Kenku

It's a shame it doesn't work that way, because it still seems an elegant way of doing things. Oh well. I think I know where to get some FC trailing coils, so I guess I'll live.

And I can't afford a 20B, so I get to keep wanting one. Or bust out the milling machine and a bunch of 12A irons and try to make a 4-rotor.

... or just build my damn 13B and con someone at the DMV into thinking it doesn't need to be smogged.

2Lucky2tha7

iTrader: (3)

I would like to bring this to light because it is almost always overlooked:
The leading fires twice per rotor face, as well as the trailing firing once. This is why Mazda used a dual post coil for the leading.
What I want to know is:
Does the Megasquirt incorporate firing the Leading plugs twice per rotor face?? Because if it doesn't, then what I'm gathering from what I've read on the forum is that alot of people who have installed Megasquirt on a rotary that only run the leading plugs are only getting one spark per rotor face, instead of the "factory" 3 sparks per face.

I don't know about you guys, but I'd rather have 3 sparks per face for a more complete and consistent burn.

Brian

2Lucky2tha7

iTrader: (3)

If anyone disagrees with me or is confused, watch the ignition cycle of a rotary engine at:
www.rotaryengineillustrated.com
Then click on "Animation Library" on the left-hand side

Brian

muythaibxr

Yes, the megasquirt, by virtue of firing a wasted spark coil is already firing 2 sparks per rotor face. When it's first igniting the mixture on one rotor, it's firing just before the exhaust port opens on the other rotor... just by firing the wasted spark coil.

Right now, as far as I know, tofuball's car is the only one using stock leading with a CAS. As soon as he's done fixing everything he broke in his accident, we'll be adding trailing. I've tested trailing ignition on the oscilloscope with the latest version of MSnS-Extra, and it looks fine and will most likely work on the car.

Jeff20B

iTrader: (13)

It shouldn't be called "wasted spark" on a rotary because the spark isn't actually wasted like it is on a piston engine (where the term came from). People often use the term "wasted spark" on a rotary in an attempt to reduce confusion when describing it to others, but I've found that the confusion level stays virtually the same regardless of what it's called. Perhaps it is because "wasted spark" piston engines are a fairly recent edition to the ever-increasing emissions controls on most modern fuel injected vehicles, and most car guys have been into cars since before wasted spark ignition systems were common.

I've steadfastly called it simultaneously firing or late leading or dual leading sparks per rotor face and other variants with some success. In the end, it all boils down to whatever term causes the least amount of confusion in others. Regarding "wasted spark", since most of us came from the piston world before we ventured into rotaries, "wasted spark" seems to have stuck, for better or for worse. I don't care for piston engines, so I don't use their terminology if I can help it. Another that bothers me is calling the eccentric shaft a "crank" shaft. lol

Kenku

Back to sort of the original topic... how much a difference is intake manifold setup going to make on tuning? I decided to make an MS setup for a streetported 12A, and am wondering if the ITBs are going to make any maps that I'd start from completely invalid.

Not like I'm anywhere near the point of tuning yet (still have to build the hardware end) but this was bugging me.

muythaibxr

yeah, ITB's are going to make a pretty big difference compared to a normal manifold. Unless you have some maps to start from that were done with ITB's, you'll probably want to just start with a generated map from megatune, and tune from there using a wideband unit. That's what I'd do anyway. Wideband isn't necessary, but makes things a LOT easier.

cafcwest

same here brotha!

It has always bothered me in the FSM to label the eccentric shaft and attaching to it is a crank pully.

Kenku

Speaking as someone with a working MSnS-E (okay, granted, its on my friend's 240SX at the moment) I now find myself curious about any word on what's up with the trailing sparks?

Not that I'm too concerned, as the FB dizzy *does* work and all...

muythaibxr

It's working now, the dwell is stable (meaning it never drops a spark), and there are no misses... I've been working on it for the past few nights. The last version I posted had a small bug when trailing first kicks in at 700 rpms, but I've fixed that in the code I'm working on at the moment.

I'm going to post a version with just that 700 rpm bug fixed for people who want to use it in their cars.

The only problem remaining is that for some reason when used with the wheel decoder (instead of on the stimulator) the trailing dwell is too long at high split and low revs... I'm in the process of trying to track down how that's happening right now, so hopefully it'll be fully "ready to go" within a week (depends on when I have time to work on it). I still consider this usable "in car" though.

EDIT: Watch for updates on rotary trailing here:

http://www.msefi.com/viewtopic.php?t=7544&start=120

Ken

Kenku

Ahh, neato. May have to pick up FC ignition bits now...

t-von

iTrader: (8)

Does anyone know how many differant maps can be programed into a megasquirt?

muythaibxr

You can program 2 maps in if you want to use switchable tables.... otherwise you just load new maps off the computer.

For most people this isn't a problem, because you can tune different parts of a single map for different things. Tofuball and I have his car tuned for emissions at cruise, and power at full throttle.

