RXV 550 with a MoTec ECU [Archive]

Mbirt

01-06-2012, 12:17 PM

You'll have to make the same choices any of us do when designing an EFI system:

-Can your ECU handle the number of cylinders and the firing order? Sequential or wasted-spark?
-Can your ECU handle the crank trigger pattern and sensor type of your stock engine?
-Can your ECU output to the actuators you desire to use?

and so on...

Really, the biggest concerns are the 77 degree firing order and the 30-2 trigger pattern read by a VR sensor. If the ECU can handle the firing order, you can still fabricate a custom trigger scheme like those of us with engines which were once carburetted.

mk e

01-06-2012, 12:20 PM

I've done a few MoTec M800 installs on Ferraris... so Italian at least http://fsae.com/groupee_common/emoticons/icon_smile.gif

What exactly is the question?

If I don't know the MoTec Forum is quite good and frequented by the MoTec engineers.

mk e

01-06-2012, 12:31 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:
You'll have to make the same choices any of us do when designing an EFI system:

-Can your ECU handle the number of cylinders and the firing order? Sequential or wasted-spark?
-Can your ECU handle the crank trigger pattern and sensor type of your stock engine?
-Can your ECU output to the actuators you desire to use?

and so on...

Really, the biggest concerns are the 77 degree firing order and the 30-2 trigger pattern read by a VR sensor. If the ECU can handle the firing order, you can still fabricate a custom trigger scheme like those of us with engines which were once carburetted. </div></BLOCKQUOTE>

Once you've plunked down the money for a MoTec there is very little it can't do.

Basically any trigger pattern
Individual cylinder tuning maps
Cam, boost, idle, etc
traction control (which works really nice)

The issue is the price.....I thin it's about $3k to get started and the M800 is $5k or so.

I'm working on an open source ECU

http://fsae.com/eve/forums/a/t...25607348/m/402100613 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/402100613)

That should be able to do anything a MoTec can for a whole lot less money - $300-$500 but we are just starting to test so there is a little ways to go yet.

Mbirt

01-06-2012, 12:39 PM

http://fsae.com/eve/forums/a/t...607348/m/35220301941 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/35220301941)

http://fsae.com/eve/forums/a/t...607348/m/29610042541 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/29610042541)

http://fsae.com/eve/forums/a/t...607348/m/39510557341 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/39510557341)

Adambomb

01-06-2012, 01:09 PM

One other thing to mention, part of the money that goes towards a Motec buys you some pretty awesome tech support. Plus you can call the Australian number in the middle of the night if you're desperate. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

Drew Price

01-06-2012, 02:21 PM

'Ey there mate, we needs haaaalp!!

Jon Burford

01-07-2012, 01:42 AM

Hi
If your unsure about ECU choice I strongly recommend
www.dtafast.co.uk (http://www.dtafast.co.uk)
either an S60 or S80 depending upon extra functionality you want.
these ECU are super easy to get up and running with new engines.
We just got our Yamaha Genesis in line twin to run really quickly. it's 180 offset, odd firing, no missing teeth with hall effect cam.
The support from Alan at DTA was and still is very quick in response and really helpful.
Also there is a great discount offered by him for FS teams.
My opinion anyway

B Lewis @ PE Engine Management

01-07-2012, 06:19 PM

Hi Noam,

We had several teams running our ECU with the Aprillia last year. The PE3 can handle uneven firing, the stock trigger pattern and uses the stock sensors and actuators. It is a VERY attractive price for FSAE teams (less than $1000) for the ECU and wiring. Send me an email if you have any questions.

Kirk Feldkamp

01-07-2012, 08:41 PM

Check out the work that Seb at NLR did on a SXV 550. They make a custom cam cover to cleanly mount a sync sensor. I believe they used a M400 as well.

http://www.apriliaforum.com/fo...otoype-Turbo-Aprilia (http://www.apriliaforum.com/forums/showthread.php?159543-NLR-Formula-Protoype-Turbo-Aprilia)

http://www.apriliaforum.com/fo...-sxv-odd-fire-angles (http://www.apriliaforum.com/forums/showthread.php?148708-aprilia-450-sxv-odd-fire-angles)

-Kirk

noam

01-08-2012, 01:43 AM

Thank you all for your responds

Let me clear the table here, we have a MoTec M400 ECU, we are considering the aprilia for our next car.

On our last engine (Yamaha XJ6) it was fairly easy to add a camshaft sensor and having a new trigger wheel made for optimal synchronization.

