Members Login
Username 
 
Password 
    Remember Me  
Post Info TOPIC: Very simple questions, could use some help getting started


Veteran Member

Status: Offline
Posts: 71
Date:
RE: Very simple questions, could use some help getting started


 

It should not matter witch version it is or that you are using a 16bit ecu  it works on an 02 gsxr 1000 it just dosnt know the defiance .

the drop down dosnt work when data  is active 



__________________


Senior Member

Status: Offline
Posts: 161
Date:

Hey, it works.

I looked at the bus loading. It actually looks like the PC is sending data as well or something like that and it causes collisions? I really wish I knew more about the ecueditor program. Don't have a lot of time to spend with that part of the project just yet.

I have most of the interface circuits designed now. I still have not chopped up the harness and only test one circuit at a time with the rest of them fudged out. I finished up the fuel injector section and was playing around when I noticed they have a multi-pulse mode. Does this only activate at low speeds and do they ever run more than three pulses?



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Also, when the coolant temperature gauge rails, the tachometer continues to work correctly. Maybe another timer problem...

Now that things are sort of working, I am seeing that the tachometer in ecueditor reads exactly double what it actually is. The cluster is correct, and matches with my synthesizer. The limiter kicks in at the right RPM. It's a serial command so not sure how this could be wrong.





__________________

You can lead a horse to water but you can't make it drink.  



Veteran Member

Status: Offline
Posts: 71
Date:

I finished up the fuel injector section and was playing around when I noticed they have a multi-pulse mode. Does this only activate at low speeds and do they ever run more than three pulses?

This is a starting enrichment thing .

Now that things are sort of working, I am seeing that the tachometer in ecueditor reads exactly double what it actually is.

strange never seen that one works good for me .

Also, when the coolant temperature gauge rails, the tachometer continues to work correctly. Maybe another timer problem... 

The tachometer has a separate wire form the ecu to the tach , not the same wire as gauge data.



__________________


Senior Member

Status: Offline
Posts: 161
Date:

>>>The tachometer has a separate wire form the ecu to the tach , not the same wire as gauge data.

Good to know that it's not on the serial link. I assume the ECU handles the engine over speed function and that happens at the right RPM. So how could it report double RPM in the serial link when it seems to know what the correct RPM is?   Strange....  


Thursday I will bring in a second PC to try this ecueditor software on and see if I have any better luck. Obviously that fact that I had to remove that timer call to get it to run is a bad sign..


That makes sense that it's some sort of enrichment thing as it can triple the amount of fuel. I can bring it up at idle, then just crack the throttle real fast. There is no closed loop engine stuff yet, so the RPM just stays constant. Then I see the three pulses. Then it stops.  Blip the plate and it repeats.   Almost like an accelerator pump on a carburetor.

 

Earlier I had posted a schematic of my ignition coil simulator.   There was a MOSFET clamp that is used to set the flyback voltage as well as short the secondary to induce ignition faults.    This picture shows the MOSFET and other parts mounted to the heatsink with a Phoenix connector.  I made a small dam an filled it with epoxy to seal it up.  

 

 

 

 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Veteran Member

Status: Offline
Posts: 71
Date:

 

I can bring it up at idle, then just crack the throttle real fast.


What you have there is acceleration transient enrichment there is a place in ecu editor to adjust that .


sorry I misunderstood you question .


What I was thinking you wear talking about was when all 4 injectors fire at the same time at start up .

 



 



 



__________________


Senior Member

Status: Offline
Posts: 161
Date:

So far I have had only one injector connected to my test box. You just saved me from asking another question once I have them all wired in...







__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Does the box ever send out more than one ignition pulse per cycle?

__________________

You can lead a horse to water but you can't make it drink.  



Veteran Member

Status: Offline
Posts: 71
Date:

There are only 4 ignition events in 720 degrees of crank rotation.



__________________


Senior Member

Status: Offline
Posts: 161
Date:

Thanks. Just wanted to make sure they did not do anything like MSD, or attempt to burn raw fuel in exhaust to heat a cat, etc.


I ran across this on NI's site. Had no idea they had such a setup.
forums.ni.com/t5/Components/Automotive-Engine-Simulation-Library-for-HIL/td-p/1015707



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Had a small setback with my simulator yesterday and shorted the primary and to the secondary in my ignition load.   It looks like I did not get all of the air out when I potted it.    I ended up changing the wire and wrapping the primary coils first with a layer of tap, then the secondary.    Then back to potting the whole thing up again.   

