I have a 327 with a WCT5, '747 and Holley 670, into which I recently installed a Comp 270 cam (218/224 @ .050", .495/.502, 110 deg separation). Fuel pressure at 15, steady. It idles (fairly smoothly) at 800 or so, in the 50 -60 KPa range. I starting tuning with an AHRT .bin (5.7l manual), where I worked out a number of problems (mostly pig rich) and got the fuel tables pretty smoothed out (between 120 and 140 in the "normal" driving areas). However, i've got a *wicked* intake backfire at high (though not PE) loads at low RPM's, 2500+ at 80-100 KPa, closed loop. Sort of a really extreme stumble. I've confirmed that the cam is aligned, valves are adjusted and quiet (though not hot - it's just too warm outside!). I fooled with the timing tables a bit in Tunercat (which is set physical at 6 deg, Initial Spark Advance matching), but I'm still getting the backfire - pulled out up to 5 deg in the general range of the problem. I've checked for vacuum leaks pretty thoroughly, though one is always possible. This is my first experience with a significant cam, and I don't want to melt anything. WinALDL is showing me a few knock counts (10/20 in 30 minutes of driving, in the heat).

I'm looking for suggestions as to where to go from here. I know this is an "odd" engine combination, but I imagine the same rules apply to a 327 as to any other SBC. I'm trying to "sneak up" on the correct tuning, but with this backfire I've already stepped in it, and I don't want to make matters worse. I'd be glad to post any other info. Thanks

 

What size combustion chambers?
The larger the more timing they GENERALLY need.

Post your timing table.

Did you do the ESC test patch?.
It's at the DIY-EFI.org site.
Might not be related, but you want to do that anyway.

 

Grumpy, the combustion chambers are Vortec 64cc, modified for high lift, with flat-tops giving about 9.5:1. Static pressure is about 185 +/- 3 psi, so that sounds about right. I'll post the timing table from the stock AHRT - I've done very little fooling with the timing.

20.0 20.0 20.0 20.0 20.0 16.2 13.0 9.1 6.3 4.2 0.0 -1.8 -2.8 -3.9 -4.9
20.0 20.0 20.0 20.0 20.0 16.2 13.0 9.1 6.3 4.2 2.1 0.0 -2.8 -3.9 -4.9
20.0 20.0 20.0 20.0 20.0 16.2 14.1 10.2 9.1 5.3 3.2 1.1 -0.7 -1.8 -3.9
20.0 20.0 20.0 20.0 20.0 16.2 15.1 13.0 10.2 7.0 5.3 3.2 0.0 -0.7 -0.7
20.0 20.0 20.0 20.0 20.0 18.3 16.2 14.1 11.3 9.1 6.3 4.2 3.2 2.1 2.1
20.0 20.0 20.0 20.0 20.0 18.3 17.2 15.1 11.3 10.2 8.1 6.3 5.3 4.2 4.2
21.1 21.1 21.1 21.1 21.1 19.0 18.3 16.2 13.0 11.3 10.2 8.1 6.3 5.3 5.3
21.1 21.1 21.1 21.1 21.1 20.0 19.0 17.2 14.1 12.3 11.3 10.2 8.1 7.0 6.3
22.1 22.1 22.1 22.1 22.1 21.1 20.0 18.3 15.1 14.1 12.3 11.3 9.1 8.1 7.0
22.1 22.1 22.1 22.1 22.1 21.1 20.0 18.3 16.2 15.1 13.0 12.3 10.2 9.1 8.1
23.2 23.2 23.2 23.2 23.2 22.1 21.1 18.3 16.2 14.1 13.0 12.3 11.3 10.2 9.1
25.3 25.3 25.3 25.3 25.3 23.2 22.1 19.0 17.2 15.1 14.1 13.0 12.3 11.3 10.2
26.0 26.0 26.0 26.0 26.0 24.3 23.2 22.1 18.3 17.2 16.2 14.1 13.0 12.3 11.3
27.1 27.1 27.1 27.1 27.1 26.0 23.2 22.1 21.1 19.0 18.3 16.2 15.1 14.1 13.0

It seems pretty conservative.

