Hey all,

I got a new to me 1989 all stock (presumably), and I am hoping to get it baseline dyno tuned sometime in end of March or beginning of April.

I found a Dyno tuner in Albuquerque NM who says he has experience with "chip based GM tuning". I want to start of with just baseline dyno runs to act as a "control" sample and to get a feel for how the process goes. This tuner appears to already have the hardware required to flash the chips, but he said something along the lines of "we need to open up your ECM, pull the chip, and order a newer/aftermarket one of the correct type in order to flash it", implying that the chip that is in there currently isn't flashable.

Does this all sound right? Does anyone already know what chip (blank) my 1989 needs? Do I need to install the aftermarket socket thing on the board to keep this process smooth? Could anyone provide a sort of overview of what this process will be like? Also the tuner brought up the possibility of upgrading to a newer ECU/ECM.

I am open to this sort of thing, assuming it is pretty plug in play, retains all my factory dash functions, and doesn't require too many modifications. Is there a product out there like that? Maybe a sort of "upgraded" ECU/ECM that interfaces with the factory harness yet that runs off of a more modern protocol and that can be flashed and logged easily by cable.

Also is there any benefit to installing a wideband AFR and is there a way to do so either in replacement or in parallel with the existing AFR sensor that actually enables the tuner to do a better job (can be simultaneously logged with other engine data during runs/tuning).

Basically just looking for a noob crash course or dyno tuning and ECM/ECU options for my 1989.

The plan is to get a baseline, then to slowly install parts while returning to the tuner for new baselines and then new tuned numbers for subsequent modifications. I like data

Thanks!

FWIW I have experience doing my own road tuning on a 2006 OBDII Subaru WRX. I logged data directly through an OBDII adapter cable, then read the ECU program image to my PC, made changes to the image, then flashed the new image to the cars ECU through the OBD II cable. I also installed a wideband O2 AFR sensor that I logged data off of in parallel. This wideband sensor essentially operated independently of the cars ECU and in that sense was just an additional passive data source. I forgot the exact hardware used, and it has been nearly a decade, but I believe the wideband had its own little harness that went to USB then to the computer and I could log runs simultaneously in parallel that included synchronized data from both the ECU and the wideband.

 

Yes that sounds correct. The Moates site explains much. http://www.moates.net/ You can if you wish install a wide band but it's fine not to. I did when I started tuning myself but it's easier to data log and look at the log than watch the gauge while driving. There are a few free data log and tuning programs but I haven't kept up with what is available. The last one I used was TTS Data Master.

 

The chip can be old and not take to flashing a new set of instructions. Some tuners used to put a piggyback module on it, some just sent new PROMs (chips) to plug in there.

If the tuner is experienced you'll maintain all the dash functions, some forget to update things and you lose the MPG readout accuracy. The computer is doing a calculation to send that number to the digidash. I had to make my tuner call Moates back in the day and have him go over it to fix that.

You do not have to tune with a WB02, but its better to do that. I put another O2 bung in the passenger side header explicitly for that, and leave the bung plugged with an old O2 sensor otherwise.

 

Does this seem like the right place to start?
http://www.moates.net/gp1-package-gm....html?cPath=64

The dyno tuner should have the other necessary hardware to log data, and to flash those chips?

 

He should have the software to log and tune, yes. What you want to do is make him ride along with you on a normal daily drive, because its the part-throttle and driveability tuning that takes more time than the WOT tune on a dyno.

If he only does the WOT, I guarantee you will not be very happy with the tune at the end. I wouldnt pay $500 for a tune unless he worked with me on the part throttle tune.

My tuner did mine on the dyno, then I drove it home, and it drove like ****. Pinged, cut off at lights, etc. I went right back the next weekend and drove around and made him fix it in the passenger seat. The timing modifiers don't come up in WOT tuning, driving from stoplight to stoplight, they definitely do.

 

The devil is in the details for sure. Definitely make sure you get street manners worked on... thats what I mostly did on mine and everything else kind of fell into place after. The worst is figuring out the cold start tuning on a modified engine. You know get a couple chances each start and you only look at 1 temperature. I still find myself tweaking things from time to time.

 

The dyno tuner that I'm considering and that I've been chatting with includes a road tune with the dyno tune. He seems solid, but of course I'll start off with just a baseline dyno pull service to get a feel for what it is like working with him. Another dyno tuner I reached out to already basically said to just upgrade to a new aftermarket computer. I don't think I want to do this, but I really do want to hear from yall.

I reached out to Moates and I'm astounded by the speed and clarity of their response.

I'm not sure how it is with NA small block chevy TPI motors, but with my boosted Subaru WRX, a wideband AFR was a critical piece of tuning equipment. The essentially binary rich/lean output from the factory O2 sensors was nearly useless for actually developing a good tune. Do NA small block chevy TPI tunes benefit in a similar way from having a wideband O2 sensor? Whoever I'm chatting with by email from Moates said there are basically two ways to get synchronized wideband O2 data logged along with the rest of the data stream.

One way is to commandeer one of the analog inputs on the factory ECU, eliminate whatever sensor was going to that input, and instead input the wideband O2 signal there. It then would show up synchronized and as part of the typical scan ALDL datalogging. This sound likes a nice way to have a wideband permanently hardwired into the data stream, and perhaps then it is possible to tune/code the chip to work fuel and ignition based off of wideband O2 information? At the very least it would make for nice integration of wideband O2 data into the normal ALDL datalogs. Has anyone done this? Is there a blank input on the ECU that isn't being used from the factory? Is there a good input that can be readily sacrificed? Perhaps something like the low coolant sensor input lol. The idea of a TPI ECU running a program image that does fuel and ignition based off of realtime wideband O2 information sounds very cool to me.

