The old cut up manifold into trumpets idea

[Dru128m]

It was Bob who thought maybe the stock injectors are tapped out. He's the only one who knows what they're being asked to do, so I just smile and nod. I was shocked when he suggested it, but the stock ones are kinda small. I haven't been able to check fuel rail pressure under load yet, so maybe it's a delivery issue, but I don't get that feeling.

[wootloops]

Originally Posted by Dru128m

It was Bob who thought maybe the stock injectors are tapped out. He's the only one who knows what they're being asked to do, so I just smile and nod. I was shocked when he suggested it, but the stock ones are kinda small. I haven't been able to check fuel rail pressure under load yet, so maybe it's a delivery issue, but I don't get that feeling.

If Bob is saying it, I know he is very knowledgeable and I'll take his/your word for it.

With that said, I would be really curious to see what you dyno. Everyone has been saying the N52 stock injectors are good for around 270-280whp on 93 octane. So either you are making a lot of power with this setup, or everyone was overestimating the power level they can support.

[Dru128m]

Maybe it's not the injectors but for some reason, starting at 4000rpm it starts to lean out, by 5000rpm I need to back out because afrs climbing too high. I hope it's as easy as an injector swap. I would really love to be able to keep my foot in it all the way to the end. It feels quite healthy on the jump, but that's all I really get right now, it's a mighty tease. Trying to keep this relatively simple so hopefully no fuel tank work is required.

[wootloops]

Originally Posted by Dru128m

Maybe it's not the injectors but for some reason, starting at 4000rpm it starts to lean out, by 5000rpm I need to back out because afrs climbing too high. I hope it's as easy as an injector swap. I would really love to be able to keep my foot in it all the way to the end. It feels quite healthy on the jump, but that's all I really get right now, it's a mighty tease. Trying to keep this relatively simple so hopefully no fuel tank work is required.

Didn't you have it running up to redline wide open in previous versions of the tune (like in your video)? What changed between then and now?

I mean, I hope you are truly maxing them out because that probably means you are making a ton of power... but I would also be kind of surprised if the open manifold really adds that much extra hp.

[Dru128m]

Yeah we've been to redline on previous iterations of the tune and it loved it. Maybe it wasn't going to full valve lift on previous tunes, I'm not really sure what's different. I don't get a lot of tune specifics from Bob unfortunately. This recent issue came with some idle adjustments and a noticeable bump in power. After that was a 20% fuel increase across the map but same issue. We'll figure it out, hopefully the problem is big power. Probably not with my luck.

[Kiddvegas]

Originally Posted by Dru128m

Yeah we've been to redline on previous iterations of the tune and it loved it. Maybe it wasn't going to full valve lift on previous tunes, I'm not really sure what's different. I don't get a lot of tune specifics from Bob unfortunately. This recent issue came with some idle adjustments and a noticeable bump in power. After that was a 20% fuel increase across the map but same issue. We'll figure it out, hopefully the problem is big power. Probably not with my luck.

Any updates ?

[Dru128m]

Turns out it was the dme cutting fuel, we flashed a patch for that, but it was ineffective. So just waiting for Bob to clear yet another hurdle, I'm starting to feel bad for him. I wish there was more I could do to help.

[Dru128m]

Sorry guys, still no updates yet. Car runs, is drivable, just can't go over 5k. Hopefully something soon.

[Kiddvegas]

Originally Posted by Dru128m

Sorry guys, still no updates yet. Car runs, is drivable, just can't go over 5k. Hopefully something soon.

Hopefully it’s resolved soon.
Have you thought about doing MILVs once it’s running properly?

[Dru128m]

I haven't really looked into the install on them, I know what the are, how they work, and what they do, but getting them in ?. Maybe if I buy a second motor at some point, I'd put them in. I'd be curious about upping the CR at that point as well.

[rjahl]

Originally Posted by Dru128m

Sorry guys, still no updates yet. Car runs, is drivable, just can't go over 5k. Hopefully something soon.

I'm sorry but something just does not sound right with this thread.

1. Bob is a really good tuner and it would surprise me If he could not overcome the apparent challenges with this set up.

2. We know the fuel injectors can feed up to about 300hp, more than you'll make naturally aspirated using stock cams.

Can you post a log for us to look at? Surely you are logging and passing that's data to Bob. Please share. Some of us have shared so much work , I don't think I am asking for too much.

There must be some mechanical issue involved.

Thanks

[Dru128m]

What's the best way to post those? I log them with torque pro. I just look at the .csv files as a table of numbers, I don't of a good way to share those. I'm concerned myself, Bob has been silent since the last update which is unusual. I'm hoping it's mechanical, I can fix that. I'm worried this is just something the stock DME refuses to do, which blows the point of this whole thing.

[mindspin311]

Upload to Dropbox or Google Drive and post a public link

[Dru128m]

https://drive.google.com/file/d/15lD...w?usp=drivesdk

I hope this works, it's not edited. There's only one spot on these logs I tried to go over 4k, but you can see it go lean and I back out.

[rjahl]

Originally Posted by Dru128m

https://drive.google.com/file/d/15lD...w?usp=drivesdk

I hope this works, it's not edited. There's only one spot on these logs I tried to go over 4k, but you can see it go lean and I back out.

Perhaps I see why Bob can't do anything.

The data is somewhat "wanting"

1. Data rate is SLOW, two frames a second is really really bad. I would not be surprised if the it's actually worse. It's possible that your program is "filling" data that's requested and not supplied by the DME. So rather than a gap for data not supplied, it actually replaced a N/A value with the last value or ?

