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.
On the lighter side:

Memes 1.jpg
 
Ordered the COP units, the connecting pigtails and an in-line spark tester. Should be here Wednesday/Thursday. The in-line spark tester is so I can check the timing before starting after I figure how to install the COPs.

Probably putting the cart before the horse as I haven't done a lot of tuning; however, get this done up front and see where the wind takes me.
 
COP units and pigtails have been delivered. The in-line spark tester as well. Have the engine idling fairly well, engine stumble is being addressed.

Fuel is sequential so this mod will finalize the install.

Did some reading and socrace used some plug wire to go between the COP and OEM plug cap. Peel the plug wire insulation back, spread out the internal wire, and fit into the COP stainless steel spring, just like in the old days of distributor caps. COP units are marked for positive and negative connection. COP dwell will be 1.5 ms with a spark duration of 2.0 ms as per socrace findings.

Have to figure out how to connect the CFI covers if this works well.

Will check engine timing before staring it.

Only other mod to this install is a baro sensor. Have the components required on order.

Had the engine started for some testing/tuning of VE and spark. Increased RPMs in increments up to 4K. MAP reading is pretty linear going across the VE table, did drop to 50 kPa. Adjusted the VE table cells, operating a bit rich. Returned to idle nicely.
 
Coil-on plug units have arrived. I have been fitting 4 of these to determine best mod fit. These COP units could be used in a coil near plug install. Have had to cut off the bolt holder:
Coil On Plug - 1.JPG
Have taken these units apart and find these are not as imposing as thought, coil, spring and a rubber boot (the rubber boot insulates the COP SS spring:
Coil On Plug - 2.JPG
The coil has a spade connector and the spring touches the spade connector on two narrow sides of the spade connector:
Coil On Plug -4.JPG
The coils are marked for polarity. You should be able to see the B+ marking on the connector:
Coil On Plug -3.JPG
The spark goes through the SS spring and to the spark plug. The fitment of the spark plug to the SS spring is not that exact, but it works. I have modified the boots on three of the COPs. Have modded one such that I used a high temp/amp spade connector connected to the spark plug wire. This allows me to adjust the position of the coil to fit best in the space available:
Coil On Plug -5.JPG
The COP on the left has the boot shortened, and uses the SS spring. The COP on the right is shortened to suit a better placement. The length of the spark plug wire can be adjusted to suit. The spade connector is a high temp/amp connector. Found out a few years back there is a difference in connectors. The length of the spark plug wire should not be an issue. Don't expect the SS spring is a well designed, calibrated spring for the voltage flow.

Will be wiring for a trial, and if all goes well, will adjust the COPs for a goos install, much like the short one I show.

Another thought is that the 1000/1100 GWs may eventually need an ignition upgrade and the products available may be hard to find/get. Installing an aftermarket ignition system, something like the Speeduino unit I am installing and just do ignition is not a far fetched idea. This is being done already with some older Ford ignition units. I would think the major difference is that once installed you can tune the ignition to suit, but may need to install additional sensors to make it work. Just a thought.
 
Installed the COP units, made one of the COP units into a #1 cylinder timing COP. Needed to extend the plug wire to use the timing light on. Checked timing and no change. Operating in sequential mode as are the injectors. Forgot to take off the fixed timing, engine did not like. Started quite well.

Was reading the Speeduino Wiki, the bible for the Speeduino. It was recommended in the section on ignition to convert from older OEM style ignition design to a more modern COP, or CNP install. This is recommended to remove the separate ignition modules, and improve system efficiency.

Now for the VE and spark tables, clean up the wiring. Will not be able to finalize the COP install until I get the front fairing back on. Probably going to be a coil-near-plug install. Want to use the CFI covers as well.
 
Last edited:
Short day today. Constructed the baro circuit board. Had to remove the resistor I had previously installed:
Baro Sensor Circuit.JPG
Had to ask the oracle (internet) why capacitors are used between signal/power and ground - got my answer. First go at a circuit board since the '80s.

Did a short trial today. Reduced the required Fuel (RF) to 5.8 ms from 6.4. More to follow on this.

Went looking for GW EFI conversion threads on this and the other forums. A few from years ago, but not too many showing up nowadays. It is not an inexpensive conversion, even if you are doing an upgrade. Trying to use the older packaged EFI system components with no spec data is not easy. Don't have to manufacture holders and such if starting with an existing EFI system such as mine. Interesting. Lots of EFI conversions being done on cars and such, don't know how many come to fruition.

