Need help with vtwin tune

[ertert]

Hi everyone,

so about a year ago I got a Microsquirt after everyone helped answer some basic questions for me in this thread:

http://www.msextra.com/forums/viewtopic ... 94&t=59618

The Microsquirt was finally installed over the Christmas break on the hyosung. Its engine is a 90deg oddfire vtwin with a 270 deg firing angle. I have gone for a fuel and spark setup with an external ignition module driving the stock dumb coils and I'm using the MAP sensor as the CAM signal for full sequential fuel and spark.

The stock setup had 2 MAP sensors, one for each intake port but since I'm using Microsquirt, I have tied the 2 intake ports together to feed a single MAP sensor as suggested in the above post. I'm currently running TunerStudio MS Ultra v3.0.18 with MS2/Extra 3.4.2 release firmware.

Good news is once it was all installed it fired up the first time really easily! Bad news is I'm new at tuning and really need some tips on how to get this running properly. I can keep the engine running but I need to hold a tiny bit of throttle open.

I'm looking at my logs and I can see that I get full-sync most of the time (status1 = 136) but every once in a while it drops to half-sync (status1 = . Can't really fine tune anything like this since fueling is pretty different for semi-seq vs full-seq injection so I think I need to solve this first.

I have a feeling my MAP sample settings are not where they should be and I'm not sure on how to go about optimizing this. Can anyone have a look at my log files make some suggestions? Also hoping someone can take a look at my msq and fuel and spark tables just to make sure I don't have a completely stupid setting somewhere.

The idle valve table wont look right since I have set them to 50 steps everywhere just for now as I'm only trying to get a clean idle at the moment. I'm targeting 1400 RPM - 1500 RPM since this was what the bike was idling at stock. I'm currently running the following MAP sample settings:



I have attached a zip file with the following logs:

running map log - high speed MAP log with engine running
cranking map log - high speed MAP log with engine running (recorded previously with different MAP window settings)

tooth logger - high speed tooth logger
composite logger - high speed composite logger

5 deg fixed spk log - log with fixed 5 deg advance
spk table log - log using spk table advance

P.S. Loving tuner studio and Megalog. Pretty powerful stuff! Especially Megalog. Beats the hell out of excel graphing and much quicker than Matlab to get useful info out of data sets. I can see this thing being really useful outside of just tuning.

[ArttuH]

Hi,
I have a Suzuki SV650 running on MS2. As far as I know the engine is very similar to yours. If you like I can send my tune file for reference.

However, I'm running it in batch mode. I have done some experiments with sequential injection as well but there seems to be something strange going on with it so I haven't got it working properly yet. Using MAP as cam trigger works quite ok for me but when it gets to sequential mode something gets wrong in fueling or ignition. It seems to wan't overly rich mixture to keep idling and engine is strugling to rev up. That needs some further investigation when time permits.

So I would suggest to get it running in batch mode first. After that it should be much easier to get sequential working.

[ertert]

Hi Arttuh,

Yes you are correct our engines are pretty similar. They have the same 270 deg firing angle and bore X stroke of 81mm x 62mm. I'm also using ignition coils, an aftermarket clutch and some other bits from an sv650.

Would be really helpful to get you tune and use it as a reference. I can check my oddfire settings and MAP sample settings against your setup. Will also be able to use your spark table as a starting point. Thank you!

Did you also connect both intake ports to a single MAP sensor?

[ArttuH]

PM sent about tune file.

Yes, I have both intake ports connected to the MAP sensor. At some point I also tried using just single port to get more consistent readings with timed MAP sampling but it didn't work any better. However, I'm using alpha-N with "multiply MAP" for fueling so MAP signal isn't that critical in the end.

The spark table is basically extracted from stock SV650 ignition box (carb model).

[ertert]

Thanks ArttuH! I replied to your PM and I will check out your tune. The spark map should be real useful.

I also played around with the MAP sample settings a bit more today and I got it to hold full sync pretty much all the time which is really helping a lot.

Unfortunately I'm still not able to get a stable MAP reading which is still throwing off the idle.

edit: I got it to idle with no throttle input! The consistent full sync really helps. Its still a little bumpy though... I think I still need to work on getting a smoother MAP reading

[R100RT]

A twin can produce pretty resonate map readings, the fluctuations in the 2 inlet tracks are amazing. Have you considered a buffer on the signal? I use a small C02 cartridge (is rounded and looks cool) with an orifice added to the inlet side, and an outlet port welded on the other end leading to my little manifold (map sensor, Hobbs switch, Boost gauge, etc.) That in conjunction with some reasonable dampening settings in TS allow flawless MAP for my Speed Density control.

[ertert]

I didn't know we had MAP dampening options! Is it this setting called "MAP Averaging lag factor"? Definitely going to try this along with the bunch of tips from Arttuh. Matt from diyautotune has also given me a few tips in regards to MAP sample window so I'm hoping I can solve this in software.

The mechanical dampening sounds really interesting though. Can you post up some pics? Would love to see your setup.

