Automobiles Forum banner
1 - 18 of 18 Posts

·
Registered
Joined
·
262 Posts
Discussion Starter · #1 ·
I currently run Siemens Deka 630's at 4 bar, and they are fine at part and WOT, at idle and startup they are OK, but I feel I could have better. It seems no matter what I do I can't get the 630's to respond like I want to at idle. I can only get the car to idle at 13.6-13.8 AFR. If I mess with idle offset and down low fueling and I can get it in the 14.5-14.8 AFR range but it doesn't like it, it runs rough and sometimes stalls. Bring idle AFR back to 13.6-13.8 and it's happy.

Part throttle low load I can cruise in the high 14's-low 15's, come to a red light and I'm back in the 13's. Am at the limits of the 630's capabilities at low pulse widths? What is a reasonable lower limit pulse width fr the 630's at idle? How does the injector pulse width clamp come into play when trying to tune idle AFR.

A buddy of mine got some Injector Dynamics EV-14 styles on a BT 1.8 t with Maestro and it idles and starts very well. It seems like the USRT Genesis II EV-14 style ones get good reviews also. If both these companies sell Bosch EV-14 injectors, what do they do to make them their " own brand" Do the EV-14's enable better idle charcteristics over the Siemens?

Anyone have any experience tuning the USRT or the ID injectors on a 034 EFI system? And for essentially the same price what is the difference?
 

·
Registered
Joined
·
6,004 Posts
You should be able to make those injectors work, you just have to understand them and do some tuning. I made some great strides in this, and tried to put up some posts that explain my findings. See here: viewtopic.php?f=2&t=48220&hilit=injector and here: viewtopic.php?f=2&t=52171&hilit=injector. The Greg Banish article I linked to in the first post is very informative about how injectors work, everything else pertains to the 034 software.

I wish I could explain better, but you just have to try it and see. Knowing how to tune the idle adjust is important, as it can help flatten out the non-linearity of the injectors at low pulsewidths. But everything else is basically developing a pattern and recognizing it. I regularly tune 4-5 034 C-level cars, so I've recognized a lot of patterns in how cars run and how you tune them; you may not have that kind of learning experience available.

Sam
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #3 ·
Thanks for the reply, I have read those posts a few times earlier, but I will try to really delve into again them this weekend, and mess around with the car. In your staged injection post you used any equation to figure out ballpark values for running the 2 sizes of injectors. In a single injector setup is a similar calculation relevant? Also it seems that I should have greater resolution as I am only running a 2.5 Bar Map. Given the factory specs for the injectors, what is the most effective way to get some ballpark numbers to start with for Main Gain, and idle adjust. It seems now I'm just tinkering and guessing. Is there a scientific way to quickly hone in on correct settings?

Specifications: Siemens Deka 630cc
Gain: 0.11ms/mg
Offset: 0.055ms
Turn on time @ 14VDC: 1.14ms
Turn off time: 0.85ms @ 600KPa
 

·
Registered
Joined
·
6,004 Posts
I might be way off as it's been a while since I tuned a car on those with the 2.5bar map, but I would try something like a 15-16 main scaler and maybe a .3-.5 idle adjust. Most cars I do seem to idle near stoich with values around .80-.85 in the idle cells, WOT at .78/11.8:1 at around 1.10ish...try some blanket values in those areas and play with the main scaler to get it right. The idle adjust has kind of a weird effect; you know you're doing it right when you can achieve consistent air:fuel ratios at low load (between 40-80kPa) without your cells varying vertically more than a point or two. I play around with the value and do some light cruising to see if it's dialed in.