Liking the sequential ignition. Have thought that staggering these instead of side by each may be a good way to install.
 
Been a while since the last post. Have had the engine operating with sequential fuel and ignition. Installed 6 ohm resistors, one per injector. Had comments on the size of the resistors. Did some numbers and the injector circuit voltages are less than with the OEM setup using one three ohm resistor for two injectors.

Have the front fairing back on and all is good however, have lost the dash tach signal. The dash tach signal is from the travel computer and the tach signal is from injectors 1/3. Went to change back to paired injectors using the 6 ohm resistor, no joy, engine would not start. Change back to sequential fuel, engine starts. Checked timing, all good.

Low impedance injectors require a good amperage jolt to open, research indicates that this should be ~4 amps. The OEM resistor pack with a 3 ohm resistor in-line with two injectors reduces the amperage to ~4 amps and both injectors will fire. Having a 6 ohm resistor replacing the 3 ohm resistor in a paired injector scenario, reduces the injector circuit amperage such that neither injector will work. To test this theory, connect the OEM resistor pack, changed to paired injectors, engine started without issue.

Still don't have a dash RPM indicator. Will be investigating this further.

Having successfully solved this issue, will be replacing the 6 ohm resistors with 3 ohm resistor per injector. This picture shows how this should work:
Injector Wiring Upgrade - 1.JPG
The injector circuit can handle up to ~8 amps. May swap out the 3 ohm for a 2 ohm resistor. Raises circuit amperage to the ECU to 3.5 amps instead of 2.8 amps. I use 14 VDC for my calculations because of the external alt mod. Alternator puts out a solid 14.3 VDC at idle.

This small change, a 6 ohm in place of a 3 ohm resistor, illustrates how changes can affect a system operation. Applicable to all components/systems.

I'm getting better at the electrical/electronic world, took courses back in the late '70s and early '80s but never really used the training.

Lesson learned, careful what you do and make sure it's correct.

Clean and the same voltage for the ECU, injectors and coils (in my case COP units) is required. Using the OEM wiring has differing voltages at these components, happens after some 40 years. With the external alt mod implemented, I changed the electrical distributionnsystem such that there is a power junction. The alternator is connected to it and all electrical feeds are from this. Installed an accessory fuse box early on. Checked the voltage at the power junction - 14.3 VDC, at the accessory fuse box - 14.1 VDC. Have connected the new ECU, COP units and injector circuits to the accessory fuse box. The ECU monitors electrical system voltage and uses the electrical system voltage to adjust the operation of the COP units and injectors. The lower the electrical system voltage requires the injector opening time to be longer to get the same performance at normal operating electrical system voltage, conversely, the higher the electrical system voltage the less time is required to get the same performance response. This applies equally to the coil dwell time. Should be good for the CFI system operation.

Have to secure the COP units. This will take some thought and experimentation. Hopefully have the bike out of the barn doing some driveway pulls, then on to the road next month.

Looking at using an android tablet for road trials. Need to load it up with Tuner Studio and MegaLogViewer to take data logs, and hopefully do VE live tuning. Not quite certain how to transfer data and files between an android tablet and a MacBook Air/PC laptop.

Cheers
 
Have a short update. Have the COP units fitted, not an easy task with the limited space available. The wiring is just about done, mostly routing and securing. Very busy wiring where the OEM RR was located. Changed where the fuel pressure regulator gets its vacuum from. Had a separate connection specifically for it, it is now connected to the vacuum hose to the ECU from the vacuum mixing chamber. Staying with sequential fuel and ignition, engine starts before fuel system prime is finished, major milestone in my books.

Corresponding with a fellow in the UK who is reverse engineering the OEM ECU. He does this for a living, but takes on little projects such as this. He has a couple of used ECUs, one is U/S - ground fault I believe and has been exposed to a 12 VDC surge or such. His second one is supposed to be good, but he could not confirm. I sent him one of mine to play with - it does work. Will get a lot of information from him on the inner workings. Found out that Honda used 16 by 16 map matrixes for this ECU, four of them. Know that one is for Ne-Pb map, SD and Alpha-n normal fueling profile - SD at idle and lower powers (uses the PB sensors for engine load), Alpha-n at higher powers above ~2800 RPM (uses the TPS as the engine load indicator). Second map is for Ne-0th, Alpha-n fuelling itself, this is used when one or both of the PB sensors are faulty - engine won't operate the best at low powers with this fueling map. One table has to be for ignition/spark timing. I suspect the fourth table will be for SD fuelling alone.

