jayroc

Do you think it would be better to reset the Trigger Angle to 60 and change the Trim Angle to -5, instead of a Trigger Angle of 55?
This is to compensate for the difference of the 12a/13b pulley markings.

muythaibxr

don't do that... you should follow what I said above:

use a trigger angle of 60, or even 65, and don't set trim angle to anything.

set the Fixed angle to 0 after you get the car running.

Readjust the CAS with the timing light so that the timing is set correctly

reset the fixed angle to -10.

This will get your timing dead-on, and get your triggers/returns correct.

with a 55 degree trigger angle, you're firing the spark plugs at -5 on crank, which might make starting a little hard. The part that compensates for the mark being at TDC is setting the Fixed Angle to 0 when you're setting the timing with the CAS and timing light. What it does is fixes the advance at the number you specify (0 in this case), so the MS thinks its firing the plugs at that advance (TDC in this case). That allows you to adjust the CAS with the timing light until everything lines up, and the MS's commanded advance is the actual advance. With a 2nd gen, we just set the fixed angle to -5 while zeroing the timing.

Maybe it'll help if I explain how I'm getting the cranking spark advance from the trigger angle. Basically when you set a trigger angle, it is telling the MS that the engine is at that angle BTDC when you hit a trigger tooth (12 or 6 in this case). on a 12 tooth wheel, there are 30 degrees of engine rotation per tooth, so tooth 12 is 60 degrees, tooth 1 is 30, and tooth 2 is TDC... If you move the trigger angle to 55, then you end up with 25 degrees on tooth 1, and -5 on tooth 0, etc... You can set the trigger angle to whatever you want (above about 50) as long as you follow the procedure above for zeroing the timing, and your timing while running will be accurate. However, once you get the timing set right while running, and turn the car off again... the cranking timing will be at whatever angle your returns are... So with 55 degrees as your trigger angle, if the CAS is adjusted right, you'll be getting a cranking timing of -5...

my suggestion is to do as above... set trigger to 60, re-zero the timing using a fixed angle of 0, reset to -10 fixed angle, and then try cranking again to see if it makes a difference.

N1XRR

Hey, would it be possible to change "Hold Ignition" to "Hold Ignition (Seconds)" in Megatune?

muythaibxr

Yeah, it wouldn't be hard at all.

EDIT: I was wrong about the hold ignition setting. It's supposed to be the number of "cycles" to hold ignition for after a stall on restart....

However, everywhere it's used in the code, it's stored in the wheel count, then the wheel count is overwritten by #WHEELINIT, meaning that hold ignition isn't being used at all with the wheel decoder. So it should have no effect whatsoever.

So this setting does pretty much nothing at all when using the wheel decoder.

jayroc

Okay, muythai, I will do that.
The reason I am running a 55 degree trig. angle is because I physically cannot adjust the CAS to its proper position. If a stud was there instead of a bolt hole, I would get to about 1-2 degrees advance and the stud would prevent me from getting to TDC. Seems like I should be able to install it one tooth off and regain some angle, but it takes a more clever man than me to accomplish that, apparently, as I always seem to wind up at the same results, or it is so far off that it's unmountable.
I believe the stock distributor has a wider adjustment potential, and there may very well be differences in how the 13b and 12a front covers allow for adjustment. Doesn't seem likely, but anyways..
Oh well, I will just dremel some material off the side of the CAS to get the room I need and redo my settings and post again tomorrow or Friday.

jayroc

Okay, I got the "trimmed CAS" installed and was able to get the movement I needed to set the Trig. Angle to 60. Didnt seem like much help..
Two things I have noticed however. One, the "idle advance wait time" begins when the CPU comes on, I believe this should be changed to count from after start because it makes hard to start cars harder, it seems like.
Also, cranking advance is taken from the "cranking advance angle (deg)" setting whether you are using "Trigger Return" OR "Time Based". I had thought it took the cranking angle through some calcuation of the trigger angle itself when using trigger return, as you explained above, but this is not the case as I see it on my MT advance gauge.
Based on these observations, I should be able to set the trig angle at 55, readjust and rezero the timing and have no adverse effects what-so-ever. Again, I believe this to be true unless there is some sort of discrepancy between what's displayed in MT and what is physically happening during cranking.

muythaibxr

there is a discrepancy between what's displayed in MT and what's actually happening during cranking. The advance that megatune shows during cranking is only accurate when using "Time-based" cranking.

