It is currently Wed Jul 18, 2018 10:02 pm



Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 132 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next
Author Message
 Post subject: Re: o5e board project
PostPosted: Fri Jan 22, 2016 12:28 pm 
Offline
User avatar

Joined: Fri Jan 22, 2016 12:13 pm
Posts: 1
Raspberry PI Foundation offers the RPI as a "compute module" which contains the basic guts of the RPI without all the interface sockets and such.
It sits on a DDR2 SODIMM type board and is meant for integration into other products where USB, video outputs and such locations need customized, and where other peripherals might be integrated on the main board.

_________________
printed circuit board assembly


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue Jan 26, 2016 9:36 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
Sorry it took me so long to realize your post was waiting approval :(

I think what you're describing is similar to the MPC5634 Development board right?


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue Jan 26, 2016 1:42 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
mk e wrote:
Sorry it took me so long to realize your post was waiting approval :(

I think what you're describing is similar to the MPC5634 Development board right?

No. The compute module is a self-contained board with the processor and support components in a SO-DIMM form factor- like laptop memory modules. It lacks the pin headers and off-board connectors (HDMI, audio, USB, SD-card, et cetera) and is intended for use in a 'final' project after prototyping and such has been done with a regular PI board.

https://www.raspberrypi.org/products/compute-module/

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue Jan 26, 2016 4:59 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
I guess I was mostly trying to figure out were he was going thought process wise. You can break a system up in any way you please and then use what connectors for the parts and all probably make sense in certain applications but when cost is a primary concern breaking things up in any way is rarely the best option...but it adds flexibility to the design.

So I'm trying to understand if he's suggesting an alternate layout or an alternate processor?


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue Jan 26, 2016 8:49 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
I think the idea of an alternate processor has some merit, i.e. one baseboard with standardized connection to a processor module. Then, if one has a penchant towards ARM, MPC, Propeller or whatever, build it and start writing.

However, a Rasp PI has more doo dads than a project like this needs, and not enough of what it does need.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Wed Jan 27, 2016 12:42 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
abecedarian wrote:
I think the idea of an alternate processor has some merit, i.e. one baseboard with standardized connection to a processor module. Then, if one has a penchant towards ARM, MPC, Propeller or whatever, build it and start writing.

However, a Rasp PI has more doo dads than a project like this needs, and not enough of what it does need.



Funny.
At the very beginning or this project I was looking to do something like the phytec SOm (system on module) with a general purpose I/O and a standardized processor board plug connector so any processor could be used going forward......connectors are too expensive, there arn't pre-made processor boards at a good price blah blah blah and it never went anywhere.

Then I wanted an I/O that the Freescale Dev board would plug onto .....easy! and when they upgrade we can upgrade easily....oh I want 4 fuel an 2 spark but others want 8/8 we shouldn't force people to take.......more blah blah blah.

So ok, then I wanted a modular system like arduino on steroids.....oh we should just use their shield which are readily available....pay no attention to the simple fact that they have no where near enough pins to build a useful stack up.

So I said, fine, we can build a main board that will hold 2 or 4 arduino shield stacks....oh those shields look cheap but once you get everything you need it's way more than a dedicated ECU I/O board would be.

That was about when I threw up my hands and focused on finishing my engine and getting it running to a known good ECU......and once it's running I'll test an o5e frankenECU unless Marcus is ready to go with his stuff then I'll try that


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Thu Jan 28, 2016 4:17 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:52 am
Posts: 304
Location: Over here, doing 'over here' things.
I'm feeling sheepish since I was in the middle of most of that confusion.
I feel worse 'cause I didn't follow through with much of anything related to it.

But in my defense, I was naïve and ambitious, at the time, and have since toned things down.

_________________
/me goes off to the corner feeling like Jerry Springer with a mullet.

My O5E candidate: 1982 Honda CX500TC motorcycle.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Jan 29, 2016 8:59 am 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
abecedarian wrote:
I'm feeling sheepish since I was in the middle of most of that confusion.
I feel worse 'cause I didn't follow through with much of anything related to it.

But in my defense, I was naïve and ambitious, at the time, and have since toned things down.


