bassn_07

I should have some info this weekend. The install is planned for this Saturday and my tech said it will be done. My tech and Rob from Needswings already started communicating and coming up with a solid game plan.

Please give me your thoughts on any of this. It seems you're very knowledgable on this subject and your input will be appreciated !

Once again here are our plans:

1. AFR (Direct)
2. EGT's (Direct)
3. Boost (Direct)
4. Fuel pressure (direct)
5. RPM and TPS (direct)
6. IAT's (OBDII)
7. Timing (Direct) I wanted to tap into the 4 corners of the cylinder but I don't have enough user inputs) Wouldn't that be great to monitor the timing on 1,4,5, and 8.
8. Knock sensor voltage (direct)
9. Engine temps (OBDII)

Mikesamg

Does that thing tell you what the winning lottery numbers will be directly as well?

bassn_07

yeah it will 60 1.5 10.80 134...get it !

Mikesamg

I'm thinking it will be more like 60 1.5 10.75 133-4

lolachampcar

It sounds like your path is going to provide some very nice insight as to just what is going on. I look forward to seeing some of the data (if you decide to post it).

My goals are somewhat different in that I am trying to provide tuners with the ability to do remote tuning. A Skype link with the programmer can allow a Tuner in Montana to do an SLR in Saudi J The only thing that is missing is a simple, reliable and consistent way for the Tuner to evaluate the dyno (or street/strip for that matter) pulls. MB proprietary data is slow but does satisfy the consistent requirement along with being simple (logger and programmer are the same tool). It also addresses the gap between the European models of tuning (owner delivers car/ECU to Tuner with expensive tool in Europe whereas in the US companies ship an inexpensive tool to the end customer) by providing a cost effective programmer/logger.

Like everything else in life, it is a compromise because of the slow speed. Luckily there are solutions like the one you are trying for providing data with much better resolution.

bassn_07

Everything is all wired up and working!!! The only thing my tech had trouble with was trying to pick up the timing. He tried picking it up at the crank sensor but it seems as though it's not going to work. Any suggestions?

This is how we have it setup right now.

1. AFR's (direct)
2. Knock voltage (monitoring both front and back knock sensors using voltage. If my knock sensors pick up any kind of activity we should see a obvious change in voltage)
3. EGT's ( wired up 3 inches from the #1 cylinder)
4. Fuel pressure (right now its showing in volts but I'll speak with Dashdaq about having it converted to psi)
5. Boost (taken off bank 1 surge tanks on the downward pipe before the intercooler y pipe)
6. RPM and TPS
7. IAT's (OBDII)
8. Engine Temps (OBDII)
9. Timing (OBDII)

lolachampcar

Not sure how the system you are working with determines the relationship between crankshaft position and cylinder firing (timing). If you must, you can always get it from the OBDii just at a slow rate.

As for using the car's lambda sensors, they appear to be narrow band units which are of no real use in doing higher power level tuning You’re on the right track with external equipment.

The only thing that looks like it will be useful from the MB proprietary data set is knock retard per cylinder at somewhere around three samples per second.

bassn_07

Lolachampcar, what are your thoughts on picking up on the knock sensor voltage? In theory if the knock sensors become active I should see a spike in voltage.

Hammer Down

Was Mark able to help?

lolachampcar

I've done a bit of research on knock sensors for a fuel control I built a few years back and then again for a friend building DP engines when Grand Am made the move to Bosch fuel controls. They are funny finicky beasts to say the least.

I’ve long since shifted out most of what I learned on them (old brain, not a lot of room) but I do recall that the ones I was looking at were frequency based devices. The interface chips had narrow band filters to allow you to “tune” the frequencies of interest then adjustable thresholds for “triggering” knock detect for a given frequency range. Apparently every engine has its own characteristic pre-ignition indicators. Most of the DP engine builders had Bosch come in and tune the knock detect for their particular engine configuration (as my friend did, I believe). The alternative was my breaking out a spectrum analyzer and going to school on the engine in question. It was cheaper to go with Bosch.

In short, the stuff I was looking at did not generate pure voltage out that was proportional to knock so you could not just hook a volt meter up to it and measure knock.

Please do not take the above as absolutes. It has been a LONG time since I looked at knock sensors. There is nothing to say MB is or was using the type of sensors I was looking at. Your data acquisition guys deal with this stuff on a regular basis and will know a lot more about it than I do.

I am a dealer for a data acquisition company but I am not trying to sell their stuff. For technical reference, here is what XYZ has to say about its knock stuff (poor spelling included)-

“To set up a reliable electronic knock control system it is necessary to determine knock levels accurately and differentiate knock from normal engine noise. Most modern engines are fitted with a piezoelectric knock sensor. XYZ’s Knock Modules use these sensors to measure engine noise and detect the knock levels by reducing the influence of background engine noise using a combination of data gating and frequency filtering.
A competent tuner needs to carry out a frequency analysis to determine the target frequency and knock module settings. This is highly specialised work and therefore the XYZ Knock Modules will only be available through authorised dealers that have undertaken specific training”

lolachampcar

Here is where I am on looking at Ignition Knock Retard as seen by the MB ECU. This was captured at about 1.3 samples per second when combined with recording mixture. I'm removing mixture as it is narrow band information and of no use to me so I should be able to get the logging rate up to somewhere around 3 samples per second.

forcefedjunkie

It'll be interesting to see if you get anything useable from the unfiltered knock sensors. I've looked at both raw knock voltage vs knock retard on Chryslers. The knock retard parameter is very valuable but the raw voltage contains too much noise to be of much use. I guess you can draw generalities from what you're seeing but I think the OBD2 knock parameters would be more benefitial even though the update is slower.

It's too bad MB stuck with the ISO protocol until 2008. With CAN you can get 20 - 30x the update rates or more.

Todd

bassn_07

I have my both my knock sensors hooked up right now (volts) and I plan on logging a 1/4 mile pass with a 100 octane. With a 100 octane pass I should have zero knock and I'll use this as my standard. Not the most scientific way but it should give me a clear indication if I'm knocking or not.

Here's a logged file from a 20-130 run.

2003 S600

I hooked up dashdaq and boost (map) psi is 5 at idle and goes up to 35-40 max psi, is this right. never hits neg psi even on engine braking. or do you have to subtract atm pressure 14.5?

2003 S600

I guess this is atmospheric pressure + boost b/c I know my kleemann tune would not allow boost over 20 psi. I have maxed at 35 psi which means 35-14.5= 20.5psi.

bassn_07

I always see negative boost when not under load, but once I apply some throttle I'll see numbers between 10-14 PSI. I've never had to do any type of additional calculations to come up with my boost numbers. If you have any question give Dashdaq a call, their customer support is awesome!!!

2003 S600

how did you get egt, fuel press? They are not on my list. I have IAT, MAP, etc but not egt, fuel press.