2003 Toxic Tacoma

[Fred]

Best post yet! :-)

- Made the mistake of blindly uninstalling EMStudio and tossing that version's installer in favor of downloading a newer copy. Turns out the newer copies won't run. My internet connection has crapped out for the past few hours and won't fully download an older archive.

I've been trying to get a good logging system going and now it's raining outside, so the actual cranking will have to wait until morning. I plan to hook the ECU up, start logging (worst case, I'll capture the crank data through "Terminal")

It's about that point that you realise what a god-send it is to have autonomous logging with no requirement for special tools, isn't it ;-)

Thanks to Fred and DeuceEFI for the encouragement, help, and occasional abuse. :lol:

You're welcome! :-)

Fred.

[ToxicGumbo]

Over the past couple days, I was able to get the latest semi-stable EMStudio for Windows 7 to work thanks to a launch argument provided by malcom2073. I needed this copy for its logging capability, but DeuceEFI provided a slightly older version that runs a little more cleanly (faster initial interrogation, remembered settings) which I used for realtime testing.

Until my EeePC is rebuilt, I'm using a beat up laptop from my sister-in-law with a crap battery. Just as soon as I was ready to test, the "charged" battery died. So I had to run about 40 feet of extension cord from the apartment complex stairwell power outlet:



Once rebooted, I forgot I had moved power from BATT to +B (switched) and was confused why EMStudio was recognizing the Jaguar, but not actually interrogating it (the LEDs were bleached out by the sun and weren't a good indicator of on/off). I'll file a feature request to see if there's some way malcom2073 can verify that the ECU is actually powered up or not (and throw an error if not) for people like me.

With everything running, I took a crank log, went back inside, and tried to analyze it. I had forgotten Fred's advice in an earlier chat to look specifically at tooth count and not RPM, so that led me to a lot of wasted time trying to find out what was wrong with the hardware, including benchtop crank tests with a power drill and junkyard crank sensor. Without a toothed wheel, I cobbled this together using a wound-wire disc along with some extras for fitment:



Here's the IRC conversation that followed which reveals my confusion and enlightenment (edited for brevity):

ToxicGumbo: After getting home from work, I created a crank log via EMStudio of my Tacoma + Jaguar A3.
ToxicGumbo: Pulled up the log in OpenLogViewer, dragged "RPM" down to the selection box
ToxicGumbo: Closed that window and saw 7 second red flatline.
ToxicGumbo: No number jumps at all either.
ToxicGumbo: Absolutely crushed.
ToxicGumbo: Brought the Jaguar back inside and hooked a disembodied crank sensor up to it directly.
ToxicGumbo: Used a drill in a creative way to test crank response
ToxicGumbo: The Jaguar diagnostic LEDs flickered in a precise, timed, manner as expected
ToxicGumbo: However, no change to RPM on the computer as seen by EMStudio: 0.00
ToxicGumbo: Doublechecked circuitry, looks correct for VR.
ToxicGumbo: Thoughts on what might be failing?
ToxicGumbo: Keep in mind that the firmware loaded without complaint and EMStudio interrogates cleanly (thanks to the com tx/rx adjust in Win7)
fredcooke: Jeff, I explicitly told you to look at primary count, not RPM, RPM will not read anything with listener....
fredcooke: will not >> can not
fredcooke: it's like a good friend, it just listens, doesn't try to think too much for you
fredcooke: i told you once you get primary count incrementing, we can swap to a firmware that will read something....
fredcooke: but we don't 100% know what firmware to swap to until AFTER you get a listener log with primary and/or secondary incrementing..
fredcooke: so
fredcooke: where are the logs???
fredcooke: you went to bed "crushed" with perfectly good data on your machine!
fredcooke: emstudio HAS those values, too, but difficult to make sense of them in real time
fredcooke: upload your data
fredcooke: prefer you OPEN it and tell me what YOU see first.
fredcooke: what does it show you?
fredcooke: tell me what's there
ToxicGumbo: 4 peaks, spaced at 1k apart, averaging 255 "primary teeth seen"
fredcooke: look for irregularities at regular intervals
ToxicGumbo: That's freaking sweet
fredcooke: look at record 6321 and 6294
fredcooke: count between those steps is?
fredcooke: 68
fredcooke: dual edges
fredcooke: 34
fredcooke: 36-2 perhaps?
fredcooke: Feeling more motivated?
fredcooke: See, told ya it was worth doing a 4 wire test first...
fredcooke: keen for an RPM reading?
ToxicGumbo: So 36-2 was accurate, only this proves it as opposed to skimmed data off the internet
fredcooke: +1
fredcooke: this proves 34 + gap
fredcooke: unlikely to be 35-1

