Finally some progress

[Peter Florance]

After a god-awful 3 and 4th quarter 2009 at work, things are coming together here and I'm been able to resume my Sequencer build instead of fighting fires.

I finally bench tested the ignition portion of my Sequencer setup; 6 BMW M42 coils and 6 IGBT ignitors. Seems to work fairly well and makes lots of ozone.

I've got the assy mounted to a spare BMW M30 head to work out ignition wire lengths, mounting brackets etc. Fabrication is not my strong point.

And at least on the stim, it goes to sequential. I did try to load The Monkey's msq and unit went off line and had to reflash it. Instead I just worked from scratch and got it running best I could understand.


Is 3.11X still the latest code?

I'm on vacation to Mexico until next Sat and hope to get it running that following week in the car. Fortunately the car has had a megasquirt installation for some time and my plan is to be able to switch from the MS2 box to this without much effort. Maybe 5 minutes or so.

I'll do a little scoping in the morning before head out.

Thanks!

[TheMonkey]

Peter-

Y, I think we are all feeling pretty beat up on the professional front right now.

Great to hear you are getting things going with the Sequencer.

Strange that my MSQ file bonked your unit. I'm trying to recall all the possible nuances - not sure what could have caused this, but there was some tweaks that had to be made to Megatune setup. Hmm.... the flash on my Sequencer was the original one that Al sent with the very first batch of beta units. This thread describes a couple tweaks made to make MT compatible: http://www.microsquirt.com/viewtopic.php?f=90&t=23163

Importantly though, just make sure that ignition is firing as expected. I set mine up to run just fuel first, then recorded on a scope a time series of the ignition output signals to make sure they were doing as expected, which they weren't until I tried different delay teeth combos.... and strangely, I found a combo that didn't cause the phantom problem. I haven't been paying close attention to Al's status on this, but not sure if he found the root cause of that. In hindsight though, it showed up on the stim, so if you have clean RPM sweeps up and down, it's prob fine.

Good luck, keep us posted, and have fun in Mexico.

Scott.

[Peter Florance]

I had some strange dwell in my first try - it would go as high as 23msec at 1400 rpm.

The latest msq seems to behave ok though

I will re-scope; without coils arcing and sparking!

[TheMonkey]

I had some strange dwell in my first try - it would go as high as 23msec at 1400 rpm.

...

rut roh... this sounds familiar to the problems i was having. i had some very long dwells, and weird 'phase shifts' from cyl to cyl. Do the BMW coils have overdwell protection built in?

perhaps you should try the delay teeth setup i ended up with. seemed like an arbitrary fix (unexplainable), but that did it for me. i'm not sure if Al found the root problem from it, i thought he did, maybe it will end up in the new code.

are you using the stock trigger wheel off of M30B35? try 20 delay teeth, -36* offset. that seemed to fix the crazy ignition stuff for me, dunno why.... but i have put loads of trouble free miles on it, and I was watching the ignition behavior very carefully.

[Peter Florance]

I had some strange dwell in my first try - it would go as high as 23msec at 1400 rpm.

...

rut roh... this sounds familiar to the problems i was having. i had some very long dwells, and weird 'phase shifts' from cyl to cyl. Do the BMW coils have overdwell protection built in?

perhaps you should try the delay teeth setup i ended up with. seemed like an arbitrary fix (unexplainable), but that did it for me. i'm not sure if Al found the root problem from it, i thought he did, maybe it will end up in the new code.

are you using the stock trigger wheel off of M30B35? try 20 delay teeth, -36* offset. that seemed to fix the crazy ignition stuff for me, dunno why.... but i have put loads of trouble free miles on it, and I was watching the ignition behavior very carefully.

I do use that wheel although I'm just bench testing

No overdwell at all on the coils. Maybe I need to rethink my choice of coils?

Seems like 36 degrees would not be enough offset, but I think I don't understand how the code works compared to MS1 extra.

[Peter Florance]

also what's the max advance you've run? I can't recall if I sent you a spark table or not, last summer.

[Peter Florance]

Oh lastly, i think I did grab your trigger wheel settings and that probably makes the difference; fixing the dwell. At a certain RPM the reported dwell still goes up by 200usec, but I really need to spend more time with it to see whether the hardware is also doing it.

I may try to use the Logic analyzer portion of the HobbyLab scope to get some data.

[TheMonkey]

Remember, offset can change to whatever you want as long as delay teeth is adjusted as well. I started with 14 delay teeth, zero offset, and Al said to try different combos - I did, and found one that works.

options of delay teeth / offset would be:

20 / -36° - works ????
19 / -30° - same timing, but does not work
18 / -24° - same timing, but does not work
17 / -18° - same timing, but does not work
16 / -12° - same timing, but does not work
15 / -6° - same timing, but does not work
14 / 0° - same timing, but does not work
13 / 6° - same timing, but does not work

for advance, I think the advace I used is pretty much based on the table you shared with me. 38 at WOT, and as high as 50 in the low-load high-rev range. car loved it... MPG shot way up (no overdrive with the dogleg trans).

if you figure out how to use the logic analyzer, let me know, it had me thoroughly confused. the data recorder on the other hand.... was the only reason i was able to sort through this. excellent tool, although i need my desktop to run it, my old laptop just doesn't cut it.

advance VEX file:

EVEME 1.0
UserRev: 1.00
UserComment:
Date: 01-22-2010
Time: 11:54
Page 0
Scale 0.100000
VE Table RPM Range [12]
[ 0] = 7
[ 1] = 9
[ 2] = 12
[ 3] = 15
[ 4] = 20
[ 5] = 26
[ 6] = 31
[ 7] = 37
[ 8] = 43
[ 9] = 49
[ 10] = 54
[ 11] = 60
VE Table Load Range (MAP) [12]
[ 0] = 20
[ 1] = 25
[ 2] = 30
[ 3] = 35
[ 4] = 40
[ 5] = 45
[ 6] = 50
[ 7] = 60
[ 8] = 70
[ 9] = 80
[ 10] = 90
[ 11] = 100
VE Table [ 12][ 12]
[ 0] [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11]
[ 0] = 270 313 350 465 500 500 500 500 500 500 500 500
[ 1] = 270 305 340 450 490 500 500 500 500 500 500 500
[ 2] = 270 290 320 420 460 480 500 500 500 500 500 500
[ 3] = 270 283 310 405 445 465 486 500 500 500 500 500
[ 4] = 270 275 300 390 430 450 471 500 500 500 500 500
[ 5] = 200 268 290 375 415 435 456 500 500 500 500 500
[ 6] = 200 260 280 360 400 420 441 500 500 500 500 500
[ 7] = 200 245 260 330 370 390 411 470 470 470 470 470
[ 8] = 200 230 240 300 340 360 381 440 440 440 440 440
[ 9] = 200 215 220 270 310 330 351 410 410 410 410 410
[ 10] = 200 208 210 255 295 315 336 395 395 395 395 395
[ 11] = 200 200 200 240 280 300 321 380 380 380 380 380

[TheMonkey]

...No overdwell at all on the coils. Maybe I need to rethink my choice of coils?...

as long as you are not over dwelling them when you get on the road, they should be fine. just be cautious through beta testing until you know that actual dwell is limited to what you are programming in. part of my problem when using the other delay / offset settings was that dwell sometimes came in way too long - even doubling up over an entire engine cycle.