When I was working out the loads for the injectors, I took them all the way to 300K ohms and it did not fault out until 6000 RPM.  With 100K I could not get it to fault out.   This is very nice as I can get away with very little power for that part. 

Room has been a wreck since I started this project.  Harness spread out everywhere.   Step in the wrong area and wires would fall out.   So it was a relief when the last circuit checked out and it was time to cut the harness.   Picture shows about all there is to the system.   

  

 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I have switched to the 32-bit controller. I went to do the mods to the harness, then realized the harness was for the 32-bit controller.

I am looking for a mechanical drawing for the crank gear, or something that gives all of the angles for the teeth.

__________________

You can lead a horse to water but you can't make it drink.  



Veteran Member

Status: Offline
Posts: 71
Date:

I posted this on this forum a long time ago

 

http://i6.photobucket.com/albums/y250/zx10bud/rotorpick2.jpg

 

 



__________________


Senior Member

Status: Offline
Posts: 161
Date:

I had seen that picture (actually it was the only one I had found) but had questions about it.

When you look at the high versus the low part of the teeth, they show the low part being shorter than the high. They use the falling edge, so does the rising edge matter?

When you look at where the cam is positioned (the missing tooth), the picture shows the tooth is sort of filled in. Does the signal see this falling edge just before the cam mark and the rising edge just after the cam mark?

Is the cam signal detected on the rising or falling edge?


I have most of my test box wired now. Ran into a snag with that 100 ohm resistor in the key switch. But now I see that the 32-bit box appears to be looking for the back EMF from the injectors. Argh.... On the plus side, it appears not to affect the output signal and just throws a code so I may be alright. My interface box is getting pretty packed and not sure there is room to be adding loads at this stage...


__________________

You can lead a horse to water but you can't make it drink.  



Veteran Member

Status: Offline
Posts: 71
Date:

When you look at the high versus the low part of the teeth, they show the low part being shorter than the high. They use the falling edge, so does the rising edge matter?  Yes

 

When you look at where the cam is positioned (the missing tooth), the picture shows the tooth is sort of filled in.

I traced around a zx12r trigger wheel just to make a quick drawing the busa wheel is not like that .

 

The busa wheel looks like my Kawasaki convershion wheel only bigger

 

.

 

Does the signal see this falling edge just before the cam mark and the rising edge just after the cam mark?   Yes.

 

Is the cam signal detected on the rising or falling edge?         Falling . All reluctor sensors trigger on the falling edge.



-- Edited by Zx10Bud on Sunday 22nd of December 2013 01:16:50 AM

__________________


Senior Member

Status: Offline
Posts: 161
Date:

Excellent. Thanks.

I made up the new waveforms and they seem fine.  

Ecuedit is running on the other PC now.    Actually, it's very useable. 




-- Edited by bigtoe on Sunday 22nd of December 2013 05:32:17 AM

__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Is there a way to get better resolution from ecueditor's tachometer?





__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1233
Date:

Try the one in WRT, it has the actual RPM in text as well as the gauge.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

While playing with ECUEDITOR and switching units, this whole other monitor screen pops up then it throws an exception.    See attached screen shot.   It looks like someone was starting to add some new graphics to it.

>>>Try the one in WRT, it has the actual RPM in text as well as the gauge.

Is yours built off of the ECUEDITOR code?  Is it open source?   And most importantly, Does it work???  

 

 

 

 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

There are a few things left to solder but for the most part the mechanics are done. As you can see, it takes up some desk space.

I am still not sure about the fuel injector errors. I hooked up my first test loads I had made for the coils to one of the injector outputs and it clears the code. It does not appear to change the on time or angle.

Next step is the model.

If anyone has some clean sound recordings of their Hayabusa, post them up.



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Veteran Member

Status: Offline
Posts: 71
Date:

While playing with ECUEDITOR and switching units, this whole other monitor screen pops up then it throws an exception. See attached screen shot. It looks like someone was starting to add some new graphics to it

 

That is the 08 and up gen2 screen the gen1 has to be used with the gen1 you must open a gen1 Bin file and use the data screen that pops up thear.

