DIYEFI.org temporary test mule (No 56k)

From DIY contraptions to sophisticated FreeEMS-specific designs! Plus general hardware development!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

DIYEFI.org temporary test mule (No 56k)

Post by Fred »

The trusty female assistant/editor/better half has given me a tentative thumbs up for some testing if I can scrape together enough money to buy some connectors and a loom.

To this end I documented all the connectors on the 7AFE Toyota Carina on sunday afternoon. I was pleased to find that it is indeed a NipponDenso 24/2 distributor setup so my code writing has been in the right direction which is good.

Here is the engine in ALL of it's glory (ie not much!) :

Image

Here is that "NipponDenso 24/2" dizzy :

Image

Image

Image

Image

Image

You can clearly see the three VR sensors that share the two wheels are.

TBC.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

Coolant sensor/plug :

Image

Image

Dizzy and plug :

Image

Image

Image

TBC
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

Injector and plug :

Image

Image

IAT sensor and plug :

Image

Image

TBC
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

TPS sensor and plug :

Image

Image

Toyota 3 wire PWM idle air valve :

Image

Toyota 1 bar MAP sensor :

Image

Image

Image

TBC
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

Rather large single ignitor and plug :

Image

Image

Image

That about covers it for now. If/when I get around to trying it on the car (even just to test for sync with the wheel) I will post here again :-)

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Tony
QFP80 - Contributor
Posts: 50
Joined: Mon Mar 31, 2008 11:47 pm

Re: DIYEFI.org temporary test mule (No 56k)

Post by Tony »

Are you going with new connectors or are you going to buy a used loom and make your own?

If your going new i might be able to help you out on the connectors (if the supplier has them in stock) they look to be AMP connectors i have a PDF with the specs on the line of connectors.

Also how hard would it be for you to get some pics of the harness connector and maybe just make up a interface to the stock harness?
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

I'm unsure at this stage. I was thinking I will probably try to get a used loom or dead ecu and do as you suggested on the inside of the car. The disadvantage there is potential grounding issues and also having to suss out all the pin out details etc. I intend to whip the cover off the ECU at some stage anyway so I can have a nosey at how they treat and combine or not the dizzy inputs. The first stage will likely be piggy backing the oem ecu and watching the VR inputs to see if my code can handle real driving conditions. After that I will likely want to try to run it depending on where the flash and serial stuff is at by then of course. This is all some time away, I just had an opportunity to crack it open and have a look, so I took it and got it done.

Thank you for your offers of help :-) We'll see where it all goes when the time comes a bit closer.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Tony
QFP80 - Contributor
Posts: 50
Joined: Mon Mar 31, 2008 11:47 pm

Re: DIYEFI.org temporary test mule (No 56k)

Post by Tony »

Sounds like a sound plan and the right way to attack it to me.

Your right about the potential ground problems associated with the stock loom i always think in terms of Jap bike stuff our ECU's are usually close to the battery and all the grounds usually go straight to the Battery but with the car you have rubber mounts grounds going to chassis and that leads to headaches that could have been avoided.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

Toyota triple VR sensor CPU interface and overall ECU design analysis :

The exterior of the case is 190 x 162 mm
And the PCB is about 180 x 155 mm

The CPU and/or eeprom chips are located at the rear of the ECU as far from the connector as possible. The vast majority of the passive components are very close to the connector. The only two to220 packages are close to the connector as well and sinked to a 5mm piece of ally bar stock that is screwed to the PCB in the centre with one machine screw that has then been solder flowed during manufacture. The bar stock is then screwed to the case from the outside to dissipate heat. There are only two other significantly sized drivers on the board

Four wires enter the ECU through four pins
A pin is a shared ground between the three VR sensors, it is labeled G- on the PCB
2 pins are the opposite sensors on the 2 tooth wheel and are labeled G1 and G2 on the PCB
The pin that carries the signal from the sensor on the main 24 tooth wheel is labeled as NE

The three VR signal pins have a capacitor each to the VR ground right at the connector and there is another from the VR ground to the main PCB ground too. I also noted that the thermistor inputs are done the same way with a capacitor straight to ground on the actual connector pin. The three caps have "S4" written on them and the one between the two grounds has A4 and is physically shorter than the other three so presumably a lower value. The thermistor ones are also labeled S4.

Each of the three lines then goes on to one end of it's own 1/4w resistor of value 6800 Ohms. The opposite end of these are joined together under the board and connected to the VR ground line. From there the traces go to another three matching resistors of value 5100 Ohms. The G1 and G2 traces cross over here, one on the bottom and one on the top of the board. The NE signal then goes to a 47000pF capacitor that is grounded to the VR ground trace. The other two are attached to a 68000pF capacitor each. All three then go to the bottom of the board.

The G1 and G2 signals pass through an smd resistor with 751 written on it in yellow each. The next stop is a pin on an NipponDenso marked single inline IC that obviously does the conditioning for each of them. There is one IC each for G1 and G2. The NE signal goes through a 751 smd resistor to one of those ICs but not the other. The matching pin on the other is grounded with a decoupling capacitor. The one that the NE signal goes to has two pins out to the CPU and the other only has one. The one that is used on the first is NC on the second. The ones that match are also joined together I guess in something like "wired OR mode".

What this says to me is that each of these ICs is a dual conditioning chip. The part numbers are SE077 9P 09

I guess this means that to run a toyota engine properly, 3 VR conditioning chips are needed on board even though only two signals make it to the CPU.

Hey Jared, best you add another LM1815 in there for all the toyota fans :-) This can be reused by others for a VSS input.

Other interesting features of the board :

to220 mounting to the board is with all three legs bent and the middle one offset further away.
There is a large 3 Amp diode going from ground to power for reverse polarity protection.
There is an even larger possibly 16V Zener from ground to power for over voltage protection. (perhaps we should clamp down harder there? Is there any reason to use a 20V part?)
The electro caps closest to the connector are 63V rated.
I believe the two large transistors in the middle are for the 3 wire PWM idle valve, the pins are labeled RSC and RSO and they have series 1 Ohm resistors and inside those diodes to the internal power rail I *think*.
The bias resistors for both air and water temperature are 2.685k Ohms

I will post more information and a bunch of pics here tomorrow.

Currently I'm thinking that the best way to run this car to test FreeEMS is by building a plugin ECU for it. I might have to track down a dead OEM one to hack and modify.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: DIYEFI.org temporary test mule (No 56k)

Post by Fred »

While I'm thinking about it...

The board has an even mix of SMD and through hole for smaller components. Most of the SMD is underneath and all of the through hole is on top.

Like Dave suggested all the passives and conditioning should be at the connector and are on this board. The CPU is right at the back too. The more I think about this setup, the more I like it.

The side of the case with the two to220 parts is about 5mm thick whereas the other side is about 1.5 - 2mm thick. Typical Japanese efficiency there I guess.

I like that you can pull both lids off of the case and view both sides of the PCB at the same time.

The connector is decent, but only has 54 pins. Though, the way the case is designed it would be dead simple to sandwich two of them together and interface on the inside.

My mate has a prius ECU that he's going to post some pics of soon, that will be interesting to see given how old this car is and how new and complex the prius is.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply