1985 GL1200 Limited ECU Replacement/Upgrade - Part 2

Classic Goldwings

Help Support Classic Goldwings:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Rednaxs60

Well-known member
Supporting Member
Joined
Dec 25, 2015
Messages
3,067
Reaction score
623
Location
Victoria, BC,
Posted on the original thread. Going with a fresh thread.

Part 1 so t speak is: 1985 GL1200 Limited ECU Replacement/Upgrade

The break from this project, the work I've done on the Limited Edition, and my understanding the CFI system since I started looking into a replacement aftermarket ECU has been beneficial.

This project is an OEM ECU replacement project, not an FI conversion. This in itself should make this project much easier to achieve.

I have started watching aftermarket ECU videos again and there are a lot of aftermarket ECUs that are a direct replacement for the OEM ECU be it auto or motorcycle. This appears to be the way forward today, and instead of replacing chips and such that was the craze - may still be, from a few years back, having a tunable aftermarket ECU installed that is non-OEM is probably preferable. Don't get into the proprietary issues. Some are splicing into the OEM wiring harness, keeping the existing OEM connectors as a fall back system. A lot has happened since I first started to look into this, or maybe I just wasn't looking at this side of the issue.

I briefly contemplated switching to a Mega/Micro Squirt, but cost is quite steep and I have the Speeduino components on hand. Get the pinout correct, populate the program with the appropriate numbers, and ensure the existing CFI components are compatible is the way forward. Need to determine if sequential firing is better than batch firing, either works, and can switch in the future. The Speeduino is more flexible In this regard. Timing correct and all would be well.

So as to not make the readership go through Part 1 to get some salient points will repost here.

Reviewing notes, forum threads, Speeduino information. Have the bones for the replacement ECU, Speeduino board and such.

Time has a good way of clearing/focusing a person's thoughts. I realize from my renewed reading, that I am just replacing an older ECU with a newer one. This does, or should not, require any wiring or component changes. There will be some reduction in the components used such as the cam sensors - have two, only one required. A new trigger wheel, thinking 16/24 tooth, no need for the missing tooth aspect because of the cam sensor.

I do lose a bit of functionality. The OEM ECU diagnostic program, as well as the dash indicator light.

The cam sensors, Gr/Gl, were Honda's answer to sequential injection and faster starts. My reading indicates that the benefit of sequential injection instead of batch firing is not that much. The Speeduino can use either.

The Speeduino does not need to use a cam sensor, but that would require a missing tooth trigger wheel.

I have updated my pin-out spreadsheet. Have the 95% solution regarding this.

Need to get the components out, get Tuner Studio and such back on line, and read the Speeduino info again. Have not been on the Speeduino forum for some time.
 
Socrace on the NG forum commented on the dual components that Honda uses in the '85/'86 FI models. Two MAP sensors - PBR/PBL, two cam sensors - Gr/Gl. These were installed for redundancy, and in the case of the cam sensors, to simulate sequential injector firing and quicker starts.

I have not found an aftermarket ECU that uses 2 sensors for the same job. Do like the redundancy aspect though.

For the PBR/PBL sensors the supplement indicates that if one sensor is faulty, the ECU uses the signal from the other sensor to operate the engine in a normal fashion, same for the Gr/Gl sensors.

I have the Suzuki inlet air pressure (IAP) installed on my '85. I have posted before that the signals from these sensors is virtually identical to the Honda OEM sensors. Will be doing a voltage check, and then will marry the cylinder bank vacuum hoses to an individual sensor. Do this for both.

Will do the same for the Gr/Gl sensors. Honda has detailed this in the Supplement, should work.

Will not remove any sensors, but keep the installation of these sensors as a backup should the one being used fail.

All being IAW the Supplement, one more step in the right direction.
 
Finished Suzuki PBR/PBL (MAP) testing.

Initial readings taken to determine the difference when first installed are:
1666825249885.png


I also checked the vacuum at the PB sensors and at approximately 3000 RPM the vacuum to the sensors is 10 to 12 inches of HG and remains steady at these vacuum values as RPM increases. At 12" of Hg - Suzuki sensor - 2.46 VDC.

Tested today regarding the redundancy of these sensors. The engine started and operated as advertised in the Supplement. Disconnected each before start, and could not tell the difference. Pass

Disconnected sensors with the engine started. The ECU takes a few seconds to assimilate the change, then the engine operated well. Pass

The FUEL SYSTEM dash indicator light comes on. Pass

Checked the Suzuki sensors for voltage IAW with the Supplement. Read in four columns:
1666824966806.png

From the initial readings above, would not expect the OEM sensors to be significantly different from the Suzuki sensor readings in this lower table. Should be good to go.

