Notices
Series I Aftermarket Performance Modifications Discussion of power adding modifications

Viability of a Megasquirt application for FI on the renesis

Thread Tools
 
Rate Thread
 
Old Aug 15, 2005 | 11:59 AM
  #176  
Sapphonica's Avatar
PingMobile
 
Joined: Feb 2005
Posts: 356
Likes: 0
From: Oakland
Originally Posted by rkostolni
Wow, Nov/Dec time huh? What is it that is going to take so long? Any way us forum members can help speed that along? Maybe donations to help pay for parts, or researching wiring diagrams, just name it if it can help get this along. I am dying to get this emanage out of control!
Likewise. I would be down with a contribution to the mighty war effort.
Reply
Old Aug 15, 2005 | 01:48 PM
  #177  
brillo's Avatar
Go Texas Longhorns!
 
Joined: Apr 2003
Posts: 1,818
Likes: 1
From: Houston, Texas
We certainly appreciate the offers for help, the biggest issue is not so much money, but getting the MS software to read the stock ignition wheel. We are going to rig a standard 36-2 wheel up to the front of my pully and install a seperate ignition sensor to test the ignition side of things, but right now even if that works, I don't have more than my one wheel that I got form Speedsource.

We want this to be as plug and play as possible, to where all you need is the unit, a wiring harness and new coil packs. having to deal with the extra trigger sensor and trigger wheel is a complication we are trying to avoid.

Getting help to write the code would be nice, but its not as simple as just inserting lines of code here and there, you really need to understand the software. If you can help with the code (which is in assembly language) at www.msefi.com that would be great. You can see there is code there for 36-1, 36-2 and a neon trigger wheel aready in the MSnS code.

The time issue is just that, we need to some once we get everything together to test and work the kinks out. Believe me, we want this done as quickly as possible, but this is my daily driver and we have to work out our work schedules.

If anyone has any spare coil packs from a 2nd gen RX7 that would help, as well as a crank angle sensor from any Mazda car(RG, I think any of them will work right?). we would be appreciative. We will certainly take small monetary donations, but thats not required, I'm going to do this anyways for the experience.

I hope we can bring what we have working to Sevenstock in October.
Reply
Old Aug 15, 2005 | 02:43 PM
  #178  
Sapphonica's Avatar
PingMobile
 
Joined: Feb 2005
Posts: 356
Likes: 0
From: Oakland
Originally Posted by brillo
We certainly appreciate the offers for help, the biggest issue is not so much money, but getting the MS software to read the stock ignition wheel. We are going to rig a standard 36-2 wheel up to the front of my pully and install a seperate ignition sensor to test the ignition side of things, but right now even if that works, I don't have more than my one wheel that I got form Speedsource.

We want this to be as plug and play as possible, to where all you need is the unit, a wiring harness and new coil packs. having to deal with the extra trigger sensor and trigger wheel is a complication we are trying to avoid.

Getting help to write the code would be nice, but its not as simple as just inserting lines of code here and there, you really need to understand the software. If you can help with the code (which is in assembly language) at www.msefi.com that would be great. You can see there is code there for 36-1, 36-2 and a neon trigger wheel aready in the MSnS code.

The time issue is just that, we need to some once we get everything together to test and work the kinks out. Believe me, we want this done as quickly as possible, but this is my daily driver and we have to work out our work schedules.

If anyone has any spare coil packs from a 2nd gen RX7 that would help, as well as a crank angle sensor from any Mazda car(RG, I think any of them will work right?). we would be appreciative. We will certainly take small monetary donations, but thats not required, I'm going to do this anyways for the experience.

I hope we can bring what we have working to Sevenstock in October.
I'm pretty sure I can help on the code side. Why are you installing an additional ignition wheel? Is it because the code is not yet written to comprehend the RX8 wheel, because it is hard to intercept the RX8 info, or both?

Assuming that to get the timing to work is an issue of interface code for the ignition wheel, what else do you need once that's in place?
Reply
Old Aug 15, 2005 | 03:02 PM
  #179  
brillo's Avatar
Go Texas Longhorns!
 
Joined: Apr 2003
Posts: 1,818
Likes: 1
From: Houston, Texas
We are installing the additional wheel b/c the car's ECU needs to have the stock wheel for it to be happy and not start shutting off things like the power steering (this is per speedsouce racing), and the MS can't read the stock wheel, therefore, we are using the new one to drive the ignition.

Ideally, we could write into the MS code a way to read the stock 36-2-2-2 wheel. Why Mazda decided to change from the former 36-2 wheel on the 3rd gen I have no idea. If we had the same wheel as the 3rd gen, we wouldn't have this problem.
Reply
Old Aug 15, 2005 | 03:33 PM
  #180  
Sapphonica's Avatar
PingMobile
 
Joined: Feb 2005
Posts: 356
Likes: 0
From: Oakland
Understood. So the MS app timing wheel interface needs to be ported to the 36-2-2-2 wheel.

Once you have that ported, do you also need to interface to the VDI & aux ports? Also, is there some way to mod the MS app so that it doesn't fry the stock coils?
Reply
Old Aug 15, 2005 | 04:00 PM
  #181  
brillo's Avatar
Go Texas Longhorns!
 
