[Rednaxs60]

Resurrecting this project after a C-19 hiatus. Started in 2019. Reason - older technology, very proprietary, very little information and no ability to configure/tune. If the ECU fails, used ECUs available, but quite expensive if found.

My thread in the Getting Started, quite lengthy is: viewtopic.php?f=18&t=2789

Bringing the discussion to the User Project forum.

This project is an ECU replacement with as little change to the OEM system as possible. Read through the manual again, and the manual mentions that the v0.4.X boards - mine is the v0.4.4, are primarily meant for an ECU replacement using the existing OEM components and wiring. Can be used for other purposes. The manual also mentions that if this is not the cas, the v0.3 family of boards may be a better fit. Interesting bit of trivia.

This is a recap from my other thread:

Time has a way of clearing one's thoughts on an issue. The break from this project has clarified a lot for me. This is actually a recap of what I have previously posted.

I am now focused on a plug and play aspect. Have read about the many aftermarket ECUs that are configured as direct drop in replacements depending on the vehicle.

I purchase an '85 Gold Wing Limited Edition FI parts bike and have the engine and CFI components on hand. Had to disassemble the engine to transport it from Ontario to British Columbia Canada in a car. The engine weighs in at some 350 pounds. Putting it back together and will be able to use it for some of the tuning and such instead of taking my on road bike apart to get readings.

Reviewing the forum threads about Gold Wing FI conversions. Most use the Micro/Mega Squirt family, but one fellow is using a Speeduino. There is good information to be had.

Most of the components are installed and useable. An O2 sensor is in the cards, as is a new trigger wheel (?). Can use the existing 8 tooth crank trigger wheel and one cam sensor.

Deciding on whether to change out the low impedance injectors that are installed with a restore pack. Sequential instead of wasted spark? This is predicated on whether there are extra components to install such as individual coils and spark igniter units. Real estate for additional items is at a premium.

Vacuum to be used for the on board MAP sensor. The Gold Wing has four vacuum ports, one for each cylinder. A fellow on another forum that converted his 1100 Gold Wing to FI married these to a single point and then took one vacuum line to the MAP sensor. A small mixing box would probably do the trick.

The crank and cam sensors are VR sensors, and need to be connected through a VR conditioner.

Upgrade to high impedance injectors - can probably get the operating specs on these, or stay with the OEM low impedance injectors - no info available that I can find other than from others who have done an EFI conversion and have info from testing, that are installed with a resistor pack. The resistor pack consists of two 3 ohm resistors. Each resistor is in line upstream of two injectors. The fellows on the Gold Wing forums may have some settings that can be used. PSIG indicated that the OEM setup is good.

Tested the CLT and IAT sensors this fall as part of my investigation into the operation of the 1200 engine. Have two data points identified, only nee a third. Will be doing this on the spare parts I have. Tested the ones that are installed on the engine against the spares, same readings. Have two confirmed readings. At 20 deg C - 2.5 K Ohm, and 80 deg C - 0.3 K Ohm. This is for both. Need a third reading - going for the freezer.

An ECU enclosure is needed. Have an OEM ECU that is faulty, has a bad ground circuit. may be able to modify this to suit and use the OEM box connector, be a nice clean install. have checked this. The OEM ECU enclosure is not suitable - will need to make one to suit. Connector is another challenge.

Spark igniters can be connected directly to the Speeduino. Need to determine if these are triggered GOING LOW or GOING HIGH.

If a different trigger wheel is installed, intend to install it where the OEM trigger wheel is. OEM installed crank trigger wheel is an 8 tooth wheel. There is a single tooth cam shaft wheel with two VR sensors. can use the crank trigger wheel and one cam sensor for operation.

The integration of the dash and travel computer happens before the ECU, or a signal from the ECU is used. Upgrading the ECU should not affect this.

The functionality that I lose at the start will be the self diagnostic that is built into the OEM ECU and the respective dash indication. I know what signals are monitored so an add on unit that does this would be nice. May be room in the OEM ECU enclosure to add a small unit. The OEM enclosure has a side window to view the ECU error codes. Need to learn how to program for this.

Rad fan operation can be controlled by the Speeduino, but is working well as a stand alone component. I also have a bypass incorporated that allows me to turn on the fan before the engine gets to the fan start temp. Good for city, start/stop driving.

