I took my car out today for a few WOT blasts to check out my new laptop and the new voltage inverter I just bought. I was running Diacom Plus.

I saved the files and when I returned home I downloaded them to a floppy. When I reviewed them on my PC I saw some very, very high timing numbers in "Spark Adv Relative to TDC" ; 70.2*, that's Seventy point two degrees, being the highest; lots of others were in the 50 and 60 degree range.

I cross checked with the last saved WOT blast and in that run, the timing numbers (in the same Spark Advance Relative to TDC section) at WOT was only 27 degrees, that's twenty seven degrees.
Any ideas as to why those number would be so high?

Thanks,

Jake

 

I just went back through the three files I saved during three WOT runs and I found another errant reading.

One single frame in the air/Fuel Ratio (commanded) area read 23.9:1, which, I think you'll agree is a little lean.

Even though I was getting some 50, 60 and even 70 spark advance numbers, I was only getting single digit knock retard but very, very few actual knock counts.

I'm going to try installed the ECS module that Gordon Kilibrew recommends that is less sensitive and see what difference that makes. Next task is to track one down (Chevy dealers are notoriously high).

Thanks for your input.

Jake


[Modified by JAKE, 11:53 AM 12/17/2003]

 

Jake, what ECS module does Grodon Kilibrew recommend? Sounds like somthing I might want to try on my 86. Thanks

 

Perhaps I wasn't clear: A data reading of 70* timing advance makes no sense IMHO.

First because most GM ECM programs limit timing (in the software) to around 45*
advance max;
Second (if you are running a distributor) distribs are built with 90* wire phasing. 70*
advance would mean the distrib finger was physically pointing at the wrong plug
wire at the time it fired.

You would probably notice an incorrect firing order on a WOT test.

Thus the data displayed by your Diacom software is (more than likely) being
corrupted somewhere.

If the car is otherwise behaving normally, the ECM is probably OK.

That means the data corruption (if that in fact is the problem) occurs either in the cable
or in the laptop. Application of Occam's razor leads one to examine the (new) laptop first
for the source of the error.

I don't quite see how the knock module could be the issue. A knock reading can only
retard the timing, not advance it.

Just trying to help. :)

 

It's part number 16038331 and should be available at your local Chevy store.

Jake

 

I agree that the timing could not have been actually that high. At over 6300 rpm I know something a little more strange would have happened if I did have over 70 degrees of timing.

In all other respects, though, the engine runs fine and the Diacom readings are all in the normal range (except for that single lean A/F command).

I suspect it's either cable or laptop too. I'll test further and see what's what.

Thanks for the responses.

Jake

 

I saw your other post and assume you want MAF traces as opposed to speed density. I've only got speed density. There are some really good tuners on here that charge for burning chips. They probably have a multitude of saved MAF scans so hopefully they will reply!

1) I agree with Occam's razor...especially when it comes down to messing around with electronics.

2) I tried opening up a handful of traces in DataMaster. Without configuration, the numbers were messed up. This was when opening up a saved GDF file.

3) What program are you using to review the data? Diacom, Datamaster, Excel, etc? Did you translate it to a DBF file or did you keep it as a GDF file?

4) Partitioning the hard drive so you can have an actual DOS partition is never a bad idea. If you ever run into problems with either your datalogging or your EPROM burner, it helps narrow some things down.

5) That GM part number crosses to a Wells brand ESC108...and like Dr J said (as long as the uh, pickup coil is correct), the ESC module will never add timing.

 

Perhaps my post was confusing. Let me try to clarify it.

I have an 86 which is a MAF system and was running DIACOM PLUS, which saves files in the GDF format.

The reason I was going to change the ESC module is not because I suspected it was contributing to the timing advance problem, but because of the knock retard I was getting. I suspect the roller rockers may be causing it. So that is a separate issue.

I made three WOT runs, all with Diacom Plus running, only the last one show the timing irregularities. The only difference between the three runs is that on the last run, I shut down Diacom Plus, then the engine, then restarted Diacom and then started the engine. I nailed it before the engine went closed loop, so I'm now wondering if going WOT while still in open loop and with the engine run time being less than one minute (according to Diacom's count) timing is effected. All the temps were normalized, so I didn't give a second thought to going WOT right after the restart.

The laptop only has DOS on it, no Windows or any other operating system and it boots directly to a "C:\" prompt.

The ESC108 is great information; I know it'll be cheaper than a GM part from Chevy. Thanks for that.

I know there's someone out there with a good GDF trace they'd like to share, right?

Thanks for the help.

Jake

 

Jake:

OK, it's becoming more clear now.

HighHopes was just having fun :) with the 'experts' who post in the expectation either
of making money for their help, or of proving PT Barnum's immortal observation; but don't
often give away useful info for the DIY tuners. Others are here to learn, or just to enjoy
the hobby. With that prologue out of the way, back to your questions:

False knock - you need to be a little careful of that. There was a fellow on here last year who
had 'false knock' blow a 'real hole' through a very expensive piston. (I think others
here have done the same thing, but this was one case I am certain of.) The only way to be sure
about the 'false' part is to pull some timing out of the spark curve & see if the knock count goes
away in that RPM/load area. That will be particularly true if you have jacked-up the cam to yield
better cylinder filling at high RPMs. Don't dismiss knock readings without some serious thought,
IMHO.

