I have the idle and cruise worked out ok on my tune. I thought I would take it out and put it down a little bit. I did and I got a nasty backfire somewhere between 3000 rpm and 3700 rpm.

You can see that somewhere between 10.5 and 11.7 seconds I had the backfire out the intake. You all know that 7747 doesn't transmit data nearly fast enuf so the logger skipped over the actual event.

I'm assuming I have an extreme lean condition since the O2 voltage doesn't break .350 mV up to that point. I think the .782 mV at 11.7 seconds was fuel left over from the run, because at that time the TPS% was 3.1.

Any tips? I have PE disabled right now. I'm thinking I should add a ton of fuel at high load, hi rpm. I know this won't be the best for performance right now but I think I need to get the fuel close before I can try that one again.

 

No PE?
You should see or at least be close to a lean sneeze, if you're kinda in it.

 

Oh yeah it sneezed...more like a heart attack. I felt it coming but by the time it registered in my head it went A-CHOO!

I 'spose I'll turn PE back on...

 

Yea, I would turn PE back on. Even if you have it set to turn on at 85 - 90% throttle, at least it will be there to provide enough gas to hopefully prevent detonation. But what do I know, I'm just a newbie!

Mike (1bad91Z)

 

It isn't real detonation, just a lean backfire through the intake which can report false knock.
Definatly put PE back on. The timing looks really mild, what do you have your tables set at? Only ~13 degrees!?!?!?

 

I've been tuning in circles for a week or two now. So I decided it was better to crap up some spark plugs with too much gas than to keep shooting flames out the intake.

That was it. It was WAY too lean.

I also had this nasty bog just off idle when going from a rolling stop. The engine would hit 800 rpm x 80 kPa. It didn't have enuf timing. I cranked it up just in that row of 800 rpm x 75-90 kPa and it did wonders for it.

JPrevost...here is my lastest bin if you want to have a look. I think you see 13d advance from the datalog. 37 * .35 = 12.95. If I read Robert Rauschers spark paper right then I have to add 8d timing to that because of my physical dizzy setting. So that would make it ~20d.

 

It is a rare day today. One post area that I have always left alone is tuning. Just too much goes on that can't be felt/heard without being there. Now you did post both a data log and a bin and asked. . .

That said, I see a couple of things that I may question. The BPC is set to 133. To me this say'th the injectors may be too small. The value of 133 is OK for a lo perf motor. Once into engines that rev higher and make more power larger injectors are required. A larger injector would have a smaller BPC in the bin (like 90-100).

I feel the same about your AE conclusion. Just doesn't look like enough fuel. This can easily be covered by increasing the VE. Been down this road. Then whenever the engine ends up at that spot (in the VE matrix) for any length of time it just goes richer and richer. This is a sign that more AE and less VE is required. (AE is the most difficult area to tune).

SA: You have the initial set to 8 deg, distributor must also be set to 8 deg BTDC. The initial is now a wash, forget about it. Now, you also have an 8 deg main spark bias value set. This subtracts 8 deg from the entire SA matrix. So at 3K RPM and 100 KPA the main SA table is 23 deg. Sub'ing off the main SA bias (23 - 8 = ) there is only 15 degs BTDC of SA. Not a whole heck of a lot. The proper value will also depend upon gears and vehicle weight (like I said I am not there).

There is a little bit of slope SA added but not much. PE SA is also set to zero.

VE: At 3200 RPM and 100 Kpa it is 73% (FL1), at 3200 RPM in FL2 it is 40%. This gives a total of 113%. Which gets truncated to 100% within the ECM. That is a loss of ~ 14% VE from what the bin has. (back to the BPC of 133 and injector size).

It looks like at about 2400 RPM and 80 KPa the VE max's out.

If I have any recommendation at all it would be to get some fuel, bigger injectors or a Holley 4bbl TBI, first(!). Then move on to the rest of the tuning.

RBob.

 

Uh oh. I was under the impression that I could have 255 in both VE tables and that was max. CRAP...

A buddy of mine had a 454 TBI with 90 pph injectors laying around. He said I could use if I needed it.

You just hit the nail on the head with something I noticed today. I've been tuning using ((VE1 + VE2) * (BLM/128)) - VE2 to estimate fuel changes needed for VE1 to achieve a BLM of 128. I had the fuel table near 128 after I changed it. It worked well so I drove it at ~3000 rpm down a empty stretch of road for a bit and the BLM ended up being ~119 after that.

If I install 90 pph injectors is there any way to roughly rescale the VE tables after I change the BPC? I have the formula to calc the BPC. Is there any way to guess how far I need to turn VE1 or VE2 down after the BPC change? I just got it "traffic worthy" today. It would be a bummer if I had to start all over.

Thanks again RBob for saving me from circular tuning hell...

