doridori-rx7

iTrader: (3)

I'm hoping that we can create a sticky with this thread. - muithai,renns,error - I leave this in your hands.

THe idea is, to incorporate the data from the previous dozen threads, into a one stop, " this is what you need, how to build it and how to set it up with the sensors in the vehicle.

I know you guys are probably getting tired of doing this and I see questions being repeated becuse of lost or poorly cross referenced data from one thread to the next..

As someone who is trying to order a kit to build I've thought i've been ready to order only to find out that thier was this one bit of info missing that blew the whole order so I've had to change my standing order 3 times now.. a definitive or at least up to date build sheet and parts list is what is needed..

This is not a ' please spoon feed me the info', the new guys just need a SOLID place to get all the right info. This was the very problem whith the NS boards.. it's ALL over the frikin place and allot of threads assume the reader has some exp. with the MS or PCB board building..

So, please, only hard data and build comments I don't want this to be a disccussion of what the thread should be or some such just solid info fromm the guys who are making this work for us 'MS noobs'

muythaibxr

I'll collect all my data and information and post an answer to this when I get home unless someone wants to do it sooner.

jayroc

I'd also like to see something compiled about the ignition stuff. It has evolved substantially in the last few months..
I was thinking of doing a writeup on converting to FC ignition on a 12a using the MSnS-E firmware and a v2.2 board.

muythaibxr

alright... the definitive answer is: "it depends"

There are many ways you can go about getting the megasquirt to run the stock rotary ignition.

For example:

You can use the stock v3 board, and cut out 2 teeth on the stock CAS, and add the 3 4.7k pullup resistors, and that's all you need for parts.

Or you can use the v3 board, add an lm1815 circuit in the proto area, and wire it up to use the 2 vr sensors in the stock CAS (and you still need the 3 4.7k pullup resistors).

Or you can use the "tried and true" configuration, which is what tofuball is running on his car:

v2.2 board with a dual lm1815 circuit, and 3 4.7k pullup resistors for the outputs. It is recommended that you use error*'s msns-extra daughterboard to do this, although I didn't use that board for any of the megasquirts I built. If you want to build your own board on a radio shack perf-board then the following parts are necessary (with digi-key part numbers)

2 lm1815 chips: LM1815N-ND
2 18k resistors: 18KQBK-ND
2 1M resistors: 1.0MQBK-ND
1 82k resistor: 82KQBK-ND
2 4.7k resistors: 4.7KQBK-ND
1 56k resistor: 56KQBK-ND
2 .33 uF capacitors: 493-1138-ND
2 .01 uF capacitors: 399-1868-1-ND
2 330 pF capacitors: 399-1920-ND

I believe error* created his own bill of materials for his daughterboard.

There are also other options that I didn't cover here, like using a v3 board but with error*'s board for example.

What option are most people going to go for?

So I guess the parts needed are those I listed above
the megasquirt partial kit
the megasquirt digi-key BOM stuff

Or you can just go to www.glensgarage.com or www.diyautotune.com and buy their kits which come with all the parts you need to build a stock megasquirt.

jayroc

I am lookin to do the v2.2 board with MSnS-E and use error's board. I haven't heard back from him yet, guess he's been preoccupied. Hope he gets back soon, tho, cause I can definitely use one of his boards..
Error's board info can be found:
http://www.msefi.com/viewtopic.php?t=9621&start=30
I believe he has a thread started somewhere around here close, also..

Hey, when you set the spark tables, what advanace are you running? I think for the 12a total advance happens around 4k and is max recommended is 24 degrees BTDC..
Do you have a spark table handy, muythaibxr?

doridori-rx7

iTrader: (3)

when you say 'cut 2 teeth out, is this leading or trailing to the exsisting event point. ? what are the draw backs to this method..

Rex4Life

I don't see the advantage to modding a stock CAS. The only drawback I see for using the stock--unmodded CAS is having to add an extra conditioning circuit. Full leading and trailing with the stock ignition system is now working--but yea you need that extra circuit.

I plan to go with the stock CAS--that way if I must revert back to the stock ECU for emission testing I can still do it. Or even IF I ever sell the car, I can throw in the old stocker.

My 3cts.

Scott

muythaibxr

The advantage is that you can use a standard v3 board (except for the +5v pullups).