A lot or blame to go around. They were all fine ideas....and all suffered the same fate in the hands of a HW designer who didn't know what he was doing and couldn't follow even simple instructions.

But that's the past. What Marcus is doing is also a fine plan and since he's actually doing it its the best solution :)

I think when I get freed up from my engine build project I'll get the frankenECU parts back and put that together because for development work and flexibility that really is the best solution and I'd like to see at least one up and running.

I also very much want to start transferring the control logic I build for my enginelab unit into o5e because I think it's more complete, more modular (which I think is critical to making it flexible) and more and just plain better....but it seems like waiting until I get it tested on my engine in a few weeks make the most sense so I'm not transferring mistakes.....just never enough time.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Jan 29, 2016 2:11 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
because for development work and flexibility that really is the best solution


I tired to do a good job of breaking up the functionality between what is on this CPU card and what could plug to it. Here is a picture of a generic proto board sitting on the CPU card. It's not a great photo but you can see 2 groups of 26 pins sticking thru the holes.

One group carries 8 injector and 8 coil signals, actually just 16 eTPU outputs. The other group carries SPI, some analog and digital pins and 5V and 12V protected and filtered power.

On the side you can see 40 pins of analog interfaces that can take a ribbon cable for breakout.

The board will be assembled with female sockets for the two 26 pin fields because they have driven pins and should be protected.

Are there reasons why the dev board is better for Franken work as opposed to this board? I can always make changes for the next spin.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Jan 29, 2016 2:43 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
Horizenjob wrote:
Are there reasons why the dev board is better for Franken work as opposed to this board? I can always make changes for the next spin.


The answer is it depends on what work you're trying to do. The Dev board has pretty much every I/O pin available and it connects easily to code warrior so right the code and click send, and just plug in the USB and it's got everything it needs....I keep one in my desk draw and can work at lunch. Easy....and easy is about my limit.

What you have is a great start on a proper ECU with an easy way to add extra stuff....but personally I'd probably still play with the code on a dev board before loading it into the ECU board just because the dev board is 1 cable and an ECU needs a test bench. Franken stuff scares me reliability wise so you have a big advantage there and if I didn't already have the dev board I may not buy one if I had your setup already....depends how easy it is to load and test new code.

The enginelab setup is even easier because the tuner is also a simulator so everything I'd want to play with can be pretty well tested before it gets anywhere near the real HW......but remember I make mostly mistakes most of the time when I code anything so my perspective may be somewhat unique.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Jan 29, 2016 4:03 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Forgot to attach the picture for my above post....


Attachments:
ECU_8b_protoboard.jpg
ECU_8b_protoboard.jpg [ 186.08 KiB | Viewed 7464 times ]
Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Jan 29, 2016 7:03 pm 
Offline
User avatar

Joined: Sat May 11, 2013 9:45 am
Posts: 729
Location: PA, USA
Horizenjob wrote:
Forgot to attach the picture for my above post....


Very cool, can't wait to see it with parts on it!


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Apr 22, 2016 12:31 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Sorry, im getting late to de discussion, any updates on this project? I was interested in testing something like this, I have some experience designing circuits for automotive applications but I been installing and programing aftermarket ECU for 15 years, mostly racing cars. Now I have 2 projects that need installation, so I start designing a new hardware but selecting a processor I get to this page. Project number 1 is a drift car all motor 4 cylinders FJ20 rally engine with ITb’s. Second project is a 24’ Pantera boat with a 2JZgte VVTI. I’m already running a Jeep Cherokee with a 2JZ vvti on a MS-3 . If you need any support just let me know.
Looking forward to try something new.

Please a update will be nice


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Mon May 23, 2016 11:24 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Porelmundo, sorry for the very tardy update. There have been a couple of things keeping me away from this project for a little while.

I need to get back on this because my car project has made progress but I do feel stalled and I'm not sure why. RIght now I'm considering just pushing thru the next version of the board. If I do that I'll send in the parts for assembly with the board files and have them do the fab and assembly in one go.

You guys can help me think this thru. I should get another assembly quote, but my initial impression is that it costs several times as much to assemble a board as opposed to fabricate it. So my initial 10 board fabrication run cost perhaps $650 but the assembly would be about $2500 not including parts. In both cases larger quantities would be much cheaper, a large part of the cost is simply setup.

