Another GL1100 Megasquirt EFI project

This is where discussions of EFI systems, Dual Carbs, Single Carb mods and plans can be discussed and shared.

If you are working on or have a problem with stock carbs then your questions do not belong here, please post them in the Goldwing Tech Discussions Forum.

Moderators: sunnbobb, Neil, Forum Moderators

Post Reply
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#31

Post by thesteve »

Yup, there are a ton of high impedance injectors that will fit in directly. These are top feed injectors, I forget the exact cc/min rating but it was in the low 200s I believe. Lots of early 90s Japanese cars share the same injector style. You just need to know how many cc/min so you can enter the correct value into tunerstudio.

So far no issues with warmup. It's such a small hole I don't expect too much of an issue. Watching the logs it's still touching 200 at long lights and not cooling excessively once moving.

That surprises me about the vr input issue. On the full size megasquirt there are adjustable potentiometers that allow lots of adjustment in the sensitivity and switchpoint/zero-crossing of the vr signal. I'd be surprised if it didn't work; it's even listed in the MS manual as a valid sensor type.
Last edited by thesteve on Tue Nov 10, 2015 11:49 am, edited 1 time in total.
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#32

Post by thesteve »

Also, thanks to both of you for your build threads. I definitely borrowed ideas and concepts from both. That's actually why I decided to drill the thermostat. I saw you had issues with the sensor going cold but I didn't want to lose the factory gauge by swapping positions.
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#33

Post by thesteve »

In case you were concerned that this project didn't have enough electronics in it, I found this online:
Image
Image

It's an Arduino (Teensy, to be more accurate) powered gauge cluster. It uses the Megasquirts onboard CAN communication system to get data. I'm thinking about building a weather resistant one to use on the bike. I have a GPS module laying around to power the speedometer, and the MS provides pretty much all of the needed info except for fuel level, indicator lights, etc.

I'll start playing around with my arduino and the basic screen I have. If I can get something neat working I'll see if I can find a bright full color LCD and fit it to the bike.
User avatar
wingrider
SUPER BIKER!!!!
SUPER BIKER!!!!
Posts: 3724
Joined: Thu May 24, 2012 12:15 pm
Location: Maple Lake, MN USA

Re: Another GL1100 Megasquirt EFI project

#34

Post by wingrider »

How's the tuning been going? Run into any hicups? Thanks again for bringing all this up! :)
2002 1800 Wing
2001 Valkyrie Interstate
1978 1200 EFI Wing with Motorvation Coupe Royale sidecar
1977 1200 Wing cafe/bobber project
1974 Suzuki GT550
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#35

Post by thesteve »

Tuning has been going well. I haven't even had to put much time into it, just a couple hours in and it's very rideable.

I took a break for a couple weeks because I was running into fuel starvation issues at high load/RPM. No matter what adjustments I made, it was leaning way out past a certain point and getting worse as time went on. I decided it was most likely the original fuel filter which I was using as a pre-strainer for the pump. Paper element filters are way too restrictive to be used in front of the pump, but at the time I didn't have anything better and I knew the amount of rust in the tank would kill the pump quickly.

I found a screen-type fuel filter on Amazon to replace the pre-filter I'm using now. It's also washable so I can clean it as needed. It should flow much more than a paper element, so the pump should be quieter and happier too. Keep in mind that this is not the main filter; I'm still running a proper automotive style filter on the high pressure side of the pump to protect the injectors. The coarser media in the prefilter is only protecting the pump itself.

Link to the filter: http://www.amazon.com/gp/product/B000VS ... ge_o06_s00 . There are several like it. I went with 5/16 hose barbs to match my hoses.

So I finally received and installed the filter about 2 days ago, but haven't ridden with it yet. My laptop's battery was dead so I didn't do anything other than start it up and check for leaks. I should have some time today to play with it.

Other than that, I'm trying to think of what I've been up to with it. I got the idle fairly locked in by:
-disabling O2 correction at idle RPMs
-tuning the map in that section by hand (no autotune or log analysis)
-backing off the throttle stop screw to allow the IACV more control
-adjusting the movement speed of the IACV.