On the aprilaia things are a bit different because the trigger teeth are on the magnet flywheel and it’s not that simple to fabricate one for my needs.
For a camshaft sensor – I am not that sure I can reliably use the cam lobe as trigger, am I wrong? As far as mounting the sensor, I think I can do much more simple design then fabricating a new cam cover….although it is really impressive what NLR did.

How does the MoTec copes with the 30_2 configuration and the 77 deg?

Jon Burford

01-08-2012, 02:28 AM

a DTA ecu would be able to handle the stock magnet flywheel assuming missing teeth straight away so no modification would be needed. we are using raised lumps on the a/c generator rotor with our engine.
I have never tried using a cam lobe for sensing.

I have never worked with motec, I tried to play with the software and I didn't like it as much, so I stuck with DTA. I can only imagine though that it would be fine with 30_2

mk e

01-08-2012, 05:41 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">

For a camshaft sensor – I am not that sure I can reliably use the cam lobe as trigger, am I wrong? As far as mounting the sensor, I think I can do much more simple design then fabricating a new cam cover….although it is really impressive what NLR did. </div></BLOCKQUOTE>

I've never done it but I've seen it done several times without any issue. Just get the sensor close (1mm) and you should be good.....and Iv'e done some pretty massive changes this way.

The custom cam cover is very slick, but I normally just TIG weld on what I want and sand smooth which is much quicker and can also look quite good.

Here's my current project, you'll find lots of examples that start with taking a sawsall to a casting then reconstructing it in a more useful shape.
http://www.ferrarilife.com/for...-conversion-v12.html (http://www.ferrarilife.com/forums/projects-rebuilds-modifications/12282-308-v12-conversion-v12.html)

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">How does the MoTec copes with the 30_2 configuration and the 77 deg? </div></BLOCKQUOTE>

You type in 30-2 and the crank rotation degrees between the cylinders (77 is either 283 or 487 depending which you call #1 and 0 degrees). Simple.

noam

01-08-2012, 08:35 AM

http://a3.sphotos.ak.fbcdn.net/hphotos-ak-snc6/190715_10150177353398126_609033125_8208572_3116785 _n.jpg

this was my solution on our last engine, i think i will do a similar solution with the aprilia just with a different triggring point to read from

Kirk Feldkamp

01-08-2012, 10:26 AM

The machined-bung-and-welded method works great too. It's not a big deal, but keep in mind many of the case covers are mag for motorcycle engines.

The big question as to whether or not you use a cam sync is if you want to do sequential injection or not. If you're only doing crank sensing, you won't know which half of the cycle the cylinder is on. In this case, the injector just fires once per revolution rather than only during the intake stroke. There are myriad effects from this... which may or may not be a problem for you. That's up for you to decide.

I think there are a few new OE systems that look at the fluctuations in RPM rate (aka crank rotational acceleration) to determine where in the rotation the engine is rather than adding a sync sensor. That said, any aftermarket ECU that's capable of doing things sequentially will have a sync sensor.

Noam, another method if you want to change your trigger wheel to a different tooth count or orientation is to shrink fit a new thin, toothed steel ring over the outside of the rotor. That said, the 30-2 should work just fine. mk e is correct about how to set it up. If you're still unsure about how the missing tooth algorithm on the Motec's work, check out the webinar section on motec.com. You will have to measure what the CRIP is on the engine itself (or at least get it close then fine tune it with a timing light once it's running). The concepts are very well explained in a couple of the archived videos.

noam

01-08-2012, 10:57 AM

I think my main concern at the moment is, if I use the original crank teeth (30_2), I am limited in the range of crank rotation from the index tooth in which the cam signal can be produced. If I am going on one of the cam lobs as index I have only 2 options to play with. I need to know if one of them falls in the desired range. If not, it's a whole new game for designing a cam triggering system.

"I think there are a few new OE systems that look at the fluctuations in RPM rate (aka crank rotational acceleration) to determine where in the rotation the engine is rather than adding a sync sensor. That said, any aftermarket ECU that's capable of doing things sequentially will have a sync sensor."

Kirk, can you elaborate on that please. How does this works if for every cyl you have the same effect of piston velocity. In the end, the crankshaft is a sum of all cyl no?

mk e

01-08-2012, 11:13 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by noam:
I think my main concern at the moment is, if I use the original crank teeth (30_2), I am limited in the range of crank rotation from the index tooth in which the cam signal can be produced. If I am going on one of the cam lobs as index I have only 2 options to play with. I need to know if one of them falls in the desired range. If not, it's a whole new game for designing a cam triggering system. </div></BLOCKQUOTE>

I don't think a MoTec cares at all where the Cam signal come in relation to the crank?