Here's Fred's capture from UltimateLogViewer:



Off to wire up more input sensor support...

Edit: I had spent multiple nights reading forum posts and comparing conflicting technical data to come up with the 36-2 decoder guess. It could very well have been wrong, so I encourage anyone else going through this to always perform and analyze a crank test.


-Jeff

[Fred]

Good work, Jeff! Onward and upward, the sky is your limit.

[Fred]

Off to bed now. You can do anything you want today, Jeff. Surprise me with #23 in the morning if you feel like it. Otherwise I'll help you with config tomorrow and you can nail it then :-)

[ToxicGumbo]

Though I skipped a meal today for the sake of progress, the 2RZ-FE isn't ready yet to be engine 23. Most inputs have been matched from DeuceEFI's assembly manual against Toyota's pin designations. The only one that hasn't yet is wideband O2 as it's still boxed up somewhere and there's no bung to hold it.


Here's a photo of things coming together:



I spoke with Fred on Skype and in his usual irreverent manner towards technology, he encouraged me to avoid staying in the lines and to get the job done first and go back later should things require refining. Sound advice. That got me back into the game and in the end, most things did turn out quite clean, except the wiring for BRV and this ugly switched 12V mess that I'm quite proud of:





He also mentioned that IAT & CHT temp sensors would likely need to be calibrated or swapped. Nevertheless, here they are for comparison--one after the car sat all day, the other after driving around town. It's also nice to see BRV measuring a voltage increase in the second snapshot.


Cold crank:



Warm crank:



To test TPS wiring, the following log snapshot reveals where I manually opened the throttle a few times from under the hood:





Though I'll need to test these sensors out more for functionality, the big problem right now is MAP. So I ate a late lunch and got into the Zone around 9:30pm. The two guys working the late shift, Zach and Jesse, took interest in the Jaguar and FreeEMS and helped me get the parts I was looking for. They also provided a great deal of information about nearby scrapyards, the local scene, and dropped names of people I should meet.



-Jeff

[Fred]

Great work, Jeff!

TPS looks steppy? Where're the logs? you should be able to get some ramp out of it as you move between closed and open and back. This could be the way you're viewing the file, though.
IAT looks soaked/useless.
CHT looks reasonable if your drive was short and/or you waited a longish time to take the reading. Usable, anyway.
Cam sync looks solid, good for testing of new code later. You can be my tester, no one else had this, yet.
BRV, unsurprisingly, looks about right. Fine tune later.

Looking forward to MAP data and RPM data (you never loaded the other fw to try RPM, should've...)

Hit me up when you're awake and/or have MAP and we'll work together on this.

Fred.

[Fred]

IAT looks soaked/useless.

This might be usable, roll with it and see. Need calibration data for both coolant and air, though.

[ToxicGumbo]

Safari just trashed my first attempt at this reply, so here's the rapid-fire version second attempt:

Significant progress was made today--and I have Fred, DeuceEFI, and malcom2073 to thank for this. Fred, especially, showed extreme patience as he walked me through different topics and made sure I was learning as we went along.

Today's primary goals were to take MAP readings off the intake, connect the fuel pump relay, and check RPM based on the new decoder firmware Fred put together. malcom2073 provided me with his latest EMStudio build which made logging significantly easier. Here's a snapshot of that data as seen in OpenLogViewer:


RPM looks clean, but there are some definitely problems with MAP output.



