1985 GL1200 Limited ECU Replacement/Upgrade - Part 3 - Road Trials

Classic Goldwings

Help Support Classic Goldwings:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
My intent is to now put together a schematic for a basic install that will work and replace the OEM ECU. I had this at the start so recreating it will not be an issue. May go back to this configuration as well. Can always upgrade later on to sequential fuel/ignition, but first to change to paired injectors and determine if I have a dash RPM indication or not. The Speeduino wiring was much less at the start as well. I will keep the wiring for the new barometric sensor. I mention this because one of my requirements at the start of this project is to be able to go back to the OEM CFI system. May have to get another wiring harness, this one has been modified a couple of times.

Going back to basics may tweak some interest in the GW community. The last piece of the puzzle will be getting a good base engine tune done from road trials, and detail how to set up the tuning software settings/parameters for others to follow. This is the challenge at the moment.

Will be heading back to Ontario for the month of July to my sister's for the last work period on the house and on return start the change/tuning process. Be off line for a while, but will monitor and comment if necessary.

Cheers
 
Back from my sister's place in Ontario. Lots of work done. My brother, Sonya and I have some 700 or so hours into the 30 year makeover of her husband's house - inside and outside. Looks good what we have done.

Getting back into the ECU replacement project. Will be reverting back to paired injectors as I have mentioned. Need to get the dash RPM indication functioning. Will keep the sequential ignition for the time being. Have to determine why the ECU cannot control the operation of the fuel pump when the key is turned to the ON position, I have a direct 12 VDC connection bypassing the ECU.

To revert back to the paired injectors have to change back to the OEM resistor pack.

More to follow.
 
Reverted back to paired injectors, also used semi-sequential as well. No joy, engine would not start. Kept at it for a couple of days, just went back to sequential injectors and engine started right up. Not going to beat the head against the wall, stay with sequential and go forward. have to visit the VE and spark tables, think I threw these out of whack but easy enough to make right.

Will be chasing the tach issue as well.

Cheers
 
Two thoughts for this post. Wide Band O2 (WBO2) sensor and cam sensors.

First, is a WBO2 needed. Not necessarily. It is a good diagnostic tool and provides information on how good your engine tune is. You put together a reasonable timing/spark/ignition table to start - lets call this the timing table. The same with the VE (fuelling) table.

The VE table is the first table to be adjusted to suit so that you can get the engine fuelling requirement where it should be. This table is paramount to getting the correct air-fuel ratio. The VE table cell values are just a number that is used in the calculated fuel requirement. The higher the number the more fuel is to be injected - rich, the smaller cell value number, less fuel is injected.

The timing table is next to be calibrated. When you increase a cell value - advancing the timing, you are in essence leaning out the air-fuel mixture because the fuel has a longer time to "burn". Reducing the cell value - retarding the timing, essentially reduces the fuel "burn" time and causes a fuel rich air-fuel mixture.

There is a sweet spot where you want the maximum fuel burn "pulse" to meet with the engine pistons for optimum engine performance and efficiency. This is approximately 10 to 15 degrees after TDC. You want the full injected and ignition to be at a point before TDC so that the fuel burn "pulse" meets this requirement.

It's a back and forth effort to get to well balanced VE and timing tables.

Once this is done, you can calibrate the AFR (WBO2) table to suit.

Do you need to use the AFR table, no. You can, but well calibrated VE and timing tables are the basis for calibrating the AFR table, and as such, is all you need.

The VE and timing tables produce an immediate engine response whereas the AFR table is an after the fact sensor for the ECU in that the WBO2 sensor that provides the input to the ECU is in response to what the engine is doing. The WBO2 signal to the ECU is based on what has happened, not what is happening or going to happen.

Honda realized this and did not incorporate an O2 sensor in these older FI GWs, and when new, emissions were probably quite good and did meet the requirement of the day. A place to install an O2 sensor didn't help.

Something to consider.

Camshaft sensor(s). These along with the crankshaft sensor provide the ECU with signal(s) to ensure correct fuel and ignition for the engine.

The crankshaft sensor is predominantly for engine timing, whereas the camshaft sensor(s) are for fuel injection.

The camshaft sensor(s) are used to indicate to the ECU when a cylinder is at TDC for engine "phasing". You can operate an engine without a camshaft sensor, but the crankshaft signal as the only crank/cam indication only allows the ECU to guesstimate what to do. It has stood the test of time, but the engine does not operate as effectively and efficiently as it could. It will operate well for most normal vehicle applications, but it can be better. Enter the camshaft sensor.

