Robert's 0.6-alpha Jaguar Build

Andy's GM DIS centric hardware design! Also works as a fuel controller for EDIS and distributor applications.
E4ODnut
QFP80 - Contributor
Posts: 79
Joined: Fri Jun 14, 2013 12:49 am
Location: Gibsons, BC Canada

Robert's 0.6-alpha Jaguar Build

Post by E4ODnut »

0.6, SN 7. This is being built as a direct replacement for my current daily driver which has been running very well with a highly modified MS1 for the past 9 years. I received the bare board and BDM from Andy a week or so ago and my Digikey BOM order arrived today. I'll put together a summary of what the particulars are of the engine and application shortly, so hopefully, as a group, we can figure out how to best modify Jaguar and existing code to make it all work. I have a custom built simulator so I can bench test everything before committing to the vehicle. This has saved my butt many times in the past.
Robert
1995 Ford E150, 300 CID I6, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1992 Winnebago 33 RQ, Ford 460 CID V8, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1993 Bayliner 3288, Twin Ford 351 CID Windsors, Converted to tuned port programmable EFI, Custom MS1-Extra
E4ODnut
QFP80 - Contributor
Posts: 79
Joined: Fri Jun 14, 2013 12:49 am
Location: Gibsons, BC Canada

Re: Robert's 0.6-alpha Jaguar Build

Post by E4ODnut »

OK. Here’s the application, the background and future plan.

I intend to build Jaguar as a direct replacement for the custom MS controller I have running my daily driver, 1995 Ford E-150 van, 300 CID I-6, E4OD transmission. The engine and transmission are completely stock except for the custom control. MS runs the engine and what I call a “semi automatic” controller of my own design runs the transmission. I have real time displays for all engine and transmission parameters in the form of highly modified Megaview units.

There is nothing wrong with what I have and it works just fine for me, but FreeEMS has sparked my interest and has opened up all kinds of new possibilities. Eventually, after I become comfortable with programming in “C”, I hope to build my own board to combine the features I now have with MS and my daughter board, as well as my transmission controller and additional features that would be impractical to add with what I have now.

In the mean time Jaguar is a starting point. It will require me to make some modifications to adapt it to what I want it to do. Fred’s code will also have to be modified. In the future I hope to be able to do this on my own, but for now I’ll need all the help I can get from the FreeEMS community, and hopefully I can contribute something in return in the way of real world testing.

I’ve been running this combination now for about 9 years and have it dialled in pretty good, but for me at least, the tuning process never really seems to end. I’ve built it so that if either of my controllers has a catastrophic failure, it is very easy for me to switch back to Ford EECIV control. For this reason many of the stock components and wiring have been left original. I’ve retained the stock TFI ignition and “bank fire” injector configuration. Ford fires the Hi Z injectors in banks of three, alternately once each crankshaft revolution. I’ve found by experimentation that I get better results firing alternately on every ignition event.

I use the on board MAP sensor in MS and the stock coolant temperature, manifold air temperature and throttle position sensors. I also use the stock relays for power to the ECU and fuel pump. I have a remote pot that is used for real time ignition or mixture trim and logic state input from a toggle switch to select which function is active. I also have a logic state input from the transmission controller to permit or inhibit manual decel fuel cut. In addition I have a logic state input from a KnockSenseMS unit so I can data log any ignition knock events. I have an input for vehicle speed, 12V square wave, 8000 pulse per mile so my display can calculate current fuel burn in litres per hour and kilometres per litre. I have an open collector output to a pulse recorder which records accumulated injector on time. I can use this information to calculate fuel burn between fill ups. I also have an input for the flyback voltage of the coil through a dropping resistor so I can calculate the actual timing angle so I can compare it to the commanded timing angle.

