1985 GL1200 Limited ECU Replacement/Upgrade - Part 2

[Rednaxs60]

Project has slowed down a bit. Have done a new tune, using a Bosch coil driver in place of the OEM spark units - more information available on newer components.

Looking into new coils for the project. Have always had a weaker spark on cylinders 3/4 when the cylinder banks (left-right) have been balanced. Changing the cylinder bank balance does affect which cylinder is not doing as much work, but when balanced as close as possible, always numbers 3/4 cylinders not doing as much work. Doesn't help that Honda specs for cylinder bank balancing specifies that if the cylinder bank vacuum is within 1 1.2 in HG all is well.

Took time out this afternoon to browse the various forums regarding coil upgrades, what is being considered and used. Dyna, Access, 1500 coils are some of the usual suspects, and mostly an upgrade to a 3 ohm coil. GWs, 1000/1100/1200 models have all suffered with varying electrical issues, and coils are no exception.

Check pricing for a new set of aftermarket coils such as these, very expensive. Honda OEM coils are not far off the expensive mark either. Aftermarket OEM style 3 ohm coils are a good alternative - approximately $100.00 CDN for a set, but the specs are similar to the OEM coils.

The FI models have another issue and that s the spark units (coil drivers). No way to test. Do these get weak over time affecting the firing of the coils? This is problematic because it makes replacement spark units (coil drivers) hard to source and when a decision is made to replace, wiring the new coil driver into the system and which one to use can be a hit/miss scenario.

Having mentioned all this, I have browsed the Speeduino forum for a suitable alternative coil driver for my project and there are a number of coil pack and coil drivers available that have settings already determined for use with the Speeduino.

Thinking that Honda used 3 ohm coils as a standard, then designed the OEM ECU programming to suit.

Now that I am replacing the OEM ECU, I am looking at automotive coil packs such as the Bosch 211coil pack (smart coil) with an integral coil driver:



Another option is a coil pack for an Opal Cross - dumb coil, no integral coil driver:


Honda designed these engines with a car design philosophy. I have mentioned on many an occasion that the '85 Limited Edition FI model is a Honda Civic on two wheels.

This project has had me question the operation of the spark units, specifically when the ECU fires the coils through the spark plugs.

I have mentioned that the spark units are probably a Darlington Pair transistor design, can't confirm. This has its merit, but is not the only solution to the issue. The OEM ECU has to have been designed to take into consideration coil dwell time, and spark duration. The new ECU allows me flexibility as to which coil pack I can use, and whether the coil pack has an integral coil driver, or not.

Review other FI threads, and when a project comes to fruition, not many revert back to the OEM design. This is much the same as single carb conversions (SCC).

Modernization of the OEM CFI system is underway. Not many components to consider thankfully.

 

[Dusterdude]

Red,some folks use the dodge neon coil too.

 

[Rednaxs60]

Yes, I have read about these since I purchased my 1200. It's a pig in the poke so to speak as to what works, and how much testing has been done. This project really accentuates what is done by OEMs when components are chosen. Thankfully there's quite a bit of wiggle room if you stay close to home.

 

[pidjones]

I did a short experiment on my '79 GL1000 with Ford TFI modules triggered by the points. They modules handled dwell and coil current, placing only tiny current across the points. I felt this system would give great performance (well controlled dweel and long point life) but did not realize that the regulator had failed and I was feeding the system at up to 18 volts! Worked great for about 100 miles, but then became erratic. After replacing the regulator with a reg/rect, charging voltage wasperfect and luckily battery and lighting survived, but tge TFIs remained erratic once warmed up. I pulled them and went back to straight points, which was quite easy as I had designed the system to permit a quick roadside swap requiring only a screwdriver to remove the left side cover.

Point being - there are many coil driver packs for cars out there, triggered by various sensors (my Ford push-start TFIs by the points opening). Some, such as the push-start TFIs control dwell themselves, others such as later TFIs and GM modules have inputs for ECU dwell control.

 

[Rednaxs60]

Pidjones - thanks for the comments. I'm one of the crowd that, if possible, will try to find the information and use what has been proven to be effective. This project is having me think I have to do testing of components to get the ECU setup as good as possible. Will take me a bit of time as I like to do the bulk work and not the nitty gritty stuff. Have food that the Bosch 601B is widely used for the Speeduino and Squirt family installs, and there are settings available as well.

The Bosch 601B coil pack with integral coil driver is widely used for these automotive conversions. Perused the web last night and lots of good information. The GW engine has a good amount of automotive design philosophy.