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">
Kirk, can you elaborate on that please. How does this works if for every cyl you have the same effect of piston velocity. In the end, the crankshaft is a sum of all cyl no? </div></BLOCKQUOTE>

Not if it's an uneven fire. The ECU can look at the missing tooth then when the rpm changes and determine which cylinder is in compression. The other option is the ECU can look at the first rpm change and then how long to the next. Neither of these will work on an even fire engine though becasue as you say, it's a "net zero" thing. H-D uses the MAP signal vs the missing tooth to make the sync determination.

noam

01-08-2012, 12:02 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by mk e:

I don't think a MoTec cares at all where the Cam signal come in relation to the crank?

</div></BLOCKQUOTE>
it does. +-70 deg from the index tooth.
Last year I had to learn the hard way. I ended up making 2 new trigger wheels because I misunderstand this issue and had some problems synchronizing the reff and sync.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">
Not if it's an uneven fire. The ECU can look at the missing tooth then when the rpm changes and determine which cylinder is in compression. The other option is the ECU can look at the first rpm change and then how long to the next. Neither of these will work on an even fire engine though becasue as you say, it's a "net zero" thing. H-D uses the MAP signal vs the missing tooth to make the sync determination. </div></BLOCKQUOTE>

Interesting sh**t, love to hear more….
Thanks

mk e

01-08-2012, 12:44 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by noam:

it does. +-70 deg from the index tooth.
Last year I had to learn the hard way. I ended up making 2 new trigger wheels because I misunderstand this issue and had some problems synchronizing the reff and sync. </div></BLOCKQUOTE>

I'd forgotten all about that.....I've been so buried in my own code (which doesn't care) that I'm losing my mind. sorry.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">
Interesting sh**t, love to hear more….
Thanks </div></BLOCKQUOTE>

There really isn't more. If you hav a 2 cylinder with un-even cylinder spacing, 1 cylinder is in compression "x" teeth form the missing tooth and the other "y" teeth so by watching the MAP signal or the cranking rpm and when the change comes from the missing tooth you know which cylinder you are looking at. With no missing tooth you can look at the time between MAP or rpm changes and know where you are.

Personally I'm not a fan of systems like this that can't detect an error condition.....with just a non-missing tooth crank signal you could lose count and never know. I'm sure they do some fancy tooth time-out checking so I'm nervous about nothing but.....

On my Open5xxxECU project we've been discussing using H-D ion sensing coils to do ion knock detection but also to use the ion signal for the sync event (the signal is different depending on cylinder pressure). This can be done on an even fire engine as long as there are at least 2 ion signal input channels to the ECU and you don't wire pairs (like 1&4, 2&3) to the same channel. We haven't done it yet though....we need to actually get an engine running first then the fancy stuff can come.

Dewi Griffiths

01-09-2012, 09:26 AM

Hi noam,

Here at Cardiff University we run an Aprilia SXV550 (We have had an RXV too) with a Motec M400. They work fine together. It's a great ECU that allows you to run the engine under normal mode or "big bang mode".

The Motec is a very versatile ECU and quite simple to get to grips with. We have only just scratched the surface! The only thing I will say is that it is rather pricey to start with will also take up a great chunk of your vehicles cost so could hurt you in that aspect of the competition. If cost is an issue for you, take Jon Burford's advice and go with the DTA.

Jon Burford

01-09-2012, 09:35 AM

It should also be noted that the DTA has all of the plain engine running functionality of the MOTEC, it too can run in two stroke mode etc etc etc. It's not a second best unit by any stretch of the imagination.

mk e

01-09-2012, 10:01 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by dewigriff:
The only thing I will say is that it is rather pricey to start with will also take up a great chunk of your vehicles cost so could hurt you in that aspect of the competition. </div></BLOCKQUOTE>

Unless the rules have changed since my day, you account on production costs not what you had to pay for 1 piece. If you look at the actual stuff that is in the ECU and the cost to build 10k of them you are looking at about $200 for most ECUs or MAYBE as much as $500 if there's tons of data logging memory and stuff in it.

