SD problems with modifications, who has them?

[Ryan@FRP]

It's not really tuner friendly because, it's relying on a sensor that measures vacuum or lackthereof to determine engine load. The engine is pulling 19 inches of vacuum (just a number I picked) in stock configuration. Now you port the heads, put a different intake on, bore the motor out some. Now it still pulls about 19 inches of vacuum, but the computer still thinks that it's a stock configuration. There is not really a decent adaptive memory on OBD-I cars (except mid 90s when they were getting ready for the OBD-II crossover). The problem with tuning Speed density is, you really don't know how much air is entering the engine, without a mass air sensor, and potentially a wideband O2. so if you were to go to all the trouble to fit your car with a MAF sensor for tuning purposes, it would be a lot less annoying to make it a Mass Air system. It's more straight forward to tune a MAF configuration than to tune a Speed Density one.

I'd love to see what you come up with though. I for one would love to see a SD tuned panther. I have the quarterhorse software and I'm running an A9P (Mass air auto) in my vic, with a quarterhorse. Though my next panther, whenever I get it, I'd like to keep Speed Density for the convenience of plug and play and not having to mess with any wiring.

Sure it is, it just requires a different approach. As I mentioned in a previous post, you have to forget everything you know about MAF, as it doesn't really apply to SD tuning. You take a tune, either stick in closed loop and datalog trims by MAP or load and rpm, or force open loop, and datalog AFR by RPM/MAP. Make changes as necessary, retune, rinse,repeat. Not too different from MAF in that sense, except the table you're making changes in is a multi-dimensional table, as opposed to just a single column MAF transfer table.

By adaptive memory , you are referring to fuel trims, yes? So what does the EEC-IV do? Just reset itself from time to time? I'm not questioning you, I just don't know much about these ECU's. I know about SD, which is why I am here.

[Ryan@FRP]

Affordable is the key. More often than not, junkyard MAF swaps are cheaper. In case you are not aware, the Ford eec-iv SD is not directly programable, which plays into Mustang MAF swaps being the norm for high performance guys. Mustang material is common. Aftermarket Lo-po SD support is nothing.

For SD mods you got things like the SCT 6600 http://www.sctflash.com/products.php?PID=31 it is a piggy back chip for SPECIFIC SD computers. So even if you wanted to do this with your cv/gm, you would still need to swap to a compatable (usually Mustang) SD computer to pull it off. Back in the "old" days, SD mods where mostly single tune piggy back chips. If your going to swap beyond a basic HO conversion, most opt to do a MAF swap, since you can usually source all the bits 'n pieces for realtively cheap.

Alex.

I have to ask- Can a swap to a mustang SD box like a DA1 be plug-and-play? From what I have seen it seems like SD ECU's (even the ones that are supported) go for dirt cheap, people practically give them away. The A9Ls and A9Ps seem to go for decent money. I'm not sure how a junkyard MAF swap could be cheaper than finding another SD box that needs no sensors, and can be slapped in. Unless you are implying that a junkyard MAF swap typically runs better without a tune.

Not sure what you mean by not "being directly programmable", unless you are referring to the lack of support for the SD bins, especially for the Panther cores.

I am somewhat familiar with SCT's stuff, as its what I use on the OBDII Mopars I tune.

[Ryan@FRP]

i would like to have someone figure out sd as ive been fighting the hesitation and surges for 3 years now with the ecu showing no codes. my other cfi car runs better than my sd wagon. the sd is so iregular as it will run like a top for hours then with the flip of the switch it runs poor at idle, part throttle issues, and idle surges. i thought for sure it was my egr system but the egr eliminator didnt fix all of the issue. every day i get closer and closer to thinking of a carb swap. i wish i didnt live so far away from most of you guys because people out here dont no much about anything.

Tunes can be sent by email, I do it all the time with SCT. Matter of fact, I do ALL my tuning by email. No dyno here, no fancy shop, I'm just a dude in South Florida with a laptop.

If the supported SD ECUs are indeed plug and play (which it sounds like they are), itd be as simple as getting one, and getting a means of datalogging (Quarterhorse, etc). The datalogs will tell the big picture.

