Code Request for Honda RC-51 and the likes.

This is a forum for discussing the development and testing of alpha MS2/Extra code. Documentation
(Runs on MS2 and Microsquirt)

Moderators: jsmcortina, muythaibxr

90turbovan
MS/Extra Newbie
Posts: 24
Joined: Sun Aug 10, 2008 9:05 pm

Re: Code Request for Honda RC-51 and the likes.

Post by 90turbovan »

I have the older 3.0.3q and 3.0.3m working with this wheel on a Silverwing parallel twin. The patterns are the same, the trigger tooth numbers are just a little different. In the older codes, the crank and cam angles per tooth are incorrect but I was under the impression that they were only used for interrupt masking/noise rejection. In any case, it syncs with no issues and runs up to 10k. I did, however, have problems with noise on the crank sensor. The compression stroke slowed the wheel down enough that it registered multiple zero crossings on a few of the teeth. I used the trigger logger and noise filters to get rid of it and it works fine now. I also had the polarity set wrong once and it wouldn't rev past 4.5k without lots of sync loss. I don't have the msq with me but I could check all of my settings tomorrow if it'll help.

Compression seemed to change the cranking timing on my engine, I set the trigger angle with the plugs out and hoped that once it started it was going fast enough to smooth things out some.

Edit: I'm running uS too.
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

90turbovan wrote:I have the older 3.0.3q and 3.0.3m working with this wheel on a Silverwing parallel twin. The patterns are the same, the trigger tooth numbers are just a little different. In the older codes, the crank and cam angles per tooth are incorrect but I was under the impression that they were only used for interrupt masking/noise rejection. In any case, it syncs with no issues and runs up to 10k. I did, however, have problems with noise on the crank sensor. The compression stroke slowed the wheel down enough that it registered multiple zero crossings on a few of the teeth. I used the trigger logger and noise filters to get rid of it and it works fine now. I also had the polarity set wrong once and it wouldn't rev past 4.5k without lots of sync loss. I don't have the msq with me but I could check all of my settings tomorrow if it'll help.

Compression seemed to change the cranking timing on my engine, I set the trigger angle with the plugs out and hoped that once it started it was going fast enough to smooth things out some.

Edit: I'm running uS too.
Oooooohw....

I have no noise filter set, don't really know how that works either, but if that's an option i'll look into it. Certainly at cranking the big twin slows noticeably...
Don't know about the noise, tried to scope with a soundcard scope and with an SSI-4, but the first says my old PIII has no sound card, which isn't quite true and the second shows pulses if you turn the crank manualy. When you crank it goes like, ah, that's 5V that goes up 'n down a bit....

The polarity, of the crank or cam sensor? Well that's a given... If it originaly works with that...

If the trigger teeth is not the same, well then imo it's not the same, yanno... This one's odd fire. 12 on the crank, 3 on the cam at, 120º, 180º, 540º.

I have a race alternator, with 12 little bolts in it. I could place that and remove one, to get a 12-1 setup. But that's not fixing a problem, that's going around it...

Could also go piggy back and use the Keihin for sparks, that would be most reliable. And use the Us as a power commander with lamda correction. But then i won't get past 10k rpm, which is why i went with the Us in the very first place.

It sparks 10 BTDC one time, then 10 ATDC the next, it fires up right away, and the next time it just doesn't, one time you have to remove the serial cable, the other time you just need to cycle the power.
I don't see how this thing is ever gonna be reliable, but that's maybe cuz i'm slightly frustrated with it at the moment.

I eeeuh, i guess i have to do something, i just dono what... :?
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

Just tried the 3.03alpha code...

It still gives no crank rpm. It does make spark, but the timing is too late. I set an offset on it. Doesn't seem to change much. So i figured i'd go extreme. I don't see a lot of difference between -90 and +90 offset. I figured i'd go even extremer, but then it goes, nope, can't do that. On the front cylinder i think i noticed it goes like, i'll spark twice and if that didn't work, i'll give up. On the rear it seems to keep trying.

Also i've noticed and this is totaly weird... I'd set staging in the old code, never got to the rpm where it worked, but it needs a change in req_fuel so i set it again in this code. It seemed to accept this, till i calibrated the thermister table. Then it goes like, you can't set the same trigger twice on the staging. I used RPM as the first, TPS as the second. I reflashed the code, since RPM and TPS isn't the same to me at all, so this was to me odd. Same thing. After thermistor calibration it won't accept those staging settings no more. So i set RPM and MAP. That is ok. Then i set RPM and nothing, that's ok too. Then i set RPM and TPS again and again it goes, no can't do that, those are the same???

