Just receive the MS, Relay board and simulator last friday.

All is mounted, and working with the simulator, now i start to make the mods for the Rotary engine.

What i plan to run in my 12A N/A.

FC CAS with 2 VR

Coils, i can do it 2 ways, the MS drive the double coil setup for leading and trailing.
or take the 2x 5v signals and use one coil pack that already have power drivers built in. PLEASE ADVISE, and whats the best way to do it

CBR600 ITB with 8 injectors.

Questions.


firmware MS2 Extra 2.0.1, ms2extra_beta20090225 is ok for this setup or better to use 029y4 or other ?

Built in VR is for the 24 tooth circuit input, the extra VR with LM1815 is for the 2º signal correct ?
Any pictures of the 2º VR biuilt on the proto area?

For both the v3 and v2.2 board, you'll want to put the 4.7Kohm resistors as detailed in my last post, and then you'll want to wire the negative lead of led 17 (the one closest to the serial port) to your leading ignitor's IGt wire (IGt-L). The negative lead of led 18 goes to the coil select wire on your trailing ignitor (IGs-T). Finally, the negative lead of led 19 will need to be wired to the IGt wire on the trailing ignitor (IGt-T).

In my case i don´t have FC coils, so i just need (+) 5v outup signal to trigger the driver, were to get this signal, direct from the 4.7k also ?


I have other questions, but is better go step by step, as more i look for info more i get confused with litle details, i know that is impossible to have one manual for each setup, but the info is like one puzzle were i still loocking for the parts.

Tanks in advance

 

The most direct and problem-free route would be to use FC coils. 029y4 is the MS1 code, not applicable. You will want the latest beta code as Ken and James have improved the wheel decode function. It is, by all reports, much less affected by noise.

 

Hello,

yes i have look and found this new version http://www.msextra.com/viewtopic.php?f=91&t=30944

i will start with this one, instead of jump to old firmware and have to do it again for new one.

About the changes i have almost done like in this PDF
http://stuff.pw.cx/Megasquirt%20MS2%...azda%20CAS.pdf

Now about the FC coils, i didn´t undrestand or found how it works the Trailing Coil Select, from what i see there is 2 independent outputs for trailing and the coil select change from one to the other, now if i can´t get the FC coils there is no way to do it? for example using one pack with 2 signal inputs and 2x2 outupts
(the leading fire at same time and same for trailing) wast spark type.

Please someone clarify this for me

 

You can fire leading in a wasted spark configuration. Trailing spark definitely CANNOT be fired in wasted spark config, that's why the FC trailing coil has the "coil select" line - the trailing spark needs to occur only on the rotor that is in it's combustion phase, usually a few degrees after leading or at the same time as leading - ONLY ON THAT ROTOR - . An option would be to use FD mode. FD mode uses one wasted spark coil (one coil, two high tension leads, two plugs, both in the leading position) and 2 separate coils, one for each trailing plug. FD mode looks great on the stim and scope and should work just fine. I have a current 12A turbo install that is using the FD scheme as described but is not completed yet. I'll advise my progress. I'm using an FC leading coil for the wasted spark leading spark and the stock 12A coils for the 2 trailing...

-Mike

 

Tanks Mike

yes now i understand better, it is only the leading that can go waste spark.
So, to go FD mode i will need one coil with 2 high tension leads ( i have it here) and 2 separate coils one for each trailing, than i assume that in FD mode
the signal output and the select output for trailing will fire each separate coil ( i can get this also) i have here 4 cbr600 coils that i can use, or like you the stock 12A trailing coils.

(Are you using the 2 stock coils direct to plugs ? i read somewere that they are made to pass first on distributor and there is others for direct fire, explain please.

To use 2 cbr600 coils i need a driver, i have here 3x BIP373, what is the better way to connect it?

 

Yes, you have described the FD requirements perfectly! I've used the first gen coils for many, many miles direct to the plug. I cannot think of a single technical reason the coil would care (or know) if it's secondary current is flowing through a distributor or not. I don't know if the CBR600 used CDI or not but if so those coils won't work. CDI is fairly common on that class of engine and it works on a different electronic principle then the standard Kettering style coils. If I were you I'd just use the first gen coils. They're already mounted, they're cheap, they're available and they just work. In any case, the BIP drivers should work just fine; Bosch did afine job of engineering a reliable, tough component.

-Mike

PS: like your sig.

 

The CBR600 coils need a driver, i don´t have CDI for it, so it must be the BIP373 to drive each coil.

About your setup, did you use the stock drivers "J109" if so how did you connect it in FD mode, there is 2 wires trailing and trailing select, one must go to front rotor, the other goes to back rotor.

Do you remember how you did it and the best way to setup the BIP373 if you didn´t use the stock drivers?
Still need to use resistor pack if using the BIP?

Any pictures?

i have google Megasquirt FD MODE but didn´t found any info about this setup

The BIP can handle more power than the stock ignitors so i think is the best way to go.