 

looks like this



-- Edited by Zx10Bud on Monday 23rd of December 2013 11:49:20 PM

__________________


Veteran Member

Status: Offline
Posts: 71
Date:

 

bigtoe

 

You have done a fantastic job on your stimulator look much nicer than the one I made 5 years ago.

 

Much more sophisticated than mine but I just made mine to do the job .

 

I could not have made my carbureted Kawasaki to gen1 hayabusa conversion kit that I sell without it.

 

 



__________________


Senior Member

Status: Offline
Posts: 161
Date:

I am not sure why this screen comes up when I change the units. No big deal, just don't change the units and the problem is solved.

Yea, that new Suzuki has been out long enough and it was time for me to finally buy one. All of my bikes have mechanical fuel systems on them. So this is new territory for me. Thanks for posting those pictures. Looks like an old early 80's Phillips on the right. Good to see someone else actually diving in.

My setup would not have helped you with your work as you have a motor that drives your custom gears and use this to drive the ECM.

With my simulator, you have all the data from the ECM live on the PC. Windows is not good for real time so I have hardware that looks at each cylinder's fuel injection time and angle. I capture all three events in hardware. The same for the ignition (except I only look for one event). The hardware is also what is creating the waveforms and driving all the I/O of the ECM, so everything is synchronous. All of the crank torsional noise is done in hardware. So, if you wanted to make one cylinder weak, that is all hardware. I measure time down to 1uS resolution and cam angles down to 0.1 degree. Everything is digital. Even all of my analog outputs are digital. Basically, one IC sitting in the PC doing it all. The hardware tells the PC when it's time to do something (like run the model) using an interrupt. So far its a mix of Labview and C. I am hoping that with the new PC, there is enough power to run the engine model in Labview. Have no idea what I am doing, so this should get real interesting.

I'll make a new video once I have it doing something.   Picture showing the front view. 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1233
Date:

bigtoe wrote:

While playing with ECUEDITOR and switching units, this whole other monitor screen pops up then it throws an exception.    See attached screen shot.   It looks like someone was starting to add some new graphics to it.

>>>Try the one in WRT, it has the actual RPM in text as well as the gauge.

Is yours built off of the ECUEDITOR code?  Is it open source?   And most importantly, Does it work???  

 


 No the code is not based on ecueditor, it has been written new from the ground up, its not open source, the open source model doesnt really work with this stuff, not enough people willing or able to contribute,  It has a record function as well so you can record data and play it back slower or jump to a certain section etc.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

That's very impressive, to develop the code with no research of the ecueditor project. You had to figure out the protocols and everything from scratch. I'm way to lazy and would have to start with looking at what was done (pretty much what I am doing now). Riding the shoulders of the giants! haha You have the advantage of not having to fix all the bugs.

>>> And most importantly, Does it work???

Does it work better than ecueditor? More compatible with PCs and other software (virus scanners, etc), more compatible with different hardware, less exceptions generated.

__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1233
Date:

The basic SDS protocol information was used but the rest of the code was written from scratch in a different language and simplified significantly. I am also one of the developers (the only remaining one ha ha) of ecuedtor..

 

 



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Guru

Status: Offline
Posts: 1233
Date:

and yes i think it is more stable than ecueditor, we have support for many more models and variations of bin files etc as well as that was the one of the fundamental reasons for developing the WRT software from scratch.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

I'll have to read up on it. My plan is to make my own mods to the ECM and I will need a way to make adjustments with the application as well. Do you offer some sort of scripting language that I can use to make my own menus, etc? If not, how does one use it to develop their own code? Are you wanting people to contract with you to co-develop new features?





__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I made a little Christmas video showing my HIL simulator. 

 

 



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Does anyone know where the older versions of the source for ecueditor are archived?





__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I spent some time looking over the ecueditor code and stripped it way down. I removed all the internet, tweeting stuff and anything that was not supported by the Gen1.    I then started looking at how it works but not knowing visual basic, I ended up starting to write my own test code.  

Looking at the checksum, for example in ecueditor:


        l = 8
        i = 0
        Do While i < (l - 1)
            c = b(i) + c
            If c > 256 Then c = c - 256
            i = i + 1
        Loop
        If c <> 0 Then
            b(l - 1) = 256 - c
        Else
            b(l - 1) = 0
        End If

Really, just read add the payload modulo 0x100 or mask with 0xff, subtract 1, then invert.  No need for all these test conditions.  