Could not get it running, not even a cough or a bang.

Don't see much difference either. At first i was thinking the PW came up faster, but you said you made it faster, so that was expected and maybe i wanted to see it.

Anyways, there's a diff of 180º between offset +90 and -90, this should be easy to see with the naked eye. I don't trust that setting at all.

Maybe it didn't run, because it was totally lean. First time i set staging i needed to double the req_fuel. Can't recall i did that this time. I'll look into that.

It also gave me a bunch of errors when i opened the diagnostic tab, i forgot what they were. Also everything but RPM was striked out in red and seemed unavailable. RPM didn't come up, so i didn't bother to log that.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: Code Request for Honda RC-51 and the likes.

Post by jsmcortina »

OB6D wrote: Anyways, there's a diff of 180º between offset +90 and -90, this should be easy to see with the naked eye. I don't trust that setting at all.
Only +/-20 deg of change is actually used.

Can you confirm with a strobe where it is actually firing and what offset you really need?

Try resetting your ReqFuel with the calculator, there have been a number of changes in the code relating to two cylinder engines so the resulting pulsewidth might be quite different now - compare your logs.

During crank is the engine keeping sync? Or is it losing and gaining sync? Did you get any sync-error logs? If so, please email them to me and hopefully I can advise what is happening.

The section I examined of the "cranking when warm" log you sent the other day looked just fine.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

jsmcortina wrote:
OB6D wrote: Anyways, there's a diff of 180º between offset +90 and -90, this should be easy to see with the naked eye. I don't trust that setting at all.
Only +/-20 deg of change is actually used.

Can you confirm with a strobe where it is actually firing and what offset you really need?

Try resetting your ReqFuel with the calculator, there have been a number of changes in the code relating to two cylinder engines so the resulting pulsewidth might be quite different now - compare your logs.

During crank is the engine keeping sync? Or is it losing and gaining sync? Did you get any sync-error logs? If so, please email them to me and hopefully I can advise what is happening.

The section I examined of the "cranking when warm" log you sent the other day looked just fine.

James
No, becuase when i turn the knob on the strobe, the mark moves even further away. It's not designed to do negative values. I can measure advance with it, not retard. The paintmarker i made on the bolt of the flywheel points to the cylinder when it sparks. The cylinder is on about 45 degrees with the horizon, when the strobe flashes my mark is about level with the horizon, pointing straight forward. Then i'm guessing, and it has to be a guess, it's about 45º ATDC when it sparks.

I've tried again, with more fuel it doesn't fire up. Can't make a guess to where the timing is when it runs as such. Spark seems a bit random to me in the way that it's there and then again not.

Also i'm looking at the PW when it cranks, it gave about 5.5 to 6ms, i've played around long ennuf with it to know that aint ennuf. It needs 8.5 to 9 with these temps and above 10 it's a bit overly rich. So that's why i raised it.

Don't know if it keeps sync James, but from the seemingly random sparking pattern i'd guess not. I bet it's loosing 'n gaining sync.

And also snatched a picture of the error i get when i open diagnostics.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: Code Request for Honda RC-51 and the likes.

Post by jsmcortina »

There is no need to guess - enable the lost sync count and lost sync reason gauges. Then you will know if it losing sync.

You'll also tell on the main screen if "Sync" flashes red/green.

That error message you posted, seems like you enabled some gauges on the diagnostics tab that don't exist any more. Redefine them to some that do exist.

I'll email you an s19 with the spark more advanced.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

Mailed you back. Tried this. With a lil more fumbling this one will run it. Then i'll update. Soon.
90turbovan
MS/Extra Newbie
Posts: 24
Joined: Sun Aug 10, 2008 9:05 pm

Re: Code Request for Honda RC-51 and the likes.

Post by 90turbovan »

First one shows the noise I was seeing, really short pulses. Second one is with the period rejection filter enabled, no noise at all. MSQ is for 3.0.3r. This engine uses the same crank and cam wheels, its just even fire. All I do is change what tooth it considers TDC and remove the "Must have odd fire" config warning. Maybe odd fire is causing some problems?

I didn't have to change reqfuel, it fired right up.
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: Code Request for Honda RC-51 and the likes.