The switchable tables require an extra circuit with a switch and some other things to switch between tables.

jayroc

I am using the MSnS-E mod in fuel only mode with the circuit that GSL-SE Addict made which allows me to run a FC coil and plugs. It works very well if you're interested in an ignition upgrade and want to run the upgraded code.
My car is a FB 12a, so I already have a good mechanical fallback. ie, distributor. So, I don't have to worry about getting ignition to work like the FC crowd.
The work you guys are doing on getting ignition going with MS is awesome. I will probably switch over and have MS manage my ignition at some point, but for now I'll wait.

muythaibxr

Link to a really big pic of the inside of a modded megasquirt with a dual lm1815 vr conditioner circuit:

http://www.speakeasy.org/~culverk/IMG_0014.JPG

Pele

iTrader: (28)

Now THAT is hard core. Heatshrink and everything... Very aesthetically pleasing.

Rex4Life

It's all bigger than I thought. Looks like you added a db25 for some external connections. I'm thinking of using the stock ECU case and connectors. Maybe remounting the stock connectors onto a breadboard and then running a db15 ribbon cable over to a db15 ribbon connector on the CAS signal board. Do those ribbon cables have enough current capability? If all I need to add is one LM 1815 circuit (for the v3.0 board), would one LM 1815 circuit + the db15 ribbon connector fit within the prototype area?

Looks like I've got some design work to do. I also want to add in the EGT circuit. I guess I'd like the ability to remove the main board and not have to unsolder connections.

Scott

muythaibxr

I don't know what those ribbon cables usually have as far as current handling capability. I think they'd be fine for signals, but not for injectors or the BAC valve, as the drivers for both of those generally handle a lot of current.

I also don't know if a 15 pin ribbon cable will fit in the proto area... you should look at the pics I posted in another thread of the v3 board... that should tell you what you need to know.

The only reason I used a db25 is because the store was all out of db9's.

Yeah, I don't think you can fit the second lm1815, another ribbon connector, and the EGT in the proto area... MAYBE the ribbon and the lm1815, but not with the EGT.

Eventually I'm thinking that if I continue selling pre-built ones for rx7's, I'm going to find a source of broken ECU's so I can put all the megasquirt components in old ECU cases.

Oh, and the pic above is actually bigger than the actual size of the megasquirt on most monitors.

2Lucky2tha7

iTrader: (3)

If you'll be selling pre-built Megasquirt ECU's for the rx7, I'll be the first in line!!

muythaibxr

I've already sold a few (the pic above is for a customer), I'm just taking a little break from it until I get a chance to test with a v3 board... and until I get some broken ECUs to put all the megasquirt stuff in (plug 'n play hopefully)

I won't be able to do it anytime soon though, right now I'm working on writing some code to make rx8 ignition systems work. (similar to rx7's, but 4 coils instead of 3, and a really weird looking vr sensor wheel) I'm also going to start porting msns-extra's assembly code to the megasquirt 2. If I were to do both the pre-built megasquirts, and the coding at the same time, I'd have to quit my day job!

Ken

rotarygod

iTrader: (1)

Ken I was thinking about doing the exact same thing to my board with the extra connector. Did you just remove the LED's altogether? Is there anything special you need to install in their place?

muythaibxr

I just looked at a schematic for the led circuit with the pullup, and realized that the LEDs aren't needed, so I cut them out and put the connector in their place. Nothing extra is needed.

trydis7

iTrader: (1)

Hi Ken;

A couple of questions on the mods you've made:

I see the 4.7k LED pullups.

Upgraded voltage regulator from LM2937 (.5A) to LM2940 (1A)?
Did you replace the Air Temp Sensor R4 with a 47K resistor?
50k resistor for the coil rpm sensor?
Are you using the flyback board/injector resistors?


Great work guys!

Cary

muythaibxr

yep, that's correct

It's not necessary to upgrade it, but it doesn't hurt... the one in the pic has a 2940, but the one I'm actually running in a car right now has the 2937, and it runs fine.

I'm using a 42.2k resistor for the people who want to use the stock air temp sensor on a 2nd gen. For 3rd gen people, the standard resistor is fine.

Nope. The standard megasquirt tach circuit is no longer used.... so rpm information doesn't come from the coil anymore. I get the rpm from the CAS. That's the whole point of the wheel-decoder in megasquirt. You just tell it how many teeth your CAS has, and what teeth you want to be triggers, and what teeth you want it to fire on for cranking, and it decodes the wheel and figures out rpm and when to fire the spark plugs from there.

The one I have running in a car right now is on an S5, which has high-z injectors. There's no need for the flyback board on that one.

For an S4 I'd recommend either getting the ms v3 board, or using the stock resistor pack.

muythaibxr

I forgot to mention another mod. If you're going to be tuning with a wideband controller that outputs a signal between 0-5v, you have to leave out or cut out D11. Otherwise your reading will be a bit off, and you'll only ever be able to see up to about 17:1 instead of 19:1 (with a techedge controller, may be slightly different for Innovate or other controllers).