Should you wish to have sequential fuel and ignition, you will need a crankshaft and camshaft sensor. If you wish to stay with a wasted spark and paired injector engine profile, you can still benefit from a camshaft sensor.

The '85/'86 FI GWs have two camshaft sensors for accurate fuel injection and engine operation. Honda has achieved a semi-sequential fuelling using these two sensors, the engine is more fuel efficient, starts extremely well.

These sensors indicate to the ECU when a cylinder TDC has occurred and when to inject fuel into the engine. The Gr (right) sensor controls fuel injection for cylinders 1 and 3, the Gl (left) sensor controls fuel injection for cylinders 2 and 4.

Once the ECU receives a signal from a camshaft sensor indicating a cylinder TDC, 4 signal pulses from the crankshaft sensor after the signal from the camshaft sensor and the ECU triggers the respective fuel injector(s).

Very sophisticated for the time, and not much has changed since.

Are the camshaft sensors beneficial for any FI conversion, yes, even if you do not use these for sequential applications. The more and more precise information that the ECU has to use, the better the engine operation and performance will be. Bottom line, install a camshaft sensor, or a sensor that will send a signal to the ECU at half crankshaft speed.

Do you need an O2 sensor, no, but it's a good engine tuning tool.

Do you need a camshaft sensor or a facsimile of such, no, but It will permit the ECU to better control the engine operation and performance.

Just my thoughts on the FI system components.

Cheers
 
Very well worded ...As I dont have any experience with FI on a oldwing ..I do have plenty experiences on carb models and there lack of adjustment ..not only in fuel but also in timing ...oldwing motors use waisted spark system witch is ok till you add 2 timing belts ...witch has plus as it very reliable contact ...but the?system provides only one setting to timing the motor as in advancing the timing ...a simple extra. pully that could adjust the cam and tension on belt wpuld have been super smart. Way to have control of this ..making the best of both worlds of engineering ...something goes for the points timing being dual points you could not advance one side of motor without hurting the other side ..I don't know hold a didn T allow enough adjustment here ..these things are why single carb set ups suffer so much ..you basically can't dial in a carb by adjusting carb only .....the same deal applies to FI you can't get good results just changing fuel only ...in my opinion ...as said I have never been down this road ..and these?are?just thoughts ...on my hooch bike the C5 IGNITION eliminated all the timing hurdles and I was able to program the ignition or Paul did the owner of C5 ..AS things turn out my Weber two stage carb only works on the highest advance setting and was the biggest?reason the bike turn into a monster of power ..the ignition was dial into the carb ...I think This has a lot to do with FI TUNING AMONG OTHER ISSUES LIKE CAM PROFILE AMD SUCH ...AS SAID NICE READ I ENJOYED IT
 
Continuing with the journey. Have been contemplating switching to the Alpha-N fuelling profile using the TPS as load combined with RPM. Started with speed density (SD), but this fuelling profile on a bike has a very narrow tuning window.

Did a lot of research and found that quite a few converts, EFI conversion, started with SD fuelling, but eventually went to Alpha-N. Not for everyone, but worth a try.

The basis for this is that there is a considerable change in air flow when the throttle plates are first opened. It's not linear. It's like throwing a bucket of water at someone as soon as you start to play, then once going, there is less and less change in the amount of water being used. This is because the rate of change - Alpha, as the throttle is opened gets less and less.

I am using a single VE and spark table presently. There are provisions to use a second VE and spark table. Found a YouTube video regarding this, but the fellow did not mention the spark table. He is using the two tables to get better resolution on initial throttle opening up to 20% TPS and then from 20% up to 100%. The granularity is quite good. This will need to be explored.

There is one additional fuelling profile that can be used and the Speeduino ECU does allow for this, that of a blended fuelling profile where you use on profile within a certain window, then switch to the other fuelling profile for a different window.

The CX 500 ('82)/650 ('83) turbo motorcycles are examples of this. The fuelling profile is SD initially, then switches to an Alpha-N profile. This makes sense in that the CX motorcycles have "boost" and there is more air being being consumed at higher powers. For the '85 and '86 GW FI models without "boost", A blended fuelling profile where Alpha-N is used initially then transition to an SD fuelling profile.