Post by jsmcortina »

90turbovan wrote:This engine uses the same crank and cam wheels, its just even fire. All I do is change what tooth it considers TDC and remove the "Must have odd fire" config warning. Maybe odd fire is causing some problems?
Could you post what tooth numbers you use as triggers and I can make the code cope with both by using the odd-fire setting as a selector.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

90turbovan wrote:First one shows the noise I was seeing, really short pulses. Second one is with the period rejection filter enabled, no noise at all. MSQ is for 3.0.3r. This engine uses the same crank and cam wheels, its just even fire. All I do is change what tooth it considers TDC and remove the "Must have odd fire" config warning. Maybe odd fire is causing some problems?

I didn't have to change reqfuel, it fired right up.
Thanks :D

My tooth 'n trigger logger show spikes, not pulses like yours does. But maybe it's my puter? PIII 833Mhz...

The msq, looks very much the same, also the enrichments and stuff, pretty much in the same zone i set it, at first glance at least.
Injector flow, you got the same even, but mine uses two sets of those. 2 cylinder 4 injectors. I think what you mean with what tooth it considers TDC is your -5 offset.
Anyways, looks a lot the same... Ohw yes, i went Alpha-N because of the low vacuum. I changed about everything, it's crazy. But it did run. Then.
You set alternating, i didn't. 7.6 per cylinder for req_fuel. That's kinda weird also. I needed around 12.5 on half the injectors, when i set staged injection, i had to go up to 25 to get the same. Anyways, when warm it idles on two injectors with about 3.5ms PW.

But,

What i don't understand is you set wasted spark.
Now i remember in the 2.0.1p release James wrote, my doubled RPM reading was fixed. If i remember correctly, James said at that time, it was running like a V4 with wasted spark. And the wasted spark stuff, was as i remember, removed.

Does yours use one wasted spark kind of coil then? Must be so, then if that works, that puzzles me. It should be running on one leg then... heh? :?
Or do you use both ignition outputs, how did you set that up?

Again, thanks for the info. And i'm glad yours works (maybe it's a typical Honda kinda trigger (the blackbird has it too)), all effort wasn't in vain then at least.
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

Just got out the garage. And with the plugs out, cranking, just looking at them it doesn't even make sense. It sparks, then a whole time not, then i swear at times it fires two times quickly after one another, then some not, then once again, then again two quickies. I must've messed something up terribly.

I don't get it. But i borrowed a scope. Now i need to learn how that works too. :RTFM:

I've made logs while cranking, a triggerlog, a toothlog and a lost sync log. Attached for who's interested.
90turbovan
MS/Extra Newbie
Posts: 24
Joined: Sun Aug 10, 2008 9:05 pm

Re: Code Request for Honda RC-51 and the likes.

Post by 90turbovan »

There's a zoom button in the bottom right of the diagnostic screen, zoom in until you can see the bars. Took me quite a while to figure that out...

Everythings in %reqfuel now, seems like the enrichments don't really need to change much between engines. I could probably do with less afterstart, but it runs acceptably. The engine has to breath through a 20mm diameter restrictor, so vacuum is never a problem. Possible turbo plans for the weekend, so SD is the way to go.

I have alternating set because I'm running sequential; with simultaneous, reqfuel is half what it should be. Idle PW is about 4ms at 13:1.

I could run single coil, wasted spark, or COP. The stock coil is a two in one deal that fires both plugs at once, so wasted spark or single coil are my options, the only difference being that MS thinks single coil is running with a distributor. I've got VAG coilpacks that might make it on for COP. The number of coil selection shouldn't affect RPM, it just tells the code how many outputs it needs and when to fire them.

You don't seem to be losing sync during cranking. I would set the timing to use a fixed value of 0 degrees, get a timing light, and verify that you get consistent spark at TDC for each cylinder. Forgetting TDC, if you can't get consistent spark anywhere and you're not losing sync, I'd start to blame the odd fire code. Can't imagine how there could be a problem there, but it works fine in even fire.

Scope won't help much at verifying timing during cranking unless its a digital storage one. Stuff happens too slowly to actually follow. You should be able to verify rising/falling edge selections though, if you watch the crank signal during cranking you'll notice that one direction crosses zero in a straight diagonal line where the other direction flattens out a bit as it goes through zero. You'll want the edge that's a straight line. Keep in mind the inversions in the VR circuitry, there's something different with uS if I remember correctly.