The disadvantage is that your CAS will only be usable with the megasquirt if you cut the teeth out.

You can cut whichever 2 teeth you want as long as you align it right when you drop the CAS back in.

After thinking about it, you can even leave the CAS in the stock position, and when configuring the wheel decoder, just set it like this:

trig/return a: 12,2
trig/return b: 6,8

with a trigger angle of 60 degrees.

muythaibxr

I think we're using 25 degrees at full throttle and it's brought in around 3500-4000.

total advance with a vacuum is close to 40 degrees with a vacuum. I'll take a screenshot of the advance table we're using in tofuball's car later tonight.

I based it mostly off of renns' table here:




I changed the idle advance to be around 2 or 3 degrees (-5 gives the best idle with a stock flywheel... we're using a lightweight one so I just did what I had to in order to make the idle smooth). Also, a lot of our lower rev part of the table aside from idle has the advance slightly higher. Last, our table is 12x12 since we're using msns-extra, and renns was using an old pre-msns-extra firmware. I'm also pretty sure we brought in full advance sooner.

Ken

petree_777

iTrader: (1)

ok so first where can i buy the LM1815N-ND chip to solder in and where do you put the pull up resistors. so now you say that you dont have to cut any teeth off the cas or what. Last you giving 40 degrees advanced at full vacume. isint that a little much. or am i lost here. this is for a full trailing and leading right.

Thanks muythaibxr

vince

muythaibxr

You can buy all the parts I listed at digikey, or mouser.. there are other places but those are the places I've used.

The pullup resistors go from the negative lead of each of the LED's to any +5v source, usually I solder them to the +5v side of a nearby resistor. I'll take a picture as soon as I can.

The table above is for leading only, there is a separate table for trailing that is specified in degrees of split between 0 and 20 degrees.

I've never hit the high rev full vacuum part of the table, so I don't know if that's ok or not, I don't think it'll be a problem though as I believe renns pulled his table mostly from what the stock computer does. Maybe he can comment here on that.

I'll post what I'm using for my table later today if I get time, or tomorrow otherwise.

jayroc

Great info, thanks, muythaibxr.

Using that map specifically, I could see you getting into those areas if you took it up to 8k or higher then left off the throttle and let the engine brake.. My 12a will get down to 20 or 25kPa on decel, especially from high RPM. The 13b would too, I imagine.
I noticed that table is going from 12 to 25 very quickly.. Wouldn't you want a more gradual oncoming? I saw where renns was talking about getting that info from the FSM as well, though..

doridori-rx7

iTrader: (3)

COUld you layout where those replacement resistors go on the V3 board ( and the 2.2 for those choosing to go with that )

One, more thing I thought of when you said this, just for clarification: the two teeth cut, are they side by side or are they at different points , seperated by other teeth.

the setting, assumes that the crank is in a specific place and a specific time.. what is that point in time TDC?

as far as dropping it back in, in the stock location is this TDC with zero deg. or can we be a bit looser with that..

12-2 I understand but 6-8 has me confused a bit.. could you explain this setting ?

muythaibxr

yeah, and we haven't had any misfires or anything else bad when engine braking from about 6000 rpm down, (aside from the fact that fuel isn't tuned there on our fuel maps yet) so I think the advance is fine where it is. I've seen kPa's as low as 15 on engine braking and high revs on the 13b.

The 12-25 doesn't seem to be an issue. The only issue I've seen so far is that with my table, the engine idles extremely smooth with an advance of between -5 and 3 degrees, but then above idle, with light throttle, there's a bit of an oscillation. I haven't yet pinned it down to an issue of advance or an issue of fuel either because we're also running like 12.2:1 on idle (again b/c that's where it idles the smoothest), but we jump to 14.7:1 under light load and low revs above idle. So we still have some tuning to do... and it may or may not be advance related.

muythaibxr

You probably want to read the short howto I've done for wheel decoder settings here: (I've posted it before in other threads, and nobody has said anything about it)

http://www.speakeasy.org/~culverk/wh...oder-howto.txt

It is assumed that you've at least been through the wheel decoder settings in megatune, so doing that would be a good idea too.

If you cut two teeth out, they will be opposite of each other, then you configure the megasquirt to read a 12-1 wheel instead of a 12 tooth wheel with extra teeth.

