Steve Saunders Goldwing Forums banner

86 SEI Running rough suddenly

7K views 48 replies 7 participants last post by  EAGLECRY24 
#1 ·
Greetings gents,

Had bike down for a bit (6 months) due to time and other mechanical parts change (back wheel etc...)

Bike starts up no problem cold. After it warms up, it runs like poop. Sometimes I can hear the injectors clicking like crazy after it dies.

No fault codes...

Also, after it warms up ( I cleaned the injectors by running Seafoam through the vacuum lines to the injectors). Unplugging the vacuum line (or lines) while warm does not change the RPM. I’ve had to turn up the idle quite a bit to keep it running. I’ve tried the ol intake leak trick by spraying starting fluid around the boots and vacuum lines but I get no rpm change. I’m at a loss. Unless the injectors are clogged causing erroneous mixtures anyway.

https://youtu.be/Rv0T415Ljy8
 
#2 ·
Have an '85 Limited that I have been refurbishing for three and 1/2 years now, do work then ride, repeat as necessary. Once these FI bikes are working well, great bikes.

First thing I would do is set the idle and calibrate the TPS. Next I would suggest balancing the right/left cylinder banks. The idle will be affected so it's a dual calibration.

The vacuum lines go from each cylinder bank up to the PB sensors (MAP). I would put a vacuum gauge on each side and check the vacuum. If the system is working correctly you will draw a good vacuum. The reed valves work quite well and draw a lot of air from the idle air control system for cold starts, and to provide a source of air for the cylinders when the throttle is shut fully such as when decelerating.

If you disconnect one PB sensor vacuum line at a time, there will be no change in the engine operation. Disconnect both at the same time and there will be an issue. The ECU uses the PB sensor signal to assist in engine idle and engine performance below approximately 3000 RPM.

If you don't have the service manuals for your bike, would recommend getting a copy.

Troubleshooting these older FI bikes without the manuals is not an easy task.

No error codes is good. The injectors on these FI bikes are quite robust. Had mine cleaned and flow tested a while back, before cleaning flowed at 64 CC/min, after cleaning 66 cc/min. Shouldn't need to run seafoam or other products through the injectors.

The fuel system light, if it comes on is the equivalent of a check engine light. Had this light come on on another '85 LTD that I put on the road and have subsequently sold. Thought it was the fuel system, but there is nothing on the fuel system to trigger this warning light. Did some other tests such as leaving the air temp sensor disconnected, and took the TPS out of calibration, fuel system light came on.

Best of luck. Cheers
 
#4 ·
( I cleaned the injectors by running Seafoam through the vacuum lines to the injectors).
You cannot clean injectors through the vacuum lines, cleaner has to be pumped through them under pressure.
You are correct sir. I was under the impression that the vacuum check lines provided air to the injectors for a better atomization (Too old of a system for that).

Seems to run worse with increased oil temperatures/higher overall engine temps. (Operating temp).
 
#9 ·
I think Dennis made a comment a while back regarding these FI bikes in that he wouldn't put fuel in one without having the manuals - good advice. PM sent.
 
#10 ·
So, I did a ride this morning to troubleshoot. It was running horribly rough (like a vacuum line is open on both sides). And she was smoking like something fierce.

After a mile or so the “Fuel System” light came on and the idle came up to 1800 or so and it ran wayyy better. I’d love to understand more on how this system works. Of course after turning the bike off and back on again clears the message and the idle drops back down to around 800 rpm and running rough sometimes dieing and starting back up.
 
#15 ·
After a mile or so the “Fuel System” light came on and the idle came up to 1800 or so and it ran wayyy better. Of course after turning the bike off and back on again clears the message and the idle drops back down to around 800 rpm and running rough sometimes dieing and starting back up.
Hello! The light comes on and it runs better. This indicates the ECM saw the malfunction and went into fail-safe mode. (Fixed values) Turning it off erased the learning and it ran like crap again.
Check the code sheet for inputs that cause locked values. You can ignore the ones that "stop" ignition or fuel signals.
 

Attachments

#13 ·
Update,

After running around this morning, letting it cool down to atmospheric temperature then running it again, it is running way better but extremely rich. I think a sensor or a vacuum tube isn’t right.