There are benefits to each fuelling profile. For now I will investigate and pursue the Alha-N fuelling.

I have mentioned in other posts and threads that there are piggyback units that you plug into your motorcycle's system to improve performance, change fuelling characteristics and the likes. Thes units are generally programmed with an Alpha-N fuelling profile. I mentioned the this has to be done with caution because if the engine tune uses an SD fuelling profile, there could be conflicts between the OEM and aftermarket - something to be aware of.

For now it's another chapter.

Have decided to install a different Speeduino compatible ECU. I have many reasons for this, but mainly this will allow me to do bench projects. I would have to buy another Speeduino interface board to do this so I will use the one I have for that. The new ECU will also come configured with tach output - have to do a mod to the existing board to get this, and a baro sensor installed. Two important sensors in my mind. This new ECU uses the Arduino 2560 Pro (smaller footprint than the Arduino Mega 2560. The case (included) is very compact.

Will post more as I go forward.

Cheers
 
Last edited:
Side trip - new ECU microprocessor is an Arduino Mega 2560. There have been some technology developments. Have been looking at a different Speeduino ECU because I want a Speeduino ECU for bench projects.

The new ECU I am looking at is a Spark Gap X4. The interface board is a Speeduino clone of the Speeduino v0.4.4 by Josh Stewart. It has a smaller footprint than the Speeduino x0.4.4. This new ECU uses an Arduino Mega Pro microprocessor. The Arduino Mega PRO performs and operates the same as the bigger brother Arduino Mega 2560. The Arduino Mega Pro has some additional features and usable pins, but essentially it's the same.

Ordered an Arduino Mega Pro and it was here today. The Arduino Mega 2560 is approximately 4" long by 2" wide. The Arduino Mega Pro is 2" long by 1 1/2" wide. The only other way to get a Speeduino ECU with a smaller footprint is to have the Arduino chip installed on the interface board. The Ocelot from WMTronics is such an ECU.

Picture of the Arduino Mega 2560 and the Arduino Mega Pro:
Arduino Old-New.JPG
Have to solder the pin connectors. Need the practice - should have ordered the Arduino Mega Pro with the connectors already attached to the board.
 
Had trouble getting the Arduino MEGA Pro connected to the Speedy Loader and Tuner Studio. Tried different cables, looked for comm/USB ports, asked for help on the Speeduino forum. PSIG replied and he was having no problems connecting the Pro to his computers. Last kick, I took the cable from the Sean headsets, used it and connected the Pro to the Mac and Toughbook SpeedyLoader and TS apps (used to call these programs). Loaded up the Speeduino firmware and all is good.

Time to order the new box. Winter project is a rewiring project. I have more wiring duplications than I should that uses the same solid colours. Will order some wire that is colour coded, striped or whatever. Be good for wiring schematic and for troubleshooting. Need to add some larger connectors to separate the wiring going into the rear trunk:

1726590056615.png

The new ECU will be a Spark Gap X4 from openlogicefi. Uses the Arduino MEGA Pro microprocessing board. Choosing this one for the size, all connectors are provided.

1726589133447.png1726589207218.png1726589381883.png
 
Hard to imagine that the Spark Gap X4 ECU has most of the features of an OEM ECU. Technology has been busy.

Edit: The size is so much smaller. Soldered the male pin headers. Not too daunting, getting better at this.
 
Last edited:
Comparison pics between Arduino Mega 2560 and Arduino Mega Pro. These controllers can operate your car! The first pic illustrates a different size perspective. The Arduino Mega 2560 (blue one) is 4 inches long by 2 inches wide. The PCB for the Arduino Mega Pro with components installed, is quite thin, second pic on right. Both do the same job.

Going to connect the Arduino Mega Pro to the engine simulator I have and connect to Tuner Studio to trial it. Wondrous world we live in.

Arduino Mega 2560 versus Arduino Mega Pro (small one).JPGArduino Mega Pro - Size.JPG

When I receive the new ECU will be doing a post on size and such. Might even try a video. Have to learn how to upload to YouTube.

Had to do a new tune for the GW. Don't know what I did, but settings were wonky and would not allow correction. Did learn how to take a table from one engine tune and import it into a different engine tune.

Cheers
 
Got a fresh engine tune installed, had an issue with the older ones - maybe too many changes over time.

Out for a road trial this afternoon. Used Tuner Studio "Tune Analyze Live" and let it update the controller as the ride progressed.

Did this because there is a very small MAP window that is used by the engine. Most times and RPM dependent, the MAP window is 60 to 85 Kpa. Going to do this a few times to get a better VE table, then assess and "smooth" out the VE table to suit. The engine ignition timing will be addressed after the VE table is stable and better tuned. The AFR table will be addressed last.

The AFR gauge in the Tuner Studio software does go lean, but for a very quick moment. Not concerned about this at this time.

A fellow in France has developed a Speeduino clone that has everything installed on the interface board. It's call the "Warthog". Has the Arduino controller chip integrated, and quite a few options. He mentioned that he may work on a "Warthog" for a four cylinder.

Short update.
 
Interesting encounter yesterday while I was out on the GW. Stopped at a traffic light beside a Victoria transit bus. Driver opened his window, a younger fellow, and we chatted real quick about my bike, he complemented me on the look. He mentioned he would like to get an 1100/1200 GW and convert it to EFI with a Micro/Mega Squirt. I mentioned that mine was EFI and I was installing a Speeduino ECU. Light turned green and we parted ways. Nice to now these older GWs get favourable looks/comments on the street.
 
Second road trial letting the tuning software, Tuner Studio adjust/calibrate the VE (fuel) table using Tune Analyze Live. I have mentioned that the GW engine(s) have a very small MAP operating window, approximately between 55 and 85 kPa. Riding and trying to see what is happening, then stopping to adjust the VE cell values is difficult, hence the change in tuning strategy to using the Tune Analyze Live, review the catalogs and make additional adjustments based on the VE, ignition timing and AFR tables. Dyed in the wool tuners may not appreciate this as tuning, but I have read many a forum thread where this is being done - especially with motorcycles. More on this as I progress.

Onto ignition timing, quite complex and lots of information to digest. Found 0n-line articles that discuss engine timing and how it developed and has been rolled into the EFI world.

Back in the day, mechanical engine timing had three components; locked timing, mechanical advance and vacuum advance. OEMs did a decent job of meeting the emission standards the day with these distributor ignition system(s), but it was a best guess. Transition this to the modern day EFI systems and a timing curve can be done nicely.

Regardless of the ignition system, the engines of today react exactly the same to ignition timing as engines of yesteryear. Engine knock, coolant temperature, fouled spark plugs and so on are all the same.

Too much fuel, not enough burn time and fouled plugs, poor fuel economy, and poor engine performance. Too little fuel, too much timing advance, maybe better fuel economy, engine knock.

Too many scenarios to think about.

My latest learning session is quite interesting. Fuel burn time, ignition timing. The OEM and we neophytes in the aftermarket EFI world should strive to have an air-fuel ratio (AFR) of 14.7:1 (14.7 parts air to one part fuel). After this, it is tuning to get the best fuel economy, performance from the engine as possible, adjusting the VE table to suit the various RPM/engine load combinations.

How does timing affect this? At idle and with a lean air-fuel mixture, you need to have ignition timing that allows the air-fuel mixture to completely burn by the time the piston passes the 10 to 15 degrees ATDC. Too soon before this, engine knock could happen, going past this point and fuel continues to burn, using the engine as a heat sink and the engine operating temperature starts to climb to where you do not want it to go. This scenario plays out the same when you are riding on the road.

I would like to have the same setup as a car, you can have someone with you who understands what has to be done, but not going to happen. Maybe need a side car :LOL:!

Best fuel economy is when you have the VE and ignition timing tables calibrated so that the engine is operating just below the "knock" threshold for a specific engine RPM and load. AFR should be trending towards a lean fuel condition, the fuel burn is complete before the magic 10 to 15 degrees ATDC, and engine coolant temperature is within the normal operating range.

I do engine data logging as well. The engine data is collected by Tuner Studio (TS) and used in MegaLogViewer (MLV) to provide you with information that can be used to adjust the engine tune. I will be using MLV to provide me with 3 matrices: RPM vs MAP vs VE, RPM vs MAP vs spark, and RPM vs MAP vs AFR.

I am looking for AFR readings in the 13.0 to 15.0 AFR range for now calibrating the VE table first, then ignition timing, then AFR. The AFR table is an AFR target table that is used when O2 corrections are enabled. When cruising, engine under light load, you can lean out the fuel mixture for better fuel economy.

The spark (ignition timing) table will be adjusted in small increments of 0.5 to 1.0.