Power is not an issue, did the external alternator mod a while back. Have a 55 amp auto alternator installed.

[Rednaxs60]

Reviewing my thread in the Getting Started forum. Made some notes as a review.

PSIG - At the start of this thread and project, you mentioned the 8 tooth crank trigger wheel, and one cam sensor would work in an 8/1 configuration. Understand this better. You mention the low-Z injectors and resistor pack is useable as well. Will peruse the Gold Wing conversions for information on what they used for these settings. Could change over to high-Z injectors using jumpers in the OEM wiring so that if the need to revert came about, can easily revert back to the original configuration. More to follow on this.

For LAV1000/theonewithin/PSIG - measured FP output at greater than 960 cc/min. OEM spec is for a minima of 630 cc/min. The injectors flow approximately 280 cc/min. Injectors ND 195500-1070.

Looked at the crank sensor - Ns. The trigger wheel is 8 tooth, no missing tooth. There is an upper and lower install position, either works. Have used the lower and upper, but not at the same time. Wonder if the Speeduino can do this?

LAV1000 - have tried to contact 14point7, no joy to date. Will find a suitable O2 sensor. Don’t have a welder and having welding done piecemeal is expensive. Did a JB weld test where I took a bracket, used JB weld to secure a nut to it. Let it cure for some 24 hours, applied 30 ft-lbs and nothing happened. Will be good to attach an O2 boss to the exhaust.

LAV1000 - have licensed Tuner Studio and MegaLog Viewer software.

PSIG - regarding sequential ignition, understand. Something like using straight rate and progressive rate springs in a suspension setup. Most would not notice the difference.

PSIG - MAP sensors. Have tested the Suzuki IAP sensors and will look into the Speeduino setup to determine what can be done. The Gold Wing vacuum system has four connection points, one at each injector holder. One Gold Wing owner took four vacuum hoses of equal length, connected these to a common point and used the averaged vacuum so to speak to connect to the MAP sensor.

Daze - regarding thee OEM ECU diagnostics. Will not be piggybacking the OEM ECU and Speeduino. Thinking a small unit that monitors the required signals, and if all is well no error code is generated. This would be active whenever the engine is going to be started, when the ignition key is turned to the ON position, and continuously there after. When an error code is generated, this unit would send a signal to the dash warning light illuminating it. There are other issues to contend with such as when the engine will be stopped because of the fault, and which ones would go into a limp home type mode. Lot of if-then type statements I would think. This would be a follow on project.

PSIG - regarding COP coils, not enough room to install.

Post #34845 on this thread, discusses ignition timing. I will be revisiting the setup according to the Speeduino manual. No joy when I first tried so I tried a different approach. Set #1 at TDC, very easy to do. When the engine is started, the timing mark moves and the resulting timing is about 13 degrees BTDC. Timing after this is approximately 15 degrees BTDC - my guess.

First tooth past the post with an 8/1 configuration came in at approximately 135 degrees. Missing something as I mentioned.

Post #35117 (mine) and #35142 (LAV1000)of this thread, looked into generic programs to convert degrees into time. The programs I used give me a cranking time of approximately 4.2ms at 700 RPM cranking, fast idle 1280 RPM approximately 3.2 ms. idle dwell time of approximately 2.3 ms at 1000 RPM, and fast idle of 1280 RPM of about 3.2 MS. The example in the manual indicated a crank dwell time of 4.5 ms. Will keep this In mind for the initial tune. Good place to start.

PSIG - going to BBQ the whole pig at the get go, but with a realistic component being to use what I have and make it work, and work well as you recommended. Can always modify as time goes on.

[Rednaxs60]

The v0.4.4 board that I have has a VR conditioner installed:

VR Conditioner.JPG (261.27 KiB) Viewed 25709 times

. Not sure if it single or dual VR conditioner, or what pin(s) to use. Thoughts?

There is a Step motor driver as well. Don't know what to do with it:

Step Motor driver.JPG (256.02 KiB) Viewed 25709 times

[Rednaxs60]

I have done another review of the documentation for the Speeduino. Marked up my copy with comments/queries.

This post is a synopsis of what I have determined. Comments appreciated.