When you drop it back in, you set the engine to the 5 atdc mark, and stab the cas back in with the trig a return tooth in line with the vr sensor for that wheel. So for my configuration, I dropped it back in on tooth #3 (the third tooth after the 2nd wheel's tooth passes its vr sensor) which corresponds with 0 degrees. After I started the car, I set the static advance to -5 in megatune, and used a timing light and turned the CAS until the -5 mark was under the pin. This is all covered in the howto I just posted, so read that, then ask whatever questions you have

Also, I'll take a pic of the pullup resistors on a v2.2 board tonight and post it. I haven't done that mod to my v3 board yet, but the location that the pullups will go in is similar to the v2.2 board.

renns

One aspect that's missing from that ignition timing map is the effect of the 'vacuum control solenoid valve' on the vacuum signal line to the dizzy. When that solenoid is actuated, the engine-side of the vacuum line is closed, and the dizzy side is open to atmosphere, effectively eliminating the vacuum advance. From the FSM snippet below, vacuum advance is disabled at idle (below 1000~1200 rpm), during decel, and when a/c is turned on.

doridori-rx7

iTrader: (3)

I think I'm ready to pony up and buy a setup.. Iv'e got a nice clean 91' that's been turboed and I think it would be a good plan to dso a complete pictorial on HOW to do this.. step by step useing the ver3 /MS1 and the two MOST common methods mentioned by muythai


NOW who in the toronto/400 corridor that's worked on these for a time is wanting to tack technical credit on this and come up and help out with this install...

renns

I'd gladly give you a hand, but I avoid the big city like the plague! My shop is an hour's drive west on 401. If you'd like to stop by and hash out the details on the bench first, I'd be happy to do that. I've got a full MS install running in my SA, and another configured as a bench test mule.

Is your plan to install the MS within an old ECU case for plug-n-play functionality? That'd be really slick.

muythaibxr

That is EXTREMELY interesting... does the 2nd gen stock ECU do that as well?

I may have to implement some code that says "if rotary is enabled, and the tps says the throttle is closed, and we're below a certain rpm, use a specific advance for idle" or something like that.

renns

Given the full 12x12 ignition map, most of those characteristics could be reflected directly in the advance table. The only one not possible is the a/c. It's interesting how turning on the a/c throws a bunch of advance in at the same time. I'd guess one of the myriad of idle speed control schemes kicks in at the same time, bumping the idle speed up as well.

I haven't studied the FC in detail, but there were ignition maps from the stock n/a and turbo FC's posted over on the nopistons forum a while back. They'd be a good baseline to use as well.

Rex4Life

I'd also be interested in seeing a stock turbo FC map. We may also be able to decipher the Microtec and Haltec base maps if necessary, but stock would be better.

For the MS do we have to zero out the timing or can we leave it at the stock marks? Seems like we should be able to go with the stock CAS setting and adjust the timing table to compensate.

renns

Here's the link to that thread. Part way down the tables are in .txt format for importing into your favourite spreadsheet. http://www.nopistons.com/forums/inde...howtopic=45742

muythaibxr

I'm not sure I understand what you're asking, so I'll just say how to set up the megasquirt software for the rx7

You just have to set up the spark settings in the megasquirt correctly, then check the timing with a timing light while adjusting CAS.

To go with the stock CAS setting, you don't adjust the table, you adjust which teeth you use as triggers/returns in the wheel decoder settings... and you adjust the trigger angle...

For the stock CAS position, I'd use these settings for the wheel decoder:

12+1 wheel

Trig/ret A: 12,2
Trig/ret B: 6,8

and use 60 degrees as your trigger angle, then start the car, and adjust the timing as normal by moving the CAS.

Oh, I'll have to take the screenshot of the ignition table we're using tomorrow since the laptop that has the table is still on tofuball's computer.

petree_777

iTrader: (1)

one quick question do you need to flash the v 3.0 with rotary code or is it cool. and you dont need to cut the teeth of the cas and just run it off the trigger teeth how you said it off of like 12 and whatever

muythaibxr

the v3 is a board, the chip is the same across the v3 or v2.2 board, so yes you have to flash it.

Also, it's up to you whether or not to cut the teeth off the CAS, You just have to build the extra circuitry if you don't cut the teeth off.