TunerPro RT with 1991 Cadillac Seville 4.9L

Discuss General Motors (GMC, Chevy, Pontiac, Hummer, Cadillac, etc) tuning topics here. Request definitions, discuss parameters, etc.

Moderators: robertisaar, dex

Post Reply
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

1. Using the ADX file for 4.9L Cadillac from Red Devil River ALDL-USB adapter package CD, which is stated to be from work done years ago by Victor Seitz (LittleNo) of Munich, who was last seen on CaddillacForums.com in 2014. Have not tried the ADS file found on this site.

2. TunerPro RT able to connect and acquire a data stream. However, either the stream is out of phase, or more than one stream is being acquired. Once in a while the correct PROM ID briefly flashes by, to be quickly replaced by a variety of other numbers. This pattern repeats rapidly and continuously.

3. Using the 1992-1993 3800 ADX file as a guide, I added in commands and a macro to send 'Mode 8 (Disable Comms).' The macro appears in the dialog box. I tried clicking the Send button both before and during stream acquisition. There was no change in the behavior described in 2.

4. I selected Record in the dropdown menu, and the Connected box indicated that a recording was being made. The data list window continued to show the stream(s) as described in 2. After a few moments I selected Stop in the dropdown menu, and a popup requested an XDL file to save the datalog to. I supplied a name and a file was created. I stopped the acquisition, shut down, and came inside to look at the datalog file. It contains one (1) packet of data, which has a seemingly random number instead of the correct PROM ID and other seemingly random data. This is not very useful. I was expecting to find a record of the continuous stream(s) of acquired data that I watched in the data list window, as described in 2.

EDIT: See below on the datalog.
Last edited by notstar on Fri Apr 23, 2021 4:23 pm, edited 2 times in total.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

I should add, I am using the latest release of TunerPro RT.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

I should also add that I was wrong about the XDL file. I played it back, and it flashes through the behavior I described. I exported a CSV file (some 50K+ bytes), so I'll be looking through that later.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

How can I post the XDL and/or CSV file(s)? Attachments only seems to want image file types.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

After looking at the datalog, correct PCM packets appear to be in phase, with some other stream(s) from some other source(s) with some other phasing(s) randomly mixed in to disrupt things. In less than 8 seconds of datalogging I acquired 5 packets with the correct PROM ID, and in these packets the rest of the data looks stable with (mostly) reasonable values. Since 2 of these correct packets are consecutive (no incorrect packets between them), the acquisition can't possibly be out of phase.

The 5 correct packets all say only Mode 1 is on, Mode 8 is off. There are 52 incorrect packets, total 57 packets in less than 8 seconds.

That's about all I can get from it.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

Leaving no stone unturned, I just tried "DeVille mode," that is, sending 0x01 instead of 0x00 in the Mode 1 command. Seemed reasonable, since the DS file mentions that there is "no master on the DeVille bus," or some such. With that I still get a flood of seemingly random incorrect packets, but get no correct packets at all. So no, that won't work, lol.

At this point, I think if I want to use TunerPro RT to datalog this thing -- and I do, it's shown me items of interest already once I fish packets out of the flood -- I'm going to have to remove BCM and friends, leaving only the PCM to talk to. Doing that requires first constructing a homebrew instrument panel, because in this car the BCM feeds the instrument panel over the same ALDL data line. But I see no real downside, since I'd just as soon yank the computer-controlled struts, A/C, ABS, etc., etc., and the factory instrument panel only shows speed, miles and fuel anyway. Time will tell.

This motor seems to have a following in the Fiero community for swaps. That's ironic, since the styling on this gen Seville is directly derived from the Fiero. So if I pull BCM and friends, I'll only be making my Seville more Fiero-like than it already is.

The only alternative to all of that would be for TunerPro RT to have a way to filter packets based on specified values at specified offsets in acquired data. I don't know if that would even be practical, but I assume the cars that are encumbered in this way are probably not popular enough to justify trying to do that anyway.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

Anything I've missed, or other input, would be greatly appreciated.
clarkent5477
Posts: 2
Joined: Wed May 05, 2021 7:12 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by clarkent5477 »

I saw a post elsewhere mention the BCM could get chatty. I have a 91 Seville as well so I'm very interested to see what happens here.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

I think the BCM does most of the talking, and the only way to shut it up is to remove it, along with a lot of other things. Major surgery. Then we might be able to datalog the ECM. But how many people actually want to remove the A/C from a Cadillac? I might be the only one.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

I just found a way to get some use out the thing the way it is. Do datalogging with TunerPro RT, then export the datalog to a CSV file. Then look at the CSV file in File Viewer Plus 4 (a program I just bought for $40, installed with difficulty due to issues getting the required version of .NET on Windows 7). FVP4 allows me to edit out the bogus rows of data, leaving only those that match the PROM ID. It's not a proper datalog, but does give a random sample of values under various conditions. I can then export the result from FVP4 in a variety of forms. It's a nuisance and not exactly the whole nine yards, but at least it's something.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

clarkent5477 wrote: Wed May 05, 2021 7:17 pm I saw a post elsewhere mention the BCM could get chatty. I have a 91 Seville as well so I'm very interested to see what happens here.
I see only one thing that could possibly happen here: I turn up some day talking about how I yanked BCM and friends out. I have no clue what else, and there hasn't been even the faintest whiff of interest from anyone associated with TunerPro, so I assume there isn't anything else.
notstar
Posts: 13
Joined: Thu Apr 22, 2021 3:02 pm

Re: TunerPro RT with 1991 Cadillac Seville 4.9L

Post by notstar »

I recently submitted an inquiry to Mark Mansur, and got this reply:

"Hi Jim,

Generally, what’s done during scanning for vehicles like this is to silence the other modules on the line such that only chatter from the device you’re interested in is being transmitted. This is done, IIRC, with mode 8 commands.

You can, however, do header matching in TunerPro. It’s not well tested, but it does exist. You just enter the header bytes that you expect in the reply. Anything that doesn’t match won’t be parsed further.

-Mark"

I'll be looking into these points. Thanks, Mark!
Post Reply