My Sig,, hehehe it was the first problem that i have 2 weeks after i buy the RX7, the headers start to glow, than later someone give it the "Paris Hilton" symptom
after other users report same problem, so i start to use it for fun.

I did solve the "Paris Hilton" symptom but the carb didn´t want to tune (or i didn´t know how to tune it, and here i´m going with Megasquirt.

 

From my limited Google, the CBR600 comes stock with CDI ignition as do many other high RPM engines. CDI takes the dwell/coil charge time out of the equation. They fire by dumping a 400+ volt capacitive discharge through the CDI coil primary; the very rapid pulse through the primary causes a very short-lived magnetic field to build and collapse thus inducing the secondary winding impulse. Standard Kettering style builds the spark charge in the coil's magnetic field then interrupts current causing the collapsing magnetic field to induce the spark voltage/current in the secondary winding. If you already knew this, my apologies. My point is that a CDI coil will not produce spark if fed with a Kettering style 12 volt dwell period pulse. If you're using the FC CAS, the J109s are not really in the picture as they are mounted to the distributor. Use the BIPs - they're certainly more robust than the J109s. Reference the trailing - I cannot remember precisely which output goes to which coil. I did start the engine on leading only with trailing plug wires going to spare plugs laying on the shock tower and used a timing light on the trailing wires to determine which was rotor #1 and #2. The #1 trailing will fire somewhere around the timing mark on the pulley - that's as close as you need to get to determine which is which. Pictures will be forthcoming. By the way, the engine will run just fine on leading only. I spent a few hundred US$ a few years ago to quantify the effect of trailing on a 12A NA engine and the power difference was within the error range of the dyno. I'm sure emissions were up, etc., and do not recommend this, just thought I'd mention that fact. A guy I know flew a rotary powered plane for 20+ hours with the trailing disconnected and never noticed a difference.

-Mike

 

Mike,

about the CBR600 or high RPM engines i didn´t know the details, but did know that it need to be different.

But in my case the cbr coils that i have are 4 RED single units (old style), i can use 1 ,, 2 or 3,, no CDI connected just the 2 pins to drive the coil and the high voltage output.

But today i talk with a friend and "maybe" tomorrow he will send to me FC coils with the 2º Gen CAS, not easy because here FB, FC and FD are very rare.

About the way to discover what is what, yes i didn´t think about it, since trailing is 6º~8º after each leading we just need to have the engine runing on leading and use the timing light
to get the correct one.

Let me see if i can put all this together this week, still missing fuel pump and few other things, and i´m shore many questions will come up.

I will try to add some pictures tomorrow

 

Keep at it matey!

What do you have Megasquirt 1 or Megasquirt 2 cpu?

 

(i think the question is for me)

Jabro, i have follow your advise
3 BIP´s, extra LM1815N
PCB V3.0 with MS2 CPU (but i have one eye on the MS3 CPU), found and installed ms2extra_beta20090413 tested on table.

I still have to had the LM1815 circuit and the BIP´s, there is still things to do that i´m learning (Mike teach me the way to know were is Trailing 1 and 2) and more to learn.

Before i get the MS2 and build it, test it, all this MS2 was very very confusing, now that i get it working on table with MegaStim and play with table etc all is much more clear, now i can say
it is easy to get to this poing, but only because i found that PDF and some people give help, but i know that i´m only 50% of the way, tuning will take the other 50% were is much more to learn

 

Ok,

get the CAS and leading coils

I think that this can go on the 12A engine, anyone to confirm?




And the coils, how to connect it?
i try to find schematics or someone that did this conversion
but still didn´t found it.

Someone to give a help on this please, i don´t know what car it come from, but look like FC pack

 

That CAS drops right in, no problem. The coil you have there is an FC trailing coil/igniter assembly. The sticky FAQ at the top of this forum has FSM scans and details as to what wire does what. Basically you'll have +12, coil trigger and coil select lines. I don't remember if it grounds through the mount or through a separate ground wire. That should work just fine! A convenient place to mount it is on the left side shock tower.

-Mike

 

Now you give me good news, tanks.

I have already and today i try again to find that FSM scans, didn´t find nothing, maybe i´m not using the correct words, if you or someone have the link please tell me.

I have see that there is some firmware out that support sequential injection with 4 drivers, is this better to rotary? or not yet supported?

 

https://www.rx7club.com/megasquirt-forum-153/official-rx7-megasquirt-faq-ms1-check-here-before-you-ask-your-question-658411/ , message #3 has the wire colors for the FC coils. Sequential for the rotary is not a viable alternative at this time; the processor timer used for trailing ignition has been retasked to perform the sequential timing duties. If you provide 4 squirts/cycle with the nonsequential code you will have similar conditions for each rotor thus negating injection timing as an influence. Sequential injection timing discussion could be the topic of a new thread. My advice would be to get it running first, get a decent, safe base tune then play with enhancements. It's good that you're looking around and staying up on things, please don't misunderstand me!

