TPI Guy: I have a DUI question for ya

[RWC]

What exactly is the best timing curve? I've had problems working mine out on my Pro-Jection, although I think I'll largely blame Holley for making it vary with RPM instead of with manifold pressure. Also, what do you suggest base timing be set for (ie., how many degrees as timed with a light and with the TPS unplugged)?

Right now I have my base timing at 10 degrees, but then almost no additional timing for the first 2000 RPM, then I kick in an extra 20 degrees for a total of 30 by the time I hit 2500.

Thanks!

RWC

[Veepster]

all EFI systems use RPM as the basis for timing...that is the only way to do it.

your base timing is OK but I would make the timing curve kick in more gradually....

[RWC]

Whoops; my fault. Of course timing varies with RPM. The Pro-Jection also has it vary with throttle-position, whereas I think it would make more sense to have it vary with manifold pressure. Apparently Holley must have thought so too, since the new version (the Commander 950) does it this way.

RWC

[Ralph]

I think any EFI system that uses a MAP sensor will have at least some, if not all, of the spark curve tailored to both RPM and manifold vacuum because these are the the aspects of engine load most easily read by simple electronic means.

[RWC]

Um...no. The Pro-Jection 4Di software that I have has the x-axis as RPM, the y-axis as throttle position, and the z-axis as timing advance. Obviously the primary fuel map uses manifold pressure, but not the timing curve.

RWC

[TPIGUY]

<blockquote>quote:</font><hr>Originally posted by RWC:
Um...no. The Pro-Jection 4Di software that I have has the x-axis as RPM, the y-axis as throttle position, and the z-axis as timing advance. Obviously the primary fuel map uses manifold pressure, but not the timing curve.

RWC<hr></blockquote>

RWC,

Using the TPS for spark advance is an acceptable way to program spark advance. But first, let me explain something. Everyone knows that spark advance is controlled from both vacuum and RPM on your non-computer Jeeps right. You've got centrifugal weights inside the distributor and you've got a vacuum motor on the side. Well, MAP (manifold absolute pressure) is just the inverse of vacuum, and in a GM EFI system, spark advance is based on a 3D table with RPM and MAP as two axes, and spark advance as the dependent variable. Look at the last table in this link for what my spark advance table looks like in my truck (www.customefis.com/GMEFI.html). However, MAP and the TPS signals are almost identical in shape as you drive. I have graphed both signals from the same run and although they are not at the same scale, the shape of the curves is nearly identical. So, all it takes is a rescaling of the TPS value and you can program spark timing from TPS and be close enough. I'll let you in on one of my tricks. Look at the table in my site again. Notice that at the 600 RPM row, the spark advance stays constant to 60 MAP where it starts dropping until it is again steady at 85 MAP. This corresponds with about 12 in Hg vacuum down to 4.5 in Hg vacuum. Since RPM is constant, this row is the vacuum only contribution. Following the same reasoning, look at the 100 MAP column and see that at 600 RPM you have 10 degrees advance going up to 23 degrees at 3600 RPM. THis is the centrifugal only contribution. If you take these two contributions, and add them in an EXCEL spreadsheet, you end up with a complete spark table with both contributiuons. I create one table with the MAP only contribution, and one with the RPM only contribution, and add each cell from both tables to arrive at my spark tables for non-GM engines. You also have to add in the distributor setting to get total degrees.

As for what I use in my Jeep EPROMs, I'll try and post the table here and you can pick out the numbers I hope.

Won't work, I'll have to post JPG images on my website tomorrow. I'll put them under www.customefis.com/jeeptime.html (http://www.customefis.com/jeeptime.html) My wife has to do it because I don't know how to use the screen capture program yet.

I will say that DUIs curve is weak at mid throttle which leads to sluggish performance, but strong at full throttle high RPM. The problem is that for most GW drivers, they never hit 3600 RPM at full throttle. DUIs curve is more suited for AMC Javelins that Grand Wagoneers.

TPIGUY
www.customefis.com (http://www.customefis.com)

[RWC]

Therefore, the slightly older Pro-Jection that I have only has the capability to directly generate the "centrifugal only" contribution, since my second variable is going to be based on throttle-position. But nevertheless, I think I understand what you're saying since throttle-position is going to vary directly with MAP, and so throttle-position is an acceptable substitute for MAP in generating the "vacuum only" contribution. I appreciate your pointing out the obviousness of that to me since I hadn't thought of it that way myself. smile.gif

I'd still like to take a look at what you use for a Jeep spark-map, at least to help me get an idea of where to start from. I'd also like you to confirm that having my base-timing set for 10 degrees is proper. My other thought was to bring it back down to five or zero and have my ECU do it all. According to your FIGURE 2 it looks like you're base timing is indeed zero.

I understand your critisim of the DUI curve being weak at mid-throttle, but I don't understand how that could possibly apply to my DUI since it doesn't use a vacuum-advance, and since I'll be building my own spark-curve in my ECU.

Thank you very much for your help!

RWC

[RWC]

I just wanted to bring this back to the top in case TPIGUY missed my last post.

Also, the link to the jpeg images isn't up (no pressure). :D My engine is STILL at the shop, ;( so it'll be at least another week before I get to play with my ECU again.

RWC