I have to "rebin" the three tables. This is where you adjust the "X"/"Y" table axis and have the software interpolate the cell values. I think this is necessary because of the narrow MAP window the engine operates in, provides better resolution for engine tuning.

I have expanded the MAP region between 60 and 90 kPa. As I have mentioned there is very little information regarding how to set up the "X"/"Y" axis on the three tables for motorcycles, lots for cars/trucks.

One last issue. Ignition timing setup. You do an initial trigger angle calculation determine where #1 TDC IS. You then do an engine timing to get the exact trigger angle. From here, you check the engine timing at idle ensure the engine timing and engine tune timing is the same.

Honda indicates that the 1200 engine idle timing is 10 degrees BTDC. This may be achievable with good results by Honda; however, Honda did some weird and wonderful engine tuning with the CFI system. Take away here is to throw the Honda engine timing out the window and start afresh.

I have mentioned that I was informed that when doing initial idle timing, give the engine what it wants, fuel and timing. This could be anywhere from 15 to 30 degrees BTDC depending on the engine in conjunction with fuelling - do not adjust the trigger angle.

The 1200 OEM ignition timing profile allows for timing from 10 to 45 degrees BTDC depending on the engine RPM and load. Lots of wiggle room.

I'm presently using an idle ignition timing value of 18 degrees BTDC along with the corresponding VE cell value (give the engine what it wants. This timing setting provides an AFR reading of approximately 14.0 at an idle of approximately 1000 RPM, good for now - will probably change as the tune progresses. This indicates to me that all other ignition timing settings need to start at 18 degrees BTDC, but should not be less (TBD).

Short update. More to follow as the project progresses.

Cheers
 
Especially interested in the ignition timing part, I feel that is the hardest part to get right without listening via knock sensing equipment and doing it on a load controlled dyno.
 
Good afternoon - ignition timing is the holy grail from what I have read for tweaking the last bits out of the engine tune. Are you dealing with EFI or carbs?

No dyno around here, road trials and seat of the pants tuning.

I have mentioned this before, Honda did a good job of calibrating carb and FI GWs, but Honda has better tools than we do. Honda cobbled together the fuel/ignition systems and made these work. Trying to duplicate this is a challenge and in reality, an exercise in futility.

I mention this because the requirement is to give the engine what it wants, and this will probably be different than what the OEM design is.

If you have a carb model GW, or another make/model, Not much can be done regarding timing and fuel. The OEM specs are the best you are going to get. You could install an O2 sensor to monitor the emissions and tune the carbs with this. You may need two if you do not have a crossover pipe between the cylinder banks. Timing is fixed with base timing, lock the distributor so that the mechanical advance does not work and plug the vacuum advance. Set the base timing with #1 cylinder at TDC. Idle timing needs the mechanical advance unlocked and you can connect the vacuum advance. Check the engine idle timing using the engine crankshaft. Once the engine timing is set, time to go after fuel (no un intended).

There are two crankshaft timing marks; "T1" to align the crankshaft and camshafts and have #1 cylinder at TDC. "F1" is idle timing of approximately 10 degrees BTDC. Once you have done these checks, that's it for timing.

OEM jetting is to deliver the best fuel economy and engine performance straight from the factory. Installing larger jets may well result in a fuel rich operating condition that at low powers and idling, can cause plug fouling. Smaller fuel jets can result in a lean fuel condition that at low powers and idling should not be an issue, but at higher powers, operating the engine in a lean fuel condition is not good.

The optimum fuel burn time is from injection (this is not literal) fuel to the cylinder, carb or FI, until the fuel is completely burned by about 10 to 15 degrees ATDC. This is the very best scenario and aids in engine performance and fuel economy.

Too much fuel and fuel is still combusting after this optimum timing, resulting in poor fuel economy and engine performance, and the plugs will probably foul.

Too little fuel and you run the risk of engine "knock". Catch 22 all round, especially with a carb engine.

The 1100 is a good candidate for an aftermarket electronic ignition control, but it is not an easy upgrade, lots to consider. If the bike you have is a keeper, a complete EFI install may be in the cards.

This is my thread on upgrading the 1200 GW carb ignition system. The 1200 GW carb models use an ignition control unit (ICU): https://classicgoldwings.com/thread...d-1986-se-i-fuel-injection-motorcycles.14941/

Just a few of my thoughts on your query. Good luck.

Cheers
 

Latest posts

Top