1985 GL1200 Limited ECU Replacement/Upgrade - Part 2

[Rednaxs60]

Been contemplating the issue with the starting of the engine with the new ECU. These are power hogs at the best of times. When the battery starts to degrade it can rob the electrical system of a lot of the power that should be used to operate the electrical system. Had this happen last year, the engine started to do all sorts of weird things. My brother's 2011 Hyundai Santa Fe had a battery issue. Door locks, lights and other items started to misbehave. Sonya's brother's Spyder, before he bought it, would start, go for a bit then resort to the get home mode - new battery and all was well.

Taking the day off to do some house cleaning and such, but will go out tomorrow and get a new battery. thinking a small car battery for the trials being done. Have the lithium-ion for when it is operating better, and starts easier. The lithium-ion battery is not for testing.

Steep learning curve still to come.

 

[pidjones]

I believe fuel consumption (instantanious) is derived from injector pulse width and rate.

 

[Rednaxs60]

I believe fuel consumption (instantanious) is derived from injector pulse width and rate.

I have a spare travel computer. I will be connecting this to the system before the next try with the Speeduino. Intend to revert t the OEM system, start and clean out the engine with the travel computer attached. Make sure all the settings work, then revert to the Speeduino ECU setup, start the engine, get it to an idle, and check the travel computer.

The injector pulse you mention may be the same as for the tach. It will be from numbers 1 and 3 injector. I have not changed any wiring with regards to these injectors.

 

[Rednaxs60]

Here's the Travel Computer wiring schematic:

 

[Rednaxs60]

Another afternoon of test and trial. Put the CFI system back to the OEM configuration, new battery, plugs. Operated the engine to get it warm and clean out some of the trial residue.

Installed the Speeduino, and cranked away. The engine did start after a few tries, and operated well in the 1500 to 2000 RPM range. Could not get it to idle no matter what I did. Turned the throttle idle screw all the way in and still would not idle. Checked the timing and it would not change, stayed locked at approximately 5 deg BTDC, the OEM timing is at "0" deg TDC. Will check this setting, could be fixed and not letting the Speeduino react to changes in the trigger setting. Checked the plugs after the trial. Dry black soot on all four - far cry from the white porcelain ones installed just before the Speeduino trial:

This is apparently caused by the air-fuel ratio (AFR). Need to do some reading and viewing about this.

Had the Travel Computer connected during the trial. Was able to go through the TC functions, everything works. This issue is put to bed.

Overall a good day with the exception of starting and idling the engine. Gots to be a setting or two, and my understanding of the TS settings.

 

[Rednaxs60]

Mike - trust you have a cold one with you when you read my posts, or use them to go to sleep at night. Cheers

 

[Rednaxs60]

Had another go with the Speeduino ECU installed. Changed a few settings and got the engine started, and a rough idle at approximately 1500 RPM. Checked the timing, and went through the Tuner Studio tables. Very interesting to view what the individual tables are doing. You can see what is happening in a table and how the engine is operating and what cells are being used.

Will be doing another trial tomorrow, and hopefully, get the idle dialled in a bit better. Have changed a few settings, but it is easy to revert back to the initial settings by loading the previous tune, or changing the settings in Tuner Studio software.

The Mrs asked why I was doing this. It's not because I have to, have a spare OEM ECU, but because I want to. Understanding what is happening and learning the various aspects of the ECU tuning needs is a great learning experience. This ECU project gets me into the nuts and bolts of the ECU.

 

[julimike54]

Mike - trust you have a cold one with you when you read my posts, or use them to go to sleep at night. Cheers

I generally am drinking coffee just have been interested in how these computerized fuel injection systems work, but have no first hand knowledge so it's a way to try and gather some more info. One of these days I may need to know more?

 

[pidjones]

I have a spare travel computer. I will be connecting this to the system before the next try with the Speeduino. Intend to revert t the OEM system, start and clean out the engine with the travel computer attached. Make sure all the settings work, then revert to the Speeduino ECU setup, start the engine, get it to an idle, and check the travel computer.

The injector pulse you mention may be the same as for the tach. It will be from numbers 1 and 3 injector. I have not changed any wiring with regards to these injectors.