The problem you're having could probably be helped with some idle adjust, and also by going to a 3bar fuel pressure regulator if you'll have enough fuel with it. It will basically make your injectors behave more linearly. The problem you're likely encountering is that you're right on the edge of the low pw/high pw slopes meeting at your ~idle pulsewidth, so the flow rate of the injector (mass per millisecond) is changing depending on what pulsewidth you're running...it gets to be a fine line between the rated flow rate and a noticeably lower one that requires more pulsewidth to keep the engine running at the same air:fuel ratio. I've found this is often the cause for idle oscillations...a .3ms drop in pulsewidth from the engine idling well causes the injector to transition to a lower flow rate, which causes the engine to go lean unless the tune compensates for it, which causes it to go to a higher pressure at a lower RPM which is properly fuelled and causes the engine to freely rev and on and on. Also keep an eye on the voltage, as if your idle RPM is dropping low enough to change alternator output, and your injector voltage compensation isn't right, you have something else altering your fuelling.

As best I've been able to tell, there isn't a way to take real injector data and some maths and stick it into the 034 software to have a setup car. Experience helps figure out what works, but it really takes a lot of playing around to get good results. It's an easy system to get a car running on, but it isn't a straight-forward system to make a car perfectly driveable on.

Sam
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #5 ·
That's interesting I am nowhere near those values Included are the screenshots the directly relate to this disscusion I can post more later with logs once I starting tuning. I might add I had this tuned by a tuner last year, with one goal...to pass emissions testing. It did pass barely, but I'd like to make more drive-able when not at WOT. This tune is basically the same as it's been for 12 months. I think a 3 bar FPR could be feasible for now, a quick look back at some logs shows about 31% on the injectors at 16 PSI, I think the 2.5 bar with 83 KPA ambient will max out around 24 PSI. I'd be happy to drive it around at under 20 PSI.
Also currently I'm just using timing for idle, and I have a throttle stop. Closed loop is also activated for EGO.
 

Attachments

·
Registered
Joined
·
262 Posts
Discussion Starter · #6 ·
Well I have made a lot of adjustments over the last week and I have the car idling pretty nicely at 14.7-15.2 AFR. At idle my injector pulse width is going from 1.38-1.51 ms (supposedly these injectors go non-linear below 1.2 ms or so, I have the min pulse width clamped at 1.15ms) EGO correction is minor at idle anywhere between -2% and +2%, I think that is a reasonable error for the ECU to take care of.

I do I have another problem that I can't quite sort out at low throttle angles, TPS values from 5%-20% in the 2000-3000 rpm range at pretty low cruising vacuum 15-25 Kpa at say 40 mph its going really lean, causing bucking. The ecu is calculating injector pulse widths really low from 0.9ms up to my minimum of 1.15 ms. So essentially I'm not injecting fuel. I know this is probably self inflicted, but I can't figure out what settings I have setup incorrectly. The cells in the basic fuel mapping screen are pretty high around 1.2 ms in this area, which I feel is too much anyway.

So what am I missing, and why is my ecu calculating these low pulse widths?
 

·
Registered
Joined
·
6,004 Posts
Because 15-25kPa is an extreme amount of vacuum to actually be running an engine under. I would verify that vacuum reading with a gauge. I've seen a couple of ECUs that have had MAP sensors replaced show very odd vacuum readings...showing idling at 15-20kPa when a gauge shows ~18-20inhg at the very most (which is what most 1.8Ts idle at depending on ignition timing). I guessed it was because of a poor solder joint between the map sensor and the board, but that is just a total guess. I was able to tune all of these without a problem, but it did cause some issues with lack of resolution tuning the high vacuum areas.

If you set the minimum pw to 1.15ms, it means that the lowest pw that will actually be used is 1.15ms, until you hit the overrun fuel cut at which point pw goes to zero (both calculated and actual). Most of the time this actually results in a rich condition when you ride the min pw. If you're getting lean bucking under extreme vacuum at the minimum pw, it may be set too low. At the same time, make sure you're not bumping into your overrun fuel cut. The parameters for that are in the injection parameters tab of the config screen.

Sam
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #8 ·
I have an Autometer gauge and it hits about 21-22 in/HG max. The bucking seems to happen in the high teens 17,18,19 in/HG I'm not sure how that works out with ambient pressure involved in the equation, because running 16 Kpa through a conversion yields 4.75 in/HG, but that's absolute. The elevation here during my test runs was about 5,600 ft. which is about 83.2-83.3 Kpa ambient before the engine starts. Idle it sits right about 33-34 Kpa.