Checked the use of one MAP sensor with the two cylinder banks vacuum hoses married. ECU did not like this.

Edit: The Speeduino on board MAP sensor is 2.5 bar. Small little item P/N MPX- 4250AP. Looked at more on line information regarding vacuum. This was discussed on the NG Forum. The 1200 has two MAP sensors, each taking vacuum from the respective cylinder bank.

One fellow married the 4 vacuum points into a collector box, then had one vacuum line to the MAP sensor. Theorized that the resulting vacuum would be an average from all four vacuum connections. Use equal length hose to join to a collector box. The hose that would be used to get to the Speeduino MAP sensor would be fairly long because I intend to mount the Speeduino in the same place as the OEM ECU - don't know if this matters.

Have to consider, but there is limited room to install more components on the '85 Limited edition.
 
Last edited:
The O2 sensor requirement is going to be interesting. The wide band O2 sensor is preferred. The O2 sensor would be connected as an analogue devise. More info to assimilate regarding the O2 sensor and how it is connected to the Speeduino. Understand it is primarily used for start and idle ECU adjustments.

Read on line over on the NG forum (have perused this forum for information as well) that two sensors may be needed. Will contact 14point7 in Ontario to discuss this. Mention 14point7 because this company is in Canada.

Installing a boss for O2 sensor install can be done a couple of ways, welded of course, or use JB Weld.

Did a JB Weld test a couple of weeks ago. Secured a nut to a steel bracket using JB weld. Let it cure then applied 30 ft-lbs of torque - could have used more. The JB Weld held. I believe the JB Weld can take the exhaust temp. This is a very good option for myself. There is no stress and very little if any vibration where the O2 sensor would be installed. Need to determine where the O2 sensor(s) would be installed as these units are proud by about 2 1/2 inches. Thinking an angle boss connector, if available, would be good.
 
Further reading, don't read this much in a year. The Speeduino can use the existing low impedance injectors because of the resistor pack. 3 ohm resistors to give a 4 amp circuit draw, and keeping the wasted spark arrangement. Thinking that this equates to approximately 2 amp per injector.

The Speeduino as with most other aftermarket ECUs now use high impedance injectors and the ECU drivers work well with these. High impedance injectors are approximately 8 to 14 ohms. With a 10 amp injector, the circuit amperage would be approximately 1.2 amps.

The OEM low impedance injectors are Denso 195500-1070 and flow approximately 280 cc/min each.

To use these injectors in a sequential injector application, would have to modify the injector wiring circuit and add individual resistors for each injector. The other way forward is to use high impedance injectors with similar flow rates, and dimensions. Bypassing the resistor pack would be the only wiring mod to be done.

The Speeduino can trigger the injectors individually, just have to ensure that the firing sequence is correct as it fires the individual channels used in a 1/2/3/4 order - channels . Firing order for the 1200 is 1/3/2/4.

Looked at the Speeduino pinout. there are 4 ignition circuits.

Would be curious to determine if high impedance injectors would work with the OEM ECU. Theoretically you would bypass the resistor pack and provide 12 VDC directly to the injectors.

Tony - Ansimp - mentioned to upgrade to newer, more modern technologies. Hmmmmm?
 
A lot of what I will post in the short term is a review of what I have in my initial post on this subject. Keep the faith, will get into the more interesting aspects of this project.

The Speeduino is a very capable software/hardware package. It rivals the Mega/Micro Squirt units. I know about these, have the manuals on both, but there are other units out there that you can research.

The components that are used in a fuel injected system are common to both, but there are idiosyncrasies that differentiate how a component is used and calibrated.

I've wrapped my head around each component, and how each component will interact as a whole. This is not to imply that I will stop reading the manual and forum(s) information, or viewing the videos that are out there. Pick up something new every time I revisit an article, video.

The Speeduino can use the existing passive idle air control circuit.

A decision has to made whether to continue using the existing thermo fan control. Can use the existing thermo valve or have the ECU control the on/off function of the rad fan based on a set of temps.

There is a flood clear level that is used to clear the engine cylinders when excess fuel has been injected into the cylinders. This requires two parameters to be met. The indicated RPM must be below the Cranking RPM setting, and the TPS signal must be above the threshold of this setting - I take this to be greater than the full throttle ADC count. ADC is Analogue to Digital Converter number. This component converts an analogue/sine wave signal into a digital/square wave that can be read and processed by a microcontroller. These aftermarket ECUs and the electronic/digital world has a language all its own.