You might get called on the software but there are 2 open source ECU projects out there right now that other FREE software

o5e - http://code.google.com/p/open5xxxecu/ (my most excellent project for a top shelf processor)
FreeEMS - http://forum.diyefi.org/ (based an a common and in expensive industrial processor)

HW - $200
SW - Free
Prototype - Motec M400, DTA, etc

I would think the really clever teams know this.....http://fsae.com/groupee_common/emoticons/icon_wink.gif

Mbirt

01-09-2012, 10:29 AM

For those interested in phase detection from rate of change in RPM, check out the BRP/Rotax SAE paper on the development of the 600 ACE 180 deg. parallel twin engine. There's at least one nice graphical representation of the logic.

mke,

The academic price for the Woodward ECM 5554-112 is pretty close to the DIY price you've quoted. Can the Open5xxxECU project be used to tune the Woodward ECM with Tunerstudio instead of MotoHawk/Simulink?

mk e

01-09-2012, 10:51 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:
mke,

The academic price for the Woodward ECM 5554-112 is pretty close to the DIY price you've quoted. Can the Open5xxxECU project be used to tune the Woodward ECM with Tunerstudio instead of MotoHawk/Simulink? </div></BLOCKQUOTE>

Yes.....but you'd need to know what processor pins are on what.

I actually talked to them a while back about being the hardware supplier for the project and there was mild interest but they weren't willing to give me the info I needed to allow the o5e code to be ported to their hardware so the o5e just team decided to just design our own.

So, if somebody can get the pinout info from Woodward or wants to cut one open and reverse engineer it, the o5e code can probably be VERY easily ported to that hardware - like 1-2 hours of work unless they are using SPI controlled drivers then it would be more like 20-40 hours because we don't yet support SPI, it's planed and we do have some SPI stuff on our board, we just haven't gotten there yet.

I did really like their 112 connector and spec'd it on our board.

I'm pretty sure that GM and Chrysler ECUs also use a MPC5xxx processor. I think the both used a 5554 until last year and now use a 5566?. Point being if you had the pinout info these ECUs could also run the o5e code and be fully tuneable with TunerStudio

Mbirt

01-09-2012, 11:09 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by mk e:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:
mke,

The academic price for the Woodward ECM 5554-112 is pretty close to the DIY price you've quoted. Can the Open5xxxECU project be used to tune the Woodward ECM with Tunerstudio instead of MotoHawk/Simulink? </div></BLOCKQUOTE>

Yes.....but you'd need to know what processor pins are on what.

I actually talked to them a while back about being the hardware supplier for the project and there was mild interest but they weren't willing to give me the info I needed to allow the o5e code to be ported to their hardware so the o5e just team decided to just design our own.

So, if somebody can get the pinout info from Woodward or wants to cut one open and reverse engineer it, the o5e code can probably be VERY easily ported to that hardware - like 1-2 hours of work unless they are using SPI controlled drivers then it would be more like 20-40 hours because we don't yet support SPI, it's planed and we do have some SPI stuff on our board, we just haven't gotten there yet.

I did really like their 112 connector and spec'd it on our board.

I'm pretty sure that GM and Chrysler ECUs also use a MPC5xxx processor. I think the both used a 5554 until last year and now use a 5566?. Point being if you had the pinout info these ECUs could also run the o5e code and be fully tuneable with TunerStudio </div></BLOCKQUOTE>
http://www.neweagle.net/suppor...CM-5554-112-0902.pdf (http://www.neweagle.net/support/wiki/docs/Datasheets/112pin/ECM-5554-112-0902.pdf)
http://www.neweagle.net/suppor...12pin/36350_0904.pdf (http://www.neweagle.net/support/wiki/docs/Datasheets/112pin/36350_0904.pdf)

Port away http://fsae.com/groupee_common/emoticons/icon_wink.gif

Really, if this can be done, I'll be the happiest boy ever. All of my experience is with Megasquirt, but I'm moving our FSAE and CSC programs over to MotoTron hardware. The current plan is to use a MPC563-based controller in the FSAE car and a MPC565-based controller for the snowmobile. The MPC5554-based 112-pin controller is a better value for the feature content, but we'd have to use MotoHawk/MotoTune, which we don't happen to have the budget for.

Dewi Griffiths

01-09-2012, 11:23 AM

@mk e

The cost of the Motec in the cost report is huge IMO and I can tell you it took a large chunk of our car cost (Competition event). It's not so much an expense to the team since it's a component that is carried over.

@Jon Burford, Ddylset ti fod yn adolygu nawr?!

Jon Burford

01-09-2012, 11:30 AM

Why make a custom ECU?
I don't see the attraction.

@Griffiths - wrth gwrs?