-Mike

 

I´m not good with color codes also i take some time to discover that TAN is BLACK (my bad English), i have one plug with 2 brown wires, and other with 4 wires
like in picture, since my colors didn´t match the one in your link i take a better look at the PDF and now i think is correct, if not please someone correct me.




Mike please explain better this to me.

"If you provide 4 squirts/cycle with the nonsequential code you will have similar conditions for each rotor thus negating injection timing as an influence"

And yes, i plan to make it run with safe base first, i´m only making plans that if i need any extra wires to let it now ready for future, i already see that if all goes like plan maybe this year we will be using MS3 MCU.

And don´t worry i didn´t misunderstand you, it is always good to have advises.

Tanks for your help

 

Just ignore the red line coming from the GRAY wire, i forget to remove it

 

Mike,

first i want to say sorry, i feel bad because i didn´t find the link to the sticky FAQ and you have to give me the link, yes first time i open your link i just look at
3º post, than much later i notice that was on top of Megasquirt forum (shame on me)
yes i have already read it some time ago, but since the colors didn´t match i didn´t care more about it at that time.


So, i have make some connections on the table, 2 injectors only and the CAS, rotate it by hand and i got RPM in megatune, the relay go ON etc (this with basic file loaded).

I will install it all next week if i manage to have to intake ready.

I´m building my MSQ file with ms2extra_beta20090413 cause i didn´t found MSQ files to load for code 2.0.1, i will post this file later.


Is there a way to verify if the second VR circuit is running correct?
Other than use oscilloscope?

 

To verify the 2nd VR circuit, i hooked the CAS up to the MS. Then, while powering the MS with the jimstim or regular stimulator, spin the CAS with a drill or something (just use tape to attach the end of the drill to the gear on the CAS) and verify that you get a nice stable RPM signal in megatune. Basically the same thing as you've already done except using a drill to spin the CAS

 

"If you provide 4 squirts/cycle with the nonsequential code you will have similar conditions for each rotor thus negating injection timing as an influence" What I mean by this is that with things set to 4 stroke, 4 injections per cycle, you get an injection every 720/4 degrees - every 180. If you're also set to simultaneous, Each rotor only ingests air/fuel each 360, so each rotor will see one injection at one particular time, the next injection at a second, 180 off particular time. The driveability and idle issues that timed injection seems to fix are, I believe, caused by inequity in what each rotor/cylinder/etc. experiences. The 4 inj/cycle gets rid of this inequity as each rotor experiences the same fuel delivery scheme. Anything much past idle and your injectors are firing longer than the port is open anyway. The US automakers use timed injection to purposely inject onto the back of a closed intake valve - this helps vaporization of the incoming charge and reduces emissions while improving idle quality. There's apparently a whole lot more to the "timed injection" thing than the obvious jump-to-conclusion that it's better to inject while the intake tract is actually moving charge uinto the combustion chamber. If this still isn't clear, let me know and I'll try again. Or just try using 4/cycle and watch how your idle improves. My ideas on this are only hypothesis but the empiric observation with a sample of 1 (!!!) is that it helped my engine run much more smoothly. As always, YMMV in more than one way!

-Mike

 

I try the osciloscope, it looks like for every 2 impulses from the CAS i have one square output from the VR circuit ( run out of battery on the osciloscope after see this).
i have to find one way to make it rotate stable with one drill or some electric engine to verify in Megatune how it works.

 

so for every two zero crossings of the VR sensor you get the LM1815 output going high, or for each zero crossing you get the wave going high? Does the output fall before the next VR zero crossing?

 

Mike i understand, since i´m going to use 8 injectors 270cc each one i was looking for one way to make it run different, so i´m going to use the 4 lower injectors on INJ1 and the other upper 4 in INJ3 and make it staged.


I think that i have collect all the parts that i need to make this run, just missing the intake buil/finish, and the idle, this part i still don´t know how to do it, my setup have one system i think is for idle, it is 4 air input near each throttle with a fine tune by hand and some VAC operated system.

My question now is, with this 4 throttle bodies what is the better for idle warm up ? Fidle, is this s simple system that the relay just activate one valve to let air enter?
Someone to give me some lights on this

 

will check this tomorrow, don´t know were i have the charger for the scope, but
from what i remember i have to see 2 impulses on the input to have one at the output, maybe it should be for every zero crossing going high one pulse at the outup, from the PDF of the LM1815 if is not correct mounted it will do different things.

Tell me how it should be pls.

 

"My question now is, with this 4 throttle bodies what is the better for idle warm up ? Fidle, is this s simple system that the relay just activate one valve to let air enter?"

Fidle in it's simplest form is on/off and binary in function. You would have to bleed air from an air solenoid into both rotors - doesn't much matter where as long as the bleeds are behind the throttle plates. I'll let you know as I progress on staged injection related to the number of injections per cycle. Yours may be easier than mine as I have asymmetrical injector sizes re primary and secondary injectors, 460 and 680s respectively.