Joined: Apr 2003
Posts: 1,818
Likes: 1
From: Houston, Texas
Originally Posted by Sapphonica
Understood. So the MS app timing wheel interface needs to be ported to the 36-2-2-2 wheel.

Once you have that ported, do you also need to interface to the VDI & aux ports? Also, is there some way to mod the MS app so that it doesn't fry the stock coils?
The VDI and Aux are nice features, but not required for FI and I think we can find a way to control them without re-writing the software.

The coils are another issue, I don't think there is a way to run the MS in non-wasted spark mode. RG knows more about this one than me.
Reply
Old Aug 15, 2005 | 04:01 PM
  #182  
rkostolni's Avatar
Registered
 
Joined: Aug 2004
Posts: 1,294
Likes: 0
From: Virginia/Maryland
So what is the next couple steps, and time frame looking like?
Reply
Old Aug 15, 2005 | 05:00 PM
  #183  
rotarygod's Avatar
Registered
 
Joined: Apr 2003
Posts: 9,134
Likes: 26
From: Houston
The only reason we are going to use the additional 36-2 wheel is because the code is currently not capable of reading the 36-2-2-2 wheel. We can't just substitute the new wheel for the old one as then the factory ecu can't read it. This is why we need an additonal wheel and sensor. This is only temporary until we can find a way to get the new code written. Basically we are doing this to so we can work on getting a full tune in the car to show different results of different tuning. We are using this to bypass the waiting so we can keep making progress. Once the new code gets written and is known to work, then we'll just wire it up normally and remove the new wheel. The ultimate goal is plug and play.

The MS can only run rotary ignition in wasted spark at this time. This is because it was designed around the RX-7 ignition systems and both the 2nd and 3rd gen RX-7's use wasted spark mode. I fear that by firing the stock RX-8 leading coils twice as often will kill them. If the code can be changed to allow then to work independently of each other, that would be perfect. However that also means there needs to be a new output integrated into the MS ecu itself. For the total cost of this sytem, new coils still don't add up to a whole lot of money.

The next step at this time is to get it tuned in well as fuel only to see what kind of potential there is with this amount of control. You never know. There might be more than we think there is hidden in this motor. As we get this step down, the next will be to integrate our temporary ignition setup onto the car and then make sure it works.
Reply
Old Aug 15, 2005 | 05:09 PM
  #184  
Richard Paul's Avatar
Registered
iTrader: (5)
 
Joined: May 2004
Posts: 2,438
Likes: 20
From: Chatsworth Ca
Geeezz Fred your avatar is making me ill. That's alot of mass to be rocking. Now I have a moving avatar myself, thanks to BaronVonBigmeat.
Just to hyjack your thread where I know I've got your attention, I need a mock up engine to fab these brackets. Can I use an older engine or is the 8 engine different as far as the mounting points go. I need the bosses for the idler pulley and the engine lifting strap to be the same. I don't know about the other points yet. I know the manifolding will be different but I'll live with that as long as the alt is the same.
Any thoughts??
Reply
Old Aug 15, 2005 | 05:36 PM
  #185  
rotarygod's Avatar
Registered
 
Joined: Apr 2003
Posts: 9,134
Likes: 26
From: Houston
As far as I know, the mounting points for everything "should" be the same. I am not entirely sure though. Dave "guitarjunkie" should know. If you are actually attaching to the housings themselves, you will probably be fine. If you are attaching to the front cover, water pump assembly, etc, you may have problems.
Reply
Old Aug 15, 2005 | 07:26 PM
  #186  
Sapphonica's Avatar
PingMobile
 
Joined: Feb 2005
Posts: 356
Likes: 0
From: Oakland
How will the timing wheel code be spec'ed and tested? I have a vague idea what needs to be written, but not nearly enough detail to make sure it gets done properly.
Reply
Old Aug 15, 2005 | 11:52 PM
  #187  
Razz1's Avatar
Mu ha.. ha...
 
Joined: Aug 2004
Posts: 14,361
Likes: 3
From: Cali
NOW I know your crazy Rotary God.

Look at that Avitar!
Reply
Old Aug 16, 2005 | 12:37 PM
  #188  
zoom44's Avatar
Administrator
 
Joined: Jul 2002
Posts: 21,958
Likes: 115
From: portland oregon
is it possible to only share control of the ignition with the megasquirt or any other ems solution. i wonder if there is any system that could control it but still allow the stock pcm to react by retarding timing due to input from the knock censor?
Reply
Old Aug 16, 2005 | 12:46 PM
  #189  
Sapphonica's Avatar
PingMobile
 
Joined: Feb 2005
Posts: 356
Likes: 0
From: Oakland
Originally Posted by zoom44
is it possible to only share control of the ignition with the megasquirt or any other ems solution. i wonder if there is any system that could control it but still allow the stock pcm to react by retarding timing due to input from the knock censor?
If you can see the knock sensor input, then you can write software to retard the timing on a knock event.
Reply
Old Aug 16, 2005 | 12:49 PM
  #190  
brillo's Avatar
Go Texas Longhorns!
 