There is a Priming Pulse Width function that upon power up , the Speeduino fires all injectors simultaneously for this fixed period of time. This is intended to clear any air out of the fuel lines that may have entered during the shut down period, it is not intended as a staring fuel load.

The fuel pump operation will be as per OEM operation.

Enough for now, need to give the brain box a rest.
 
I'm answering all my questions as I dig deeper into this replacement/upgrade. Question I had was how many coils and spark igniters do I need with sequential fuel injection. The answer I have been able to find is 4 coils and 4 spark igniters. Can get away with two coils, but the charge time between firings is considerably less and this can/could affect the firing of the cylinder. COP is at the top of the list for doing sequential fuel injection because of this.

I have read that sequential fuel injection is beneficial at lower powers, idle and environmentally, but the lines get blurred as the engine operation and motorcycle speed is increased significantly.

The spark igniter are a different kettle of fish because these only change the grounding of the coils and as such just do double duty. Would have to look at the wiring schematic to come up with a game plan. I have a set of spare igniter units that can be installed. Real estate is getting very scarce.

May have a source for a 36-1 trigger wheel sourced that can be installed where the existing trigger wheel is located. May have to modify the sensor position to allow for space between the sensor and trigger wheel, or file down the teeth a tad - the wheel I have seen is 76.6 mm instead of 75.5 mm. This wheel that I have found has a smaller centre hole that would have to be enlarged. Would be able to install the Speeduino without using a cam sensor. Thinking about this. Have seen some trigger wheels made with a grinding wheel. Also an option.

More reading and research regarding Gold Wing EFI conversions.
 
Have attached photos comparing the Speeduino against an early model ECU by Honda for its 1980s FI motorcycles. Quite a substantial difference. The only issue with this ECU is there is a ground circuit fault.
1985 versus 2019 ECU.JPGCross Section of ECUs.JPG

Looking at it there is a black box/pot with an atmospheric connection. Don't know if this would be for parametric pressure or was intended to be an on board MAP sensor. The enclosure has a port for attaching tubing to this sensor. It is a Sensor P1 P/N 949940-0481.

The connector ribbon between the two boards is a paper ribbon.

Every component is labeled without exception.

The external connector has every pin labelled for what it is used for, VBAT, GND, GR, GL, and so on.

The diagnostic lights are part of this circuit board:
Error Code Lights.JPG

A fellow on the CX500 forum has gotten into the maps, but that is way above my pay grade.

Thought this would be of interest.
 
Decided to load the firmware on the Arduino and connect to the Speeduino board v0.4.4. Downloaded Speedy Loader and installed firmware 202207.1. I believe I now have a Speeduino ECU that needs to be configured for the '85 Gold Wing system. Have Tuner Studio so next step is to look at the various calibration issues and populate the initial tune for my motorcycle. Need to get an O2 sensor - actively looking.
 
Dusterdude - There are some choice words to describe your ailment!:cool:

Short update on the project. First rule when doing a project - READ everything you can, assimilate as much as possible, then repeat. This has been my focus the last while. Information that was foreign to me when I first started this a few years back now makes sense.

I have identified the pinout, decided on the enclosure I will use at the start. The circuit board I am using Speeduino v0.4.4 is intended as to be integrated into existing OEM wiring with the aim that the interface can be easily connected and achieved through the on board IDC40 connector. This is exactly what I wanted as I am not doing an EFI conversion, but a replacement ECU using as much if not all the OEM CFI system. This is to me a key element in the success of this project.

I have to reassign the fuel pump pin, but as it stands now, that is one of the few changes to be made.

Mulling over whether to do a trigger wheel change and only use a crank trigger wheel. It would be a 36-1 tooth. This would allow me to use a missing tooth configuration at the onset, and have the cam sensors in reserve.

A 12 or 24 tooth wheel to replace the OEM 8 tooth trigger wheel would be best, more data points for the new ECU to work with.

I believe I mentioned using the 1200 carcass engine to investigate the timing issue. Will be progressing engine reassembly this week - won't take long.

Have to wrap my head around the barometric issue. Know what it is for, but how to achieve and plumb it into the system is the issue.

Going to instrument an injector and use my small oscilloscope to determine if I can identify the initial dead time - time required to open the injector before fuel flows. Have read about this and viewed a lot of videos, and it comes in around 3 ms or so. Reading about this and the conversions that have been done on Gold Wings for additional information.