The camshaft sensors have always been a bit of a mystery, called pulse generators (PGs) like the ones on the carb models. These cam sensors do not do the same thing. The PG seniors on the carb models is for ignition timing. The camshaft sensors on the '85/'86 GW FI Models is for fuelling and determining a cylinder TDC location. Gr (right side camshaft sensor) controls the fuel event for cylinders 1/3, Gl (left side camshaft sensor) controls the fuel event for cylinders 2/4.

When the ECU receives a TDC signal from a camshaft sensor, 4 pulses from the crankshaft sensor (Ns) after the camshaft sensor signal (180 degree rotation of the crankshaft - 90 degree rotation of the camshaft(s)) and the ECU fires the fuel injectors for the corresponding cylinders as mentioned above. Ignition follows fuel injection as appropriate. The engine will operate well with just one camshaft sensor being faulty, both faulty engine fuelling and ignition events are stopped - engine will not operate.

This has been mentioned before and in various different posts, but I now have a very good "handle" on this aspect.

From the above, thinking I will look into using a blended SD - Alpha-n fuelling profile for my GW project. Will try to find out the RPM at which time the MAP sensor is at maximum, then switch to an Alpha-n fuelling profile. This new ECU has individual cylinder fuel trim (ICFT) capability that allows for fine tuning of the fuel event for each cylinder to balance the cylinders. This is percentage based fuel addition/subtraction. Have more to read on this. Individual cylinder ignition trim (ICIT) is not available, may in the future, but I don't believe it is as useful as ICFT.

This project has become more than just a component replacement project, it's a modernization/upgrade project. Didn't see this coming.

Just some additional musings. Cheers
 
Short update. I mentioned that the mig/stick/lift tig welder came in. Great for the small projects I have. Bit rusty but welds are getting better. Grind the welds flat and paint black, can hardly notice. Bought an auto darkening helmet as well. New tech is great. Have all the small brackets finished and installed and have the new components such as the coil-on-plug (COP) (in reality coil-near-plug (CNP)) arranged so as not to impact on the plastics and other covers.

Bought a Panasonic Toughbook CF-20 - older model but operates on Windows 10. Finished installing MegaLogViewer and TS, and transferred the tuning files over. Thought of an android tablet, but transferring files and using a third application isn't in the cards as of yet. Will probably fashion a holder for an android tablet, or some other device so that continuous monitoring can be done, RealDash, or similar.

Will be installing a temporary holder for the Toughbook in place of the Travel Computer for initial road tests. Will be easy to swap back and forth as I go.

Have one wiring issue with #3 cylinder coil, no spark - swapped spark plug and use a different COP unit to confirm. Have checked the wiring for all four injector circuits and the other three coils and all is well continuity wise. No wiring continuity between the ECU and the coil driver for #3 cylinder. Will connect a temporary wire to confirm. The four fuel and ignition lights on the interface board flash like they should.

Been reading up on individual cylinder fuel trim (ICFT). Interesting feature, seems you can adjust the fuel percentage at the lower end to even out the cylinders so each is doing the same work and operating as close as possible. Compensates for individual fuel injector being slightly different.

Now that I am close to moving the bike out of the "barn", thinking about initial trials and what to bring back on line, when and in what order.
 
Thanks for the likes. Still a novice with this new ECU, the EFI world and the physical change.

This is like going back to college/university after a prolonged period for the course "1985 Honda ECU Replacement"; however, the different classes that you must take at the same time to get to the certificate are varied and required.
 
Thought I’d do an update before the New year.

Lots has been happening. Have the bike ready for road trials, but have to do some maintenance on the front brakes, seem to be dragging – next week before end year. Want to do a trip around the block before the New Year.

Have had to do some mods to make components fit such as the COP units, vacuum mixing block and other issues, but have now resolved almost everything. It’s a little busy in the rear trunk area with vacuum hoses and wiring, clean up required.

Have been corresponding with a fellow in the UK regarding the OEM CFI ECU. He has managed to reverse engineer the programming code, and has posted it on GitHub. The programming software is not available, but this gives more insight into what Honda did.

