GS1000G mods- big bore and fuel injection

[posplayr]

It then reported a version mismatch from the Microssquirt to the tune I had saved, so I tried writing a new tune. Now it no longer connects to my laptop, so 1 step forward and 1 step back. I've updated the USB to serial drivers to use the recommended D2XX drivers, apparently these are as touchy as a dry leaf in winter. It connected momentarily and then nothing. From reading posts in the Tuner Studio forum, I am going to buy a $10 PCMCIA to serial adapter and bypass the entire USB to serial debacle. Now I remember why this was on the back burner for so long.

That really sounds like a step back.

 

[GS1000G Shopper]

I found that the later versions of the Bowling & Grippo firmware are useless for me. I cannot get the laptop to connect via the USB port or the express card (my laptop is not PCMCIA) interface. The support forum for it is a ghost town. I returned to MS2 Extra, using the latest version which is 3.3.3. It installed fine and the support forum is great.

In the process of doing this, apparently I missed a dwell setting and fried one of the coils.

The next step is to replace the coils and plug wires. I'm going to try a pair of used Accel 140403 3 ohm coils with a set of 8.8mm wires. All of the reviews I read about them seemed positive.

I found Emgo makes aftermarket coils for around $30 each street price but could not find much in the way of reviews on them. I'm going to try a pair on my 1980 GS1000G along with a set of 7mm plug wires I had on the shelf and a new set of NGK plug caps.

 

[Thumper88]

I have been working on my own GS1100E fuel injection project, this thread has helped quite a bit with planning the project--it is now coming to fruition.
I started with Arttu's crank trigger, as it seemed like a no-brainer to me...after seeing all the trouble you had with various setups.

One note on frying the coils, when you loaded the MS2 Extra firmware, did you remove the fuse to ignition coils, as the loader says to do? Otherwise it is very possible they were being powered during the flash.
I also had purchased a used set of Accel 140403 coils, but ended up sticking with stock--couldn't find a proper way to mount them...they are oddly configured.

 

[GS1000G Shopper]

That's probably what did it, although one was within specs. I also found one of the plug caps was out of spec, I think that was just due to old age. One of the 140403 coils was bad so I returned them. While I never tried to test-fit them, I think with some bracket work they could be fitted OK.

I've done a lot of reading about coils the past few days and am going to see if the Super Coil version of a Neon coil will fit in there. If not, I found a dual coil for wasted spark from a 90's model Subaru (Standard Motor Parts # UF-159) that's very compact. The plug wire terminals are different that what I have but Rock Auto sells matching wires cheap.

I'll try & take some pics of whatever I end up with.

 

[ArttuH]

One note on frying the coils, when you loaded the MS2 Extra firmware, did you remove the fuse to ignition coils, as the loader says to do? Otherwise it is very possible they were being powered during the flash.

The stock coils should be quite tolerant for "wrong" use like staying powered during firmware load. 3 ohm primary resistance limits current to low enough level to prevent them burning or melting if they stay powered for short period (like some minutes). I think even the stock ignitor may keep coils powered when the engine is stationary in some situations. Of course it isn't advisable to leave them for long periods as current still heats them and that may cause damage eventually.

In contrast coils with low primary resistance, like many car coils and some performance coils, may burn or burn the ignitor components within seconds. They are designed to run with precisely controlled dwell time in milliseconds range. If you exceed maximum dwell time the current will rise rapidly to harmful level.

 

[GS1000G Shopper]

The Super Coil and the aftermarket Subaru coil are under 1 ohm each for primary resistance, the SC is about 12K secondary while the Suby is about 21K. I'm looking at 3-3.5 ms for dwell, which calcs out in the 5-6 amp range. I found the connecter needed for the Suby coil tonight, and at a good price. If either works I will update the thread. I'm also going to try a pair of Emgos on my other 1000G with some new wires and caps.

 

[posplayr]

The Super Coil and the aftermarket Subaru coil are under 1 ohm each for primary resistance, the SC is about 12K secondary while the Suby is about 21K. I'm looking at 3-3.5 ms for dwell, which calcs out in the 5-6 amp range. I found the connecter needed for the Suby coil tonight, and at a good price. If either works I will update the thread. I'm also going to try a pair of Emgos on my other 1000G with some new wires and caps.

You realize that 1 ohm and 12.8 volts is 12.8 amps right? Basically you are talking about enough current to run your whole bike with normal coils including charging the battery. You won't be able to leave that on very long at all. Even 3 ohm coils only average about 3 amps while running, but will get upwards of 4 amps is sitting still.

 

[GS1000G Shopper]

Yes. As a stock replacement, a low-ohm coil would certainly cause those problems. The Microsquirt allows dwell (and accordingly amperage) to be fine-tuned and limited. Default dwell is 3 ms. From this page:

To control your coil for maximum performance and reliability, you must set the maximum dwell parameter for your MicroSquirt? controller. A couple of points are that few automotive coils are designed for more than about 6-6.5 Amps, and reduced coil resistance has been used to shorten dwell times as engine rpm limits have risen over the years (shorter dwells means higher rpms before the dwell is cut back), so as a result most coils are between 2.0 and 4.0 milliseconds, with a general trend towards shorter times for more recent designs and longer times for older designs (wasted spark and coil on plug can be different, of course, as they fire less often and thus don't get cut back until a much higher rpm).

That page has a calculator that shows the amp draw based upon input data. I do not have inductance data for the Subaru coil, so I will have to try it and see. If I recall correctly when I laid out the wiring and amp draws, I budgeted 5 amps for the coils. Using 3 ms of dwell with the Super Coil will keep me under that (4.63A). If it turns out the draw is too high to charge the battery with both, then I'll have to try the Emgo coils.

EDIT
This page has some detailed and relevant info. The author tested a number of motorcycle coils using resistance in ohms (O) and inductance in milliHenries (MH) and listed specs for each. Using their numbers with the calculator, I can determine amps (A) drawn at dwell settings:
Kawasaki points coil:
O: 4~5.1
MH: 13~16
Using the calculator and average values, this returns 2.99 MS of dwell @ 1.59 amps.

Kawasaki electronic ignition coil (should be similar to GS OEM & the Emgo replacements):
O: 2.4~3.2
MH: 7~8
Using average values, this returns 3.22 MS of dwell @ 3 amps. To get 3 MS, it would draw about 2.88 A.

Accel (motorcycle) Super Coil:
O: 3.1~3.8
MH: 15~18
Using average values, this returns 9.21 MS @ 3 A. At 3 MS, it draws only 1.62 A.

As a side note, the Accel Neon Super Coil is rated at .54 O and 7 MH, so it would draw 4.63 A for 3 MS of dwell.

Dyna gray coil:
O: 2.2~2.9
MH: 8~9
Using average values, this returns 3.33 MS @ 3 A.

Neon aftermarket coil:
O: .6~.7
MH: 4.98~5.03
Using .65 ohms and 5 MH, this returns 1.36 MS @ 3 A.
To get 3 MS, the draw is right at 6 A. At 2 MS, the draw is about 4.25 A.

Dyna mini-coil:
O: .5~.7
MH: 3.3
Using an average ohm value, this returns 3 MS @ 8.4 A.

Looking at the end results, I'd say that the Dyna min-coil and the Neon aftermarket are excluded. The Accel motorcycle Super Coil gives the most dwell for 3A of current. The reviews I've read about it for performance seem to be favorable but like a member commented here mounting is the issue. From a power usage standpoint, the Kawasaki points coil uses the least- but does it provide optimum performance? Wasn't there a reason why the manufacturers (Suzuki & Kawasaki) lowered the primary resistance when going to an electronic ignition? I'm sure there is no free lunch here.

 

[GS1000G Shopper]

The stock coils should be quite tolerant for "wrong" use like staying powered during firmware load. 3 ohm primary resistance limits current to low enough level to prevent them burning or melting if they stay powered for short period (like some minutes). I think even the stock ignitor may keep coils powered when the engine is stationary in some situations. Of course it isn't advisable to leave them for long periods as current still heats them and that may cause damage eventually.

In contrast coils with low primary resistance, like many car coils and some performance coils, may burn or burn the ignitor components within seconds. They are designed to run with precisely controlled dwell time in milliseconds range. If you exceed maximum dwell time the current will rise rapidly to harmful level.

I'm not using the stock ignitor beyond the box that originally housed it- it was gutted and a new board installed. The BIP373 transistors I'm using have a built-in thermal limiter and are highly-regarded by Megasquirt folks.

 

[ArttuH]

Looking at the end results, I'd say that the Dyna min-coil and the Neon aftermarket are excluded. The Accel motorcycle Super Coil gives the most dwell for 3A of current. The reviews I've read about it for performance seem to be favorable but like a member commented here mounting is the issue. From a power usage standpoint, the Kawasaki points coil uses the least- but does it provide optimum performance? Wasn't there a reason why the manufacturers (Suzuki & Kawasaki) lowered the primary resistance when going to an electronic ignition? I'm sure there is no free lunch here.

Why you think those two coils are not suitable? If you are worried about current draw keep in mind those are just peak values of the current. Average current during dwell period is roughly a half of peak current. And shorter the dwell time longer the coil stays unpowered over an engine revolution. So let's assume the engine runs at 6000 rpm and dwell time is 3 ms. One engine revolution takes 10 ms so current is flowing through coil only 30% of time. So if peak current is 10 A and average over dwell 5 A then total average current is only 0.3 x 5 A = 1.5 A. This is for one coil though so total ignition current draw would be 3A. So high current coils won't cause significant load for the charging system. In addition there is no need to stick to that default 3 ms dwell value. You can use (almost) any dwell, whatever your coil requires.

On the other hand risk of electrical interference increases within coil current so it isn't good idea to go extremes without good reason.

Coil energy available for spark depends on inductance of the coil and primary current at the moment when spark is released. The formula is (I*I*L)/2 where I is current and L inductance. So if you want to increase spark energy you can increase either current or inductance. But increasing inductance also slows down current rise during dwell and therefore increases required dwell time. So you can't get far with inductance as there is only limited time available to charge the coil. Another approach is to increase current and even decrease inductance to get higher current in shorter dwell time. Since current affects to energy in power of two it easily overcomes effect of decreased inductance. Since 80's engine manufacturers have been increased spark energy to get more reliable and consistent burn process. That's the reason why modern coils tend to have relatively low inductance and resistance and high current.

 

[coombehouse]

Why not keep it simple & use Dyna green coils. I use them & they work perfectly with Microsquirt.
Not keeping It simple is why this thread is so long & the bike still doesn't work.

 

[GS1000G Shopper]

Arttu,
Thanks for the detailed information.

 

[GS1000G Shopper]

Coil option fitment

Coil option fitment

I am going to try several coils for fit. I received the quad-outlet Neon Super Coil ($45 shipped from Rock Auto) & Subaru coil ($25 shipped on eBay, 90's model Legacy, Standard Motor part # UF-159) along with a pair of the dual-outlet Emgos ($30 shipped each on eBay, Emgo part #24-37810) today. All physically fit in the room under the top rail and do not stick out past the side rails. The Emgos of course would bolt-on since they are nearly identical to the OEM units. They use .250" male terminals for primary wiring, and have threaded collars for the secondary side, so plug wires are replaceable.

I checked the primary & secondary resistance and here is what I saw:
Neon-
Primary: .7~.8 ohms
Secondary: 13.25 K ohms

Subaru-
Primary: 1.0 ohms
Secondary: 17.2 K ohms

Emgo-
Primary: 2.5~2.6 ohms
Secondary: 14.3 K ohms

It looks like I can make a pair of .5 x 1.25" aluminum rails to mount the Neon or Subaru coils using the OEM mount points and have enough room left over for a bonded washer (rubber/metal) to allow for some vibration isolation. I'll upload some photos once this part is complete. Here's a photo of all 4 coils:

 

[GS1000G Shopper]

As a side note, on my blue 1980 GS1000 I installed the Emgo coils, new NGK caps, and fabricated the plug wires from a new in box 7mm V8 set I had on the shelf for a number of years. I had to cut off the old coil primary wiring with connector and crimp/solder some .250 terminals, which I then covered with some heavy gauge heat shrink. They measured 2.8 ohms primary and 32-33K secondary. It had been hard starting, but that is no more, it lights off right away.

 

[coombehouse]

One other part I'm looking at now is a timing ring so I can verify when the #1 plug is firing using a timing light. I played around with the e-Machine Shop software and created a 2.5" ID / 3.0" OD ring with timing marks from 0-45 degrees. While it would be cost-prohibitive to have this made via a machine shop, I found Shapeways is an online 3D printing service and they are much less expensive. They offer several materials, to include brass or bronze. It looks like this could be made without the numbers out of metal for around $35, or under $5 from plastic, but the max temperature for the plastic is around 180 degrees. I'm not sure how hot the timing wheel area gets, but I'd rather spend a little more for metal.

Once created, the ring would be cut in half and the upper part aligned for TDC and then mounted to Arttu's plate by drilling a couple of small holes on each end.

Here's the drawing:

I am afraid you are overcomplicating this. I just printed the graduations on some paper & glued it onto the trigger plate. There is next to no heat in this area.
Nick

 

[GS1000G Shopper]

The Neon coil mount is complete. Here is the info:
Tools:
Drill or drill press
5mm drill bit
6mm x 1.0 tap
Sawzall

Materials/hardware:
Aluminum bar, 1.25" x .5" x 12" cut to size.
Bolt, coil to mount: 6m x 1.0 x 50mm (4)
Bolt, mount to frame: 6mm x 1.0 x 20mm (4)
1/4” bonded rubber/metal washers: 4 (These are available at Lowes or a hardware store)
6mm flat washers: 8
6mm lock washers: 8

Once cut to match the existing coil brackets, I marked it for the top mount holes and also for cutting away excess material by the front mount. While it would likely clear the tank, it was not too big of a job to cut the front part down. A small piece of aluminum was left over after the first cut. The overall width to mount the Neon coil is 3.75", so placing the remaining piece of aluminum between the other two allows the exact width needed (1.25" x 3) for marking the holes underneath. I aligned the coil with the rear of the bracket and used a 1/4" drill bit to mark the holes. After drilling & tapping the holes, I installed the mounts.

The coil is a tight fit, but can go in with the bracket installed. The trick is then to place the bonded washers under the mount tabs while inserting the 6mm bolts from above.

The tank cleared just fine.

All I need to do now is wire the coil pigtail (should be here later today) and measure/cut/assemble the plug wires.

Here are some pics:
Mounts


Top view installed


Side view

 

[GS1000G Shopper]

I am afraid you are overcomplicating this. I just printed the graduations on some paper & glued it onto the trigger plate. There is next to no heat in this area.
Nick

I ended up doing the same thing for now, and found once the bike started I had reversed the scale direction. The timing marks need to be to the left of TDC, or looked at like the face of a clock, from about 10 o'clock to 12 o'clock instead of 12 to about 2. I have deleted the incorrect posts and will re-publish once I have the correct setup.

I'll disagree about overcomplicating something as basic as being able to check timing. The factory setup used timing marks here for a reason. The MS manual is very clear about being able to verify timing with a light, and since I may at some point sell this bike I'm sure the next owner would appreciate being able to do so w/o any extra effort. If I can get a brass plate made for under $20, I'll consider this a good investment.

 

[Ecklund]

I ended up doing the same thing for now, and found once the bike started I had reversed the scale direction. The timing marks need to be to the left of TDC, or looked at like the face of a clock, from about 10 o'clock to 12 o'clock instead of 12 to about 2. I have deleted the incorrect posts and will re-publish once I have the correct setup.

I'll disagree about overcomplicating something as basic as being able to check timing. The factory setup used timing marks here for a reason. The MS manual is very clear about being able to verify timing with a light, and since I may at some point sell this bike I'm sure the next owner would appreciate being able to do so w/o any extra effort. If I can get a brass plate made for under $20, I'll consider this a good investment.

Agreed that you have no over complicated this process or made the thread too long.

I've wondered what it might take to add FI to an GS and your thread has been very instructive and interesting.

I appreciate the fact that you are willing to post both successful progress and things that need unforeseen corrections.

Can't imagine you would sell this bike after all of this. But great thread/post either way.

 

[GS1000G Shopper]

Thank you. I've gone through the thread today and tried to cut out posts that have outdated info or where better options exist.

I sorted out the latest issue after updating to MS Extra 2 version 3.3.3. I found I had to reverse the pickup wires to the Microsquirt. The bike starts easily, but idles around 3K RPM.

I modified the ignition timing with little effect, so I think my fast idle is due to my messing up the plate alignment on the throttle bodies. I'm going to replace the one I modified & use coombehouse's center linkage setup. I have updated the first post with a link.

You're right, I'll likely not sell this after all of this work.

 

[GS1000G Shopper]

Timing ring

Timing ring

Here's the latest timing ring design. Shapeways estimated about $20 and a couple of weeks. If this works as designed, and someone should want one, I can sell it via their site.