When I was experimenting with RT Tunercat and the Romulator, I'd take a couple of degrees out of the affected columns (from 60 up) to no or little affect; I also tried pulling timing by row at lower RPM's; then I tried both (the thinking being that it was "safe" experimentation). There was little affect, other than the change in the "feel". To tell the truth, I've been quite nervous about inducing these backfires - they're quite violent - blew out the paper type header flange gaskets a couple of times.

And no, I haven't applied the patch - I thought that it was for the '8746, though if I *can* apply it to a '7747 it would seem like a good idea given my problem. BTW, I'm running a 350-type ESC and knock sensor.

 

You said you took out fuel to correct a pig rich condition. How did you do this?

If you leaned it by injector constant then go back and use the correct constant and lean the VE tables.

From what you describe it sounds as if the engine may be on the lean side and as you open the throttle it goes lean and misfires. As the computer brings it back rich by closed loop action the raw fuel lights in the exhaust pipe creating your backfire.

If this is the case then you need more pump shot or more timing. You timing seems really conservative. I would think with those heads that a good place to start would be to set it up like a mechanical and move away from there I would start with 16 at idle 28 total at WOT and 8 or so for the vacuum. Then you can go line by line adjusting as needed.

The catch 22 here is that if you have to little spark advance the engine will be sluggish and require more fuel. You need to figure out whats going on fuel wise(make sure you always have a burnable mixture.)

HTH

John

 

The '747 ESC patch is called ESC_747.zip (mixed case) @ ftp://ftp.diy-efi.org/incoming/

With the Vortec heads the cast iron headed LT1 timing table is a good match (Caprice ran iron heads).

With that cam/heads you are going to need some fuel. This is both pump and injectors. Find out what you current injectors are and get them to flow at least 100 #/hr each. (fuel pressure or larger ones). You may need to go a little more of the flow. This you will find out later as the tune comes together. Your BPC should be at around 90 or so.

The backfire you are getting could be either SA or fuel, or both. At 2500 RPM the cam is starting to really wake up, so it is going to want some fuel.

BTW, what intake are you using?

RBob.

 

I set the BPW constant to 120 at the very beginning, using the formula BPW=1461.5*(vol/rate) with the volume of one cylinder being .67 liter and the rate being 8.2 g/sec. That's for the Holley 670's 65 lb/hr injectors. No! wait! that's for *one* injector! Should I calculate two? Total of 130 lb/hr? That would make my rate 16.4 g/sec, for an BPW of 59.7. Could this be correct??? I've been taking fuel out all across the VE tables, is this making sense?

 

Your first calc is correct (one injector). I would expect your engine to be making ~ 350 HP. You need more injector. Either bigger or more fuel pressure. Do your injectors half a black plastic hemispherical cap on top? Or do they just have a plug coming oout of the top front?

RBob.

 

No hemispherical caps - I believe that makes it the "later" (GM?) injectors. And, sorry I didn't post it earlier, I have the Edelbrock vortec single-plane. Also, the exhaust ports have been polished a bit.

My "target" power is in the 350 range, torque a bit higher than that, so I'm glad to hear it sounds right to someone else.

I installed an external regulator and on-regulator gauge, so increasing the pressure is zero problem - would that be 15*(100/65)=23 psi (assuming flow proportional to pressure)? Very do-able. That would change the BPW to 77.7. (BTW, is the unit for that milliseconds?)

So I'm thinking I'll do this: pretty much start at the beginning, put the BPW at 78, reset the pressure to 23 , spark table stock, get it to idle; rework the VE tables as necessary (downward) to compensate for the increased pressure, leave the pump-shot numbers at stock (as a baseline) until I clear the backfire, then tune for smooth transition to PE using the pump-shot. Once that's all done, sneak up on the correct timing. Finally full PE.