Have ordered a Bosch 601B with the integral coil driver. The integral coil driver can be removed/replaced, or you can remove, use the coil pack as a dumb coil with an external coil driver. Will install in the same space as the OEM coils, only need approximately 3" height to install.

 

[Rednaxs60]

Proceeding with phase 2 of the project. Part 1 did not deliver the expected result(s). All sorts of issues trying to accommodate the OEM design and components, my lack of understanding regarding some of the issues, and there is so much to learn. Have prepared a new way ahead document.

Phase 2 of the ECU Upgrade/Replacement Project

Starting on the second phase, first didn’t work quite as well as I thought it would. Will be incorporating the lessons learned, recommendations from the collective, and settings that have proven to be acceptable from first time round. This is for an ’85 Honda GL1200 Gold Wing Limited Edition Fuel Injected model.

Won’t go through the rationale for the project, have documented that in the previous threads.

Lessons learned from the first go round:

1. Wheel decoder system. Wanted to keep the OEM dual wheel setup, unfortunately this did not work as planned. Most GW EFI conversions use the 36-1 crank trigger wheel. Not many go to a dual wheel system after this to do sequential fuel injection. Recommendation: switch to a 36-1 missing tooth trigger wheel at the start of the project.
2. The variable reluctance (VR) crank sensor (Ns) can be used with the 36-1 crank trigger wheel.
3. Making an older OEM FI system and components work with a new aftermarket ECU is not as easy as I thought. This has been more difficult than expected. Specs are not available for the FI components, and the components, in this case, are designed for a specific FI system. These components can be made to work; however, upgrading the original OEM FI components for more modern, and tested FI components may/will assist in bringing the project to fruition faster.
4. Coils - the OEM GW coils are good for the OEM purpose built system. These coils have been found to be acceptable, but are not equivalent to the more modern coils. I base this on the fact that coil upgrades are being done quite regularly by GW owners. Some of the preferred coils for an upgrade are Dana, Access, Dodge Neon coils, and GL1500 3 Ohm coils. The GL1500 coils are also used as an upgrade for GL1200 coils. Recommendation - look into changing the OEM coils for more modern, and proven to work with the new ECU.
5. The spark igniters are specific to the OEM CFI system. How these operate can be ascertained, but these spark igniters (coil drivers) are some 37 years old and probably well past the best before date. Recommendation - change to more modern coil drivers that compliment the new coils you will use. Chose a coil driver that is proven to work well with the Speeduino ECU and settings are known that can get you very close to the desired result. Most GW EFI conversions use more modern coil drivers such as the GM 7 pin HEI coil drivers even when OEM coil drivers are available.
6. You can use the OEM injectors - low impedance with a resistor pack, or upgrade to high impedance injectors. Generic settings for injectors are available with some research, and once in the ballpark settings can be adjusted to suit the tune.
7. Coolant and air temperature sensors can be used. Generic sensor settings work with these.
8. TPS sensor can be used.
9. Needed to install a WBO2 sensor and controller. Will be used for additional tuning, just need a spot to install the sensor.
10. MAP sensor - use the on board MAP sensor. You can use the OEM MAP sensor, its output is very linear. I would recommend using an OEM MAP (PB) sensor for a real time Baro sensor
11. Idle air control - the OEM system is passive and works well, no need to upgrade or change.
12. Pay attention to the recommendations from the collective. The same recommendation(s) are reiterated over and over.
13. Read as many forum threads as possible. Answers to various questions that you may have can generally be found by doing this.
14. OEM CFI system design is an issue that must be rationalized. The OEM CFI system is designed to provide excellent performance within the OEM design parameters. Is it better to wrestle with the OEM design or change/modify the system to suit the upgrade/replacement. I would submit that, in hindsight, a modernization of the OEM CFI system components to components that have proven to work well with the new ECU is a good way ahead.

The recommendations above are for my project and may apply to others. I’m fortunate that the FI system is in place, don’t have to go through a design and fit process. Lots to consider without doing this.

Will get all upgrades/changes in place in the next two weeks.

The ECU and component setup is:

Josh Stewart v0.4.4 interface board with an Arduino 2560
Coil Driver - Mitsubishi J723T
Coil Pack - Opal Corsa coil pack
Trigger wheel - 36-1
Crank Sensor - OEM VR
MAP sensor - on board MPX4250AP
Injectors - OEM low impedance 284 cc/min
Coolant/air temperature sensors - OEM
TPS - OEM or after market 3 wire rheostat
Idle Air Control - OEM passive system
WBO2 Sensor/controller - 14point7 Spartan 3 Lite with LSU/ADV sensor

Snap shot of the system. Now to make it work.