I'll start another thread for the Silverwing data, don't want to interfere with this one.
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

Alrighty.

Someone who wishes to stay anonymous had made the crank 'n cam sensor loose ground here. When that was discovered and quick fixed, the battery was almost empty, but the engine was kicking again...
So, i'm again, close...
Electricity is to put it mildly, not exactly my specialty. I bet in the end that'll turn out to be a major part of the problem...

I have to admit, i didn't know you could zoom in the diagnostics screen.

The scope is a Panasonic 561 OP or 5610 P it's 200Mhz, four channels and it looks like a challenge on it's own with all those buttons.
Wanted to use it, to determine the trigger edge and to check spacing between crank 'n cam pulses. Detect noise. And then maybe mess around a bit with +/-, to see what'd give best results.
But i think i'm just gonna give it back, before i smoke it.

Got a picture of your bike turbovan? Can't seem to find it. Guess it's not a European model.
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

It just started first time on the new code. :D

Looks like i forgot to recalibrate the MAP sensor (running Alpha-N), but should fix that.

And according to the log it idled well under 1000 RPM. 900 ish for the most part and 800's, which may be (hardly) possible though i don't think so. And 500 ish in the end, which i find just very hard to believe. I don't trust that at all, but may be too early to draw conclusions there... If i had to draw a conclusion, i'd say it was twice that.

Edit:
Nah, that recorded idle speed, that is just NOT true. nope.

Edit again:
I remember on the very first version of the code it had doubled RPM readings. Later that was fixed.

I bet it'll turn out to have halfed RPM's this time. Seriously, it's just wrong.
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

The RPM's have halved, but Obsessed's now running the 303r Alpha on the RC51. Now i'll sort the crank sensor, somehow...
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: Code Request for Honda RC-51 and the likes.

Post by jsmcortina »

I have tested this on the scope and I agree that the rpms displayed are half the actual. Working on it.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

jsmcortina wrote:I have tested this on the scope and I agree that the rpms displayed are half the actual. Working on it.

James
Yup, they're looking correct now. Thanks :D
I also revisited your 70% pulse rejection suggestion, which got me up to 5000 RPM without problems. :D
Need to tweak it and then i'll see if i can double that. (if the neighbours don't mind, hehehe).

BIG thanks to everybody (James in particular), that helped me get this far. :yeah!:
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

Am currently remaking the wire harness some. Making it neater and i'll make some taps for the SSI-4. Will set the SSI-4 up for RPM, TP, WheelSpeed (km/h) and brakelight with a devider on CH4 maybe (or i'll buy me a pressure transmitter thingy some day). Then i'll run the serial out of the LC-1 through the SSI-4 and with the LM-2 that i also have laying around, i'll be able to log all that and road dyno tune it. :idea:
OB6D
Experienced MS/Extra'er
Posts: 150
Joined: Wed Feb 04, 2009 6:54 am

Re: Code Request for Honda RC-51 and the likes.

Post by OB6D »

OB6D wrote:Alrighty.

<...>

Electricity is to put it mildly, not exactly my specialty. I bet in the end that'll turn out to be a major part of the problem...

<...>
Pin 33 VRmin (coax shield) is this supposed to be a ground for the VR sensor, or is it just a shield and does the VR sensor ground go elsewhere, with the shield wrapped around it and not connected to it.

My shield is my sensor ground. Is this where i f.. up?
24c
Master MS/Extra'er
Posts: 847
Joined: Tue Jan 20, 2009 10:21 am
Location: Lancashire UK
Contact:

Re: Code Request for Honda RC-51 and the likes.

Post by 24c »

OB6D wrote:Pin 33 VRmin (coax shield) is this supposed to be a ground for the VR sensor, or is it just a shield and does the VR sensor ground go elsewhere ....
The coax pin 33 is the shared ground for VR1- & VR2-, AFAIK using a cased MicroSquirt.

PS For datalogging without a laptop I only have the Innovate LMA-2 & LM1, but that SSI-4, LMA-3 & DL-32 stuff looks good. :)
Yamaha GTS1000 v2 MicroSquirt, B&G 2.891
Yamaha GTS1000 v3 beta MicroSquirt, B&G3.760
Yamaha GTS1000 MSExtra 3.1, Dual VR Board
Yamaha YZF1000 MSExtra 3.1
Post Reply