Regards

mk e

01-09-2012, 11:32 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by dewigriff:
@mk e

The cost of the Motec in the cost report is huge IMO and I can tell you it took a large chunk of our car cost (Competition event). </div></BLOCKQUOTE>

It don't think it should be was my point. You could be treating that just like you treat everything else and looking at the cost of making it in production quantities. A commercial aftermarket ECU is a development piece not a production piece....a factory flashed production version should be a couple hundred dollars.

mk e

01-09-2012, 11:40 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jon Burford:
Why make a custom ECU?
I don't see the attraction.
</div></BLOCKQUOTE>

Cost

If the team does a custom ECU their is no question you can cost it based on the production cost - $200

There is the secondary benefit of having any feature you please, but cost and how you can account the cost is the biggest driver in my mind.

I know accounting the production cost of an off the shelf ECU was accepted in the cost competition at least once, but that was some years ago now so I don't know that they would still buy that argument, but with a custom ECU you are rock solid calling the ECU a $200 item.

Jon Burford

01-09-2012, 11:47 AM

It's an impressive cost saving for sure.
I suppose if you have the guys that want to do and can do it then it's one thing.

mk e

01-09-2012, 12:13 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Jon Burford:
It's an impressive cost saving for sure.

I suppose if you have the guys that want to do and can do it then it's one thing. </div></BLOCKQUOTE>

Yeah......the downside obviously is the RISK of burning resources and still having no ECU.

That's changing though. With 2 opensource projects rolling it's about to get pretty frikin easy to have a "custom" ecu.

The o5e code is in testing now so not something an '12 team could KNOW they'd have it on time (alhtough I THINK the SW will be working by the end of the week, it's really close), but for '13 it will be done. There are multiple HW options for o5e right now...h*ll you could use a $99 5634 demo board jumper to a MS driver board and and stuffed in a case and you'd have a working ECU for under $200.

FreeScale is getting ready to release an ECU demo unit that is a complete ECU designed and tested by FreeScale that will probably be nearly free to universities and about a $250 production cost (about $500 list). They will also be releasing demo SW to run it, but the o5e SW will also run on this board and has more features.

The FreeEMS project is running engines today. Fred (the guy running the show over there) can be a bit of a D...challenge to get a long with, but the ECU is working and open source and under $200 to build right now today.

With super cheap options out there now I'd think it's going to get nearly impossible to not go with a "custom" ECU and still hope to do well in the cost competition.

Mbirt

01-09-2012, 12:25 PM

Did those Woodward data sheets show what you need to see for the 112-pin controller? The flash version is less than $250 for those of us in academia.

mk e

01-09-2012, 12:41 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:
Did those Woodward data sheets show what you need to see for the 112-pin controller? The flash version is less than $250 for those of us in academia. </div></BLOCKQUOTE>

No, that's just the connector pins. I need to know what processor pin the connector pin ends up at.

If they are using simple drivers we could just load software that sends a signal to each pin 1 at a time and map the thing out......but if the drivers need an enable signal of some kind we'd never see any out put.

it's either get them to share the info or open 1 up and see what there is to see.

Dewi Griffiths

01-10-2012, 04:16 AM

I must agree that making your own ECU is a large cost saving to make however I personally don't see it as a viable solution.
Purchasing one gives you a guarantee, technical support and also an ECU in the post almost immediately.
Although, having an in house designed ECU is something design judges at comp will get very excited about.

@Jon Burf, Gobeithio aeth yr arholiad yn dda. Mae'n bryd amser I ni ddechrau siarad Cymraeg fan hyn! Twll tyn pob sais. 'Nuff said.

mk e

01-10-2012, 05:26 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by dewigriff:
I must agree that making your own ECU is a large cost saving to make however I personally don't see it as a viable solution.
Purchasing one gives you a guarantee, technical support and also an ECU in the post almost immediately.
Although, having an in house designed ECU is something design judges at comp will get very excited about.
</div></BLOCKQUOTE>

You are completely correct.....which is why you just don't see teams making there own ECUs.

An ECU is a HUGE project and IMO well beyond the scope of what can be done by pretty much all formula teams. The o5e project currently has about 6-8 senior to retired engineers working on it, 3 have been basically working on it full time for the past 6-12 months (because they're retired). There is some real talent, like the guy doing a lot of the low level code also did a lot of the code for the Apollo guidance computers and some modern fighters, we have a guy that did a lot of the work on Ford's first ECUs, a guy who's an ECU design consultant for a living and a bunch of other really talented guys and it's still a hard project that takes a long time.