I have mentioned that the CFI system uses a Speed Density (SD) fueling profile using the PB sensors as the engine load, and Alpha-n fueling profile using the TPS as the engine load indicator. There are four 16 by 16 VE fueling maps used in the CFI system. Two for the SD fueling and two for the Alpha-n fueling profile.

If you read the CFI system error code document, you may notice that when one or both PB sensors are failing/faulty, the ECU transitions to the Alpha-n fueling profile(s). When the TPS is faulty/failing, the ECU transitions to the SD fueling profile(s).

The CFI system uses the SD fueling profile(s) for idle and low engine powers up to about 2800 RPM where the PB sensors are maxed out. The exact transition point is not known.

This is a logical expectation in that when the PB sensor(s) are at max output to the ECU, the SD fueling profile engine load requirement will not continue to change and fueling can be affected. Changing to an Alpha-n fueling profile for higher engine loads using the TPS as the engine load indicator will allow for the fueling profile to continue to change for optimum engine operation up to full power. I would expect the TPS signal to the ECU at the transition point will only be about say 30% of the full TPS signal to the ECU. Lots of TPS signal room left for fueling.

With this in mind, I will be doing a “blended” SD/Alpha-n fueling profile for the GW, or using the second fueling table to accommodate.

EDIT: The Gr/Gl camshaft sensors have been a bit of a mystery to me, but have finally sorted these out. It has been mentioned in other forum posts, be it here or other, that these are used for fuel injection timing. This is what happens; however, the Gr camshaft sensor controls fuel injection for numbers 1/3 cylinders and Gl camshaft sensor does fuel injection timing for cylinders 2/4. These sensors provide the ECU with a cylinder TDC indication.

Once a signal is sent to the ECU from one of these sensors, and the ECU receives four pulses from the Ns crankshaft sensor after the camshaft signal, the ECU fires the appropriate fuel injectors. Ignition timing follows fuel injection as appropriate.

The fuel injectors are fired once every engine cycle of 720 degrees - 2 revolutions.

This scenario makes sense in that the firing order is 1/3/2/4. Fuel injection is 1/3 and 2/4. Ignition is 1/2 and 3/4.

I have always meant to conquer a CAD type application for wiring and schematic changes. This project is making this a must have requirement. Too many wiring and component changes to keep track of. Have started to look at and use KiCad. This is a freeware, open-source application. It is generally used by the PCB crowd, but can be used solely for schematic/wiring diagrams. Too many hoses and wiring changes in the past 8 years. Looking for wiring that is multi-colour (striped) and will start a major rewiring project next winter.

It's been a great learning experience and I’m glad I have done this.
 
Last edited:
Had some time today to make some progress. Front brakes have been serviced, front wheel turns freely. Will still need a good maintenance period before the summer.

Good news, bike is out of the garage:
GW out of garage .JPG

GW out of garage .JPG (77.52 KiB) Not viewed yet
Will have to wait until after Boxing Day to insure the bike. Did some small runs up the driveway in gear. Gives me some time to get the tablet mounted and maybe look at the mess under the seat.

Contemplating the tach issue. Think that there is only one injector pulse to the travel computer for the dash tach. Something will come to light regarding this.

Have some rhetorical question(s) regarding first hitting the road. First issue is to have TS data logging. Second is regarding VEA/VEAL. Thinking I can start VEA and just log results, do not have to have it adjust the ECU tables at the start.
 
Thanks for the "(y)". Staring the final document regarding this project, lessons learned. The key issue with this project was not being as patient at the start as I should have been. Thought it would be simple and pushed ahead at the "high port". Not a good idea. Slow and steady wins this race.
 
Not really patience, but definitely not letting go.

Road trials will be starting soon. Have the tablet platform installed and trialed:
Tablet Install.JPG

Tablet Install.JPG (77.37 KiB) Viewed 1 time
New tablet battery on order, used tablet to be expected. Have a car charger coming as well. Think the battery will be good for a go round the block.

Took the spare wiring harness apart for the different coloured wire. There are a @#$% load of splices and all in good condition. Realize there are a lot, but was not ready for the number that were glommed together, five wires into one - especially the grounds.
 
Good news and a milestone. Went for a 7-8 Km ride today. No sync losses, but the O2 sensor is indicating a rich mixture. Had it up to about 80 KPH. The engine operating temp went to exactly where I thought it would. Four bars on the temp dash temp gauge when on the road. Noticed the rear brake was a bit soft (non-existent) so some maintenance required. More to follow.
 

Latest posts

Top