[zoomie]

I think I can get a definition for a D9S, I have a LUXO (for a DA1) here on my laptop. If you can get a datalog for me, do you want help?

that would be awsome, it will be a bit before i can get back to the shop to datalog, but definatly

[Crownvicman289]

Lots of misinformation floating around here, but Ryan has a handle on it. When I spoke to actual tuners, I was told the SD EECs were extremely limited, like there were clamps on the values that could be plugged in which severely limits how far you can go on a SD even with a tune. I don't think a GT40P setup with stock cam and bolt ons exceeds this, but that's merely speculation. A wideband is an absolute must for this sort of thing, no question about it. So many people underestimate the importance, they'll blow a couple grand building a motor, but won't even spare $200 for an AEM wideband, so even a mail order tune here would only be partially effective. I'm assuming the cheap way to do this would be to datalog with WinALDL?

There isn't a terrible limitation on the learning curve of SD EECs, the limitation is how they learn. They learn engine load (MAP) vs ECT. At low RPM and part throttle, you can easily hit atmospheric pressure. That adaptive value also applies to WOT within the respective temperature cell, so that's why adjustable FPRs don't work out long term. There is one way I think I've found to trick the MAF EECs though: Apexi SAFC. I have an unused one and could trigger it at a TP high enough to be open loop, so the EEC would never learn what I was doing there. All it would be good for is WOT tuning, and that's assuming closed loop did a good job of cleaning up the lower load areas to include transient fueling, which I believe gets absolutely nothing from adaptive learning.

[Ryan@FRP]

Thank you for the vote of confidence. I admit that I am still learning as a tuner and am open minded enough that someone within this thread can teach me something, even if it's specific to Fords.

I'm perusing a DA1 bin/LUX0 strategy at the moment. This appears to be the SD ECU of choice. Obviously there is a difference between changing values in the tune versus applying said changes on a running vehicle (being from the Mopar world I have all too often encountered the saying "the software doesn't always work as it should"), but I don't see anything that would hinder the ability to tune, within reason.

On older vehicles like this, I would agree that a wideband is important. Tuning via open loop may very well be best on these. On newer vehicles, I can trust the O2 sensors to tell me accurate trims, and I have a special trick (at least it works on DCX stuff, hopefully Ford is the same) to hold AFR in place in open loop. As in, acheiving targeted OL AFR with minimal deviation. When I know the trims are where I want them to be across the map, the wideband is not really needed. That's not to say a wideband isn't a good idea to have, of course it is. On boosted builds, a wideband or dyno AFR reading is an absolute must.

Map vs. ECT? Hmm. I would suspect that the reason for fuel pressure changes not working long term is because of fuel trims. MAF or SD, if your trims are off, you're going to have issues with part throttle drivability and WOT AFR regardless. Correct the trims, and you correct the problems. If you dial in VE enough for there to be nothing to adapt, then there is nothing to...well...adapt.

Acceleration enrichment doesn't get affected by adaptive learning.

Lots of misinformation floating around here, but Ryan has a handle on it. When I spoke to actual tuners, I was told the SD EECs were extremely limited, like there were clamps on the values that could be plugged in which severely limits how far you can go on a SD even with a tune. I don't think a GT40P setup with stock cam and bolt ons exceeds this, but that's merely speculation. A wideband is an absolute must for this sort of thing, no question about it. So many people underestimate the importance, they'll blow a couple grand building a motor, but won't even spare $200 for an AEM wideband, so even a mail order tune here would only be partially effective. I'm assuming the cheap way to do this would be to datalog with WinALDL?

There isn't a terrible limitation on the learning curve of SD EECs, the limitation is how they learn. They learn engine load (MAP) vs ECT. At low RPM and part throttle, you can easily hit atmospheric pressure. That adaptive value also applies to WOT within the respective temperature cell, so that's why adjustable FPRs don't work out long term. There is one way I think I've found to trick the MAF EECs though: Apexi SAFC. I have an unused one and could trigger it at a TP high enough to be open loop, so the EEC would never learn what I was doing there. All it would be good for is WOT tuning, and that's assuming closed loop did a good job of cleaning up the lower load areas to include transient fueling, which I believe gets absolutely nothing from adaptive learning.