[R100RT]

I didn't know we had MAP dampening options! Is it this setting called "MAP Averaging lag factor"? Definitely going to try this along with the bunch of tips from Arttuh. Matt from diyautotune has also given me a few tips in regards to MAP sample window so I'm hoping I can solve this in software.

The mechanical dampening sounds really interesting though. Can you post up some pics? Would love to see your setup.

C02 Cartridge.jpg

Buffer.jpg

The little cartridge has a tip that can be cleaned up with a pipe thread die to present 1/8mpt.
I drilled the round end and joined a 1/8fpt mini scully in place, then fitted with a 1/8 male/brass female fitting closed with solder and then drilled to a rather small orifice size (I believe was a mini bit 0.030"). This became the inlet.
The outlet is depicted with a 4 way 1/8 cross fitting to allow my various "buffered MAP" reading ports. The result is that the small volume beyond the orifice serves as a buffer zone to average out inlet ambient pressure.
Sorry, photos limited and being a fared bike everything hides where it may.
This was rather simple to add on and has provided very acceptable results.

[lutorm]

I noticed you have the MAP sampling set to 15 degrees. That seems like a really weird number to me. I'm pretty sure that value is BTDC just like timing angle and 15 BTDC is at the end of the compression stroke, sampling the MAP at that point doesn't make sense. On my V4 I have it set to 180 and that pretty much matches the min in the MAP signal.

If you do a MAP log you can plot the MAPsample window along with the MAP trace in MLV and see where your window is. I couldn't open your MAP logs but eyeballing the csv your sample window is nowhere near where the MAP minimum is.

[ertert]

@R100RT, hey I know those CO2 canisters! They are from a motorcycle puncture repair kit right? Love the creative use of materials. If I can't get a decent map in software I will be following you down that road. Thanks for the pictures.

@lutorm, Yep the sample window uses the same reference point as timing. I had it set at 15 deg because that's where the dip in my map was that corresponds to the peak on my the power stroke. I'm trying to use the phase detect feature to get it running in full sequential so I needed a map window that will capture the signal difference he intake and power strokes.

Because of my setup I originally had the map sensor tied to both intake ports and the map signal looked like this:



You can see there are multiple peaks and dips over 720 degrees. So after much stuffing around with that I swapped over to a map signal from only the front cylinder only:



Good news is I get a really stable map signal from this but it does not want to get full rpm sync for some reason.

[ertert]

I took a really long map log with this setup and found that when connected to a single map sensor the dip moves around ALOT with rpm. I have a giant sample window so it always catches the dip but some times i found the dip would be in the power stroke and sometime it ended up in the intake stroke.! I'm guessing that's what confuses the map phase detection.

You can see it here:

[R100RT]

I always thought that the "Map Phase" detection feature only took advantage of the reading at initial sync/ start up, and then was redundant, but perhaps otherwise.
Surprising how the reading is shifting around from where you would think it should be reflecting a predictable inlet state, but perhaps that is the resonance and hysteresis pushing the waves around.
Not to promote the handy little buffer device ( ), but perhaps that might help buffer out the shifting as well so that it remains more or less at the same point of crank rotation for you?

[ertert]

I always thought that the "Map Phase" detection feature only took advantage of the reading at initial sync/ start up, and then was redundant,..

From what I understand I think you are correct. I found an interesting post:

http://www.msextra.com/forums/viewtopic ... 94&t=57979

The guy was having a hard time getting full rpm sync as well. He had very different map signals when cranking and when the engine is running much like my situation. Also in that post they mention that the map phase detect will give up trying for full sync after 100 revolutions.

In my case If i set the phase detect threshold high, I can get full synch during cranking. But when my map signal moves and I drop to half synch, the phase detect threshold needs to be low so it never rpm syncs again.

I think for my setup I'm trying to do too much with a single map sensor! I'm going to give up on full-sync for now and get it tuned in semi-sequential mode. Later on I will try the CO2 capsule and will also probably try 2 map sensors to detect engine phase and then process the signal using an arduino to spoof a cam signal.

[R100RT]

Exactly, I believe I need more "Arduino" on my machinations (beginning with the ability to pronounce it correctly )

[lutorm]

I thought that once you had full sync it would never drop back to half sync unless you have a sync loss.

[lutorm]

I always thought that the "Map Phase" detection feature only took advantage of the reading at initial sync/ start up, and then was redundant,..

I think for my setup I'm trying to do too much with a single map sensor! I'm going to give up on full-sync for now and get it tuned in semi-sequential mode. Later on I will try the CO2 capsule and will also probably try 2 map sensors to detect engine phase and then process the signal using an arduino to spoof a cam signal.

Have you considered the QuadraMap sold by dontz125? I'm using it to combine my 4 MAP signals, and it also has a sync output. http://efi.ttrignition.com/quadramap.html

[sebou29]

Interesting message!
I started to use studio tuner for my gen2 sv650.
This is the first time I try to tune an ECU.
I managed to start the motorbike.
I have used Power Commander before.
I'm interested if you have a base map, to begin.
Regards.