2. Your fuel trims are crazy. both rich and lean, does not matter. Do you have headers? What type?


The data makes no sense at all, I'm seeing strange stuff like really lean commanded fuel mixtures at moderate load and miss matched data, see point #1.

Look at the data around rows 141-145, Engine load jumps to 100 percent but you see no change in engine speed for three frames or 1.5 seconds, not possible. Data is junk.

You should be able to log about 8 frames a second with around 25 parameters. Check out my posts regarding data logging with Test-O.

[Dru128m]

I've had the same question as far as log quality, and I've asked Bob if he needs anything better/ different, but he never mentions it. Headers are some AA knockoffs, no leaks, good fitment. I have confidence Bob will crack this egg, I think he's just busy frying bigger fish right now. I get it.

[jonwilli]

Originally Posted by Dru128m

What's the best way to post those? I log them with torque pro. I just look at the .csv files as a table of numbers, I don't of a good way to share those. I'm concerned myself, Bob has been silent since the last update which is unusual. I'm hoping it's mechanical, I can fix that. I'm worried this is just something the stock DME refuses to do, which blows the point of this whole thing.

I've been waiting on a reply from Bob for 3 weeks or so. I guess he might be taking holiday or has issues at his garage?

[rjahl]

Originally Posted by Dru128m

I've had the same question as far as log quality, and I've asked Bob if he needs anything better/ different, but he never mentions it. Headers are some AA knockoffs, no leaks, good fitment. I have confidence Bob will crack this egg, I think he's just busy frying bigger fish right now. I get it.

I'm not a professional tuner and I would not consider my self in the same league as Bob but I would find it difficult if not impossible to tune an engine with this data. Probably would not even try. If I was busy with other well paying projects and had someone sending me this, I would probably put them on the back burner.

I'm not sure what Bob asked for on a log but we get the best full load data by a rolling third gear start and hold the throttle down full until the next shift. Traction control needs to be turned off.



The light throttle conditions in your log, with the log latency yields nearly no useful data.

First data picture shows about the most useful data in the log, Almost full throttle for a very short period of time but the data latency kills ruins it.

Check out the second picture, notice the repeated data. probably the software filling in blank data with the last value given. This is why I gave up on some of the early logging tools available to us. I think Bimmergeeks logging tools has overcome this and I know Test-o can do better.


Edit: I just noticed I circled the AIT field, those are normally pretty stable. Plenty of other places in this log with "frozen" data.

[Dru128m]

I'm totally with you on the hokey data replication. I may not have torque pro totally optimized for the highest sample rate, I'll look into that. That data was just driving through town to my 3rd gear pull spot, when I thought I'd try to up the revs in 1st or 2nd to see if was still cutting fuel, it was so I just drove back home. Not good data at all. I normally edit most of that out unless something interesting happens. I'll try to to optimize the sample rate and grab some proper data this weekend for you, if your still curious.

[rjahl]

Originally Posted by Dru128m

I'm totally with you on the hokey data replication. I may not have torque pro totally optimized for the highest sample rate, I'll look into that. That data was just driving through town to my 3rd gear pull spot, when I thought I'd try to up the revs in 1st or 2nd to see if was still cutting fuel, it was so I just drove back home. Not good data at all. I normally edit most of that out unless something interesting happens. I'll try to to optimize the sample rate and grab some proper data this weekend for you, if your still curious.

I'm not sure torque pro has the correct protocols to poll data from the MSV80 at good data rates. This requires certain Idiabas function settings that few people know about.

If I look at only fuel trims, it looks like you are running lean at very light loads only. they get better with load and relative throttle?

It would surprise me if Bob was having a problem with the no MAF tune, perhaps the 02 sensor locations in the headers are the issue. Do you have a picture of your headers? in particular where the 02 sensor is located. Should be in the collector and not near the head in the center exhaust pipe. Does Bob know you have headers?

[Dru128m]

Bob is aware of the headers, and the 02 sensors are just downstream of the collector. Car ran fine with the headers before when the stock intake was still installed.

There was a bit of revelation yesterday. I took her out for some exercise, still cuts fuel under partial throttle around 4500k, but I mashed it to the floor just to try a proper 3rd gear pull and she ran up all the way no problems! Some data still looks off, but it's promising.

I did what I could to increase the log sample rate and sent it off to Bob, see what happens from here.

https://drive.google.com/file/d/1ARC...w?usp=drivesdk

[rjahl]

Data is better but it's still very hard to use.

It appears the program is polling at a good frequency but the DME is simply not supplying the data. Program then just fills in the missing data with the last value received. Just bad.

Take a look at the raw log that I have attached here. Its the amount of data and the frequency that a tuner can use. Data frequency is every 125 thousands of a second and there is a crap load of data to review. Everything from pedal position, speed, water temp, knock sensor voltages, eccentric cam position, relative air flow. Target lamda, actual lambda both banks. Cam positions. etc. All done with Free software a Dcan cable and a laptop.

I leave all the values in German even through Testo has transition features. Just learned to read German after a while.

Glad to see you have found some correlation to your lean conditions and load. You may also have something going on with your engine timing. Same conditions, timing is way low for those loads and RPMs. Light throttle 2.5 K and above I could see 40 degree on level ground, constant speed freeway stuff. Your data appears to be lean + lazy timing. Could be related or it could be crappy data.

I only know two programs that can poll data from the MSV70 / MSV80 at a reasonable rate. Bimmergeeks app and Test-o. Of course the Bimmergeeks app will be easier to set up.

In theory Torque should be capable, but may not have the correct data fetching routine for these DMEs. You need to pull data with single a request with multiple fields.