 

[Rednaxs60]

Have been doing research on coils and coil drivers. It appears that the automotive industry uses two types of coils, one with an internal resistor and coils requiring an external resistor. The coil(s) that have an internal resistor are rated at approximately 3 ohm, coils requiring an external resistor have approximately half this value.

The reason I went looking for answers to this question is that I am going to replace the OEM coils with more modern coil packs. Was thinking about the change out and how the coil packs would affect the coil driver, external or integral. Do not want to "fry" the coil driver prematurely.

The OEM coils are approximately 3.3 ohms. This gives me a circuit current of approximately 3.5 amps. If the coil is only 1.5 ohms, this value is doubled and can cause some grief. It is an issue with a distributor points system, but more problematic with an electronic distributorless ignition system (DIS), the amps in the circuit can cause issues. My thought is that if the new coil packs are less than 3 ohm coils, I should put in an external restore or use a purpose built resistance wire to bring the circuit resistance up to approximately 3 ohm. Have seen a lot of coil installations, old and new and just expect the design to be the same, apparently not so.

Good info on this web site (read several others as well): Automotive Ignition Coils
Liked the summary of what may be an issue depending on which way you decide to proceed with: "Interchanging these two types of ignition coils can have one of two very different effects on the ignition system. The use of an “internal resistor” coil in an ignition system equipped with an external resistor will cause excessive resistance in the system, leading to weak coil and spark output. In a system requiring an “internal resistor” type coil, the use of the external resistor style coil can lead to heat failure of the ignition coil. In vehicles equipped with breaker points, this can also cause premature wear/burning of the points. Using the same formula as before, with a 12V supply and only 1.5 ohms of resistance, the resulting current flow is doubled to 8 amps.

Something to consider and ponder.

 

[Rednaxs60]

With regards to the ignition, and thinking about coils and such, have been thinking about the coil/plug wire/plug caps and plugs, mostly about the plug caps and plugs.

Have resistor plugs, goes without saying. There's a 5K Ohm resistor in the plug cap.

Using Beldon Metallic plug wires, have used copper core as well. Don't know if these are suppression spark plug wires.

Question is why use a resistor plug and resistor plug cap. Understand the noise issue, but most vehicles I've worked on, early/late models, have suppression type wires and resistor plugs, but not resistor plug caps.

Have read several forum threads where the person has replaced the 5K ohm plug cap resistor with a length of copper rod.

Is there a consensus here regarding this query?

 

[Dusterdude]

Welcome back red

 

[pidjones]

You only need and want one resistor in the chain. Plug, cap, or wire. I use non-resistor plugs and wires with resistor caps, or non-resistor wires and caps with resistor plugs. Resistor plugs seem to be the most commonly available these days, so I have replaced the resistor in my caps with solid copper on most bikes. For your application, you definitely want a 5k somewhere to prevent noise generated that might mess with the computer. If it were a non-ECM ignition for something like a race bike, you would not want the resistance limiting the current flow to the plugs.

 

[dan filipi]

With regards to the ignition, and thinking about coils and such, have been thinking about the coil/plug wire/plug caps and plugs, mostly about the plug caps and plugs.

Have resistor plugs, goes without saying. There's a 5K Ohm resistor in the plug cap.

Using Beldon Metallic plug wires, have used copper core as well. Don't know if these are suppression spark plug wires.

Question is why use a resistor plug and resistor plug cap. Understand the noise issue, but most vehicles I've worked on, early/late models, have suppression type wires and resistor plugs, but not resistor plug caps.

Have read several forum threads where the person has replaced the 5K ohm plug cap resistor with a length of copper rod.

Is there a consensus here regarding this query?

I replaced the resistors with a solid copper wire because I couldn't see a reason to have resistor plugs as well as resistor caps. There wasn't any real difference in any performance that I could tell.

 

[Rednaxs60]

Dusterdude - haven't gone anywhere, just taking my time with the next phase. Spring here in Victoria and have to pay some attention to yard work.

Dan/Pidjones - thanks for weighing in. The plugs are NGK DPR8EA-9 and have a resistive rating of 5K ohm - plug caps are 5K ohm. The combined resistance is 10K ohm. Thinking I'll try the copper rod in the plug caps.

The other ignition change will be to install a newer style coil pack. Looking at using the UF-503 coil pack. Had a Bosch 106B but one side was U/S - sent back. May order another Bosch 106B with integral coil driver - pondering.

The issue with this is the specs for the UF-503 coil pack. This coil pack appears to be one that is well used for Speeduino projects, as is the Bosch 106B with integral driver.