I understand the motto "Perfect is the enemy of good enough", but am struggling to understand why I feel stalled on this. Part of what happened is that I continued work on the schematics after I had the boards fabricated. So the circuitry is largely correct on the original, but the part numbers and all their gory details where not filled in until later. Unlike software though it is very difficult to "diff" a pair of schematics or board layouts.

So after a couple of attempts I see I stall at the level of cross checking every part number both for the BOM and the pick and place file. If I had a minion I would be happy to spend even a few days in the lab mentoring them for this task, which oddly is enough time for me to do it... Go figure...

The fixes I remember on the new board are the following:

- Some diode packs to clamp the VR signals before they reach the Maxim chip. Perhaps at this point I could also add jumpers for enabling Hall sensors.

- My protection circuit for the analog inputs was not well conceived. I applied external Schotky diodes on the inputs, but they are not isolated from the on chip units so cannot be expected to provide protection ( thermal runaway etc. ). I would leave the parts off and rely on the internal diodes. The new board would use a two stage filter and apply the protection in the middle, much nicer.

- My header pins for the external drivers board are slightly off grid from each other. With some difficulty they fit but it would be nicer for them to be right. Being able to easily plug not only the driver board but various test and proto boards based on standard "Vector" boards ( like in picture above ) would be better.

- A small number of caps with 0805 footprints instead of 0603.

- Little silkscreen problems.

In addition to those fixes the board grew just a bit at the end of the design phase to fit a slightly larger enclosure. That extra space allows a few more analog channels and possibly a few other small changes. I don't really want to touch the bulk of the layout however.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Mon May 23, 2016 7:10 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Excellent, provably i can help with some stuff, but first i think that the price for prototyping that you pay is to expensive when provably can be done for less with a Chinese manufacturer, perhaps i can assembly a board for testing if i have a stencil.
also you can share the board layout ans schematics to see if i can do it for less.
I'm working with mike to see if i can test the code on my rally car.

Regards,
Rob


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue May 24, 2016 1:53 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Hi Rob, lots of stuff to talk about. I'm not sure that going offshore is the best thing at the moment. The people I'm working with have been very helpful and I think their work is first rate. I am able to spec everything about the boards and their turnaround is quick.

On the list above we should add that I think it makes sense to transition to working with panels of boards instead of single boards like I did the first go around. That should also drop the assembly price substantially.

I don't have stencils, the board has components on both sides, but just small ones on the bottom. We could get stencils. It would be a lot of effort to do manually though. Depending on what you want to get to, you could also just leave a lot of parts off the board, I don't know.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Sun Jun 19, 2016 9:17 am 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Any chance to see those KiCad or Eagle Files? main and driver board.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Mon Jun 20, 2016 10:24 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
We can work something out. Where are you located? This work is all in KiCad.

My original plan was to fully open source this, but I've had some second thoughts because it's not clear what the legal status of something like copyright is on hardware. My biggest concern is actually the board layout. I don't think the schematics themselves have much of anything unexpected in them.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Mon Jun 20, 2016 2:00 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Yes im looking forward to work something out and I am located in Puerto Rico.

I don't think you going to have any issues with legal stuff but i not a expert in that field nether so lets keep it like that for now until we have it clear.

My idea about all this is to collaborate in some way and to speed up the process of research. trying to make a plan on connectors, pinout location, connectivity to the I/O board and enclosure. is not all about my car, i want to be testing something by the end of the summer. Also i have the opportunity to try it on some pretty fast cars that i already have in schedule so for that i will need some support.

Rob


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue Jun 21, 2016 1:38 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Puerto Rico is good, we are in same time zone and shipping should be quick.

Quote:
I don't think you going to have any issues with legal stuff


So if you can just say you won't use the Gerber files to produce units, I'll send you the whole set of files. We can try email and if that doesn't work I can send you a pointer to the Google drive file for a backup.

Quote:
My idea about all this is to collaborate in some way and to speed up the process of research. trying to make a plan on connectors, pinout location, connectivity to the I/O board and enclosure.