Joined: Apr 2003
Posts: 1,818
Likes: 1
From: Houston, Texas
I might help things for those interesting to log onto the MS forums in the MegasquirtnSpark section

Here: http://www.msefi.com/viewforum.php?f...7c594650b483f7

and voice your interest in getting the code written to read the RX8 trigger wheel pattern of 36-2-2-2. I don't want to bombard them to death, but there are quite a few rotary folks over there and I think if they really see the interest someone can step up and help alter the code. Please show/share your support for our project in the MS forums as well
Reply
Old Aug 16, 2005 | 01:10 PM
  #191  
er0sion's Avatar
Registered User
 
Joined: Nov 2004
Posts: 13
Likes: 0
i have some experience with robotics and a motorola hc11 cpu... hmm this is pretty similar .. i was reading the megasquirt manual and ran across this line...

Many OEM and popular aftermarket EFI systems use older processor technology (like the MC68HC11 or Z80) which operate at 1 or 2 Mhz internal bus speed

interesting...
Reply
Old Aug 16, 2005 | 02:49 PM
  #192  
rotarygod's Avatar
Registered
 
Joined: Apr 2003
Posts: 9,134
Likes: 26
From: Houston
Since the factory ecu will no longer have control of fuel or spark when all is said and done, it will also have no way of controlling a knock sensor. Fortunately I know which wire to tap into on the harness. The MS has the ability to control knock. This is easy. For a nonturbo car that is properly tuned, you won't need it.
Reply
Old Aug 16, 2005 | 03:23 PM
  #193  
zoom44's Avatar
Administrator
 
Joined: Jul 2002
Posts: 21,958
Likes: 115
From: portland oregon
itd still be nice to have a "safety net"
Reply
Old Aug 16, 2005 | 09:52 PM
  #194  
rotarygod's Avatar
Registered
 
Joined: Apr 2003
Posts: 9,134
Likes: 26
From: Houston
For anyone that is interested in a possible MS application for the RX-8, please create a name and password on the Megasquirt forum www.msefi.com and add your name and wish for this to become a reality. I have basically started a plea for help from those guys since they have alot of guys very familiar with writing code for MS. If I can get the backing from many others, I hope one of them will write the code for us. The thread that I started over there is:

http://www.msefi.com/viewtopic.php?t=10809

The more support I get the better. This also means the faster I can get everything else worked out and a functional plug and play harness available to everyone. At the very least I can get a complete diy on how to do it.
Reply
Old Aug 16, 2005 | 10:32 PM
  #195  
brillo's Avatar
Go Texas Longhorns!
 
Joined: Apr 2003
Posts: 1,818
Likes: 1
From: Houston, Texas
Please show your support guys, you wanted to know how to help, here is your chance. Lets get this done.
Reply
Old Aug 17, 2005 | 07:35 AM
  #196  
TexasKyle's Avatar
Beeeep
 
Joined: Sep 2004
Posts: 130
Likes: 0
From: Katy, Tx
Posted at MS guys. Also, since I am in Katy, if you guys need anything let me know. I wouldn't mind popping by for a visit sometime just to see you guys working on it. Beers all around. GF is an ex bartender, we can set up a lil bar and set her to being busy! haha Best of luck with this endeavor.
Reply
Old Aug 17, 2005 | 12:06 PM
  #197  
olddragger's Avatar
Registered
iTrader: (3)
 
Joined: Apr 2004
Posts: 10,828
Likes: 40
From: macon, georgia
Megasquirt II is due to be released very very soon. Much more power. Would that make things any easier?
Olddragger
Reply
Old Aug 17, 2005 | 08:07 PM
  #198  
BaronVonBigmeat's Avatar
Senor Carnegrande
 
Joined: Jul 2003
Posts: 871
Likes: 0
Originally Posted by olddragger
Megasquirt II is due to be released very very soon. Much more power. Would that make things any easier?
Olddragger
The processor is from the Motorola HCS12 family. It has plenty of excess capability and is a 16 bit machine allowing as much timing resolution as anyone can practically use. Since it is in the Motorola family the MegaSquirt embedded code will port over quite easily, however the MegaSquirt-II code has been converted to the C programming language since there is a free GNU cross-compiler for this CPU. This will make it much easier for others to customize the algorithms to meet their needs.
Hmm, sooo...does this mean that any old programmer familiar with C could code this thing?
Reply
Old Aug 17, 2005 | 08:15 PM
  #199  
olddragger's Avatar
Registered
iTrader: (3)
 
Joined: Apr 2004
Posts: 10,828
Likes: 40
From: macon, georgia
That was part of my question because it seems so to this old man.
Olddragger
Reply
Old Aug 17, 2005 | 08:16 PM
  #200  
brillo's Avatar
Go Texas Longhorns!
 
Joined: Apr 2003
Posts: 1,818
Likes: 1
From: Houston, Texas
Currently, all the "rotary specific" code is written for MS I only. I'm sur they will port it over at some point but right now, MS II won't really help us.
Reply


You have already rated this thread Rating: Thread Rating: 0 votes,  average.


All times are GMT -5. The time now is 03:01 AM.