Reading the book I have reads to clean the throttle valve and sync every 8k miles just like carb syncing...
 

Attachments

#14 ·
Black smoke, white smoke, blue smoke. Once you get the manual, locate the pressure regulator. Check the vacuum line to it. Inspect the line for fuel. The pressure regulator may be ruptured allowing fuel into the intake un-metered. Check the fuel pressure if you have the tools available. Generally speaking, removing the vacuum hose from the regulator will increase fuel pressure @10psi. Pull the hose and plug it with your finger. Note the change in rpm. Increase means your running lean. Decrease would indicate mixture was already close to correct. Leaving the vacuum line disconnected during this test will speed up the engine. More fuel. More air.
 
#16 ·
So... I’m not really sure what’s going on anymore... I checked everything I know to check and put it back together. Ran it for a bit and the idle slowly came up to 2400rpm and I adjusted it down to spec (1000 +- 50). The thing is purring like a kitten for now...
 
#17 ·
Update,

Turns out that seafoam in the gas saved me a heartache of extraneous and unnecessary maintenance. This thing is purring like a kitten now. The link below is a video of the idle (a little low) smooth. After pouring half a can of Seafoam, 3/4 of a big bottle of Marvel and a full bottle of Mechanic in a Bottle on around 1/2-3/4 tank of gas and running it hard and shutting it down till it cooled completely then running it again hard. The brief ride this evening was smooth and powerful. The acceleration was smooth and absolutely no hiccups at idle or high RPM (5-5500). Not sure what exactly cured it but it’s good now.

https://youtu.be/UfP3bv6Hkzs
 
  • Like
Reactions: AZgl1800
#20 ·
Have to ask. How did you adjust the idle. Did you use the idle screw that is on the throttle linkage midway between the right left cylinder bank, or the screw under the CFI cover on the right cylinder bank?

Looked at the YouTube and the engine does sound good. Noticed you have the dash on the oil temp/pressure settings as well. I'd bring the idle closer to 1000.

Cheers
 
#21 ·
Way up in the middle there’s a large flathead adjustment screw. It is the typical location for idle speed adjustment on carb’d models as well. I will bring it up to 1000 +-50 today. It’s gotta have some gas since it’s nearly empty and I’ll adjust it while it’s warm. Thanks for the help everyone! I learned way more about this system now. I’ll be riding it down to meet someone today and I’ll probably post one more time the results of that ride.

Thanks!!
 
#23 ·
Bad news... it’s back. When the fuel system light came on it indicated the TPS.

Still this though: https://youtu.be/Rv0T415Ljy8

I can’t wrap my head around why the injectors would be firing like this when the engine isn’t running.
The injector click is part of an "onboard self test". Ignition on, open the throttle wide and close.
As far as the TPS code goes, it can be an open or short concern. The tps is also a major input to the control unit and affects injector pulse and duration. To eliminate the possibility of dirty connections, I suggest you start all measurements at the control unit to check the entire circuit. Got a book yet?
 
#24 ·
The control unit sends out a five volt reference signal from pin 29 to the tps. It's just a rheostat. It's grounded at pin 3, along with several other components as well as the ecm ground. Since you have injector click by opening and closing the throttle, we can safely assume you have a complete circuit. The third wire goes back to the ecm at terminal eleven. This is the one you want to pay attention to. This is the return voltage from the tps and tells the ecm what position the throttle switch is in. Generally speaking, with the ignition on, and the throttle at rest, the return voltage should be @ one half volt. (point 5 volts). As the throttle is opened, the return voltage should rise to around 4 to 4.5 volts. It should be linear and SMOOTH with NO drop outs. What happens over the years, is the tps develops open areas that the ecm can't read. Since it doesn't know what the throttle position is, it can't deliver the correct amount of fuel. I suspect you have a bad spot or two in your tps. Possible water in the connection(s). Notice there are two screws that can be loosened to adjust the tps, but in your case, a simple adjustment won't help. Verify your results and check back.
 