This is all good. In return I'm willing to provide you with some free units. prototypes for testing and development. You get to say what you think would make these desirable units.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Tue Jun 21, 2016 6:39 pm 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Quote:
if you can just say you won't use the Gerber files to produce units, I'll send you the whole set of files.


I won't. i have no intention of take advantage on this i just want to study it and think ahead, do my best to make it possible maybe people start to be more interested in this project, I'm willing to contribute in any way i can.

Quote:
In return I'm willing to provide you with some free units. prototypes for testing and development.

That will be nice, as soon as we can make it work, more people will join us and do something before it get obsolete.

Quote:
You get to say what you think would make these desirable units


It's been provably more than 10 years that i say to myself that I'm going to design a ECU, i never have the time to do it, but instead i so another better opportunity doing a TCU that i already made. with the past of the years i seen a lot of failures and missing things in other ECU's most of the problems related to Drivers, and the most important thing Power Supply that's why always say "i can do it better" but never do.
Some of this open projects are overdesign i can work with less, and that's what manufacturers use simplicity over quality.
What I'm looking for is flexibility, improvement capabilities and learning overall.

I will send you a PM with my contact information.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Wed Jun 22, 2016 12:38 am 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
Quote:
What I'm looking for is flexibility, improvement capabilities and learning overall.


Me too.

When you get the files maybe the power input is a good first thing to look at. I tried to protect from reverse battery and to limit the startup current to something reasonable like 10 or 20 Amps. The box will need an external fuse. I looked for a resettable fuse "polyfuse" but didn't find something that looked right and could take a lot of heat.

I used a large TVS diode also, but will move up another size to the one rated for "load dump". This would blow a fuse though if you try to jumpstart with 2 batteries. Maybe that's the right thing to do, I don't know. Sometimes I spend too much time worrying about these things, so I look forward to another vote. :-)


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Wed Jun 22, 2016 1:10 am 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
Quote:
The box will need an external fuse. I looked for a resettable fuse "polyfuse" but didn't find something that looked right and could take a lot of heat.


serious Automotive racing applications use resetable circuit breakers same as we use on airplanes they will stand a lot of heat,

Quote:
I used a large TVS diode also, but will move up another size to the one rated for "load dump". This would blow a fuse though if you try to jumpstart with 2 batteries. Maybe that's the right thing to do


Very good idea.


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Wed Jun 22, 2016 10:02 am 
Offline
User avatar

Joined: Thu Apr 21, 2016 9:25 pm
Posts: 51
A tvs like the TPSMD22A will do the job. Some of the older ECUS use a MR2535L but is obsolete.
For the polyfuse, what you looking is something like the
1812L150/24MR


Top
 Profile  
 
 Post subject: Re: o5e board project
PostPosted: Fri Jun 24, 2016 2:46 pm 
Offline
User avatar

Joined: Thu Mar 26, 2015 11:55 am
Posts: 112
Location: Massachusetts, USA
I missed your last update, Rob. I have a TPSMD-17 in the schematic now. I was thinking of moving up to an SLD20-018. That's might be too much for the board and it's wiring though.

Maybe the 1812L150 would work. I just wish they were more perfect devices! The real world sucks. At 70C it's minimum trip is 0.75A which is maybe marginal, but I am not sure yet how much current the board will draw. Then it takes 1.5 sec to trip at 8A, but it is only rated for 20A. I'm not sure what would pull 8 amps when it breaks on the board and how much collateral damage there would be, during that 1.5 sec. The 12V part trips much quicker. It's just frustrating.

There are power control chips that wuld do the slow startup and over current shutdown but I got discouraged with them too for some reason, but maybe just cost so $5 for a part to do that may be reasonable.

I'd rather make progress than agonize so we'll make a choice soon and move on.

I have a weird sense of humor, but maybe it makes sense to do this with the CPU? You would use a current sense resistor that you monitor with an A/D channel and drive an And gate or flip-flop to turn off a MOSFET.


Top
 Profile  
 
Display posts from previous:  Sort by  
Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 132 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next

Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Theme designed by stylerbb.net © 2008
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
All times are UTC - 5 hours [ DST ]