Cygnus or others, what do you think of this FMU idea?

[Guest bastaad525]

Okay, first let me say, I was wrong... very wrong... about the FMU not affecting the fuel curve at low RPM when boost first kicks in, and about it not having anything to do with the super rich midrange I ended up with when I tuned it at the dyno. I had assumed, since the fuel curve I ended up with was so similiar to how it was before I added the I/C and upped the boost 2psi, at which time it still had that super rich midrange... I figured the FMU was not to blame. But the truth was staring me right in the face...

 

 

This is the sheet that shows all of my six runs, on the left are the torque curves, and on the right, the a/f ratios. On my first runs, the FMU was having almost no effect, fuel pressure was barely going up over what it was without the FMU. You'll notice it still dips kinda rich to about 3500rpm, about 11:1. But then as I turned the FMU up and up, trying to get the top end down into the 12:1 area, that rich dip in the midrange gets worse and worse, until, by the last run, it was getting as rich as a high 9:1. You can also see on the torque curves that the initial torque, as the turbo just comes on, takes a big hit from this initial richness, losing almost 40ftlbs from when the turbo first kicks in. Overall torque comes down as fuel gets richer as well, but doesn't suffer as much as that first hit after 3000rpm.

 

So, I sorta borrowed one of Cygunx1's ideas, though with a bit of a twist, and wanted to ask what you guys thought of what I did, and if you guys think it will work.

 

I'm using one of the do-it-yourself, grainger ball valves as my boost controller. Cygnus' idea was to use that same kind of adjustable ball valve, inline with the boost signal line to the FMU, and use that to keep the FMU from doing anything until a certain boost point is reached. I had planned on getting a second ball valve, and setting it to kick in at maybe 8-9psi of boost. And hopefully, this would go a long way towards getting rid of that super rich spike in the midrange. Well... being the cheap bastard that I am I thought maybe there is another way I can do this w/o ordering another ball valve? So what I did was disconnect the tee that was supplying the vac/boost signal to both the FMU and stock FPR, hooked the stock FPR back to the intake manifold. Then I ran the tee inline with the boost controller, splitting the boost signal from after the boost controller, to the wastegate actuator and the FMU. Hence, keeping the FMU from doing anything until 13-14psi of boost is reached.

 

First let me say, I definately noticed a difference in the responsiveness of the car after doing this. At part throttle/partial boost, or even up to full boost, which can be hit before opening the throttle fully, the motor doesn't bog or surge at all like it used to (which was only minor, but still noticeable). It transitions onto boost smoother and faster, and the minor boost spiking I always have isn't as noticeable... though it's still spiking the same (up to about 15psi) you don't FEEL it, whereas before it would surge noticeably). It was very noticeably when shifting... it had always felt a little sluggish right after shifting for just a fraction of a second, what I thought was a tiny bit of turbo lag. Now it hits power strongly enough after shifting to chirp 2nd and 3rd gear easily.

 

My worry is that now there may be a lean spike instead of a rich spike? Not so much if I just floor it from a stop light, but in transitional driving situations... say I'm accelerating at mid-pedal, running maybe 4-5psi of boost at like 4000rpm, and suddenly snap it wide open, maybe it will go lean for that split second before boost hits full and the FMU goes open? Things like this were why I had always planned to set the FMU to kick in at a lower boost level. Well.... at any rate, I did a lot of experimenting after setting it up like this, trying scenarios like the one I just mentioned, and didn't once hear it ping, so I'm relieved there. I think it should be okay like this, and the benefits are many. I know the only way to really know is to hook it up to a wideband... which is still at the top of my xmas list, but was thinking some of you more tech minded guys might be able to formulate a good idea of exactly what is happening with me running it set up like this?

[Guest bastaad525]

On a side note, I've also been thinking about/planning on following up on another one of Cyngus' very good ideas, which was to run an RPM operated switch in between the FMU and intake manifold, that would not allow boost signal to hit the FMU until a certain RPM is reached... this may be the better idea, or run the two combined even?

[cygnusx1]

I am glad to see that both my ideas make sense. I really like the next step you took it to by eliminating another grainger valve. I would just be wary that the FMU wont do anything until the wastegate opens. So, in theory, you could be 1psi shy of max boost at 5700rpm and still be lean (no FMU action). I think that if you use the RPM switch, you can put the FMU back on the manifold without a grainger valve but I like your idea nonetheless. I once told my boss that a lazy engineer is a good engineer because we find easier ways to do the same thing. He didn't laugh.