A few observations:

• I corrected my mistaken piston firing order on this thread's first page to 1-3-4-2, paired for ign/inj as 1-4, 3-2
• The Toyota intake manifold has quite a few rubber covered locations for taking pressure readings
• The stock (pink) Toyota injectors should be able to handle a 300cc flow rate
• The fuel pump was as easy as hooking up ECU B-6 (FC) the Jaguar's "Fuel" tap (and is responding as expected)
• By performing a logged "suck test", the MAP sensor was cleared of any problems or leaks with respect to the results we're seeing
• The A3 board will need to be slightly modified so that all four outputs of the CD74AC86E chip can be taken advantage of. 100 ohm resistors will sit between it and the ignitors. The TC4424 IC was mentioned for possible consideration at a later date.

Here's a photo of the PCB changes I'll need to do (but forgot to account for the polarity change):



Some useful links:

• TI's CD74AC86E datasheet
• Toyota's fuel injection system datasheet
• Toyota's TPS data sheet
• Microchip's TC4424 high-current IC

Tomorrow: Finish up the igniter support mods, find my timing light, and get correct MAP readings.


-Jeff

[Fred]

TPS looks steppy? Where're the logs? you should be able to get some ramp out of it as you move between closed and open and back. This could be the way you're viewing the file, though.

Orrrrr, it could be my/the firmware's fault! Sampling is slow while engine isn't spinning. Fixed in future versions.

Safari just trashed my first attempt at this reply, so here's the rapid-fire version second attempt:

Don't use a boy's tool for a man's job :-p Firefox crash or OOME would have come back with the typed content ready to roll.

Significant progress was made today--and I have Fred, DeuceEFI, and malcom2073 to thank for this. Fred, especially, showed extreme patience as he walked me through different topics and made sure I was learning as we went along.

Make me proud, Jeff! :-)

Today's primary goals were to take MAP readings off the intake, connect the fuel pump relay, and check RPM

Also prove TPS was fine: Success!

RPM looks clean, but there are some definitely problems with MAP output.

Specifically, MAP values are unrealistically high. The shape and behaviour is correct, but the range is just 6kPa. There is a blockage or leak in the path from manifold to sensor as proved by Jeff's lungs.

The stock (pink) Toyota injectors should be able to handle a 300cc flow rate

I strongly believe these are 325 - 330CC injectors. As per links given in IRC.

The A3 board will need to be slightly modified so that all four outputs of the CD74AC86E chip can be taken advantage of. 100 ohm resistors will sit between it and the ignitors. The TC4424 IC was mentioned for possible consideration at a later date.

Should work, Hentai ran like this or worse. Not recommended, though, not by me, anyway. I know some other idiots who do recommend this :-) Fail. Upgrade before boost.

Tomorrow: Finish up the igniter support mods, find my timing light, and get correct MAP readings.

Looking forward to hearing that you were able to configure the timing correctly. MAP may behave better once running, though I doubt it. So perhaps wait and see. If you can improve it beforehand, though, all the better. As anecdotal evidence, my first MS2 start up and drive was with the map hose completely off, and it worked OK; pig rich at idle, good under load (where reality and sensor matched). You're in better shape than that. ;-)

Fred.

[em_knaps]

Jeff,

Fred wanted me to post this info regarding the auto trans in your Tecoma.

If it is the normal auto that comes behind the jap 2RZ's we get here in NZ then it should operate
without electronic control of the solenoids.

you will not be able to get 3rd gear and maybe 1st (I can't recall if 1st worked)

but it will function (albeit crudely) if you manually shift it through the gears.

this is my experience of the Aisin Warner A340 family of transmissions.
If you have a different trans I cant draw from my experience but I still may have and idea or two on how to make it function.

in future you can either use the shift position indicator on the side of the trans to operate the solenoids,
or alternatively there are ECU's specifically designed to control these transmissions.

Spitronics is one the comes to mind ( http://www.spitronics.co.za/products/tcu.php )

may the 4x4 gods smile upon you.

Daniel.