In a nutshell, the TFI ignition works like this. A hall effect sensor and shutter wheel in the distributor generates a 12-volt square wave signal with the rising edge occurring when any piston is at the default timing angle, usually set at 10 degrees BTDC. This signal is called the Profile Ignition Pickup signal and is sent to the TFI module and the ECU. The strategy is for “next cylinder” firing. When the ECU receives this signal, it calculates the delay time required to fire the coil at the desired timing angle for the next cylinder in rotation. It doesn’t know or care which cylinder it is because the distributor takes care of that duty. Both the TFI module and the ECU use the rising edge of the PIP signal only, the falling edge is ignored. When the correct time to fire the spark occurs, the ECU sends a 12-volt square wave SParkOUTput signal to the TFI module. The module fires the coil on the rising edge of this signal, the falling edge is ignored. The module also takes care of dwell time based on RPM. If the TFI module does not receive a SPOUT signal before the next PIP signal, it fires the coil on the PIP signal. This allows for a “push start” condition and “limp home” mode if there is no SPOUT signal from the ECU.

The coil fly back voltage, through a dropping resistor, generates the Ignition Diagnostic Monitor signal. I’m not sure what Ford uses this signal for, but I use it to calculate the actual timing angle so I can see if it matches the commanded timing angle.

I have a custom built simulator so I can do any testing of hardware, firmware and configuration/tuning software on the bench. The first thing I want to do is to get the board built and modified to accept my inputs and outputs, download whatever firmware may be available and see what I can do with the configuration/tuning software. When I get to the point to where I’m satisfied that everything is working well on the bench I’ll plug it in to the van and do some real world testing.

So, as you can see, I will have to make more than a few modifications to hard ware and firmware. Any suggestions will be greatly appreciated.
Last edited by E4ODnut on Thu Nov 14, 2013 6:28 am, edited 1 time in total.
Robert
1995 Ford E150, 300 CID I6, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1992 Winnebago 33 RQ, Ford 460 CID V8, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1993 Bayliner 3288, Twin Ford 351 CID Windsors, Converted to tuned port programmable EFI, Custom MS1-Extra
User avatar
ToxicGumbo
LQFP144 - On Top Of The Game
Posts: 474
Joined: Tue Jan 03, 2012 8:37 pm
Location: Manhattan, KS. USA

Re: Robert's 0.6-alpha Jaguar Build

Post by ToxicGumbo »

Image

Subscribed.


-Jeff
E4ODnut
QFP80 - Contributor
Posts: 79
Joined: Fri Jun 14, 2013 12:49 am
Location: Gibsons, BC Canada

Re: Robert's 0.6-alpha Jaguar Build

Post by E4ODnut »

For those who may be interested
Attachments
Ignition.doc
The best documentation on TFI that I have been able to find
(276.5 KiB) Downloaded 886 times
Robert
1995 Ford E150, 300 CID I6, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1992 Winnebago 33 RQ, Ford 460 CID V8, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1993 Bayliner 3288, Twin Ford 351 CID Windsors, Converted to tuned port programmable EFI, Custom MS1-Extra
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Robert's 0.6-alpha Jaguar Build

Post by Fred »

Thanks for the write up, Robert! :-)

A few things:

1) I see you edited the big post, but I don't see what you did, in future clearly indicate what changed as reading that entire block again just to find a change is not fun. I locked the post so it can't change anymore for this reason.
2) Thank you for the ignition .doc file. Not only does it cover TFI, it also covers EDIS, and both in good detail. Thanks to your post and this document it's clear how it works, now.
3) You say the board and firmware will need "more than a few" modifications, however I disagree. Your approach is dead wrong. You're trying to swallow a whole roast dinner in a single mouthful ;-) There is, as far as I can tell, exactly ONE modification that is required and was on the TODO list already anyway. The Jaguar shouldn't need any modifications at all to get up and running. What might need a bunch of mods is all of the *optional* features that you currently have. None of which you actually need to get up and running. They're all optional. They can all wait until later. They can all wait until after you're running. Once your running your perspective is likely to change a bit.
4) I'll help you form a short itemised list of things your setup needs, then we can address them point by point.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
E4ODnut
QFP80 - Contributor
Posts: 79
Joined: Fri Jun 14, 2013 12:49 am
Location: Gibsons, BC Canada

Re: Robert's 0.6-alpha Jaguar Build

Post by E4ODnut »

Fred,
You're welcome.

The edit was to correct a typo that I missed proof reading.