When I looked at the BAUD rate, they run it at at a non-standard time.   If I just reprogram the Boost by Smith adapter to that rate, and look at the 8 bytes being sent, the last byte is indeed the checksum and calculates out correctly.   The last two bytes of the payload appear to be the tach.  What is interesting is that it is not even close.  

On the plus side, there is no need to do all the power cycling to sync and it has no effect on the cluster.  As suspected, ecueditor is asynchronously writing to the port which I suspect is why the cluster stops working.   

It appears the first two bytes are the coolant temp.  What's interesting about this is the values range from about 70 counts to about 50,000.   16-bits of resolution.  Not sure where you come up with the 10bit analog inputs?  



-- Edited by bigtoe on Saturday 28th of December 2013 11:57:46 PM

Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Looking at another freeze problem with ECUEDITOR, problem happens when selecting the ignition maps. Again, running inside of VS, the app runs fine. Run it stand alone and it locks the PC. In an attempt to isolate the problem, I would just exit the program. After several reboots,


IgnitionMap.vb

Public Sub SelectMap(ByVal map As Integer)
'System.Windows.Forms.Application.Exit() ' makes it here
LoadMap()
System.Windows.Forms.Application.Exit() ' Locks up before here !!!!!!!!!!!!!!!!!!!!!!!

Public Sub LoadMap()
IgnitionMapGrid.SelectionMode = DataGridViewSelectionMode.CellSelect
'System.Windows.Forms.Application.Exit() ' makes it here
End Sub


It appeared the stack was getting messed up. I commented out the entire Loadmap routine, then started adding code back to see if I could find the offending problem. After spending the day, in a nut shell, all of the problems I have been having with ECUEDITOR on this PC are caused by the virus scanning software:

WEBROOT SecureAnywhere

I am not sure why it causes the PC to hard freeze. Turning off WEBROOT and the original program runs fine!!






__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I saw the notes about how to request the data from the ECM. 

http://www.romraider.com/forum/viewtopic.php?f=3&t=3406

Changed my program to pull  88 locations from the ECM.  Then adjusting inputs to the ECM and seeing how things change.   I can see the two locations for the tach.   However, for the coolant I only see the one byte.   Strange when you power the system up and just sniff the coolant, you get a 16-bit number.  

I have been unable to find any posts about where the 16-bit numbers are stored.  Any help?

 

 

 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Looking at the data the ECM normally sends to the cluster,

D0(7..0) - Coolant (10..2)
D1(7..6) - Coolant (1..0);  D1(4) - Set when service jumper active; D1(3) Set when extended faults present; D1(2..1) 0 when not in service, otherwise reports status of TPS alignment;  I am not sure what the other bits are used for.
D2(7..0) - Fault code Extended (6..3) Injector faults
D3(7..0) - Fault code 
D4(7..0) - I am not sure what this byte is used for, it appears to always be 0
D5(7..0) - Tach(15..8)
D6(7..0) - Tach(7..0)
D7(7..0) - Checksum

When I attempt to playback this format and indeed the gauge will remain active, however the ECM hangs. If I wait for a few seconds, it comes back and will respond to requests again but by then the cluster pops up CHEC.

After some testing, it appears there is a work around that will prevent the ECM from hanging. I have it working with my software now and am still able to keep the cluster refreshed even when pulling down all 88 locations.

It appears that there are 256 locations that may be read.  Requesting data beyond this causes it to wrap and show the first location.  

Looking at all the locations, I do not see where the the status of the service jumper can be requested.   So while most of the cluster could be made to work normally when ECUEDITOR is running, there would be a few small differences.  I was thinking I could just set this bit when any errors are present.  This would at least get the coolant gauge working along with the fault codes. 

If everything were available, the Boost by Smith analog GPS indicator could have been all digital just by tapping onto this data bus.

Also, I attempted to install and run the program from WR.  Interesting enough, it also locks up my PC when Webroot is installed and active.  I have installed the free version of Microsoft SE and both of these programs seem to work fine with it.