[gadget73]

The ECM will learn and counter any changes made to fuel pressure to a point. You can get the pressure so high that its outside the ECM's ability to adapt to it, and then it just runs fat across the board which is also useless. The small handful of people that I've run across that understand the ECM learning strategy, or at least some of it, all say that the AFPR is a useless investment for this exact reason. It made sense to me anyway and when my AFPR failed, I replaced it with a stock OEM one.

[Ryan@FRP]

The ECM will learn and counter any changes made to fuel pressure to a point. You can get the pressure so high that its outside the ECM's ability to adapt to it, and then it just runs fat across the board which is also useless. The small handful of people that I've run across that understand the ECM learning strategy, or at least some of it, all say that the AFPR is a useless investment for this exact reason. It made sense to me anyway and when my AFPR failed, I replaced it with a stock OEM one.

Not sure if this is implied in your post, but yes, the fuel trims will correct the changes whilst in closed loop. It can only correct so far, as you've stated.

It would be a little different if the ECU was forced into open loop- but, if you went that far, you may as well tune the ECU the right way, since it is indeed possible.

How much of a pain would it be to make a DA1 work in one of these cars?

[86VickyLX]

Not sure if this is implied in your post, but yes, the fuel trims will correct the changes whilst in closed loop. It can only correct so far, as you've stated.

It would be a little different if the ECU was forced into open loop- but, if you went that far, you may as well tune the ECU the right way, since it is indeed possible.

How much of a pain would it be to make a DA1 work in one of these cars?

Is that a Mustang Speed Density computer? It is plug and play. Granted you gotta configure the engine to the computer (different cam, injectors and such), or you repin the injector harness, but then you gotta tune the computer to accept that the firing order is different. You will also have issues if it's an 88+ panther with lack of cruise control (computer controlled, mustang didn't have cruise).

[Crownvicman289]

Map vs. ECT? Hmm. I would suspect that the reason for fuel pressure changes not working long term is because of fuel trims. MAF or SD, if your trims are off, you're going to have issues with part throttle drivability and WOT AFR regardless. Correct the trims, and you correct the problems. If you dial in VE enough for there to be nothing to adapt, then there is nothing to...well...adapt.

The trims are applied via MAP vs ECT on the DA1, TMoss posted a snapshot of the map over on the corral. The trims are only off if A: they max out or B: the O2s are junk.

I've got a DA1 in my car right now. With the 289, the only issues I noticed was a slight tip in hesitation around 1200ish RPMs if I lugged the motor, but even then it wasn't there 100% of the time. I haven't run the new motor much, but at 39psi there was a massive tip in hesitation and at 41psi it went away. Since there were few temp installed items, I don't really count the way it ran as accurate just yet. Another week or so and I should have a bit more info on it.

[86VickyLX]

The trims are applied via MAP vs ECT on the DA1, TMoss posted a snapshot of the map over on the corral. The trims are only off if A: they max out or B: the O2s are junk.

I've got a DA1 in my car right now. With the 289, the only issues I noticed was a slight tip in hesitation around 1200ish RPMs if I lugged the motor, but even then it wasn't there 100% of the time. I haven't run the new motor much, but at 39psi there was a massive tip in hesitation and at 41psi it went away. Since there were few temp installed items, I don't really count the way it ran as accurate just yet. Another week or so and I should have a bit more info on it.

What voltage do you have the TPS set at closed throttle? Should be at no higher than .900v

[gadget73]

the ecm should self-zero the tps value at each key-on as long as its between like 0.7 and 1.2 volts or something like that. It doesn't need to be at some magical voltage point to work right.

[86VickyLX]

I have found that the cars that I set to .9 volts have better throttle response than those not set there.

[1990LTD]

Saw them mentioned in another thread, and now that I think of it, my engine will be almost identical to a Thunderbird HO engine since it'll still have the E6s with the Cobra cam. Those came with speed density from the factory...I shouldn't have any problems when it's actually up and running.

[Crownvicman289]

I have found that the cars that I set to .9 volts have better throttle response than those not set there.

Mine is set to .98. 1V is the magic number to not exceed from all the reading I've ever done. I've never heard of .9V to be the magic number.