 

To the guys that have already sucessfully switched to the ultimate solutions like programmable setups and the later ECU's all this may sound like petty stuff, but hey, it works. I am going to try hooking my FMU after the wastegate grainger as well. Good thinking Baast.

 

EDIT: On my way home from work today I was thinking about how you have the FMU in the wastegate line. I would be a tiny bit concerned because the FMU allows boost signal to leak via that bleed screw (onset gain). That, in effect, is a signal leak to the wastegate which can change the opening behavior of the wastegate. I dont really think its a big deal but it is something worth bringing to attention.

[Guest bastaad525]

LONG POST ALERT! LONG POST ALERT!

 

 

EDIT: On my way home from work today I was thinking about how you have the FMU in the wastegate line. I would be a tiny bit concerned because the FMU allows boost signal to leak via that bleed screw (onset gain). That' date=' in effect, is a signal leak to the wastegate which can change the opening behavior of the wastegate. I dont really think its a big deal but it is something worth bringing to attention.[/quote']

 

 

I thought about that as well, I figured if it were going to happen it'd probably show up as either 'boost creep', more exaggerated boost spiking, or, the boost would just settle higher than I had it set before swapping around all the signal lines. I did quite a few third gear pulls after changing everything, and the behavior of the boost didn't seem to change at all. Still the brief spike, still settling at 13-14psi, and no creep all the way up to 5500rpm. As I continue driving it like this I will definately keep an eagles eye on that boost gauge though for any changes.

 

I am glad to see that both my ideas make sense. I really like the next step you took it to by eliminating another grainger valve. I would just be wary that the FMU wont do anything until the wastegate opens. So' date=' in theory, you could be 1psi shy of max boost at 5700rpm and still be lean (no FMU action).

[/quote']

 

Very good point. Here are a few of my ideas about that. #1 is that, in your example of not being at full boost but at 5700rpm, this would mean that I would not be at WOT. When accelerating I can hit full boost holding the throttle at something like 2/3's of the way down, so in your example I'd be around there, maybe even less throttle at such a high rpm. With the throttle butterfly only half open, even at 12psi of boost in the manifold, isn't the air flowing little enough that the AFM would still be well within it's measuring capabilities and providing enough fuel on it's own? Remember the AFM basically just measures the airflow and supplies enough fuel for that airflow until the flap goes fully open and cant measure anymore... I guess it's one of those things we'll never really know for sure (and gets debated often), but I"m betting that at 1/2 throttle the AFM is not fully open. At such a high RPM it's hard to say, BUT... take into account my next point, about the rpm:

 

#2 I never drive in a way where I would do as you listed in your example. For one, I never shift any higher than 5000rpm... maybe 5500, but I only ever rev that high if I"m going flat out WOT... I can't stand the sound of the engine revving so high when just cruising or moderately accelerating. With the non hydraulic head the valvetrain sound at this RPM, w/o the full throttle intake/exhaust wail to drown it out, just sounds PAINFUL (heheh Cygnus not like YOU would know eh?? )! Highest I ever let the engine rev if I'm not flat out flooring it is about 4000rpm. I consider 3000-4500rpm my 'sweet spot'. But that leaves the question, what about 4000rpm, 12 psi (about 1psi short of max boost), 1/2 to 2/3 throttle? With the FMU doing nothing at this point, is the AFM still within it's measurement range and providing enough fuel? Again, I'm betting yes. Look at the first few dyno runs on that sheet, with the FMU set to do nothing (bleed screw was all the way out on the first run/the red line), at WOT, full boost acceleration the ECU still provided more than enough fuel on it's own up to just shy of 4000rpm (12.7:1 a/f ratio), which is where I"m betting the AFM 'maxes out' and isn't measuring anymore. This makes sense to me, because when I was running 10psi before, it didn't start running lean until almost 5000rpm. And even as far as it now going lean at 4000rpm, that's at WOT... at 1/2 - 2/3's throttle, I'm betting the stock curve would stay richer, higher into the rpm range. But then...

