MS2/Extra on MicroSquirt

saboteur · Post by **saboteur** » Wed Nov 08, 2006 2:58 am

Are you guys planning on supporting the MicroSquirt? I'm guessing it will have some limitations over the non-SMT version as you can't hack it about as easily (hardware-wise).

I just hadn't seen the MicroSquirt mentioned at all on the MS2/Extra section, so I thought I'd bring it up!

Bill Shurvinton · Post by **Bill Shurvinton** » Wed Nov 08, 2006 3:29 am

I don't believe that the code has been released for microsquirt yet.

Up until now Al has preferred to develop a stable initial code and then pass it on for the group as a whole to work with. I am sure that there will be a number of areas where code X-fertilises between them. However uS is very IO limited, because it was designed with a particular purpose in mind, so very little 'extra' will be able to be supported.

Post by **jsmcortina** » Wed Nov 08, 2006 12:42 pm

We can still do wasted spark wheel decoding, overrun fuel cut, Ken's Enhanced Accel Enrich (alternative to X-Tau) and the like.

We'll need to check over what I/O is being used on the Microsquirt and modify the code as appropriate to support those pins. To my knowledge the board will feature two VB921 so 4cyl wasted spark with any wheel should be relatively painless.

My understanding was that Al was supporting 4cyl W/S for twin VR (bike) setups by using two spark control interrupt routines, I'm not sure if he is doing 4cyl w/s wheel decoding.

James

newtyres1 · Post by **newtyres1** » Wed Nov 08, 2006 3:50 pm

There's also some relevant and interesting info here

http://www.microsquirt.info/uscode.htm

and the 2nd paragraph here

http://www.microsquirt.info/dualspark.htm

Ian.

KGB · Post by **KGB** » Sun Jan 07, 2007 11:56 pm

Just noticed that the MicroSquirt would be a good platform to upgrade to for my engine (4 cylinder wasted spark), but with the MS2/Extra sticking to bit-banging the ignition, what would happen to the 2 ignition channel and the extra VR-in? Are they now moved to become some GPIO pins? Personal preference is that there is an option to wire up the MicroSquirt as is but it's probably still too early to call since it's only in Beta form

Post by **muythaibxr** » Mon Jan 08, 2007 4:55 am

We also don't have microscquirts to test with (yet).

Ken

Post by **jsmcortina** » Mon Jan 08, 2007 6:31 am

Aside from having no hardware (never stopped me before...) the MS2/Extra code should be adaptable to work with the Microsquirt hardware and utilise the two on-board VB921 when running in typical wheel decode mode.

Using the twin VR pickups as per 4cyl bikes would require some new code and isn't likely to be immediate as the 2.8 MS2 code should cover that now (I think.)

James

KGB · Post by **KGB** » Wed Nov 21, 2007 3:14 pm

For my own info on pin-mapping MS2/Extra usage to HC12 on Microsquirt:
JS0/JS1 => PT7 (IAC2) (not wired)
JS2/JS3 => PT6 (IAC1) (not wired)
JS4 => AN07 (AD7) (not wired)
JS5 => AN06 (AD6) (SPAREADC)
JS7 => PE0 (FLEX)
JS10 => PT5 (IGN) (IGNOUT)
JS11 => PA0 (not wired)
D14 => PM3 (InjLED) (TACHOUT)
D15 => PM5 (WarmLED) (WLED)
D16 => PM4 (AccelLED) (ALED)
FIDLE => PM2 (Idle) (IDL)

Too many pins unwired or wired into different types of circuits and yet this is the only Squirt box that has built-in dual VR....

Post by **jsmcortina** » Wed Nov 21, 2007 3:44 pm

I have been testing on the Microsquirt. The existing 2.0 beta code should mostly work as I've added Microsquirt support in there - although "trigger return" does not appear to work correctly. I believe this is a bug in the code in general and not specific to Microsquirt.

To use the Microsquirt version you need to use the right downloader as we build a different s19 for Microsquirt.

James

KGB · Post by **KGB** » Wed Nov 21, 2007 5:14 pm

Assuming the code works fully and ignoring the un-wired outputs:
FLEX - Table Switch
FIDLE - Boost Control
IGN1 - 5V pull up
IGN2 - 5V pull up
VRIN+ - Crank Wheel
VR2IN+ - Cam Wheel
VRIN- - GND (Not sure which ground to use because my stock VR sensors only provide the + side)

The only problem I'm left with is having an MAT/CLT installed since I cannot remove the bias resistor and re-use the stock ones.

Slightly OT - if the AMPSEAL grew by a few pins and have all these extra outputs wired out
JS0/JS1 => PT7 (IAC2) (not wired)
JS2/JS3 => PT6 (IAC1) (not wired)
JS4 => AN07 (AD7) (not wired)
JS11 => PA0 (not wired)

You could make a board for sequential 4cyl by having PT6/7 go thru injector drivers. As for the rest of the outputs, it's just code
JS5 => AN06 (AD6) (SPAREADC)
D14 => PM3 (InjLED) (TACHOUT)
D15 => PM5 (WarmLED) (WLED)
D16 => PM4 (AccelLED) (ALED)
JS10 => PT5 (IGN) (IGNOUT)

Don't know if there is room for jumpers, but if Microsquirt can jumper out the VB921 to straight pull-ups as well as bypassing the bias resistors for MAT/CLT, it would be a few extra circuits away from having all the functionality of MS2/E built-in (they seem to share a few common circuits) which could technically be built-in and jumpered to the proper pins. The microsquirt would grow, but at the same time, it's a stepping stone to MS3 as right now you would need your own circuits outside the board to handle Boost/N2O/Table Switch and all sorts of goodies

I'm dreaming...

wes kiser · Post by **wes kiser** » Thu Nov 22, 2007 3:31 pm

jsmcortina wrote:I have been testing on the Microsquirt. The existing 2.0 beta code should mostly work as I've added Microsquirt support in there - although "trigger return" does not appear to work correctly. I believe this is a bug in the code in general and not specific to Microsquirt.

To use the Microsquirt version you need to use the right downloader as we build a different s19 for Microsquirt.

James

So there is a possiblity the reason I couldn't get the micro to work with extra before on the KTM engine may have just been the trigger setup? I had already convinced them to make a real missing tooth wheel, and it is fabricated and mounted.

Should it work as long as it is just "standard" stuff (24-1 wheel, single coil, single cylinder). If so I will try it this week.

Post by **jsmcortina** » Thu Nov 22, 2007 3:41 pm

Well, it _should_ but it might be better to test in private!

James

wes kiser · Post by **wes kiser** » Thu Nov 22, 2007 3:54 pm

I will let you know. It likely will have an audience (but I won't tell them what's going on