I use a CAD program that I downloaded from Sunstone Circuits. It works good and it's free. Now I just have to get all the tools loaded onto the new machine with the Core i7 3.4 Ghz processor. I can get boards made by Sunstone.

 

That's great. Ill have to check that program out.

If you need anyone to test I would be interested.

Also how does one tune the controller?

 

There is a USB Interface. I currently use it for programming the controller. I will need to create a Windows program for changing the shift tables. At this time I modify the code and download a new program to change the tables.

I don't need anyone to test it now, but I may take orders for Beta Versions when I am ready to order PC Boards.

 

Are you going to have a Linux program? I may be able to help you with that, as I have quite a bit of experience with Linux. An increasing number of people use Linux for custom tuning stuff and that'd be a great feature to have!

 

I plan to convert my old computer to Unbuntu Linux. That program would be a good way for me to get familiar with Linux.

 

I'll just warn you that Ubuntu is a bit buggy, always has been, but for a newbie it would be the best one to use.

I helped develop their new interface actually...

But back on topic.

Are you making the tuner a GUI application? If so you might want to do it in Python or Java. Both of those are cross-platform languages and it'd save you a bunch of time getting a separate program working on Windows and Linux. Or if you made the Windows program right it could run under WINE in Linux, eliminating the need for a separate Linux version altogether.

 

It will be a GUI application. I did a similar one in Windows for another box, but it's premature to get into that now. I need to solve my problem first.

 

Fairly new-bee to this forum.. I REALLY hope this thread isn't dead!! Really good tech and dead on applicable to my project.

I have a 1980 F150 with a 300 I6 that I'd really like to add a E4OD to. I currently have a Megasquirt III and a II that I've been considering combining for engine control and an trans controller but I bought a AOD trans that I planned on using instead. However I'd much rather have a E4OD and a DIY controller like this. It just leaves a bad taste in my mouth to have to buy a after market black box and I have no desire whatsoever to adapt a factory ECU/PCM to do it.

Danlee, I've been programming and interfacing micro-controllers to all kinds of devices for years and I would love to lend a hand getting your controller done if there's anything you would like help with. It's obvious that you are extremely competent and really don't NEED help but if you would like help, I would be thrilled.

Keep up the good work and please keep going!!

Lonnie

 

I set it aside, because I had other more pressing issues. I built another box, however. It has a 4 position rotary switch, and one SPST toggle switch. The 4 position switch selects the gear by grounding the proper solenoids. The toggle switch locks the torque converter.
This does box not control the fluid pressure. It works like a standard trans, you have to shift into each gear. You also need to select 1st gear when you come to a stop.

I do have to get back to working on the controller. Maybe you have given me some reason to get back to it. All I need to do is to count the Interrupts from the Speed Sensor & convert that to RPM. The problem that I have been having is not being able to get a stable output from that algorithm.

The real difficulty is that my development system is on my desktop computer, and every time that I need to make a change to the program, I have to run upstairs to my computer, recompile & burn. Then I go back to the truck Reconnect & test. I have to put in a printf statement so a message or parameter will display on my terminal(Laptop). It was cold last winter and I got tired of running in and out of the cold. I could set up the laptop as the development system, but that is a job all by itself.

I'll give it another try this week.

 

Did I understand from your previous posts that you're using a ARM processor? I've done a little bit of work on those but lately I've mostly been using AVRs.. I love those things.. What kind of development tools are you using and are they windows based? One thing you might want to consider is (assuming a Windows development environment ) is to install VNC on the two machines. You could then run a client on the laptop to your development environment (assuming you have wireless) just transfer the compiled binaries.

I'm really impressed as to how far you've come if you don't have any more technical information on the E4OD than I've been able to find. The totally manual controller might actually be a workable mode. How do you set it up to get the truck started?

Like I said, I'm sure you don't NEED any help but if you would like some, I would be thrilled to help. I'm a EE with about 30 years of doing hardware and software development for all kinds of servo control, camera systems and motion control systems and I LOVE this kind of stuff.

Keep up the good work..

Lonnie

 

I am using Notepad++ as my editor on my Windows 7 machine. I compile using a program called MakeitC, supplied by Coridium, the processer maker. The processer is a Coridium SuperPro, and I write the code in 'C'. The program is simple, it is only a switch/case that uses the information from the trans Range Sensor and sets the solenoids for the proper gear based on the tailshaft RPM, the throttle position, and the current gear. All the gears are mechanically selected in all positions except Drive. This is the normal operation of the E4OD. The program runs all of the time, but is only active in Drive. I use look up tables to solve the TPS/RPM. I also output 6 I/O lines for LEDs (PRND21), and another for OD. I don't have an OD cancel input. RPM is calculated by measuring the time between Interrupts, and calculating the rate.

The manual controller is only effective in Drive. The trans lever selects all of the other gears. That is a normal function of the E4OD.

 

I spent some time today debugging the controller. It's getting spurious interrupts from the differential receiver. On the desk, if I short the inputs to the receiver, I get no interrupts. In the truck I get no interrupts until I turn on the ignition key, then I get continuous interrupts. The truck is not started, just ON. I have some filters in the program, and they filter most of the garbage out, but it only takes one spurious value to get through the filter to mess up the RPM calculation.

It appears to be a hardware problem. I'll have to check and isolate the wiring for the VSS. Maybe some better shielding will help.

 

I looked at the truck wiring today. The transmission cables run all over the place. I have to shorten the VSS cable. I'm going to order some 16 AWG Twisted Pair Shielded cable from Ebay. I'll run direct wires from the Vehicle Speed Sensor to the Controller. That should help quiet the noise.

 

One thing of note about these models is that the VSS picks up metal shavings and is prone to sending dirty signals to the ECU. Might want to pull it & clean. Also if you don't want to invest in a huge roll of wire, I might have enough if you give me a needed length. I'll eat the postage just to contribute to advancing the "cause".

 

This is the type VSS sensor that I am using.

I don't think that sensor will attract metal shavings. I used Bid Nip to bid on a 9' section of 16 AWG Shielded Twisted Pair. I didn't have the right password in my BidNip account, so I didn't win that cable. I'm trying again with a 10' section of 14 AWG Shielded Twisted Pair. I hope to get it for $11.00.

What kind of cable do you have? I'll send a PayPal payment for the shipping, if it's suitable.