I tried running RomRaider with Webroot and it seems to run however, I have no idea how to get it to talk with the Gen 1 ECM.
-- Edited by bigtoe on Thursday 2nd of January 2014 01:31:10 AM



-- Edited by bigtoe on Thursday 2nd of January 2014 01:40:12 AM

__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I am trying to figure out how the timing works.   The basic problem I am having is that my fuel and timing numbers don't match with what ECUEDITOR shows.

On the ECUEDITOR, they show Fuel PW.  What the heck is this supposed to mean?   Is it pulse width?

The attached picture was taken from the EFIC-101 manual from Schnitz.   Notice that this shows TDC 1,4 at the falling edge of the 8th tooth after the minus 1.  The other picture shows it at the rising edge of the 8th tooth.  Any idea where TDC 1 really is?

When I look at the ECM signals with a scope what I measure appears correct with where I place the true TDC1.  For the ignition timing, I am looking at when the output of the ECM goes high, or when the ignition coil would fire.   For the fuel injector, I look at when the driver first turns on (pulls low) for the start of injection.   For both fuel and timing, I measure the time that the drivers are active with a 1uS resolution.  I have verified what I read against a scope, again pretty much dead on.   

It appears the CAM location has nothing to do with the timing.  I can move that all over and it seems to have no effect on the operation. 

ECMEDITOR is all over the place for ignition timing.  It can be off both directions and it seems be quite a bit.   The ECM I am using has never been flashed. 

With the bike in neutral...

At 100 RPM, I read 0 crank degrees,  ECUEDITOR shows 8

At 500 RPM, I read -1.0 crank degrees,  ECUEDITOR shows 8 (- ATDC),    9 degrees of error

At 600 - 800 RPM, I read -1.0 crank degrees,  ECUEDITOR shows 19,    20 degrees of error

At 900 RPM, I read 12.4 crank degrees,  ECUEDITOR shows 26, 13.6 degrees of error

At 5,000 RPM, I read 40.2 crank degrees,  ECUEDITOR shows 43  

At 10,00 RPM, I read 34.4 crank degrees,  ECUEDITOR shows 44  

While the actual timing seems smooth, EDUEDITOR shows harsh switch points.   

 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1233
Date:

ecueditor gets the ignition timing value from the SDS data packet from the ECU itself, so it appears this is not particularly accurate for ignition timing.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

I can't make any sense of the fuel PW.

RPM = Int(b(2)) * 100
'FUEL = CInt(((b(8) * 256) - 1024) * 2.47 / 100) * 2 ' if 48, then *1, if 24 then*2
'FUEL = Int(((b(8) * 12.8) - 64))
Fuel = Int(((b(8) * 12.2) - 48))
IGN = CInt(((b(9) * 2.56 / 10) * 1.31) - 3.5) + (0.75 * RPM / 1000)
'IGN = CInt(b(9) * 2.56 / 10)

It looks like they were playing with both the fuel and timing. Maybe they just never sorted it out?



__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1233
Date:

I didnt have much to do with the original Engine Data Screen in Ecu Editor. That was developed by Petri.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

Any idea what happened to the original developers? Seems they just stopped posting and closed the no-ip link down.



__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1233
Date:

Petri got a new Job and became very busy and stopped active development of ecueditor a couple of years ago, the site was hosted on his personal server which was extremely slow and would often go offline, so i took over the hosting of the ecueditor site and pay for this out of my own funds to keep it available. There were one or two other developers but there were not really major contributors.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

That's too bad that he has no time for it. I missed out on the gold rush. Would have been nice to have been involved during that time.

Vacation is over and its back to work here as well. Managed to get quite a bit done and planning to put together another clip to show my progress. Would like to show some of the software running but my camera doesn't do too well when looking at the screen and the instrument cluster at the same time.

__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Me explaining some of the hardware features in my simulator.   There is quite a bit to it and I don't cover everything I am doing, but you will get a fair idea.

 

 



-- Edited by bigtoe on Tuesday 7th of January 2014 06:47:56 AM

__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

In this video I explain some of what I have learned about ECUEDITOR over the last month and how I am planning to address some of the problems with it. 

 



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Obviously making videos is not my strong point.

This next video I go over the software that runs my simulator as it stands today. In the other videos, I have talked about the signal generator side of things. This video will show how I plan to capture the data from the ECM that will eventually be used to run a closed loop simulation. This is all work in progress but things are starting to come together.