Had feedback on the Speeduino forum regarding my thought on the Gr/Gl cam sensor being a hall-effect sensor with an analogue output. Apparently I was spot on. I looked at a hall-sensor with a digital output and these generally are a three wire sensor. the sensors on the Gold Wing are 2 wire.

Have to consolidate all the info, pictures and such that I have collected and bring everything into the same folder. Difficult trying to find something when it is scattered all over the place.

Feeling better about the project and the probability of success.
 
I finally got the ignition timing and fuel injection for the OEM CFI system figured out. This was mentioned a while back, and It took a few years to digest.

Info on the 8 tooth crank trigger wheel, and 2 cam position sensor with a single tooth cam trigger wheel. From Clymers: This 8 tooth trigger wheel sends out 8 pulses per revolution. The ECU then triggers the fuel injector at the 4th pulse after a TDC pulse is received from the camshaft angle sensors. This information is used to control the ignition timing and fuel injection volume accurately. Good explanation of the correlation between the crank and cam sensors, and is a feel good indication that the 8/1 trigger wheel setup will work with the Megasquirt unit. The Megasquirt unit apparently only requires one cam shaft trigger, not the dual OEM setup.

Just read this passage again after the hiatus, and the drawing I made now makes sense. This information indicates that the standard timing advance is 15 degrees BTDC. Everything happens in 15 degree increments. This drawing demonstrates this:
Timing Marks.jpg
 
Update - I will be installing a 14point7 WBO2 sensor - this is the only new component in the system. I bastardized a faulty ECU for the enclosure and enclosure connector. This way I can utilize the existing wiring harness connector and if necessary, revert back to the OEM ECU. The Speeduino will be installed in this enclosure. I have the Speeduino assembled and firmware installed. Will be using a dual wheel, crank and cam sensor. The '85 CFI system has two cam sensors but only one required. It will be an 8/1 dual trigger wheel install - this is what is fitted. Eight tooth crank (Ns) primary trigger wheel, single tooth cam shaft (Gr/Gl) wheel. The crank and cam sensors output an analogue signal that will be converted to a digital signal through a VR conditioner on the Speeduino. The cam sensors are Hall effect but designed to output an analogue signal.

I mention in a previous post that I would consider a 36-1 tooth crank trigger wheel. This would allow me to not use a cam shaft sensor. Remember the 1500 crank trigger wheel being a 12-1 trigger wheel that would fit nicely where the 8 tooth trigger wheel is on the 1200 engine. This is an option.

Timing is expected to have a trigger angle (TA) of approximately 180 degrees +/- depending on the firing. This reminds me of the old Dodge/Chrysler distributors that had the distributor drive fit into a slot. If the engine did not fire, take it out and turn it 180 degrees. Simple.

The injectors are low impedance - low-Z, and can be used because of the OEM resistor unit. This resistor unit has a 12 VDC input, through 2 - 3 Ohm resistors to the injectors - one 3 Ohm resistor for two injectors. The 3 Ohm resistors spec out at 3.5 Ohms, the injectors at 2.8 Ohms. Can change to high impedance - high-Z injectors but no need to at this time.

The CLT and IAT sensors can be used. these would be custom configuration that requires three resistance points be found. These should be at 100 deg C, ambient room temp and -10 degree C. Think close will do well. The test procedure for these sensors is the same.

The Speeduino can control the rad fan if so desired; however, it can be left as per OEM design.

The fuel pump requires the ECU to ground the fuel pump relay. Speeduino has a couple of spare grounding points that can be reallocated in Tuner Studio for the fuel pump control.

Ignition - wasted spark, batch, semi-sequential, and sequential are all supported. Will be using wasted spark initially. if the overall performance is good, won't be changing.

The GL1200 4-wire spark igniters can be used.

The ECU diagnostic, dash CFI fault indicator light will not be available, but I'm thinking a piggyback unit to do this function is possible. The travel computer should not be affected.

Have to determine injector timing characteristics.

About all for now. Cheers
 
Just off the CX500 forum. Joined that forum because of the early model Honda turbo motorcycles, the forerunner CFI to mine. One of its eminent members - Pim, is looking into the Speeduino. He mentions that he has tried to repair a few turbo ECUs lately, and not having schematics or design info, no joy. The Speeduino has a lot of traction in the auto world especially for older vehicles. It's not as popular in the motorcycle world, but is being used successfully for FI projects.
 