 

#3 What if I suddenly punch it at that point? Boost suddenly jumps, spiking up to 15psi, but surely the FMU does take a split second to react, especially with the restrictor in the signal line? This is one idea that does have me kinda worried. But... then I think back to a question I had asked a few weeks ago, regarding the placement of the AFM after I installed my I/C. I was figuring that, since air goes thru the AFM first, that when suddenly punching the throttle, the turbo spins up, the additional air has to go thru the AFM first, causing a momentary rich condition... as, the AFM measures more air and provides the extra fuel for this almost instantly, thinking it's only traveling the short distance to the turbo and thru the stock J-pipe, when it's actually having to get thru all the extra pipe, as well as having to pressurize the I/C itself, before actually making it to the cylinders. So I'm thinking that for that first instant before full boost is reached, the AFM will do the enriching I need to balance out for the FMU not coming on instantly. Maybe THIS is why there is the rich dip in the a/f ratio on the dyno sheet, even with the FMU set to do nothing. Maybe this is the pressurized air playing 'catch up' with the AFM that saw it coming from a mile away. But again, I do wonder how this whole theory plays out when suddenly flooring it from a much higher rpm, where boost hits much faster.

 

#4 what if I AM running lean? Well... I've never once had a car, any of the cars I've owned, ping under anything less than WOT acceleration. So if I'm at 4000rpm, 12psi of boost, but only 2/3 throttle, am I gonna ping? I did test this out, on the street, but didn't hear a peep. Then again, it was a relatively cool morning... what if I did this same test on a hot day, climbing up a hill?

 

 

So... the worry is definately there, and I'm going to be very very vigilant for any slight sound of ping. On hot days or climbing hills I'll just need to be very careful especially, maybe change my 'driving style' even further in those situations to basically 'ignore' any boost level between say 8psi and 13psi... in other words, never let boost just 'sit' in that area if there's a large load on the engine or it's hot out. Either floor it, or keep throttle low enough to not pass 8psi of boost, if I'm at higher than maybe 3500rpm. Or... I'll chicken out like I always do and just put things back the way they were until I can verify all of this with my own W/B O2 sensor.

 

 

Well... I can always take consolence in the fact that I at least have a reliable back up car now, if I do break the Z. And if I just blow a head gasket... well lord knows I"ve gotten PLENTY of practice R & R'ing the head on this thing!!!

[Guest bastaad525]

I stand corrected again... I didn't get any difference on the street, but a few full pulls in 4th gear on the freeway and I notice the boost just settle at just a little bit higher than it used to. So the FMU bleeding off some of the pressure is in effect raising the boost, about 1psi. But no boost creep.

 

Well I did some more experimenting, cruising at 4000rpm, varying throttle to see how the boost level reacts. At 4000rpm in fourth, I hit full boost before even giving half throttle. I also tried various combinations of cruising then nailing it and such, and thought I heard a ping once or twice but man it's really so hard to tell, this damn car so rickety and creaky it's always making noises when I push it hard, the sound I heard was very faint and seemed to emanate from the rear of the car so I'm thinking it wasnt detonation. Knowing myself, I will end up putting it back the way it was tommorow and try this again once I get my W/B... but man it does drive a lot better this way I'm kinda reluctant to switch back.

 

Blew the doors off some kid in a Beemer 325 while I was out testing, come to find out he's got nitrous, he actually knew what my car was which cool, and I GUESS my turbo or BOV make a lot of noise because he knew what I was running w/o asking I always figured they weren't that audible.

[Guest bastaad525]

After much thought (and two days more than I thought I could hold out) I ended up putting it back the way it was. Something finally occured to me that put a bit of a scare in me. Two things actually. One is that, when the pressure threshold of the grainger valve is reached, I started thinking... you know the thing probably doesn't just go wide open, rather, it sorta 'oscillates' open and closed very rapidly. Further, it obviously is not allowing the full boost pressure that enters is to get thru to the other side... in the case of the wastegate actuator, it only allows enough pressure thru to partially open the WG... I'm honestly not fully sure how all of this works, but in the end I got a chill when I realized, that fully pressure was probably never reaching the FMU either, and the FMU was probably functioning way less than I had set it to if at all, and I'd be back to running too lean. So I put it back the way it was. Some day when I either get an in car fuel pressure gauge or WB 02 I'll try this experiment again.

 

In the meantime... the more I think about it the more I think that I'd really be better off following Cyngus' other idea and running some kind of RPM switch, which could operate a solenoid or something that would block boost signal from the FMU until a given RPM point and then go wide open allowing full boost pressure to hit it. This way I could set it not to come on until about 4000rpm and completely get rid of the super rich 3000-4000rpm spike. This project is probably more complicated than I'd be really up for (really uncomfortable tinkering too much now that the car is running so well), but I'll definately look into it.