GL1100 with throttle bodies (Fuel Injection)

[socrace]

Here's my cold fuel settings:

warmupwizardmenu.PNG

primingpwmenu.PNG

crankingpwmenu.PNG

aftstartenrichmenu.PNG

aftstartenrichtapermenu.PNG

Also, here are my ve and spark tables. Just copy and paste into a text file for each, give them a name and .table extension, then they can be imported into tunerstudio from the table editor screen:
FUEL TABLE:
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<tableData xmlns="socrace's vetable for gl1100 v1.0">
<versionInfo fileFormat="1.0"/>
<table cols="12" rows="12">
<xAxis cols="12" name="rpm">
500
1000
1500
2000
2500
3000
3500
4000
5000
6000
7000
9000
</xAxis>
<yAxis name="kpaix" rows="12">
30
50
55
60
65
70
75
80
85
90
95
100
</yAxis>
<zValues cols="12" rows="12">
58 58 59 60 60 61 62 63 64 64 64 62
62 62 63 64 65 66 67 68 69 69 69 67
65 65 65 66 67 68 69 70 71 72 71 69
66 66 67 68 68 69 70 72 73 73 73 70
67 67 68 69 70 71 72 73 74 75 74 72
68 68 69 70 71 72 73 74 75 76 76 73
71 71 72 73 74 75 76 77 79 80 79 76
75 75 76 78 78 79 80 82 83 84 84 80
81 81 82 83 84 85 86 87 88 90 89 86
86 86 87 88 89 90 91 92 94 95 94 91
89 89 90 92 92 93 94 96 97 98 97 94
91 92 93 94 95 96 97 98 99 100 99 96
</zValues>
</table>
</tableData>
SPARK TABLE:
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<tableData xmlns="socrace's sparktable for gl1100 v1.0">
<versionInfo fileFormat="1.0"/>
<table cols="12" rows="12">
<xAxis cols="12" name="rpm">
500
1000
1500
2000
2500
3000
3500
4000
5000
6000
7000
9000
</xAxis>
<yAxis name="kpaix" rows="12">
30
50
55
60
65
70
75
80
85
90
95
100
</yAxis>
<zValues cols="12" rows="12">
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
13 13 13 21 30 38 38 38 38 38 38 38
</zValues>
</table>
</tableData>

[Casper]

Good work, thanks for posting the detailed information. Both you and icb are way-way ahead of me. I hope to make some EFI engine noise in the Mar/Apr time frame. Riding season starts in mid May in these parts.

Cheers,

[socrace]

There's one other major calibration item needed, what microsquirt calls the "required fuel". This is the estimated pulse width needed, calculated by tunerstudio, probably using something close to this formula:

pulse_width_seconds = engine_ci *.156/(cyl_num * 14.7 * inject_lbhr)

This number is used for main fuel calculations and also gets passed to other places in tunerstudio and the megasquirt code (like cold enrich, afterstart enrich), so it's important to get it entered and approximately correct. The tunerstudio menu for the GL1100:

2c9d05a2-ebc1-48f0-a6ba-5b976431effe.JPG

To use these variables as shown, the trigger used by microsquirt needs to be one pulse every 180 deg crankshaft. This can be done by combining the pulses from both ignition coils using 2 diodes or by using the voltage drop on the ballast resistor (method I'm using).

[JBz]

Hi Bob.Can you print the complete algebra formula with the actual numbers you are using. Are Your injectors 24lbs. I went to the algebra calculator online but this is where I get stymied the math. I used 74*.156/4*14.7*24

Thanks jb

[socrace]

Hi Bob.Can you print the complete algebra formula with the actual numbers you are using. Are Your injectors 24lbs. I went to the algebra calculator online but this is where I get stymied the math. I used 74*.156/4*14.7*24

Hey JBz, here's my engine:

6.7 ms = .0067 sec = 67*.156/(4*14.7*26.7)

What I get for yours:

8.2 ms = .0082 sec = 74*.156/(4*14.7*24)

These agree with what TunerStudio calculates for "required fuel", which is supposed to be the actual pulse width needed when the engine is running at 100%VE.
What I have found is I have to reduce my calculated number of 6.7 ms down to 5.5 ms, in order for the VE table brought over from the Weber ECU to run right. The megasquirt documentation does say to use "required fuel" as the global fuel adjuster, there is no other way.
Not sure why it has to be adjusted so much. One answer is that these injectors are actually flowing quite a bit more than 26.7 lb/hr! I remember when I first set up the Weber, that converting from a chevy small block gave unexpected rich results, so this megasquirt result is consistent with that.
Using the global fuel adjustment works to get it running right, although would be interesting to do a timed injection of one of these injectors into a measuring cup and see what the thing actually flows.

[socrace]

And here's yet another update configuration page. This one sets up accel enrichment (aka "pump shot" on carbs). In particular, the tpsdot driven number should be set to 100% (can't remember what the default was, maybe 50%) and the tps threshold set to 30% (default was 200%). This page brought to us courtesy of the 65 deg weather today!

accelenrich.JPG

[irishcarbomb]

Nice. I was going to mess with those pre-turbo but figured I'd wait.

I did notice some lean spikes with throttle application on the datalog and wide band without the accel turned on.

The default values are set high so you can tune the fuel table without the accel enrichment messing with the values.

[Casper]

This is really fun to watch, I had hoped I would be closer by now, but as we all know life does have a way of getting in the way of your hobby time. But its great to watch others sorting out the EFI issues. If you think about it, the accelerator pump on a lot of carbs, was based on a 'manual' TPS and how quickly you you were 'mashing down' on the throttle. So a combo of actual TPS and the rate of change of the TPS, I think mainly a function of the rate of change to the throttle position. Any EFI system needs to somehow emulate the same characteristics.

Cheers,

[JBz]

Got mine all hooked up and getting one or 2 sparks off the plugs when I first turn it over and that's it. No spark after that. I had my laptop hooked up and I thought I had the 2 talking to one another on the black page. After about 4 hours I didn't get anywhere programming. Computers //// I saw a video on utube how to set up the programming but just too fast for my noggin. . Ill try again soon........jb

[socrace]

Hey ICB, Casper and JB, been following your efi projects, nice work.
I think the accel enrichments shown above should work with boosted efi as well, since the effects of accel enrichment are basically done above about 3000 rpm.
JB, the fact you're getting sparks while cranking is a good sign.
To anyone reading this, feel free to PM questions.

[Whiskerfish]

I can talk about Turboshaft Jet engines: Compressor Bleed Air (P-3) and Intake Bleed air (T-2) and EGT, OAT, power turbine speed verses free turbine speed, ESCL position and torque sharing and all that but you guys are speaking Greek to me!!!!

Keep going I am trying to keep up

[socrace]

Hey Whiskerfish, installing a turboshaft engine in a Goldwing would be fun, especially for when those tailgaters get too close!

Thought I'd also post about installing a 36-1 crank wheel (finally) and getting microsquirt's spark control working on my project. Watching irish car bomb's posts and experience on this really helped, thanks icb!

Didn't take many pictures this time, but here's one of an early mock up of the wheel and sensor mounting.

Image0277.jpg

The final version looks nearly the same, except turned down the teeth a little to clear the housing mounting screws (and used a larger wheel mounting screw). The sensor finally used was a Hall sensor from DiyAutotune, mounted in about the same location as the Ford inductive sensor shown in the photo. The original vacuum advance mount was cut off and turned 90 deg to hold the sensor. Of course, the swing arm has to be pulled to get at this thing, so a fair amount of wrenching needed.

After making a few runs, decide to set ignition to emulate the stock mechanical advance of a '81 GL1100 for now, will add some vacuum advance to it later:

SparkAdv361.PNG

My previous plan to use the stock dual pickups with microsquirt running in "dual spark" mode did not work out. Problem was that microsquirt cannot read a single tooth vr sensor, it can only do so with a hall sensor. Too bad, would have been an easy hookup with no need to pull the swingarm. Am still using the B&G code, and the bike is running very well. Idle is noticeably smoother now, probably because the 33 year old mechanical advance is gone!

[irishcarbomb]

Interesting. I tried to run my trigger from there but was getting all kinds of noise.

That's about how my spark table looks. 10 degrees at idle and 38.5 at 3000+ rpm. Then i let tunerstudio interpolate the rest of the values. Seems to work pretty well.

[JBz]

Just curious Bob if the previous fuel charts might be compatable with ms2 extra and the 36minus1 wheel. Im thinking the the standard injection chart might be different with the 36-1 rather than running off the coils?? If so ill burn them . Thanks JB

[socrace]

The engine is running about the same, using the previous maps with the new ignition. The maps are on the rich side, so should be pretty close to what your 1200 needs. As for using ms2extra with these maps, I'll have to check into that and get back to you, but I'm near certain they will.
Been thinking about switching to extra also, since it has more features and is open source, unlike B&G. Maybe this weekend.