TFI works well, but if I had to do it all over again, knowing what I know now, I'd have gone EDIS/waste spark on the boat engines. The only downside with EDIS is that the modules are very expensive and not every scrap yard has them. This situation will only get worse. It would be preferable for the firm ware to handle the duties that the EDIS module does. Come to think of it, if the firmware could handle TFI duties as well it would be preferable because those modules aren't cheap either and it's the same situation at the scrap yard.

I've started to populate the board. This is my first crack at SMD and it is a major challenge for an old dog like me. I just did a few to sort of get acquainted with it. I'm going to move on to the power supply section first and prove that. Then I'd like to do the communications section and prove that if possible. How can this be done?
Robert
1995 Ford E150, 300 CID I6, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1992 Winnebago 33 RQ, Ford 460 CID V8, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1993 Bayliner 3288, Twin Ford 351 CID Windsors, Converted to tuned port programmable EFI, Custom MS1-Extra
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Robert's 0.6-alpha Jaguar Build

Post by Fred »

Robert, most of the ICs are fairly tolerant of heat (but don't abuse them!), however the FT232RL IC is very fragile when it comes to being overheated. This is obvious to the touch because the body of the package gets hot fast from heat contact on the legs, unlike all of the other ICs. So be careful with that one.

In terms of testing comms, you don't need the power supplies for that. The power supplies should NOT be the first thing you assemble. The FIRST thing you assemble should be the all-SMD voltage clamp circuit. You should then use a low value resistor (an old defunct low-z injector?) to pull the 5V rail toward 12V and confirm that it stays put at 5.15. Only AFTER that should you do the power supplies. I was hoping Andy would have done a quick "rough steps" guide by now, but he's busy. This is a key step, though. Or maybe he did and I forgot?

Testing comms as much as you can without a CPU = installing all FT232RL related parts and monitoring the output of the isolator IC with a scope or logic analyser.

The comms subsection is powered by the USB port of the PC and the ground is isolated too.

Once you have the clamp nailed and tested to 5.15V, you can assemble the power supplies and verify the CPU rail is 5V +/- 0.1V. Once you've done that, and you must do it per Andy's photograph of bent legs, then you can install everything else except those things listed in the "options" thread and the ignition outputs (which should be listed in that thread).

If you screwed up the build order and didn't test the clamp first, worst case is the clamp doesn't work, and you burn whichever ICs you've installed while attempting to prove it good. Not a huge deal.

Andy, feel free to correct anything that's wrong.

Re TFI, EDIS, DIS, etc. They're all well and good, and all there for a technical reason, but yes, the firmware can control the engine directly, and in the case of your 6 cylinder, would do a much nicer job of it if you were doing it that way. On the other hand, it's nice to support such antiquated systems simply for the principal of minimal change. IE, one change at a time, in this case, the ECU, nothing else. Once comfortable with the ECU, you can then go ahead and swap hardware/wiring IF you so choose.

Gotta go, back later!

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
E4ODnut
QFP80 - Contributor
Posts: 79
Joined: Fri Jun 14, 2013 12:49 am
Location: Gibsons, BC Canada

Re: Robert's 0.6-alpha Jaguar Build

Post by E4ODnut »

What's a clamp circuit and what components does it consist of?

I see two 5 VDC supplies. 5vdc- analog and 5 dcv-cpu
Robert
1995 Ford E150, 300 CID I6, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1992 Winnebago 33 RQ, Ford 460 CID V8, Custom MS1-Extra, Custom semi-auto E4OD trans controller
1993 Bayliner 3288, Twin Ford 351 CID Windsors, Converted to tuned port programmable EFI, Custom MS1-Extra
User avatar
DeuceEFI
LQFP144 - On Top Of The Game
Posts: 578
Joined: Thu Feb 25, 2010 3:57 am
Location: Gosport, IN USA
Contact:

Re: Robert's 0.6-alpha Jaguar Build

Post by DeuceEFI »

I'll post that info in the morning, I have a list of the component IDs in the order of installation that Fred mentioned along with what the measured values of the voltages for each step so you can verify everything is ok before going to the next set of components.
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Robert's 0.6-alpha Jaguar Build

Post by Fred »

Thanks Andy, I look forward to it.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply