Bandit 1200 Turbo EFI can't get out of it's own way

[rider384]

Hi everyone, first post here. Sorry if this is the wrong forum, but it seems like most EFI talk is focused here. 

Long story short, I have a Bandit 1200 turbo, factory sealed engine, K03 at 7psi, probably 160ish whp, nothing crazy (for now). It was running alright on the factory carbs, but I could never get it perfect and I deal with altitude change so I swapped it to EFI. 

EFI specs:

-Microsquirt 3.0 running hybrid alpha-n

-K1 GSXR600 throttle bodies and injectors 

-Stock Bandit coils and plugs run by MS via 2x Audi 2.7t ignitors

-Inline fuel pump being run at 50psi

The problem I'm having is that it's dogshit slow. Unironically, my '75 CL360 would smoke it in a straight line. AFR is acceptable, not perfect (12:1-14:1 at WOT, needs work but isn't far enough out to cause it to only have what feels like 20 horsepower). Timing has been verified with a timing light, but I think something funky is going on with it; the manufacturer of my trigger wheel told me that 1st tooth BTDC is 348*, but I had to set it to 4* to make it fire at actual top dead center (found with TDC tool, 0* cranking advance to check timing marks I made). Stockish Bandit timing curve, 13* at idle ramping up to 35* by 4,000ish RPM. It will build boost, but I currently have the wastegate hanging open so I can tune the alpha-n portion of the map. 

I'm kind of lost on what to do from here; I've verified that all cylinders have spark, all cylinders are getting fuel, plugs look good and consistant, throttles are opening all the way, spark plugs are tight, grounds are good, battery is charged, etc. It sounds like it's dropping cylinders under load, which points me towards ignition issues. 

Have people run stock Bandit coils on MS without issue before? Everything I've read has painted it to be a mixed bag, and I'm close to throwing a COP ignition system at it to see if it solves the issue, but I'm hesitant to throw money at the issue without knowing what it is. 

[Reinhoud]

Does it idle alright? How's the throttle response?

 

Did you measured the ignition timing on the proper side of the TDC? You wouldn't be the first who did that.

 

[rider384]

Idle is fine and throttle response is great when free revving. It feels like it's dropping cylinders as soon as I demand a load from it. Wondering if one of my coil packs is going bad. 

Considering I don't know what you mean by measuring ignition timing on the proper side of TDC, I would say that there's about a 50% chance that I did . I'm running a custom trigger wheel with no timing marks, keyed so it only sits in one position. Found TDC with a TDC tool, marked the trigger wheel and a spot on the engine with a sharpie, and timed it from there at 0* spark advance until it was dead on. 

EDIT: Do you mean TDC on the compression stroke? If so, I didn't even check to be honest, I didn't think it would make a difference since the wheel and crankshaft turn at a 1:1 rate. 

Edited October 5, 2020 by rider384

[Arttu]

One thing that came to my mind from your 1st tooth angle wondering. Did you check ignition timing at different rpms? If you have the crank sensor wires wrong way around or input polarity wrong in the settings the timing might be ok at some rpm but drift when rpm changes.

[Blubber]

7 hours ago, rider384 said:

Hi everyone, first post here. Sorry if this is the wrong forum, but it seems like most EFI talk is focused here. 

 

Hello and welcome to OSS,

You have found the correct sub forum for your question and... 2 of the resident guru's have already chimed in too 

 

A request: 

Please take a minute and start an extra thread for your introduction.

The place for that is the General Chat section.

Its nice to know with who we are sharing the info with.

TA

 

[rider384]

Update: Verified that the coils are good; primary circuits were 3.5ohms, secondaries were 34.6k and 34.8k ohms. Verified that the VR sensor is wired with the correct polarity. Verified that the spark plugs and wires are tight. Verified timing again just for kicks. Verified that the throttles are all opening evenly. Battery charged to a full 13.8v. Still the problem persists.  

I'm wondering if it's a problem with one of the junkyard ignitors I'm running - perhaps one is too weak to spark properly above a certain RPM? Not sure how to test that. 

Still no progress other than checking things out. I'm tempted to convert the MS to run fuel only temporarily to see if the problem persists. Still at a bit of a loss as to what to do. Probably going to order some VW stick coils or LS1 coils soon and swap them out, just to rule out weak spark. But I really hate to throw money at it when I don't know what the problem is. I'm wondering about stupid things now, like the MS spark output failing above a certain RPM and things like that.

Edited October 6, 2020 by rider384
Formatting

[Reinhoud]

What brand sparkplugs you have in?

 

The manual for my GS1000 says Denso or NGK, but my bike runs like absolute crap on Denso. Took me while to figure that out..

 

[Reinhoud]

Do you have resistors in the plugs and caps?

[rider384]

NGK JR9B, stock spark plugs for the Bandit 1200. Not sure if there are resistors in there, but it's stock spark plugs, stock wires, stock coils. If they're supposed to be there, they're probably there as I didn't change anything except the control method. 

As an aside, the bike ran great on the stock ignition system and carburetors, the problem only started when I converted it to MS. Hence why I'm tempted to swap back to the stock CDI box and run fuel-only MS for the sake of troubleshooting, maybe something broke in the swap. 

Plan for tomorrow is to verify fuel injector flow rates, check for spark above 3k rpm, and if that checks out, order some LS1 coils so I can eliminate the ignitors and coils from the list of suspects. 

[Arttu]

So the timing was staying correct also at higher rpm?

Did you measure voltage at the coils? Preferably under load. For loading you can use a jump wire to ground the negative terminal of the coil while measuring voltage on the positive terminal.

Checking the injector flows is definitely a good idea. Second hand injectors can be in whatever condition, especially after longer storage period. And uneven injector flows could be one potential reason why it doesn't run properly even when the measured AFR is ok.

Switching back to stock ignition and running it as fuel only for troubleshooting is a valid idea. Only problem is that sometimes it can be slightly tricky to get reliable rpm signal from the stock ignition system. So while you are eliminating some potential problem you may bring in some new ones...

[rider384]

Good, albeit slightly embarrassing news to report:

Poked around today, found one of the coils was installed with reversed polarity, and found two of the fuel injector plugs swapped. Reversed both, and the bike idles, revs, and does low-end like an absolute dream. AFR is spot on, it's smooth, and up to a certain point, power delivery is almost factory. I also tested the spark output with the MS test mode and it does spark on all cylinders, even at high RPM (in the test mode, anyways). 

However, the primary issue still persists. With the now smooth-running bottom end, it has become very clear that the bike is dropping cylinders when I demand a real load from it. Seems to get worse as the bike gets hotter, but that could just be my imagination. Tested the voltage drop at the coils and it was fairly significant; battery was putting out 12.4 volts, and the voltage at the coils was around 11 with a bit of fluctuation. That wasn't even under load, just static voltage at the coils, engine off. 

As such, I'm guessing that my coil pack wiring has seen better days, and is likely the culprit. Since I'm going to be re-wiring anyways, I think I'm going to bite the bullet and just order some LS1 coils.

Any thoughts? Does static voltage not matter, only load voltage? I have a feeling I'm tantalizingly close to fixing the issue. 

Also, Arttu, I have not tested the timing at higher RPMs. My downpipe is poorly designed and sits right in front of the timing cover - I would have to run completely open dump off the turbo at 5,000rpm to test. I just moved into a new house last week, don't want to piss off the neighbors quite yet. 

[Forced2]

I had the same issue once with a similar setup. Ended up being the alternator. It would produce power at idle to just before quarter throttle but just as the engine would get a load on it  the alternator would fail.  When it was failing there wasn't enough energy to produce proper spark on just the battery alone. Especially under boost at the same time. Took me weeks to figure it out.  

[rider384]

28 minutes ago, Forced2 said:

I had the same issue once with a similar setup. Ended up being the alternator. It would produce power at idle to just before quarter throttle but just as the engine would get a load on it  the alternator would fail.  When it was failing there wasn't enough energy to produce proper spark on just the battery alone. Especially under boost at the same time. Took me weeks to figure it out.  

I'll get a multimeter on the alternator when I get a chance. It would make sense as my battery has needed a lot of charging, but I always assumed that's because I have to have the ignition on a lot for diagnosis, probably more time with the ignition on than the bike running. 

[Arttu]

5 hours ago, rider384 said:

However, the primary issue still persists. With the now smooth-running bottom end, it has become very clear that the bike is dropping cylinders when I demand a real load from it. Seems to get worse as the bike gets hotter, but that could just be my imagination. Tested the voltage drop at the coils and it was fairly significant; battery was putting out 12.4 volts, and the voltage at the coils was around 11 with a bit of fluctuation. That wasn't even under load, just static voltage at the coils, engine off. 

As such, I'm guessing that my coil pack wiring has seen better days, and is likely the culprit. Since I'm going to be re-wiring anyways, I think I'm going to bite the bullet and just order some LS1 coils.

Any thoughts? Does static voltage not matter, only load voltage? I have a feeling I'm tantalizingly close to fixing the issue. 

Also, Arttu, I have not tested the timing at higher RPMs. My downpipe is poorly designed and sits right in front of the timing cover - I would have to run completely open dump off the turbo at 5,000rpm to test. I just moved into a new house last week, don't want to piss off the neighbors quite yet. 

Well, static voltage without load also tells something. Voltage under load can't be any better than without load. But it doesn't tell how low voltage on coils may drop under load. About 1.5V drop is already quite significant but not totally bad if it doesn't get any worse. But if your charging voltage is already low and then coils get that 1.5V less it starts to hurt spark power quite significantly. And if voltage drop gets bigger under load resulting voltage might be low enough to cause spark drops even without boost.

You can use the test mode to generate load for voltage measurement. Set it sparking at some relatively high rpm, 6000-9000rpm, and check voltage on coil terminals. Additionally you can also enable the fuel pump and repeat voltage measurement. Depending on how the pump is wired it can cause significant voltage drop too. 

[Gixer1460]

8 hours ago, rider384 said:

 My downpipe is poorly designed and sits right in front of the timing cover - I would have to run completely open dump off the turbo at 5,000rpm to test.

You don't want to be removing that cover with a running engine at ANY rev's unless you like an oil bath!

When you noted the voltage my 1st thought was alternator / regulator / rectifier as others have said and if the battery has needed lots of charging, this sort of confirms it. For future, if working on electrics a lot, hook the battery up to a charger / battery tender so it never drops to low levels and you don't have to stop for re-charging. Also, it may help if you do the alternator charging mod to ensure a constant charge voltage to the battery and also do a direct, battery to coils connection via a ignition switched relay, to avoid loom resistances and ensuring best volts to the coils - which when running should be / will be whatever the battery is receiving - usually 13+ volts.

[Arttu]

8 minutes ago, Gixer1460 said:

You don't want to be removing that cover with a running engine at ANY rev's unless you like an oil bath!

Done that several times Yes, it will be a bit messy but not a complete disaster if you lean the bike. Unfortunately there aren't much better options for timing check. Making a windowed cover is one option but based on my experience that doesn't work too well either due to splashing oil.

[rider384]

9 hours ago, Gixer1460 said:

You don't want to be removing that cover with a running engine at ANY rev's unless you like an oil bath!

When you noted the voltage my 1st thought was alternator / regulator / rectifier as others have said and if the battery has needed lots of charging, this sort of confirms it. For future, if working on electrics a lot, hook the battery up to a charger / battery tender so it never drops to low levels and you don't have to stop for re-charging. Also, it may help if you do the alternator charging mod to ensure a constant charge voltage to the battery and also do a direct, battery to coils connection via a ignition switched relay, to avoid loom resistances and ensuring best volts to the coils - which when running should be / will be whatever the battery is receiving - usually 13+ volts.

It was all parking lot work before I moved, looking forward to being able to hook up my battery tender while working on it now that I have a garage. 

I ordered LS1 coils last night, I'm going to run a relay off the stock wiring to power them directly from the battery. What is this alternator mod you're referring to? 

[Gixer1460]

2 hours ago, rider384 said:

 What is this alternator mod you're referring to? 

Refer here : https://oldskoolsuzuki.info/archives/category/how-to 

It makes for a much more stable alt. output.

[mikeyd]

Basic questions, What did you gap the plugs at? Is your fuel system dead head or do you have a return to the tank? Do you use a rising rate fuel pressure regulator?

[rider384]

Updates: Swapped to LS coils wired directly to the battery through a relay with NGK DCPR8E resistor plugs, it actually exacerbated the issue. Dwell times and other settings were adjusted accordingly, yet it still runs worse.

Battery was charged and after a short ride measured at 12.4v. With ignition, accessories, and fuel pump on, it measured 11.7-11.8v (with a low-power LED headlight). Measured the voltage at battery when it was running at idle and it was 12.0v. Voltage drops down to about 9.5v when cranking. Just checked some recent datalogs and the system is sitting at 11.8-12.2v when it's up at 4,000-5,000rpm. Datalogs also show that the system voltage actually goes up after the bike is turned off. 

I think forced2 and Gixer1460 might be bang on the money in their above replies. 11.8v at 5,000rpm seems extremely low, and voltage drop of 0.7v while running accessories seems fairly extreme, as does 11.7-11.8v at 5,000rpm, as shown in one of the logs. Beginning to think it's both the battery and alternator on their way out. It would also explain why the bike initially runs well and only starts running like shit when I try to get on it; the alternator can't keep up, and the weak battery runs out of power for everything. When I turn it off and let it sit for a minute, it fires up and runs fine again until I get on it, when it falls on it's face again. 

Going to get the battery load tested tomorrow. If it checks out, going to start going through the charging system. If it doesn't, going to replace the battery and see what happens. But based on all the symptoms, the fact that it got worse with the 4 ls1 coils, and the fact that the system voltage goes up when the bike is off, I believe it's some aspect of the charging system combined with a bad battery. 

[Gixer1460]

She may be turning but she sure as hell ain't charging fella! A swapped alternator , for a known good one - will realign the planets and bring calm to your world, and with the charge mod a steady 14.5v output - happy dayz!

[Reinhoud]

Is it an idea to make a wiring loom, a simple one, yourself?

Simplicity usually rules..

[Reinhoud]

How does it run on a fully charged car battery? Disconnect alternator and let her run on the car battery.

[rider384]

11 hours ago, Reinhoud said:

Is it an idea to make a wiring loom, a simple one, yourself?

Simplicity usually rules..

Ha, at this point, everything except the gauges and lights are running on my own harness. 

 

11 hours ago, Reinhoud said:

How does it run on a fully charged car battery? Disconnect alternator and let her run on the car battery.

I've had this thought myself, but it'd be near impossible to test as it only starts running poorly when I demand load from it. 

[rider384]

Not sure if I tested this correctly, but the alternator with no load was putting out 0.2v at idle and around 3v at higher RPM. 

Assuming the alternator doesn't need a load on it to be tested correctly, it's definitely the problem.