I think it would be pin 5 with a PWM signal. I believe duraton of the pulse is what determines the amount injected per pulse. Pin 11 for the tach would be the rate.

 

[Rednaxs60]

I generally am drinking coffee just have been interested in how these computerized fuel injection systems work, but have no first hand knowledge so it's a way to try and gather some more info. One of these days I may need to know more?

It is a great project. Very similar to the carb world, just different components. The Speeduino Project that this is based on is open source, you get to view the program code, modify, test, trial if you want to. You can view how each parameter and sequence of events are put together. I have looked, but this is not my focus at this time. I do have thoughts on follow on projects.

Went looking for information regarding what the numbers mean in the main tables. The VE table numbers are used in the fueling equation located somewhere in the program code. The higher the number the more fuel is injected - rich. The spark table numbers are degree of timing advance. The AFR table is a percentage used to calculate fuel mixture - can be a different value depending on the ECU being used. These tables need to be adjusted so that the values give best engine operation.

It's an iterative process. You make a change or a few small ones - must keep notes, connect and see what happens. Something like changing the jets in a carb. One of the difficult parts is getting to the actual values needed when using older components. First requirement is to get to a good Strat and idle. This is where I am and it is testing the thought process. Have to get the fuel and timing correct.

These Speeduino/Arduino units are being used in quite a variety of vehicles, and items such as lawn tractors. Very versatile.

Have another afternoon of tests and trials to do. Hoping for some better results. Have adjusted a few settings, small changes, but have done a change to the VE table that I would consider a leap of faith and quite detailed for my skill level.

Cheers

 

[Rednaxs60]

I think it would be pin 5 with a PWM signal. I believe duraton of the pulse is what determines the amount injected per pulse. Pin 11 for the tach would be the rate.

You're correct. The input required is the same for the tach and this comes from numbers 1/3 injectors. Honda did a great job on the TC.

 

[Rednaxs60]

Have progrès to report. Have changed the VE/Spark table settings and have managed to get the engine started without a lot of fanfare and idle at 1400 to 2000 RPM. Still a bit jerky regarding timing but heading in the right direction. Did what I would consider a radical change to the settings, but sometimes have to go further one way and come back to the centre.

There are a lot of fuel settings that add fuel to the standard fuel load. Have eliminated or toned these down. The VE table has been toned down, don't fill the garage with bluish fumes at this point, like much better.

One of the fellows on the Speeduino forum looked at the tune and data log and mentioned there appears to be a sync loss that is consistent. Will be addressing this today.

Going to modify the spark table. Have got it in the ball park, now to do some additional, hopefully fine tuning.

 

[Rednaxs60]

This afternoon should be good. Had a wee set back this morning, got it corrected. Be damn fine if a consistent idle can be achieved.

 

[Rednaxs60]

Good news actually. The engine fired and I had to keep the throttle open but it did operate. Did this for a bit then shut down and logged the data. Checked the plugs. First time these were practically the same, black carbon soot - the changes I made reduced the fuel load to numbers 3 and 4 cylinders, progress. Know the timing is not quite on, but working on smoothing this out. Did some work on the VE and spark tables and changed dwell settings.

The second go round, took some time off for lunch, installed Speeduino, as well as clean and dry plugs, and it started and idled at about 2400 RPM. Almost started like it was normal. Idle was high and engine operated well. Was able to start reducing the idle by adjusting the idle screw.

The running dwell can be changed on the fly as can the values in the VE, spark and AFR tables. The injector open time cannot be changed without a roboot of the ECU. Adjusted the running dwell, had started at 2.5 ms and adjusted this upwards. Settled at 2.8 at this time. A running dwell setting of >3.0 ms is not accepted in the software. Will be doing more trialing tomorrow, and will adjust the running dwell lower than 2.5 ms to determine how the engine will operate.

The spark table for timing is still a work in progress. I know the premise behind the spark table, it is the ignition advance for timing. My understanding is that the faster the engine spins the more timing advance needed, conversely, the slower the engine turns the less timing advance required. At this point I'm looking for an idle at 1000 RPM so my focus is on this RPM as the engine Strats to have timing issues as the RPM decreases. Have been reducing the ignition advance in the spark table, adjusted it for tomorrow's trials. Keeping a separate spreadsheet with all my changes to the various tables.