Attachments

#28 ·
The control unit sends out a five volt reference signal from pin 29 to the tps. It's just a rheostat. It's grounded at pin 3, along with several other components as well as the ecm ground. Since you have injector click by opening and closing the throttle, we can safely assume you have a complete circuit. The third wire goes back to the ecm at terminal eleven. This is the one you want to pay attention to. This is the return voltage from the tps and tells the ecm what position the throttle switch is in. Generally speaking, with the ignition on, and the throttle at rest, the return voltage should be @ one half volt. (point 5 volts). As the throttle is opened, the return voltage should rise to around 4 to 4.5 volts... Verify your results and check back.
Voltage into the sensor is 4.99 (with a non-calibrated voltmeter) measured at the connector. The sensor wire measured at the connector shows 4.5v with it disconnected. Should it measure anything at all at this point? The sensor wasn’t connected.
 
#25 ·
Update:

Ran down about an hour or so today and about halfway she started running rough again. I noticed the fuel economy was low at around 22 mpg at 80mph. I let her sit for a few hours then fueled er back up. She was spitting and cutting out like something fierce. Then otw back home just now about 40 mins into the ride it finally went into limp mode. I recorded the code before shutting er off for the night.

https://youtu.be/xVLqYFEGajg

I noticed the instant fuel economy jump up to 35 or so and she smoothed out again with an idle RPM of around 1200.

I plan to tackle the TPS connection and see if there’s a fault there and idk where else to go from there.
 
#26 ·
Budd, there are Greater Minds than mine helping you here but I do have a question? Do you still have the original 30amp dogbone fuses in this Beastie? If so bear in mind that they can crack and "Possibly" cause this... One is on the starter solenoid and the other is an inline fuse behind the starter solenoid... You might wanna check them just for grins and giggles....
 
#30 ·
I don’t see any way to adjust the TPS. I did find water in the cover... I’ll investigate more. See picture and video.

https://youtu.be/gv2ninXGIog

There’s that Click again. Un-commanded. I monitored the voltage of the TPS output and saw it jumped from 0.48 to 0.7. The resolution of the multimeter doesn’t allow for very quick changes.
 

Attachments

#36 ·
Listened to your YouTube videos. I've been through my '85 LTD with a fine tooth comb, and I would find an interrelation between the injectors and TPS such as what you have unlikely. Stranger things have happened. Looking at the wiring schematic there is no connection that the TPS and injectors share. Wiggling/moving the TPS wiring may be affecting the injector wiring in the wiring harness, something to check.

The TPS can be an issue. You can use a multimeter as has been mentioned, but a scope is much better. I have a large scope that I am learning to use, but purchased a DSO 138 Oscilloscope (assembled) to check these types of issues, you can get one assembled in an acrylic case. Tried it out today on 4 TPS units I have and found some cannot be used. The QA on these units is not the best. Heat also affects the TPS in that it makes it degrade faster. If the TPS gets too bad, you can have the engine have hard misfires that feel like the engine will stop - BTDT.

I changed the TPS on my '85. I used a Dremel tool to cut a slot in the bolt, and then used a screwdriver to extract the bolts, works well. Once you have the TPS loose you can also try a calibration as well. Attached a picture (poor quality) but you will get the gist of it.

You can use aftermarket TPS units such as this one on eBay: https://www.ebay.ca/itm/TPS-Throttl...ash=item2eeba71487:g:pqMAAOSwe7BWvT6K&vxp=mtr

You have to a slight modification to make these work. You drill a small hole in the rheostat arm near the end and insert a small screw. For added strength, use JB weld and build up the area around the screw. Keep the JB weld narrow around the screw. There are three pics showing what I did to the aftermarket TPS. The last pic shows the OEM TPS.

The aftermarket TPS mod worked well on an '85 LTD that I brought out from Ontario last June, and put over 8000 Kms on it before I sold the bike. The JB weld gives excellent structural support to the small screw.

If you have the external alt mod, there is more you have to do to make and aftermarket TPS fit, but that is for another day.

Good luck. Cheers
 

Attachments

#37 ·
Listened to your YouTube videos. I've been through my '85 LTD with a fine tooth comb, and I would find an interrelation between the injectors and TPS such as what you have unlikely. Stranger things have happened. Looking at the wiring schematic there is no connection that the TPS and injectors share.
The connection would be the ecm. If it thinks it sees throttle opening, it will trigger the injectors. Page 10-20 Ltd supplement. If there is water in the connectors or the throttle switch, there would be a reaction. Provided it only happens with the key "on", no start.
 