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

Here are a couple sweeps using more data points.   You can see the difference between the 1/4 and 2/3 fueling.  

 

 



Attachments
__________________

You can lead a horse to water but you can't make it drink.  



Guru

Status: Offline
Posts: 1247
Date:

Very nice bigtoe.

What software are you using to program the fpga and develop your gui? What hardware chipset are you using for the fpga?

Once you get the data you are looking for all documented in a stock ECU, if you you are interested, there are a few items that are a bit unclear with ECU Editing. back in teh beginning, things like ignition dwell mapping were developed, but not really tested to a great extent. Simple tests such as changing ignition dwell values in ECU Editor and checking to see if they actually change teh requested amount, as well as seeing where the dwell change is added (i.e. how does it affect the ignition timing with a dwell change) are the main ignition questions.

I am very impressed with your work. Great to see interest in this.

I am a tuner and tinkerer who tries to follow the programming side of things. I can hang, but never furthered my abilities to decipher or create code. I do work with data aq a bit both on the automotive and industrial side of things, although not so much at core programming. A bit of labview or simple things here and there.

John

__________________


Guru

Status: Offline
Posts: 1233
Date:

wow, i am really impressed at what you have done, send me an email when you get a chance (and if you want to) as i would like to discuss some stuff with you woolichracing at gmail.com



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Senior Member

Status: Offline
Posts: 161
Date:

Thanks. 
"What software are you using to program the fpga and develop your gui? What hardware chipset are you using for the fpga?"
John, I use both Xilinx (ISE) and Altera (Quartus).   I use VHDL.  This design is not using any non-standard logic beyond the devices RAM.   Currently it is  running on an Altera device.   

"Once you get the data you are looking for all documented in a stock ECU, if you you are interested, there are a few items that are a bit unclear with ECU Editing. back in teh beginning, things like ignition dwell mapping were developed, but not really tested to a great extent. Simple tests such as changing ignition dwell values in ECU Editor and checking to see if they actually change teh requested amount, as well as seeing where the dwell change is added (i.e. how does it affect the ignition timing with a dwell change) are the main ignition questions. "

I think we are on the same page.   My goal is to make changes to the ECM's code but before I start looking into that, I need to make sure I understand what the tools are doing.  There is a lot to learn and right now, I could not even tell you where TDC is! 

"wow, i am really impressed at what you have done, send me an email when you get a chance (and if you want to) as i would like to discuss some stuff with you woolichracing at gmail.com"

I am not sure how much help I could be at this point.   Feel free to PM me if you have questions that you don't want in the public domain. 



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I can't seem to get the kill feature to work. I thought it may be the way I simulate the GPS so I am back to testing with a rotary switch and a pot for each gear. This has no effect.

I tried changing the value of the resistor (thats included in the harness) by +/- a few K and this had no effect. Voltage at the pink wire does drop as expected.

Clutch is released, tested all 6 gears, RPM in the 5 - 10K range, all had no effect.


I bought a harness from Boost by Smith and just replicated what he has inside my test box. No big deal. I could not find a manual for it that explained the software or programming and I wonder if there is something I need to do when flashing the ECM to enable these hacked features?

The real fun begins....



__________________

You can lead a horse to water but you can't make it drink.  



Senior Member

Status: Offline
Posts: 161
Date:

I have the kill working now but I suspect I have uncovered a problem with the harness I purchased from Schnitz (made by BBS). Looking at the code for the BKING:

#define ECU_AD_GPS *(volatile unsigned short *) 0x00804318 // tested to work with 220 ohm resistor


After hunting around on the net, people wrote about having to use the 220 ohm to ground to enable the kill. The harness I purchased has a 41.2K ohm from the red to ground when the horn is pushed. I assumed this harness was built correctly and people just did not know series and parallel and when a 220 was used it would be in series with the red wire to kill.

How does the ECM like the flyback from the shift solenoid? Even if I used a diode for the flyback its still 12ish volts feeding into it. Just seems like there is so much wrong.... Fill me in!




__________________

You can lead a horse to water but you can't make it drink.  

«First  <  1 2 3 4 5  >  Last»  | Page of 5  sorted by
 
Quick Reply

Please log in to post quick replies.

Tweet this page Post to Digg Post to Del.icio.us


Create your own FREE Forum
Report Abuse
Powered by ActiveBoard