Being able to view the various tables is quite interesting and allows a person to see what is happening. This lets you ponder a situation and possibly come up with a game plan.

It's coming together. Cheers

 

[Rednaxs60]

Had a good afternoon, longest trial yet, engine up to operating temp, taking data log. Had made a few changes first off to the VE and spark tables for RPM that would settle out.

When I started the engine data log, operated it for quite a while and got some good data. It would idle at 800 RPM - didn't keep it at this RPM too long.

Have the program tweaked so that all four plugs look the same when removed, dry, black soot. Will be changing the Required Fuel (RF) setting to try to compensate. Unlike carbs, air-fuel mixture is changed by changing the settings in the program.

Will be changing the throttle plate position. Will reduce the idle screw adjustment until the TPS voltage gets as close to "0" VDC as possible without completely closing the throttle plate(s).

Have a sync loss issue that is hampering a good idle. Trying to isolate the issue.

Started a document detailing how I have setup the Speeduino ECU for the GW. Thought it might be a small document, couple of pages or so, but not to be. Going to be something of a mini manual. The best part of this project is that the Speeduino ECU will operate the GW CFI system. This is huge. More to follow on this.

More to come on the adventures in the ECU world.

 

[Rednaxs60]

A good afternoon trial. Best trial to date.

Hard to start and it is timing, but can't seem to get to the reason.

Have mentioned the sync issue, have to find it. the sync issue appears to be cyclic, happens once a cycle - 720 degrees. The cam/crank/spark igniters/injectors are all contenders.

Was able to get it to idle above 1500, and keep it going. This allowed me to adjust the spark and VE tables to get the best idle at the RPM settled on.

The spark settings are quite high for the RPMs being looked at, below 2000 RPM. The closer The RPM gets to the 1000 RPM level, the more the engine stumbles and will stall.

The plugs are looked at after each trial. Most times these are coated with a dry soot. This has been reduced. The change in spark and VE settings today is promising in that #4 cylinder plug has a white electrode - progress: Pic of the plugs, #4 on left to #1 on right:



Plugs 1 to 3 do not have a lot of carbon, #4 plug was a pleasant surprise.

Have fuel system work to do. May try smaller injectors 240 cc/min instead of the OEM 284 cc/min injectors. Found a web site that have approximate open/lag/dead times for voltages from 10 to 15 VDC.

Cheers

 

[Rednaxs60]

Lots to do and get right. Hopefully next trial will be more progress.

As an aside, have started a document regarding this project. I want it to be a primer/concept document that gives general information to achieve the aim.

Thought it would be a quick couple of pages, but I'm finding out this is not to be the case. Working on the section regarding the Tuner Studio (TS) settings and where necessary, an explanation of what the nuance is. A section on terminology, and what some of the acronyms mean. Tuning the engine after first start, and recommendations for the Speeduino forum members. It's a growing document, and will hopefully be useful.

Will be separate thread. More to follow.

 

[Rednaxs60]

Starter has not been cooperating. Took the OEM starter off today, thought I could use an '84 starter for the short term, nope.

The OEM starter that I took off is 38 years old - rebuilt/cleaned it some 6 years ago. It does turn when off engine, but thinking there is a short inside - sparked when I tested it on the bench. The '84 starter did not spark when tested on the bench - worked well. Will take to electrical shop to determine what can be done. Worst case, have to get a new one.

I think I've localized the sync issue to the cam sensor. Checked the cam sensor wiring IAW the manual. had continuity where it should not have been. Have removed the wires from the cam sensor connector between the cam sensors and the wiring harness. Checked for continuity and all is good.

Will do scope traces of the cam and crank sensors when I get the engine going. Good to have an initial reference.

Something has to give.

 

[julimike54]

"Something has to give" - hope it's not you

 

[Rednaxs60]

I'm higher on the food chain, it will be assimilated. Always something when doing a significant upgrade on an older vehicle. Vintage car enthusiasts probably encounter issues on a regular basis.