Attachments

#38 ·
Good catch - my bad. Hopefully not the ECU. The connector under the shelter on the left side of the steering stem can be disconnected to also aid in troubleshooting.
 
#39 ·
The issue I had with the '85 LTD FI I put on the road and rode from Ontario to Victoria BC, a hard misfire is now making sense thanks to eaglecry24 thread and the various posts. Was thinking about this at 0300 this morning - go figure.

I installed a new aftermarket TPS in the '85 LTD prior to the trip across Canada. I was looking at fuel as the culprit, but a member over on the Classic forum mentioned that the symptoms were pointing at the TPS as being the culprit. Still did not change out the TPS because it was new, couldn't be defective. Rode it for some 3000 Km in Ontario and across Canada where, in Brandon Manitoba, I relented and changed out the TPS. Had a spare with me for whatever reason.

The bike would run well above 3800 RPM and in the lower 2000ish range. It did not like the heat of the day either. It ran better when it was cool out, or first thing in the morning, and because of this I was hoping for cool weather, even rain.

I never checked the injector firing as eaglecry24 has done because I was not looking in that direction.

Fuel economy was affected as well.

The TPS failure affected the injector firing and consequently coil firing. After the TPS was changed the difference was dramatic; however, was still gun shy so to speak and waiting for the other shoe to drop, but the rest of the ride to Victoria was great - not counting the severe cross winds that can be encountered crossing the prairies into the mountains.

Back to the issue at hand, TPS replacement.

Cheers
 
#40 ·
Just found this on GoldwingDocs:

“I do think you're onto something with the PB's. If you have the sup manual go through the steps in it for the PB system carefully. It's time consuming but can show a lot of small problems, especially if you do the calculations for voltage based on altitude and temperature. some other things to think about though... Do you get high idle when the bike is cold? There is a thermostatically operated air valve which acts as a choke while the bike is cold. There is a sensor on this valve which tells the computer when it's open (cold) and in turn lengthens the duration of injector opening, thus richening the mixture. As the coolant temp rises it closes the valve and 'closes the choke', the computer recognizes this and gives less fuel (the coolant temp sensor has the same effect so that's something else to check). If this valve is stuck or the sensor has failed it could be telling the injection system that it needs more fuel. On my LTD this valve was stuck open, as a result I removed the valve and installed a small plastic ball valve inline which now acts as a manual choke and I have no problems. Also with the TPS, I had to modify first to the honda car mod (which ended up not holding up and I switched to the BEI/Mouser tps) I set the throttle body opening to the spec in the sup manual and set the tps voltage within book spec. I found that I was still running fairly rich. Now understanding the CFI system it is a fairly redundant system. It has no idea how the engine is actually running, all it knows is what fuel curve to send it as the voltages change on specific sensors and doesn't actually recognize the effects on it (no o2 sensor to know how it's burning). Because of this the tps can actually be set down to .400 vdc before the system will throw a code for it being out of range. This means we can trick the system into running much leaner by increasing the opening on the throttle body and shortening the injector duration by reducing the tps voltage which in effect shortens the fuel curve. I found that if I set mine down around .440-.450 vdc I get good plug burn with proper timing curve on the electrode and get about 3-4 mpg more. You can play around with it to suit your bike, just make sure you check plugs regularly until you find a happy spot for it.”

It’s good info. I’ll look into the air valve as well.
 
#41 ·
The throttle switch doesn't have a code for "out of range". It's simply open or shorted. Setting the throttle switch voltage lower does indeed change the injector duration throughout the entire band. Lean isn't necessarily good. Especially with today's fuels. Sometimes it's like flushing two toilets, when you only use one........... I've never been a big fan of trying to cheat the system. If you have 1400-1600 rpm cold and it settles to the correct rpm warm, there's nothing wrong with your aux air regulator. The only way I know to get 3-4 more miles per gallon, rests with your right hand.
 
#42 ·
I guess running all those cleaners through the fuel system isn’t a good idea since my fuel shutoff valve is leaking now and I can’t find one. Anyone wanna buy a bike?

I’ll be removing the TPS and disassembling it to clean and test. Then put it back together for testing.

I was gone on vacation for a bit and came back to a mess on my driveway... (see attachment)
 

Attachments

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top