As far as I know the MAP hasn't been replaced, I've owned the ECU since new in about 2006, I don't think I've ever opened it.

My injection parameters are:
Battery offset=60.8 usec/V
Engine Decel=28.0 Kpa
Max TPS for decel=3.5 %
Max cut/pressure=220 Kpa

I'm going to include a log from the last run of the day, after reading some of your older posts, I saw that you went all the way to 1.5 on the bottom row of the basic fuel map, so in this run I bumped some of those lower row numbers into the 1.4's, I haven't really analyzed this log yet, but it was still bucking a little, but I also noticed some spots are now too rich.
 

Attachments

·
Registered
Joined
·
6,004 Posts
I would drop your engine decel pressure to about 20kPa. It's probably bucking because you're just slightly dropping the throttle below its threshold while the map is already well below it, and the stumbling is actually the fuel cut suddenly coming in.

Sam
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #11 ·
With all the flooding I have been really busy, but I finally have been trying to do some tuning as of late.

I was wondering if guys could take a look a my setup and give advice/critique my setup.

I have minimized some of the low load/low throttle angle lean bucking, but it is still there. I think some of it could be O2 sensor lag during this condition, going from fuel cut to low pulsewidths / low throttle and back again.

I have added screenshots of my setup and 2 logs one short one long.

PS I'm fairly sure I have developed a small vacuum leak on injector 3 which causes idle to be high sometimes i'll have to get some O-rings and take a look at idle later.
 

Attachments

·
Registered
Joined
·
2,564 Posts
From my limited investigation on your log, you still are hitting the fuel cut while free revving below what looks like 20kpa.
The engine is not immediately starving during the fuel cut, which is confusing.
Your fuel cut needs to be put even lower yet.
I do wonder if the map type has been set correctly, can you determine that the MAP pressure reads what the BARO does when the engine is not running?

As for the fuel cut being laggy, it may be that your intake is wetted by the fuel and it is taking a while to clear it.
Could be you are overly rich causing the wetting. This is hard to say.

The section of your log that lets me say this begins at line 510 on the "D" logfile.
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #13 ·
Thanks for the quick reply, Im glad I'm not the only one confused. The screenshots in the post above show Manifold @ 82.8 and Barometric @ 83.6 with the car not running.
 

·
Registered
Joined
·
2,564 Posts
Raceface said:
Thanks for the quick reply, Im glad I'm not the only one confused. The screenshots in the post above show Manifold @ 82.8 and Barometric @ 83.6 with the car not running.
Low 80s is about right for 5000' alt.

Your engine is making quite a bit of vacuum at light load.

Put the fuel cut pressure at 10 and see if that fixes it.
Increase it later when you get a better idea what you really need.
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #15 ·
TheArchitect said:
Raceface said:
Put the fuel cut pressure at 10 and see if that fixes it.
Increase it later when you get a better idea what you really need.
I'll do that and take it for a spin, and post a log when I get back. As for what I really need, I dunno. My Vac/boost gauge never goes below(above) 22 inches HG.
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #16 ·
Well I hunted for vac leaks, couldn't find any blatant ones. Changed the injector o-rings, serviced the BOV( looked at everything in the intake tract) and kept the decel cutoff around 28 KPA. Drove it around and I wasn't getting the lean cruise bucking condition anymore. Decel seemed to work better also. I'm rich now in a lot of places I wasn't before, so I'll have to do some re-mapping.

Here's a new log
 

Attachments

·
Registered
Joined
·
2,564 Posts
I dont think you have/had a significant leak and if you did, the indication was a "strong" vacuum, not a weak one (like with a big vacuum leak).

You should probably drop your decel fuel cut pressure or its gonna bite you again (soon).
 

·
Registered
Joined
·
262 Posts
Discussion Starter · #18 ·
Sounds good, I'll run it down to 10 Kpa. Looking back at logs it looks like 13.8 Kpa is the min that it produces.
 
1 - 18 of 18 Posts
Top