 

Start all over again. Not the answer you would want but you'll spend just as much time trying to guestimate the new BPW and VE1 and VE2 tables. Since you now know that VE1 and VE2 can't total more than 100 else it's seen as 100, then do the following. For the VE2 table, set everything at 3200 down to 400 to 0, leave the 0rpm cell at 25 or whatnot. Then just tune the VE1 table for driving around town. Use winaldl's correction tables to correct the VE1 tables, put em back in and it should be as close to perfect after only a couple runs. Also set your BLM min to like 95 and max to like 155. So long as you're driving around with a light foot the BLMs should be accurate enough.
I've found it best to set your chips initial SA to exactly what your dizzy is set at. I've also found it useful to visualze the Main SA table by setting main spark bias to the same as your initial SA. I don't have the time to look at the bin but RBob has and if there's anybody I trust with TBI it's him (and Grumpy).
PM me and I could probably give you a good SA map to start with, at least better than what GM has.

 

It's not. But if it's the RIGHT way then that's the way I'm gonna do it.

I got the TBI from my friend. Unfortunatly I won't be able to use the 2" body because of all the differences between it and the Astro TBI...which had a few small differences. Another project for another day. Anyway...it has the 90 pph injectors in which I will use. I got the pn's off them and crossed them with the list at diy-efi.

To rough the VE tables back in do you think I could "math" my way thru it? If I took 91 / 133 (new vs. old BPC) I get .6842. I then add VE1 and VE2. If I get 70% could I do 70 * .6842 = 47.89. Or is that just a bad idea?

 

There may be a misunderstanding on how the main spark bias works. This value (main bias) is subtracted from the final SA value. So in order to visualze the main SA table the main spark bias needs to be mentally substracted.

This can be tedious. If the main SA table shows 23 degrees, then the real timing is 23 minus the 8 degrees of bias. So unless the engine needs retarded SA somewhere in the table, setting the main spark bias to 0 makes it a whole bunch easier. Now a main SA table value of 23 degrees is 23 degrees.

RBob.

 

No need to do this as the BPC takes care of it. In theory when the injectors are changed and the BPC changed to match then the VE table has no need to change.

In practice it isn't this exact. Injector linearity, flow differences, etc make it so that the VE table will still end up needing to be tweaked.

Note that the AE calc is not based on the BPC. So this will change with a change in injector flow.

RBob.

 

Oh good. Tweaking isn't bad. So what do I do about my VE values that are greater than 100% ? I gather from your previous post that you were saying it needs more AE and less VE.

I 'spose for now I'll just limit the VE to 100% and crank the AE up and keep on tuning.

The AE flow makes sense. If the AE hold the injectors open for an addition X amount of msec's, then a 90 pph injector would HAVE to flow more than a 55 pph injector because more fuel went thru in the same amount of time.

Do you think it will be necessary to change the sync / async fuel mode settings?

thanks

 

Right, but what makes it even easier to visualize is when you've got your base timing set to 8 degrees, you then have your initial SA in the chip set to 8 and the main spark bias set to 8. Then what you see on the main SA table is what your engine is running. This is what I was trying to imply.

 

It does not work this way. Separate the initial SA from the bias SA. They are two different items.

The initial SA term is subtracted from the final SA value. It is then added back in by the initial (or base timing) of the distributor. So the initial term is a wash. It does not need to be taken into account when looking at the main SA table. What the ECM sub's the distributor adds (as long as they are both set to the same value, as they should be).

The main spark bias is (also) subtracted from the final SA value. There is no adding of it back in anywhere, it is gone. If you want/need to use a main spark bias value there is no reason it should match the initial SA. If anything it would help to use a value such as 10 degrees to make the math easier.

As an example: set the bin initial term to 6 deg BTDC, then set the distributor to 6 deg BTDC, set the main spark bias to 0 deg. At this point the values in the main SA table are what you get at the crank. The two initial settings match each other and the bias term is zero (no sub'ing).

RBob.

 

That's what I just did a minute ago I figure if I end up needing negative values in the main SA table then I'll use the bias. I turned the bias down to zero and then subtracted the same amount from the main SA table. Leaving me with the same values...they just look nicer

 

It does work the way I explained except I assumed you were using TunerCat. In his tdf files all of the Main SA tables are referenced by changes in the main spark bias, not the initial. So I was talking about TC. I should have made that clear.
An example of what I was talking about if my breaf explination wasn't clear enough; Look at TC main SA table. Open the constants and change the initial SA, go back and open main SA table, notice it hasn't changed. Now go back to constants and change main SA bias, open main SA table and there ya go, it's changed according to the bias.
I know it's weird and I probably should change that with the tdf editor. Oh well.

 

OK, I knew I had to be missing something (I PM'd you earlier). How TC handles these values is correct. If the initial is changed in the bin and changed in the engine, then the main SA table is the same.

RBob.