?

 

Yes, it works quite well.