During trigger-return cranking, there is no code to calculate what angle your return teeth are at, so it just displays whatever advance is in the Spark table or is set in idle advance. I could probably make the idle advance countdown from start, it wouldn't be that hard, and is probably the correct behavior, so I'll add that... the advance that's displayed won't matter during crank, but after you start, it'll matter.

So on trigger-return cranking, it just fires the leading plugs whenever the return tooth passes the VR sensor, even if megatune is displaying something different.

One thing that can help with getting more stable starts is setting the cranking rpm higher. I have mine set to 400 even though the engine cranks at just below 300.

I'm surprised that setting the trigger angle to 60 and rezeroing the timing didn't help though. Did you try cranking after you rezeroed?

Also, make sure that when you switch between trigger return and time-based cranking timing, you reboot the MS... turn it completely off and back on again. I don't know if the settings there will take immediate effect without a reboot of the MS. (some settings do and some don't, and it's not always clear which ones are which, so I recommend when changing most constants you reboot the MS).

Finally, don't be afraid to experiment, It's possible that advancing the cranking timing is the wrong way to go... Maybe it needs to be more retarded... I'd try again and go as high as 3-5 degrees on cranking (so 63-65 trigger angle), or as low as -10 degrees (50 degrees trigger angle).

I know that the leading plug location is different on the 12a vs S4 vs S5... Tofuball's S5 fires up pretty quick once warm, and would probably fire up pretty quick cold too if I had time to help him with the cold cranking pulse-widths. My S4 doesn't fire up as fast, even using the same settings, and required me to add a lot of fuel compared to the S5 to get it to fire up quickly.... I think because the plugs are located further down in the housings (more advanced), which means on my S4, I might experiment with a lower trigger angle and different tooth configurations to get a cranking angle of -5 or -10, or as high as 5-10 degrees.

Anyway, since there aren't that many people who have this stuff working, it's worth-while to experiment. I'd experiement on my own S4, but I've been too busy writing code and building megasquirts to work on my own car. I'll get to my own car next week for experimenting (after I fix the stock grounds).

renns

jayroc,

I haven't made my way through the thread, but here's the fuel settings I used on my old stock FC. I just dug them out for another guy, and thought they may be of interest. My FC started great on these settings even down to sub-zero temps, and was daily driven for almost two years, including two Canadian winters. Take a look at your cranking and warmup numbers. Switching over to time-based ASE decay, rather than ignition cycle based also helps by giving you a much longer ASE period. Also, cranking pulse widths need to be quite high, as you should be firing only the primary injectors during cranking.

Roger.

muythaibxr

I think it might be a bit different though since he's on a 12a

renns

Of course, but simply scaling by displacement ratio as Mike suggested a ways back should eliminate that effect.

muythaibxr

hopefully, but I think that the locations of the leading plugs in the housings and other similar small differences might be adding to the cranking problems that jayroc is experiencing....

Also I've noticed some pretty large differences just between S4 and S5 with the same settings related to ignition timing, so I think maybe that might also be part of what jayroc is experiencing...

Even still though, it'll be nice for him to have your settings, then he can rule out fuel if he still has problems after using your settings

jayroc

Well, I drove around some tonight, but didn't have a chance to make some setting changes. I got a 13b ceramic coated manifold (it's modded for a 12a ) to go on tomorrow, so after that's installed, I'll tinker some with these settings some more.

Actually, muythai, I take it back. That DID help some. It kind of struggled before to get started, now it's starting with slightly less effort, but it was harder to notice with hot starts, because my cold starts are easier than my hot starts. Weird.
Honestly, I have been hesitant to change my cranking angle settings because of potential backfiring.. I will go out to my shop in the country to do some real testing. A couple weeks back when I was first installing the CAS and getting the stab procedure down, it backfired like a gunblast a couple of times. Seriously it was that loud. My neighbor called the cops, but fortunately they didn't know it was me, I just saw them out there after I took a break.. heheheh

I'm real close to figuring this out..　Mike is sending me his settings also, so with that and these new infos you guys have posted I should be able to get it starting like it should.