Alternatively, Moates said that they had/have a product, the APU1, that appears to be discontinued, that has additional auxiliary inputs available so that you can send the wideband O2 sensor output to one of these inputs, and pass the ALDL datalogging through the device as well, then send all of that data synchronized through USB to the PC to TunerPro, and TunerPro can then present log data with both the ALDL data as well as the synchronized wideband data.

Is this sort of wideband O2 data/tuning worth the work/effort? On my boosted WRX is was basically necessary.

I plan to keep my vette all motor. No juice, no blower, no turbo. Exhaust and intake to start. Maybe heads and cam down the road. Maybe new intake (FIRST TPI or Extrude Hone Arizona Speed Marine and TPIS setup) down the road as well. Long term is a 383 TPI Stroker.

 

Wide band makes it stupid easy in any case. I can't comment on the moates but I am running an EBL with a wide band and it will make the fueling adjustments to a bin via a datalog and then you just write the new one to the ECM. Even if it didn't, having the wide band data stream right there would also make it very easy. The fine tuning you pretty much have to do manually anyway.

 

Dynamic Efi
l went thru the moates tuning which works but the EBL is so much easier with self learning fuel maps

 

Is this an aftermarket ECU with flash memory?
Does it have improved sample rates and better ability to tune for transients than the older ECU? How is the install and integration into existing wiring harness? Does it retain factory dash functions?

 

What sample rate do you think you need? Scanning at 8192 baud should theoretically yield up to 15 samples per second. 67 bytes total with 63 byte payload plus check sum. In reality, its typical to get 9-10 samples per second which is sufficient for most purposes.

;======================================= ======
; MODE 1 DATA, 8192 BAUD
; CODE $80
;
; The ALDL tester must transmit the following message to the ECM,
; to receive the fixed data stream:
;
; MODE 1 (Transmit fixed SCAN data stream)
;
; ALDL REQUEST:
; - Message ID = $80
; - Message Length = $56 ($55+1)
; - Mode = $01
; - Sum check (= $29)
;
;
; The ECM Will respond with the following message:
;
; - Message id = $80
; - Message length = $95 ($55+64d)
; - Mode = $01
; - Data byte 1
; .
; .
; - Data byte 63
; - Sum check
;
;======================================= ======

example log (including WBO2 data) that has been exported to csv and uploaded to datazap: https://datazap.me/u/tequilaboy/160-...64&mark=76-325

I can only get about 6-7 samples per second with my Evo X logging only a few signals directly to an SD card with a tactrix open port 2.0: https://datazap.me/u/tequilaboy/evo-...mark=2268-2368

9-10 samples per second is pretty solid. The stock 165 ecm is really quite capable with supporting hardware, software and external tuning tools.

 

Count me as a fan of the EBL flash, too (www.dynamicefi.com) for TPI cars. I can't comment on whether or not the sample rate is better than the 8192 of the stock '165 ecm, but I can say it's certainly adequate. It does have better ability to tune transients and does retain the dash functions. It is a little more involved to install as you have to depin and move 3 or 4 wires around in your ecm's connectors. It converts you to speed density; so, you have to add a MAP sensor and do away with MAF. I made a plug-n-play harness to do this for temporary testing because I wasn't sure I wanted to commit to it. The easiest install has you hijacking the MAF connector and tapping into the TPS wiring for 5v reference signal. That's all there is to the install.

I highly recommend it if you plan to do your own tuning - you'll save a ton of money in the long-term and get better results. Get the EBL flash and a wideband o2, and don't bother with the dyno. I don't recommend tuning with narrow-band. It's OK for tuning cruise loads, but the narrow-band basically just tells you if you're lean or rich right around 14.7:1. You can't, for example, tune an accurate AFR of 12:1 if that's what you want for a wide-open pull.

The main benefits of the ebl flash are: it's flashable (no chip burning/emulating), comes with nice tuning/monitoring software using a usb connection, supports multiple analog inputs for things like wideband, fuel pressure sensors, etc. (and they show up on the logging/tuning console), can auto-learn fueling from your wideband o2 under all driving conditions, supports boost, etc, etc. You can read a lot about its development over on thirdgen.org - the guys that developed the system and wrote the software are still very active over there.

 

I suggest you get yourself a cable and start logging data on the road first so that you will be prepared when you hit the dyno. Be sure to record the dyno pulls if you ever get that far.

Here's another old log of mine which happened to have a sample rate of 10.4 samples per second:

https://datazap.me/u/tequilaboy/1st-...6757-6496-6297

This was recorded way back in 2010 with a vintage HP laptop (w-i7 processor) and might have even been through a serial port/cable. I don't recall when I switched to a USB cable for logging. The time signal correlates well with respect to the timeslip for this pass, so the sample number and time signals are trust-worthy and verifiable.

It seems that there is some hardware/software dependency on the sample rate that you will get, Forrest. I've noticed that my recent logs that were obtained with ALDLdroid on an old Samsung Galaxy Tab 2 (wired connection) only achieve around 6.6 samples per second. It would be nice to understand what is really limiting the sample rate and how to maximize for a given setup. Always something more to learn. YMMV.