'7060 to '7427 conversion - remaining fuel trim issues

[Daveo91]

In a previous post (http://gearhead-efi.com/Fuel-Injection/showthread.php?10092-Code-43-Knock-Sensor), I mentioned I was taking a break from spark advance tuning and 10s of thousands of fake knock counts. Now have knock retard turned off and am running stock L31 timing tables which are more conservative than what I had before. I'm now working on getting all fuel trim related items right before I go back to tackling spark and knock. Remaining issues are below. Any help or advice on any of these would be much appreciated!

1. Cold start issues. Actually starts fine, minimal cranking. But after a couple of seconds of normal operation I get a swinging idle and it will almost stall. WBO2 pegs lean when it wants to die, then swings rich when it recovers. I haven't verified, but I think stall saver is kicking in? I can resolve this a little by raising the idle with my foot, but until it warms up it will go through these cycles. In drive is worse and reverse is by far the worst. All these problems are resolved when it goes into closed loop EXCEPT reverse. Even fully warm, reverse with zero TPS it will want to die.

2. Hesitation at slow takeoff from stop. If I punch it from stop, no issues. But a slow gradual take-off will produce a noticeable (even to passengers) hesitation. I'm guessing this is due to a problem in transition from idle VE table to off idle table, but I'm struggling with how to resolve it.

3. Rich Deceleration. Most of the time taking my foot off the gas completely results in AFRs on the WB in the 11s or even high 10s. This continues all the way to almost full stop, but at 5 mph and slower AFR is going back up to 13s and 14s. At full stop it idles nicely in the 14s usually. As far as DFCO is concerned it's either never coming on or it's very seldom. I will play with DFCO but it seems like I should be resolving the rich decel first, without DFCO.

4. And now a general question for a NOOB to WBO2 tuning. I have the spreadsheet that works with the '7427 (Dave W) and I understand the general principal of making adjustments to get AFRs around stoich across the board. But the part I don't understand - we're making the changes to the VE table and we're using CL to tune. Won't the computer be making adjustments all the time based on the NB and theoretically the WB would just be seeing the the already corrected values? It doesn't seem like it would be capturing much to adjust? Would using OL to WBO2 tune work? It seems like you'd zero in on the corrections needed to the VE tables much more quickly that way.

[dave w]

Interesting Pictures Below.

Same data logs BLM vs. AFR

Approximately 170,000 data points!

dave w

17181

17182

17183

17184

17185

[Daveo91]

Interesting Pictures Below.

Same data logs BLM vs. AFR

Approximately 170,000 data points!

dave w

17181

17182

17183

17184

17185

Thanks for the comparisons, Dave. I think what I'm supposed to be seeing is the wider range of info the WB data logs are providing? I guess what I'm still not understanding is if we adjust the VE tables based on that info, will the NB closed loop system just wash them out and do what it thinks is correct? Another thing I'm not understanding is: how do we make adjustments to the areas that we DON'T want AFR to be 14.7, like decel, highway cruise, and PE? Aren't the adjustments we're making skewing everything towards 14.7?

[dave w]

Thanks for the comparisons, Dave. I think what I'm supposed to be seeing is the wider range of info the WB data logs are providing? I guess what I'm still not understanding is if we adjust the VE tables based on that info, will the NB closed loop system just wash them out and do what it thinks is correct? Another thing I'm not understanding is: how do we make adjustments to the areas that we DON'T want AFR to be 14.7, like decel, highway cruise, and PE? Aren't the adjustments we're making skewing everything towards 14.7?

I look at AFR's and BLM's as measurements. I consider AFR's more accurate then BLM's. Adjusting closed loop with AFR's is " same / same" as adjusting closed loop with BLM's. AFR's and BLM's are measurements. When closed loop AFR's are lean, add fuel to the VE Table When closed loop BLM's are lean, add fuel to the VE Table.

Course Tune / Fine Tune. Course tune is basically correcting the VE Tables. The computer bases the areas that we DON'T want AFR to be 14.7, like decel, highway cruise, and PE on an accurate Course Tuned VE Table. Fine Tune adjusts the areas that we DON'T want AFR to be 14.7, like decel, highway cruise, and PE.

dave w

[Daveo91]

I look at AFR's and BLM's as measurements. I consider AFR's more accurate then BLM's. Adjusting closed loop with AFR's is " same / same" as adjusting closed loop with BLM's. AFR's and BLM's are measurements. When closed loop AFR's are lean, add fuel to the VE Table When closed loop BLM's are lean, add fuel to the VE Table.