Diacom Data - you mentioned turning up to 6300 revs on your test. I'm personally skeptical
of ALDL data at high RPMs. It has a fairly slow update rate on the older ECMs, so I take those
readings with a LARGE grain of salt, especially on a fast, high-rev transient. I have no hard info
to back up that opinion, so use your own judgment (and maybe some plug readings). My thought
is that GM intended the ALDL stream for sort of 'run of the mill' driving diagnostics rather than
all-out race tuning. The ALDL baud rate is way too slow for that. I plan to build a test rig later
this winter so I can look at just that area of the ECM/data stream - but that is only going to apply
to a 727 ECM, and is cold comfort to you now anyway.

Example Data - you could also ask the Q over on the 3rd Gen PROM board - they get more
hobbyists with older cars over there. You might also find a place to post your files on the web,
so others could download them for review - that draws a response sometimes. If all else fails,
e-mail them to me & I'll take a look for anything obviously strange.

Start up timing - Other GM programs add some timing & fuel to the start-up event, which decay
out based on time, temp, and number of crank revs. Without looking at your particular mask I
don't know what all specific things are done to spark & timing on your initial start-up. As a
general rule it's best to give the computer a minute or two to settle down before you nail it -
unless you're the getaway driver, in which case you gotta do the needful...

Hope that helps.

DrJ

 

Gret response, thanks for taking the time.

I'm running 89 programming with some other changes (injector size, CID, EGR, etc.) I received a couple of traces but were from 95's which show up a little differently than the traces from my 86. they have two BLMs, two Integrators, two Injector pulse timing, etc., but only a single ignition timing.

I need to see a trace that has two ignition timing lines; TDC and Ref.

Your point about start up fuel and timing may be the key. Could be that all I needed to have done was wait a few to allow the engine to run a bit longer before trying to saw it in half.

I'll do some more experimenting and see if that makes any difference. I'll keep you posted.

Thanks again,

Jake

 

I've got a friend with an '89 TransAm. These were initial runs we made when he wanted to swap over to the ARAP BIN.

In ROGGE.DBF, notice starting at frame 2333, he hits some heavy knock retard when the throttle goes over 4.0 volts. Like I mentioned, these were initial runs with the ARAP first installed. I didn't review the other files. ARAP has been known to have agressive spark curves, and the carbon on top of his pistons with the stock long block probably doesn't help.

They aren't GDF files though. I translated them while in Diacom to a DBF file, which can be read by a program like Excel. [Note: THat's why I asked what format you view the files as] HTH
4 '89 Runs

Matt


[Modified by HighHopes85, 12:59 AM 12/20/2003]

 

Jake,
I had a similiar problem with crazy timing indications in diacom. I thought it was speed related(communication) and I ignored them for a while thinking it couldn't possibly be. Then I started noticing that my tach would climb funny in the rpm window of diacom.i.e 1000, 1200, 1600, 1400, 1600, 1800, 1700, 2000, 2200, 2000, 2400, 2500, 2300, etc...Other data was effected as well. MPH, AFR, etc...
I started looking at my tach in the car and it would do the same so I knew it was really happening. What I found after searching a long time was my spark plug wires were on top of the distributor pickup & module wires to the ecm behind the distributor. This was introducing voltage to the ecm messing up the signal. I replaced the MSD helicore wires with Accell 8.8 Perf+ wires and rerouted them away from the harness behind the distributor. This corrected my timing issues and the data collected in diacom. All is well now.
Not saying this is your problem but something to look into.
Oh by the way when I was struggling with this a couple of years ago you sent me a copy of your diacom run so I could compare runs. I still have it. Your correct about the 27* total.
Good luck finding it.
John

 

Thanks for the files, Matt. Later this evening I'll do a comparison of the ones you send me with the ones I saved.

Maybe I'll translate mine too so I'll be making an apples to apples comparison.

Jake

 

Thanks for that.

I took your advice and checked all the spark plug wires and found that none were close to any of the distributor wires.

I went a step further and pulled the distributor cap and rotor and using GM's shop manual compared the color of the wires of the pickup coil and ignition coil.

According to GM I had the wrong coil installed. GM says that if the PU coil has a yellow wire the coil must also have a yellow wire. My coil wire was yellow but the pu coil wire was green.

Had my original coil which has red and white wires, so I swapped coils.

Haven't had a chance to do any more testing to see if the coil change makes a difference.

Then, again, it may be just as someone already posted, that ALDL mode isn't accurate at the rpms I was at.

I'll post any new developments.

Jake

 

Yes I have come across that note as well. My colors matched though. I hope that fixed it for you.
John