Eagle7

Roger, could you also post the fuel map that goes with those settings?

Thanks,

renns

Marty,

Attached below are the two VE maps that match those settings. Stock n/a FC, no mods other than MS. The two VE maps are for primary and secondary injectors, as I was running dual table at the time. By fattening the upper RH corner of the primary map it could be used with the current staging code. I used these settings to build a replacement ecu setup for a local guy a while back. He was having fits with the stock system. He dropped in the replacement and ran it as-is for months on his daily driver.

Roger.

pmrobert

Jayroc (and whoever else wants to look), I have posted screen caps for all user entered data from my well running rotary. Some of the caps are not pertinent,like the second VE and timing tables, I don't use . This is for an Atkins street port 12A, RB header, decent exhaust, 2x50mm Weber pattern TB, RB wraparound intake, firing leading only, 2x680 injectors positioned 17" or so from the irons mounted in the TB. I use MAPdot exclusively and use a GM HEI to fire an FC leading coil in wasted spark mode. I've been having local net problems lately and cannot get them to post here so you may find them at http://smrmicro.com/Gallery/ . One HUGE word of caution - I reside in South Florida so the warmup stuff is only known to work well down to 40F - that's the coldest I've had to start in and the numbers to 40 worked fine.

-Mike

muythaibxr

Mike, I thought you were using a CAS with 2 teeth cut out running leading and trailing with the wheel decoder, and using FD style trailing...

Did you switch back? Was there a problem with the wheel decoder setup?

pmrobert

There is absolutely no problem with the 12-1/your code/FD setup as far as spark is concerned. I ran that code for probably 1200 miles or so with no concerns or unexpected behavior. I'm back to the MS2/2.2 board combo in the car because I'm in the process of redoing everything in anticipation of the MS2 trigger wheel & GPIO/Router/CANbus stuff coming out and need the v3 board back on the bench for buildup testing as it provides access to the CANbus provisions of the 2.33 MS2 code - and- I've only one 12-1 CAS and that needs to be on the bench for now for testing. I've acquired new cases, cabling, etc., for a complete reinstall of everything. Your algorithms are solid, Ken, have no fear! Way OT, but, have you done any more work on the RX-8 trigger wheel stuff? I have one of those that is just begging for an optimized EFI system. They run way too rich from the factory... :-(

-Mike

muythaibxr

I have the rx8 code done, and it's in James's hands waiting for him to have time to test it with his wheel simulator that he wrote for his C32 MS2 chip.

If he doesn't get to it soon, I'm probably going to port his c32 simulator code to the c64, and test it myself.... after I'm done making the msns-extra wheel decoder use all the teeth on the wheel for more accurate spark instead of doing all that crazy prediction stuff.

With all my software projects and my own cars to work on, I'm backed up about a month.

For the rx8, since you'd have to run it piggyback anyway (throttle body and fuel pump issues), you could just install the MS fuel-only piggybacked and get the tach signal from the stock ECU somehow.

jayroc

Okay, big breakthrough today.. I reset my ASE to 5 across the board and this resolved 90% of my starting issues.
Hot starts were definitely almost impossible. WHY ASE would have some effect on hot starts (ie, 160+ degrees), I have no idea, but it somehow fixed my problem..
I did replace my plugs as they fouled out yesterday, yet again.. However, I had new plugs before and similar settings with nowhere near these good results. Also, I set the priming pulse to 0, and prime only when >0 instead of always. Not sure if that had an effect or not. I am also using Time Based and -5 cranking advance. I may switch back to Trigger Return, but so far my best results are with Time Based. Even though I tried all these settings before, changing ASE had the biggest effect. I think ASE was coming on and flooding it out JUST as it was starting.. I REALLY should have looked at it before, but I was always trying to begin at hot start problems, and my ASE setting was 0 at 160 degrees, so I paid it no mind.. ばかだね、俺は。