Course Tune / Fine Tune. Course tune is basically correcting the VE Tables. The computer bases the areas that we DON'T want AFR to be 14.7, like decel, highway cruise, and PE on an accurate Course Tuned VE Table. Fine Tune adjusts the areas that we DON'T want AFR to be 14.7, like decel, highway cruise, and PE.

dave w

Thanks Dave. Starting to make sense. But what happens when the NBO2/BLM doesn't agree with what the WBO2 is saying? I'll explain with an example. I took my last data log and copied the BLM (off idle) and WBO2 output tables from TP into a spreadsheet. I then normalized both: BLM I divided by 128, WBO2 I divided by 14.7. Then I combined those ratio values into a single table so I could easily see where they differ. In the attached, the first number is the BLM ratio, second number is WB. It actually looks better than I thought it was going to, but there are still a number of cells where one is slightly lean and the other slightly rich. If I tune based on one of the values, won't it make the other one worse? 17243

[dave w]

Thanks Dave. Starting to make sense. But what happens when the NBO2/BLM doesn't agree with what the WBO2 is saying? I'll explain with an example. I took my last data log and copied the BLM (off idle) and WBO2 output tables from TP into a spreadsheet. I then normalized both: BLM I divided by 128, WBO2 I divided by 14.7. Then I combined those ratio values into a single table so I could easily see where they differ. In the attached, the first number is the BLM ratio, second number is WB. It actually looks better than I thought it was going to, but there are still a number of cells where one is slightly lean and the other slightly rich. If I tune based on one of the values, won't it make the other one worse? 17243

I have also noticed the differences between BLM vs. WBO2. I consider WBO2 to more accurate. I tune with either BLM or WBO2.

I recently helped with a California Emission tune. BLM failed, WBO2 passed.

Data Below is California Emissions Pass. BLM's are very lean with AFR's looking good.

17245

17246

17247

17248

17249

[brian617]

Not sure what brand o2 you're using, but I like Denso or NTK heated o2s.

[dave w]

For WBO2 I like the AEM 30-4110 https://www.aemelectronics.com/products/wideband-uego-air-fuel-controllers/digital-wideband-uego-afr-gauge/digital-wideband-uego-afr-gauge

The AEM 30-4110 is plug-n-play with 16197427 WBO2 Hack.

dave w

[Daveo91]

Not sure what brand o2 you're using, but I like Denso or NTK heated o2s.

I'm using a Delco AFS74 heated sensor. When I replace I may consider one of these two.


For WBO2 I like the AEM 30-4110 https://www.aemelectronics.com/products/wideband-uego-air-fuel-controllers/digital-wideband-uego-afr-gauge/digital-wideband-uego-afr-gauge

The AEM 30-4110 is plug-n-play with 16197427 WBO2 Hack.

dave w

I have an AEM WBO2, but I think it's a different model. It integrates fine with the 7427; no complaints other than it doesn't match perfect with the NBO2 in some areas, but no reason to believe that's the WB's fault. I know the placement of both can make a big difference too. Your post that you have this disagreement between WB and NB sometimes too helped a bunch. It gave me more confidence that I'm not chasing my tail and I've started to worry less about BLMs being off in some areas.

I've actually made some good progress in a number of areas. Still some fine tuning to do in the lower RPM/higher MAP areas of the off idle table. CL idling doing well - after about 10 seconds it stabilizes at high 14s on WB. Has good "butt dyno" feel in pretty much all areas right now. Until today I was having a lot of problems with acceleration from start. It would always swing way rich (WB pegged at 10) and then sometimes swing way lean. Often the rich swing would cause a noticeable bog, especially when cold. I know the rich swing was wrong AE settings so I searched a number of different threads, including this one (http://gearhead-efi.com/Fuel-Injection/showthread.php?7782-TPS-vs-MAP-AE-7427-0E/page3), specifically post #34, username = dud. His engine specs sounded similar to mine so I thought I'd just try out his AE settings. Wasn't expecting much, but was pleasantly surprised to find the rich swing and resulting bog is gone. It's still too lean at tip-in (in the 16s as soon as I feel movement, for about 2 seconds) so I know I have to increase something. Should I increase MAP by 10% as a start? Across the board, or just to part of the table?

[cgw]

Is there a conversion pinout somewhere to go from 7060 to 7427?

[dave w]

Attached are my notes from a '7060 - '7427 conversion I did several years ago. . . 1 Ton High Cube Van 7.4 liter / 4L80E.

dave w

[Daveo91]

Is there a conversion pinout somewhere to go from 7060 to 7427?

Here's the thread with my notes. Pinout attachments are in the first post. I've tried to keep it updated with a few revisions I've made since I first posted.

http://gearhead-efi.com/Fuel-Injection/showthread.php?9763-upgrade-to-7427-from-7060-ready-to-pull-the-trigger

[cgw]

You guys are awesome, thanks!!