It now starts right up and settles into a very steady idle. The bike rides very well when not under heavy load at high RPM. I'm pretty confident that the new filter will take care of that. I haven't taken it more than about 10 miles from home yet; there are still some quirks of the bike itself that need to be worked out. I finally got the brakes to be not-terrifying, it has a headlight now (no rear turn signals though), and anything I forgot to tighten has already let me know about it one way or another. Before taking it further, I need to be sure it won't leave me stranded (pretty confident here) and I'd like to do something about the clutch. The pull on the clutch is one of the heaviest I've ever seen on any motorcycle. It's smooth and doesn't appear to be catching or sticking, but it's hand-numbingly heavy. I'm hoping some rerouting of the clutch cable will solve it. Putting the lower bars on the GL has caused a bunch of little issues like this since the original bars are so tall.

By far the biggest issue is that there just aren't enough hours in a day.
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#36

Post by thesteve »

Nearly forgot about the gauge cluster project. I'm still in the tinkering phase of that. I rigged up a basic screen to display variables sent to it via serial communications, and it works well. Serial is used between my PC and the Arduino/Teensy, but I need to order some parts to get CAN communications working (which is how the Megasquirt communicates to a gauge cluster). So I haven't yet done much with that other than some bench testing.
User avatar
wingrider
SUPER BIKER!!!!
SUPER BIKER!!!!
Posts: 3724
Joined: Thu May 24, 2012 12:15 pm
Location: Maple Lake, MN USA

Re: Another GL1100 Megasquirt EFI project

#37

Post by wingrider »

Good deal! As far as the clutch...pay close attention to how the cable is run, and it probably wouldn't hurt to lube it. Have heard stories of very difficult pulls because it wasn't routed correct.
2002 1800 Wing
2001 Valkyrie Interstate
1978 1200 EFI Wing with Motorvation Coupe Royale sidecar
1977 1200 Wing cafe/bobber project
1974 Suzuki GT550
socrace
Titanium Member
Titanium Member
Posts: 334
Joined: Sun Aug 16, 2009 11:46 pm
Location: Springfield, Ill.

Re: Another GL1100 Megasquirt EFI project

#38

Post by socrace »

My efi build had the same high rpm problem, also caused by a rusty tank clogging the prefilter. Spent a few years messing with different filters (and limping home one time at 1200 rpm). Finally, last year, loaded up a prefilter in only 10 miles. So pulled the tank and filled it with 4:1 water and muriatic acid, 2 hours later rust all gone. That was last year, no problems since. I guess I thought that the filters would gradually reduce the amount of rust in the tank, but actually the rust seemed to get steadily worse over time.
1981 Goldwing Standard - efi'd, other stuff
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#39

Post by thesteve »

I actually stripped out the majority of rust with the electrolysis method followed by pressure washing what I could reach, but since there was over a month in between doing it and filling the tank with fuel it lightly rusted again. The big chunks are gone, but I didn't want to risk the pump so I started with the original paper filter. Even clean, the paper type elements are just too restrictive. I think the new prefilter should last a year or more in between cleanings now that the initial gunk has been washed out.

I do have a bunch of muriatic acid here as well, but containment and cleanup is a pain so that's the last resort.
socrace
Titanium Member
Titanium Member
Posts: 334
Joined: Sun Aug 16, 2009 11:46 pm
Location: Springfield, Ill.