BTW, I wonder if there would be a way to fire a priming pulse only at lower temps, or at least at a settable range of temps?
How much of a prime pulse would it take if the rail was completely depressurized with only two injectors and a fuel pump real close to the rail? Or a better question, how can I attempt to calculate this for my setup, or should I just guess?

jayroc

Oww, even better idea.. I will be installing a fuel pressure sender within a few weeks.. Is there any way to code in a fuel pressure option for the priming pulse width setting? That would be SWEET...

pmrobert

jayroc, the additional fuel potentially added by the priming pulse is extremely small - it's usually in the vicinity of 1 or 2 combustion events worth of fuel at most. Don't worry about it. You can just make it 0 and have the engine crank for another .2 seconds or so if there any air or vapor at the injector - the commanded FI events will prime it within 1 or 2 events. Glad you're making progress! It's all worth it.

-Mike

muythaibxr

I would VERY strongly recommend not using time-based cranking. It will work fine if your starter and battery are in good condition, but the first time you get a speed variation on crank, you're asking for trouble. You'll potentially get backfires at the least, and if you get some kickback, you could potentially kill an apex seal or worse....

I just don't want to see something like that happen. If you want to crank at -5, just set the trigger angle to 55, reset the CAS, and use trigger return. IT's much safer and more accurate on crank.

Congrats on getting the car running better.... For ASE, I use 0 on hot starts... changing it to 5 helped you? I guess you are using a different engine, and your intake is a lot different, which could also be changing things a bit.

jayroc

I'll reset it to trigger return. You mentioned that before I know, but I was getting desperate for some results!

Actually, at 160, my ASE was always 0, I just reset the lower values (60-130) to 5 and the -40 to 40 degree values to 10. Somehow it's cranking better on hot starts.

On the priming pulse, I was just thinking that for cold starts with 0psi on the rail, maybe it would help more, but if the difference is that small, I won't worry about it then.

jayroc

Yet another update..
I was wrong earlier. It must have been switching to time based that got me going earlier..
I did need to lean out my ASE for cooler temps with no problem, but I also did need ASE for hot starts it seems.
The settings I found to work well are: 0 prime pulse, 5.7ms cranking pulse@160 degrees, Time Based with a hold ign of 3, and ASE of 10 at 160degrees. This has effectively solved my hot start problem. Starts well in less than 2 seconds.
I tried trigger return, and muythai, your advice has me worried, but it works so much better. Trigger Return won't start or takes lots of effort, sounds like it struggles, backfires, pops.. Exact same settings as Time Based, but.. I will take my chances with Time Based unless somebody can think of a way to make Trigger Return work better for me. I'll risk motor troubles given the success I've had thus far with it.

Apologies for the differing diagnosis on my part, but it seems I was facing numerous problems at once.. Oh well, it is all coming together thanks to everyone's input!

muythaibxr

The trigger return problems you're having can be from the advance you're seeing on the returns like I said before... I'd mess with the trigger angle and resetting the CAS until you get something that starts quickly for you. The settings I've got posted in the FAQ work pretty well for my S4 (I haven't really had much of a chance to mess with them since my grounds on the stock harness are messed up... once I fix them I'll tweak my S4 settings and post in the FAQ), and work great for tofuball's S5.

I went back and read about that time-based problem. The guy that had the worst problem ended up with a bent rod on his piston engine due to kickback.

If you're still cranking with timebased at -5, it doesn't necessarily mean its definitely cranking at -5, it's hard to hit an exact timing on cranking with time based... a trigger angle of 55 will give you -5 timing on crank with the wheel settings I have posted... Also, you may want to work with resetting the trigger/return teeth to 1,3 and 7,9 and restabbing the CAS with tooth #3 to the right of the tooth on the 2 tooth wheel lined up with the Ne vr sensor to see if that helps your trigger return problems. Make sure when you restab it that you set to TDC first (just making sure). I just went back and looked at the settings I used for tofuball's car, and we're using those settings on his car, with a trigger angle of 60 degrees.

I've had no problems with 12,2 and 6,8 as the triggers/returns, but if this ends up helping you, I'll change it in the FAQ, and add instructions/pics on how to restab the CAS to compensate.