....but we're doing the heavy lifting for you and giving it to you. Free. It's a completely open source/open design so any one can use or modify it in any way they please. I can't tell you how many times I've heard people say "I'd get a MoTec if I could afford one". MoTec USA quoted me $12k-$15k for the stuff I need to control my engine (ok, nobody really NEEDS a 800-900 V12 street engine with 24 individually mapped and controlled injectors so this might be an extreme example http://fsae.com/groupee_common/emoticons/icon_smile.gif )....but o5e can do this and will only cost a few hundred bucks.

FreeEMS is working today (although I see it as inferior to o5e in every possible way http://fsae.com/groupee_common/emoticons/icon_wink.gif ) and also open source, but its GPLv3( which IMO is a virus of a license that infects everything thing it touches but that won't affect an FSAE team in anyway as long as your work is also GPLv3 or shared only with your team you'll be in compliance with the license).

Today a team would be taking a huge risk in heading down the custom ECU path, but by this time next year it looks like there will be 2 solid and fully tested code/processor bases available to choose from. That will make it easy to have a "custom" ECU.....and I think that will make very hard to win at the design & cost competitions with an off the shelf ECU.

mk e

01-11-2012, 10:54 AM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:

http://www.neweagle.net/suppor...CM-5554-112-0902.pdf (http://www.neweagle.net/support/wiki/docs/Datasheets/112pin/ECM-5554-112-0902.pdf)
http://www.neweagle.net/suppor...12pin/36350_0904.pdf (http://www.neweagle.net/support/wiki/docs/Datasheets/112pin/36350_0904.pdf)

Port away http://fsae.com/groupee_common/emoticons/icon_wink.gif

Really, if this can be done, I'll be the happiest boy ever. All of my experience is with Megasquirt, but I'm moving our FSAE and CSC programs over to MotoTron hardware. The current plan is to use a MPC563-based controller in the FSAE car and a MPC565-based controller for the snowmobile. The MPC5554-based 112-pin controller is a better value for the feature content, but we'd have to use MotoHawk/MotoTune, which we don't happen to have the budget for. </div></BLOCKQUOTE>

What I need to know is what connector pin goes to what processor pin......so I still doesn't have enough info to start the port.

Do you have or can you get 1 of these that we can open up and trace the signals?

I started a conversation with neweagle this morning so we'll see where that goes, but as it sits right now it would cost me $500 to order 1. I wouldn't mind ordering 1 if I actually needed it, but I don't. I'm reaching out to a couple friends who I know are waiting for hardware to see is they want to plunk the money down and also trying to negotiate a special o5e price. If you can get me 1 @ $250 for development work I'll buy it from you but at $500......

I'm also aggressively looking at finding an OEM ECU that can be erased and turned into an open5xxxECU. it looks like all new GM and Chrysler ecus are 5566 and would work (if they aren't locked somehow), but I know the last gen of Chrysler ecus are 5554 and will also work and can be had for $100-$150 on ebay.

mk e

01-11-2012, 12:57 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:

Port away http://fsae.com/groupee_common/emoticons/icon_wink.gif

Really, if this can be done, I'll be the happiest boy ever. </div></BLOCKQUOTE>

Start Getting happy. The New Eagle guys are getting one. They said it could take a month or so, but it's coming and shortly there after it will be running http://fsae.com/groupee_common/emoticons/icon_smile.gif

JWard

01-12-2012, 08:26 AM

Mbirt

01-12-2012, 11:55 AM

mk e

01-12-2012, 12:15 PM

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Mbirt:
Awesome! Andy Weiss at New Eagle is a Kettering FSAE alum and is a big reason we're switching over to their controllers. He wanted me to go with the 112-pin controller in the first place, but only the 48 and 128 units are compatible with ProEFI's software. We don't have the money for MotoHawk, of course.

If you need anyone to stop in down there, let me know. I'm on a first-name basis with several of the employees. </div></BLOCKQUOTE>

Our lead programmer, Jon, is in Anarbor so right down the road if you were to want to try out o5e.

I've Been talking to Rich who I think is the owner.....I honestly I keep getting what he's telling me all mixed up which I guess is part of the beauty of email.

Apparently what he was offering was a different brand control he'll be getting in soon, not a woodward. He said he's got 100's of the woodward controllers but thinks we can't use them because it's got a proprietary boot loader that only works with MotoHawk software.

....but all I keep hearing is "Mark, you've got to open the case up and flash in the new firmware using a spring loaded adapter to grab the processor jtag pins directly".

So, before I just call and order one for $500, can you get me one at the $250 price to test on? The part I'm after is "ECM- 0554-112-0904-F".