I have another question, though it probably should be a separate thread, concerning the VE tables: it has to do with how the system transitions from Table 1 to Table 2. Table 1 "runs out" (or becomes linear) above 3200: Table 2 continues to 6400. If I wanted to incorporate T2 into T1, I understand I'd simply add the numbers (keeping them below 100), and "0" T2. However, what do I do above 3200? Just leave the values alone? So T2 might look like this:

RPM % Vol Eff.
0 25.0
400 0.0
800 0.0
1200 0.0
1600 0.0
2000 0.0
2400 0.0
2800 0.0
3200 0.0
3600 33.2
4000 27.3
4400 27.3
4800 31.3
5200 33.2
5600 33.2
6000 33.2
6400 33.2

Then, if I scale T1 10%, I'd also scale T2 10%. Or would I "0" the whole table? I've tried to study the hack, but I'm just not a machine-code kind of guy.

 

You'll need to go to 28-30 psi for 100#/hr. It is good that you have the new design injectors, the old ones wouldn't take the pressure (why i asked). Flow change is the sqrt(newspi/oldpsi)*oldflow.

sqrt(28 / 12) * 65 = 99.3 #/hr

At 30 psi the flow is 102.8 #/hr

You will want the BPC set a tad on the high side, hence the 90 I mentioned previously. Also make sure you have a high pressure fuel pump, not a TBI fuel pump.

The two VE tables are always added together. Once engine is greater then 3200 RPM the 3200 RPM row is used. You don't want to do what you have above. Once you go past 3200 RPM you will be adding a lot of fuel. Along with no way to reduce the fuel below the 3200 level.

It comes down to how much lower will the VE be at 6400 RPM vs. 3200 RPM. You need to have at least that much room to remove fuel via the VE table.

This booger is going to scream once it's tuned up

RBob.

 

OK, pressure to 30 psi (it's a high-pressure pump and a recirculating regulator), and I'll kick down the BPW to 90. While I'm at it, I'll incorporate the VE2 table into VE1, now that I understand the algorithm. The only thing that concerns me is being able to get the idle comfortable; there's going to be some mighty short pulses in those injectors. Oh well, that's the least of my worries. If it's a problem, the cost of some larger injectors is pretty minimal, compared to the money "invested" (yuk yuk) in this thing.

It strikes me that this forum has been, not just a valuable resource, but has entirely provided me the knowledge and skills to pull off this project. I started this project with *zero* knowledge of fuel injection systems, and almost zero knowledge of high-performance smallblocks. If something like Thirdgen hadn't been available (along with DIY-EFI), I'm certain I'd have had to dump a carburetor on this thing and be done. And not only that: the amount of detailed, appropriate, and freely-shared knowledge contained in the forum archives (especially DIY PROM) is staggering. I researched this project by starting at the first posting in the forum ("sites for newbies to checkout" and Traxion's introduction) reading every posting up to the present. Printed things that were keep-worthy, now I have a pile of papers literally ten inches high, arranged by type of problem, sub-sorted by thread name, and cross-referenced. At moments, it seemed like a waste of time, but in retrospect it saved me literally *hundreds*, if not *thousands*, of hours of tedious, tiresome, and expensive experimentation. Heck, this little thread has likely saved me weeks. Even if I mostly lurk, I know all the names, and have seen all the hours and hours the core group here has spent patiently typing responses for people who are as needful of a bit of assistance as me. Really, you guys have *way* more patience than I do. So thanks, RBob and the others who hang out here and help, for Who-knows-what reason, guys like myself and all the others.

 

Your setup is similar to mine, I could send you my .bin if you want to have a look at it, or even try it. I'm running the 747 too.

 

Low, I'd appreciate it. The more I study these things, the more sense they make. Go ahead and send it to
scotttro@sbcglobal.net. Is there anything I should know that'd help me understand why you did what?