The issue I have is the resistance of these different coil packs compared to the OEM coils. The UF-503 coil pack primary is at 1 ohm, secondary at 5K . OEM coils are 3 ohm and secondary at approximately 13K ohm without wires attached. Looking for information on what I should do. More to follow.

 

[pidjones]

Dan/Pidjones - thanks for weighing in. The plugs are NGK DPR8EA-9 and have a resistive rating of 5K ohm - plug caps are 5K ohm. The combined resistance is 10K ohm. Thinking I'll try the copper rod in the plug caps.

Simple to do with just a piece of #12 or #10 copper cut to the same length as the resistor. The hard part is getting the resistor out on some.

 

[Rednaxs60]

Thanks for the recommendation. Now to get a piece, not something I have lying around.

Working on the fuel system as well, pressure drop is disconcerting. Ordered some 1/4" non-return valves, but like the new 2X4 size, 1/4" doesn't mean what it did. Will have it buttoned up on Wednesday when the right sized ones arrive.

Have prepared the new UF-503 coil pack. Researched it to the Nth degree, and just going with it. Going to use the Mitsubishi J723T coil driver.

 

[julimike54]

 

[pidjones]

Thanks for the recommendation. Now to get a piece, not something I have lying around.

Really? No bits of Romex or other building wire lying around? Check building sites (although they probably use 14 gauge now).

 

[Rednaxs60]

No Romex around here, but happy to say that when I was phoning around, an older gent mentioned house wiring. It's off today for a chunk of #10. You old guys know stuff!

 

[pidjones]

No Romex around here, but happy to say that when I was phoning around, an older gent mentioned house wiring. It's off today for a chunk of #10. You old guys know stuff!

Yes, "stuff" is a kind way to put it.

 

[Rednaxs60]

Got some solid core Romex 10 GA - plug caps changed, ready for use. Non-return valves for fuel system came in, install this afternoon. Coil pack and coil driver will be connected and be ready for testing. Timing will be set this week.

 

[Rednaxs60]

Have been working to progress the next set of trials.

Have a new 36-1 trigger wheel installed. Had the centre hole and keyway done at a local machine shop. Have 0.028" clearance between sensor and teeth.

Have the vacuum chamber installed. Hose lengths are pretty much the same. Once everything settles out, will connect a set of vacuum gauges and determine the cylinder vacuum variance and what I need to do to have all pulling at the same value:


Connected the new coil pack and coil driver - Bosch 211 to the system. Using the Opal-Corsa 3 pin coil pack - similar to the UF-503 coil pack I believe - dwell setting should be 3.0 to 3.1 ms I'm thinking. It has been mentioned that the coil driver setting is GOING LOW. Will have the voltage jumper on 5VDC


New spark plug wires - 7 mm copper core. Have researched the need for suppression plug wires, resistor caps, and resistor spark plugs. Most information calls for suppression plug wires and resistor plugs. Motorcycles use the resistor plug caps and resistor spark plugs, but on the GW forum, many have removed the resistors in the plug caps and replaced with a #10 solid copper wire. Could have used non-resistor spark plugs if I could find, but will use the plugs recommended but the OEM.

I have one set of OEM coils with Beldon steel core plug wires and an aftermarket set of coils with 7 mm copper core plug wires to use if necessary.

Working on the fuel system as well. Have installed a non- return valve between the fuel pump and fuel filter. Still chasing the elusive pressure bleed down. This causes excessive crank time to get the air out of the fuel rail and fuel into the injectors. Have found fittings that will allow me to install a fuel pressure gauge direct to the fuel rail.

Will be setting the timing this weekend/week. Don't need the fuel system to do this. Will disconnect the fuel pump and injectors so as not to prime the fuel system (yet - want to install the fuel pressure gauge before priming the system) and the injectors don't need to be firing dry. Honda has two timing marks on the flywheel as I have mentioned. The send is to verify the timing with numbers 3/4 cylinders. Going to have the cranking timing set to 10 degrees BTDC, and lock the running timing at 15 degrees BTDC for first start as recommended.

Will be checking the plugs to make sure the coils are firing properly.

TPS has to be set.

Have been thinking about the possibility that I may have damaged an ignition circuit. Using 1/2 ignition circuits. Have found forum reference(s) to changing the pin settings to use 3/4 ignition circuits as primary if necessary. Haven't found the file I would have to modify, but still looking. Have been through the Speeduino.ini and the project .ini file with a generic text editor. Lots of code to view.

Had ordered and received the Bosch 106B coil pack with integral driver. Had to send back because one side was U/S.

Taking my time for this next go round, not to mention the usual start to spring requirements. All things being equal, should get a fire when the fuel system put back in service.