Hardware

Inputs

Dual Wheel setup - Use as per OEM install. 8 tooth crank trigger wheel, primary. Single tooth on cam - secondary. OEM components, no need to change out at this time.

Crank Sensor - VR sensor, use as per OEM install. Connected to the Speeduino through the VR conditioner board installed

Cam Sensor - VR sensor, use as per OEM install. Connected to the Speeduino through the VR conditioner board installed

TPS - 3 wire variable potentiometer, OEM install

MAP - OEM install uses two MAP sensors for redundancy. One fails, the other operates the system. Both fail, goes into a get home mode. Suggested that I use one as the MAP sensor, the other as the barometric pressure sensor. Have to modify the vacuum hose setup to suit a single MAP sensor.

CLT and IAT sensors - use as per the OEM install. Have two resistance points need third for Tuner Studio calibration.

WBO2 sensor - to be purchased and installed.

Outputs

Injectors - use existing wasted spark arrangement - low-Z injectors with resistor pack. Can use OEM injectors, but no tech data available. May be beneficial to change to high-Z injectors - tech data may be available.

Coils - use as per OEM install.

Spark Igniters - use as per OEM install. Determine Tuner calibration GOING LOW or GOING HIGH.

Aux Outputs

TBD

V0.4.4 Board

VR conditioner ancillary board installed.

Jumper Configuration:

JP1 set to 12V
JP2 set to 12V
JP3 set to 12V
JP4 set to OFF
JP5 set to OFF


Injector connection - using wasted spark arrangement. Use injector channels 1 and 2. Firing sequence is 1/3/2/4. Injectors 1/3 on channel 1. Injectors 2/4 on channel 2.

Tuner Studio

Have not started a project in Tuner Studio, but have some thoughts that I will list here. More to come.

Determining tooth #1. Tried early on. Will be revisiting. Trigger angle previously determined was at 135 deg ATDC.

Trigger edge to be determined as I get more info on it.

Engine Constraints

Settings

Control algorithm - Speed density to start. Honda uses Speed Density and Alpha-N mapping.

Injector staging - Alternating

Engine Stroke - 4

Number of cylinders - 4

Injector Port Type - Honda has ITB. TBD for best configuration.

Engine Type - Even Type

Injector Layout - Paired

MAP Sample Method - Cycle Minimum

Injector Characteristics

Trigger Pattern - Dual Wheel (?)

Primary Base Teeth - 8

Primary Trigger Speed - understand what is required - TBD

Missing Teeth - N/A

Secondary teeth - 1

Trigger Angle Multiplier - TBD

Trigger Angle - 135 (?)

Trigger Options - Re-sync every cycle looks interesting

AFR/O2 - WBO2 sensor and controller to be purchased. PSIG mentions that unless you are doing a very custom fitment, just about any WBO2 sensor can be used. Thes are not inexpensive in Canada.

AFR/O2 - Algorithm will probably be PID

AFR/O2 - Ignition Events Per Step - what is meant by an ignition cycle?

Rev Limiter (is this required)

Soft Rev Limit - set at redline - 7500 RPM

Hard Rev Limit - set at 8000 RPM

Spark Settings

Spark Output Mode - Wasted Spark

Cranking Advance - approximately 5 degrees BTDC - will look at other installs for comparison

Spark Output Trigger - determine whether it is GOING LOW or GOING HIGH

Dwell Settings

Cranking Dwell - 4.5 ms

Running dwell - 2.3 ms

Spark Duration - TBD - initial setup 0.5

Overdwell Protection - TBD if needed - initial setup may be 8 ms

Cranking

Cranking RPM - initial setup at 500 RPM

Flood Clear Level - 100 RPM

Fuel Pump Prime Duration - 3/4 seconds

Priming Pulsewidth - Like this. The OEM ECU may be doing this too, don’t know. Initial setup at 1 second

Cranking Enrichment - Thinking not required

Cranking Bypass - Thinking not required.

Fix Cranking Timing With Trigger - Thinking not required.

Warmup
TBD

Idle

Using the OEM installed stand alone IAC system.

Stepper Driver Board

In manual. Will peruse for more info.

Fuel Pump Control

More to follow.

Thermo fan

Will use passive OEM install.

This is what I have come to grips with at this time. More to follow.

[Rednaxs60]

With all projects, one of the key elements is to read, read, and read again. Taking this to heart.

The '85 Gold Wing FI model is quite a design achievement back in the day. One of the Computerized Fuel Injection components was the Gr/Gl cam sensor arrangement. There are two sensors that allow for operating redundancy, quicker starts, and it emulates sequential injection. These bikes have no O2 sensor.

The interesting issue is there are magnets mounted on the Gr/Gl sensor plate. Many have wondered about his, as have I.

This plate:

GR-GL Sensors.jpg (347.25 KiB) Viewed 25638 times

has a magnet placed under each sensor mounting position, close to the sensor.

I think I have found what may be an answer to this. Reading the Motorcycle Fuel Injection Handbook by Adam Wade and his blurb on hall-effect sensors caught my eye.

These sensors are solid state, and are used instead of a VR sensor for certain applications. Apparently you can build signal conditioning circuitry around onto the silicon wafer that is in this sensor, but it has limitations.

Hall-effect sensors can have a digital or analogue output. The Gr/Gl sensors on the older Gold Wing FI bikes have an analogue signal.

A hall-effect sensor requires a magnetic trigger wheel, and the fingers/teeth need to be fairly wide compared to the slim fingers/teeth for a VR sensor. I surmise that the magnetic wheel in the case of the Gr/Gl sensors can be either a rotating wheel, or where the magnets are located on the Gr/Gl sensor plate. Don't need a lot of magnetism to develop a signal, just enough to activate the sensor.

This makes this type of sensor a good candidate for a small number of triggering events such as the cam sensor.

This is the only reference I have found that could explain the magnets on the Gr/Gl sensor plate. It's as good an explanation as any I've seen.

[PSIG]

Correct, and most automotive Hall sensors in the last 30 years or so have been back-biased, placing the sensor between the magnet and target (wheel), so the sensor detects the flux changes in the the magnetic field. This allows a simple ferrous target, without "flying magnets". They also are typically digital output sensors rather than analog, and why the default setting for Speeduino is Hall/opto/digital with no analog signal conditioning. For yet more trivia, do a search here for "back-biased", or read the datasheets below for some interesting info.

The MLX90217 is a random example of a gear tooth sensor designed for back-bias, which may be mounted in whatever fashion is suitable for the purpose. Most commercial sensors will have a device like this inside with the magnet:

MLX90217-Datasheet-Melexis.pdf

Automotive back-biased Hall gear tooth sensor
(594.2 KiB) Downloaded 1741 times

[Rednaxs60]

PSIG - thanks for the reply. Amazing what you learn from a review of previous readings. Missed this on the first time round.

Now focused on a 36-1 trigger wheel, may be easier to setup the system. Will contemplate this.

Pinout to be finalized, what goes where. Found another thread, cx500t used an early model Honda OEM ECU enclosure with the OEM connector. Think I will investigate this further.

ECU Enclosure and connector pics:

OEM ECU Enclosure Connector 2.JPG (238.92 KiB) Viewed 25611 times

OEM ECU Enclosure.JPG (163.64 KiB) Viewed 25611 times

OEM ECU Connector 3.JPG (144.59 KiB) Viewed 25611 times

OEM ECU Connector 1.JPG (214.98 KiB) Viewed 25611 times

[Rednaxs60]

Went at the pinout comparison and requirements.

Requirements for v0.4.4 board:

Need common sensor ground for the TPS, MAP sensor (only if not using the on board map sensor), CLT, and IAT - OEM ECU pin 3.

Need common 5 VDC sensor voltage for TPS, Map sensors - OEM ECU pin 29. Speeduino pin 22.

Need spark igniter unit input/output (?) - OEM ECU pins 15 and 33. Speeduino pins 34 and 7 respectively. Need to determine GOING LOW or GOING HIGH.

Looking at theOEM schematic, looks like it should trigger the event by GOING LOW. Mention this because a system test for the spark igniter units is to test the three wires to ground and each should show battery voltage.

Need 12 VDC signal to Speeduino. From OEM ECU pin 1 or 19. TBD

Need TPS Input - OEM ECU pin 11. Speeduino pin 22

Need MAP input - OEM ECU pin numbers 27 and 9. Speeduino only requires one MAP sensor - pin 11

Need CLT input - OEM ECU pin 28. Speeduino pin 19.

Need IAT input - OEM ECU pin 10. Speeduino pin 20.

Need Crank Sensor (Ns) - OEM ECU pin 6. Speeduino pin 25 (VR1+). There are two VR pins - numbers 26/27. Are these used and go through the VR conditioner?

Need Cam Sensor (Gr/Gl) input - OEM ECU pins 7/8. Speeduino only uses one input. Speeduino pin 24 (VR2+).

Need Fuel Pump Relay ground - OEM ECU pin 32. Speeduino pin ???.

Numbers 1/3 Injectors - OEM ECU pin 17. Speeduino pin 1.

Numbers 2/4 injectors -OEM ECU pin 35. Speeduino pin 2.

Grounds

Grounds for sensors can be common and not need to go through the ECU. Could tie the TPS, MAP, CLT, IAT, Crank and Cam sensor grounds to a common non-ECU ground. This is predicated on how the ground point work in the Speeduino. The OEM ECU has specific ground pins for Cam, Crank, MAP, TPS, IAT, CLT, Fuel Pump relay, and FUEL SYSTEM dash indicator light. The FUEL SYSTEM dash indicator light ground not required at this time. The OEM ECU has 4 pins to the motorcycle ground system. Pins 2/16/20/34.

Questions to the collective:

There are 4 ground pins for the v0.4.4 board:

Ground pin 12 - is this a ground pin to the motorcycle ground system, or is it configured for a specific signal input? Same question for pins 9, 10, and 23.

2. Is there a specific ground wire for the Speeduino that is not shown on the board. I would relocate this to the 12 VDC signal needed for the Speeduino. Thinking this ground connection at the blue connector on the rear of the v0.4.4 board. These are the 12 VDC supply and found for the Speeduino. Looking for clarification. I can connect the Speeduino 12 VDC and Ground in at OEM ECU pin 1 or 19, and pins 2/16/20/34 respectively.

3. Is there a starter signal connection? The OEM ECU has pin 30 that indicates to the OEM ECU that an engine start has been initiated.

4. Do Speeduino pins 26/27 - VR1/VR2, relate to the VR conditioner board and how the jumpers are positioned?

The fuel pump relay requires an ECU ground. I expect that the fuel pump initial start and subsequent operation would have the ECU operate the fuel pump through the appropriate ground. Have read about this previously, and have been reviewing it lately. Did an initial programming change back when I first started this project.

Having mentioned the fuel pump grounding requirement and using board v0.4.4, making a pin allocation change as mentioned in this forum would require making pin 35 the fuel pump ground pin and the boost would be allocated to pin 16 - proto area.

If I understand PSIG post regarding this is that when the ignition key is turned to the ON position, Speeduino pin 35 would ground and the the fuel pump relay would be energized and the fuel pump active. The fuel pump would prime the system and time out. The fuel pump would restart immediately when the engine starts.

Have to get familiar with using utils.ino, and Arduino IDE. I know, read the manual, and various threads. It will work out.

Chime in with comments, have thick skin, and lots to do.

[Rednaxs60]

Got the OEM ECU connector off the faulty OEM ECU board. This will allow me to use the OEM ECU enclosure - hope I have enough room.

OEM ECU Connector rear.JPG (131.09 KiB) Viewed 25590 times

OEM ECU Connector.JPG (117.1 KiB) Viewed 25590 times

This will also allow me to revert if necessary.

[Rednaxs60]

Did some further reading regarding grounding the fuel pump relay. Printed out a thread regarding Thermo Fan and Fuel Pump Outputs/Control started 30 Jan 2017 - it's been in the binder for some three years.

I am using the v0.4.4 board. apollard made a post and it was corroborated by others about pin selection for the fuel pump. Won't be changing the rad fan operation.

Apparently apollard uses pin 35 for the fuel pump relay. In Tuner Studio he mentioned that a fuel pump pin change to pin 4 in Tuner Studio is needed. As long as this pin selection provides a ground for the fuel pump relay when the ignition key is turned on, and the fuel pump cycles and operates correctly, me likes.

Hoping this is the case, simple is very good.