Re: Another GL1100 Megasquirt EFI project

#40

Post by socrace »

Sounds like you're good to go with the fuel tank.
One final thought on filter clogs. It's easy to forget that on most efi systems, fuel is being pumped at maximum flow levels needed for full engine power all the time. Excess fuel is routed back to the tank. So fuel is constantly churning round and round at high flow levels and running through the filters at high flow levels, even at idle. If a filter gets clogged, it will show up immediately as a drop in fuel pressure, it doesn't matter if the engine is at high rpm or not, since the flow is the same no matter what the engine rpm is. So a fuel pressure test can tell you if a filter is restrictive, even with the engine off, as long as the pump is running with full voltage.
1981 Goldwing Standard - efi'd, other stuff
User avatar
82aspy
True Blue Steel Biker
True Blue Steel Biker
Posts: 2279
Joined: Sun Jul 01, 2012 8:12 pm
Location: canuck

Re: Another GL1100 Megasquirt EFI project

#41

Post by 82aspy »

:lol: action1 :orange
thesteve wrote:In case you were concerned that this project didn't have enough electronics in it

Image
Image
"there is a Darwinian element"
Yoda said, "There is no try or not try; there is only do or don't do."
=====================
ngw quotes anim-cheers1 - \ "I will strip the beast" \ "the swingarm is just a hunk of metal" \ "gonna maybe ...dunno yet" \ "It's never done, thats the fun." \ "something about this site brings me back"
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#42

Post by thesteve »

Quick update. I took the bike out for a ride today and attempted to retune the top end of the table. No luck. I'm reviewing the datalogs and I see a ton of tach spikes causing issues. The Megasquirt has an adjustable fuel cut for a rev limiter, and the tach spikes were telling it to cut fuel due to excessive RPMs. Disabling the fuel cut would bandaid this, but the computer would still try to correct for the high RPM so the real solution is to have a look at the tach input circuit and see if there's any way to smooth it out a bit. The tach spikes were hitting 30,000 a few times.

What's interesting is the spikes don't appear to necessarily be correlated directly to RPM. Sometimes I can run at 5500RPM with no issue, other times it's bucking like crazy at the same RPM. High load seems to affect it a bit, but again it's not a definite thing. I'll have to see what I can do to smooth it out a bit. If I can't, I may be better off triggering from the VR sensor on the pulse generator.

The good news is the fuel pump is much quieter now with the new filter. And the rest of the tuning is looking great. Everything's nice and smooth until I hit those tach spikes. I had one stall while pulling in the clutch while coming up to a stop sign, but after it warmed up I could not reproduce the issue.
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#43

Post by thesteve »

For those unaware, this is a tach spike:

Image

The white line is RPM. At the blue marker, I'm running 6060RPM. The top of those spikes are 11794RPM. Obviously, it's not possible for a Goldwing's engine to bounce from 6000 to 12000 RPM and back again in the fraction of a second covered by each spike. Unfortunately you can see that the spikes tend to happen in clusters. This leads to 4-8 completely missed injection events, which causes huge rideability problems.

The green line is throttle position, and the red line is vacuum. You can see the RPM signal is mostly smooth until about 6000 RPM and I closed the throttle after the first few misfires. Even after the engine slows a bit towards the right side of the chart, there are still a few tach spikes when I begin rolling back on the throttle.

The only other interesting thing is that rideability at high speed seemed to get worse as time went on. This may have been autotune trying to compensate for the tach spikes and messing up my table more and more as the ride continued.
socrace
Titanium Member
Titanium Member
Posts: 334
Joined: Sun Aug 16, 2009 11:46 pm
Location: Springfield, Ill.

Re: Another GL1100 Megasquirt EFI project

#44

Post by socrace »

The Microsquirt code has some "noise cancelling" features that are known sensitive to high tach signal duty cycle. Duty cycle (aka coil dwell) hits a max around 5k rpm on the stock GL1100 system, right where those spikes start. Could try adjusting those noise features using Tunerstudio, might get lucky!
1981 Goldwing Standard - efi'd, other stuff
thesteve
Zinc Member
Zinc Member
Posts: 40
Joined: Fri Sep 04, 2015 10:25 am
Location: Phoenix AZ

Re: Another GL1100 Megasquirt EFI project

#45

Post by thesteve »

That sounds like the issue. I'll see what noise filtering options I have in tunerstudio.
Post Reply
  • Similar Topics
    Replies
    Views
    Last post

Return to “Unique Carburetion”