Jumped into the deep end last evening. Connected the Speeduino to Tuner Studio, input all my data, and hit the yes please button - didn't work. Something is not quite right, obviously. Will try again today. Think I did the GIGO scenario.
 
Update required. Progressing well. Reading a lot of information, and viewing videos to aid in this project.

Installed a new fuel pressure regulator. Working well. Static and dynamic fuel pressure at 37 PSI. Does leak down but not to zero PSI. may be through the fuel pump and shut off valve.

Have put the Speeduino together, programmed it through Tuner Studio.

Bought a WBO2 sensor from 14point7. A Spartan 3/Lite With LSU ADV sensor. Should be here soon, then time to look at location for install. Not much "real estate" available.

Have an engine simulator coming from the UK. Designed to interface with the Speeduino board I have:
Engine Simulator 2.jpg
Have been instrumenting the injector(s) to determine injector settings. Using a small DSO scope. Finally got a good injector trace, just upside down:
Injector Wave Form 2.JPG
PSIG - a fellow on the Speeduino forum inverted the pic and annotated it:
Injector Trace.jpeg
This is the injector voltage test rig:
Injector test Rig.JPG

Going to instrument for current flow. Will be using an ACS721 current sensor module:
ACS 712 Current Sensor Module.jpeg
Use this a shunt then measure the voltage going through. Ordered a 5/20/30 amp module. Costs about the same as two beers at the pub.

Will be modifying the vacuum system. 4 equal length hose to a distribution box, one hose to the Speedduino and use the MPX4250AP MAP sensor mounted on the board.

Spark igniters will verified for how the ECU fires these. The two settings are GOING LOW - 12 VDC to 0 VDC or GOING HIGH - 0 VDC to 12 VDC. Honda is notorious for going against the grain. Most spark igniter units use GOING HIGH, Honda apparently uses GOING LOW - to be verified. This terminology for how the spark igniter units are used is specific to the Speeduino project. Other aftermarket ECU manufacturers may use different terminology.

More to follow.
 
Had bought a DSO138 clone scope a while back:
DSO138.JPG
Thought I had connected it wrong and fried the little unit. Decided to dust it off and check it again and it's alive. Connected to the injector I have wired for testing and it worked. Might have to get a case, have two scopes to use now.

Have been viewing videos of the Speeduino and notice that this little scope is used a lot in testing circuits and such.
 
Quick update.

Have the Mega programmed. Was having issue(s) with this so I removed all TS files regarding the tune. Tried loading the initial firmware manually, loading Arduino IDE and such - no joy. Went back to SpeedyLoader, and did an install with the Speeeduino interface board, v0.4.4, disconnected. Install of firmware went well. Then on to TS. Downloaded the associated software through TS. Burned everything to the MEGA. The issue I was having with the MEGA recognizing the O2 sensor from 14point7 not installing is fixed. Note to self - when using the MEGA with TS, disconnect interface board.

Waiting on the O2 sensor, and engine simulator. Don't need the O2 sensor to connect new ECU to the bike, but have other fish to fry so patience is the name of the game.

Have to do a vacuum hose mod. The engine has 4 vacuum ports that have to be considered. Looking on line at vacuum blocks. Least expensive purpose made is approximately $25.00 CDN with fittings and such. Doing a DIY version would cost this easily with all the searching and running around. Wondering if the length of vacuum hose is an issue. The engine vacuum hoses will be approximately 2 feet long with another say 4 feet to the rear of the bike where I intend to install the new ECU - under the rear trunk where the OEM ECU is installed. The other option is to use one of the OEM MAP sensors and connect to the Speeduino through the wiring harness. Will compare the MPX4250AP against the OEM MAP sensor.

When the new scope comes in will instrument the sark igniter units and determine, if I can, whether these are GOING LOW or GOING HIGH. My bet is GOING LOW.

Working on the Speeduino enclosure. Bluetooth HC-10 came in.

Mulling over the barometric issue - it has been mentioned that the second OEM MAP sensor may be used for this. Connecting to the Speeduino is the issue and what specs are needed.

In my readings and looking at videos regarding the Speeduino, found information on how to understand the medium current (MC) and high current (HC) connections. I was thinking this meant a positive voltage signal, but it does not. It is a grounding point for high current applications. For example, so I have my story straight, you have a circuit that operates at 10 amps. This current would fry the a common ground point. The speeduino has several ground connections specifically designed for a high current such as this, be it the MC and/or HC connections. I don't know the specs for these MC and HC grounding points, but suffice it to say that understanding terminology used is everything.

Cheers
 
Top