Gauging interest for PnP Megasquirt solution, MS anyone? Options

[JamesM]

For my most recent Megasquirt install i went a different route and had some mini-MS boards produced. It did not make a lot of sense to just have one PCB printed so needless to say i have more then a few empty boards laying around. Now i am trying to decide if it is worth my time and money to build these boards out and sell them as a plug n play solution for d-jet cars.

For those who don't know, mini-ms is electronically compatible with megasquirt 2.2 however the components are arranged tighter on the board and it integrates the relay board all on a form factor smaller then the original MS. My first couple of installs i used a standard megasquirt 2.2 with relay board and DB-37 cable but the mini-ms seemed like the better way to go so that is what i am playing with now.

The setup i a running right now looks almost stock, the mini-ms is enclosed completely in a gutted d-jet brain and plugs directly into the d-jet harness. the only difference from stock is a MAP line running to the brain and the 2 injector ground wires being run back as well. You have to look close to tell its not stock, however i have a feeling i could make it even more stealth by hiding the MAP sensor in an old MPS in order to get rid of the vac line to the ECU. I just have not cared enough to cut up an MPS yet.

The great part about this system is it makes troubleshooting way easier and eliminates the need for stock trigger points, MPS. ECU, decell valve, and cold start injector, though they can all be left in place if desired in order to appear stock. It also allows for cool things if you desire like closed loop operation with a wideband O2 sensor, spark control, rev limiters, traction control, and a bunch of other stuff depending on how far you want to take it. It can be tuned around just about anything, so its great if you want to change cams or go bigger on your motor and still look stock.

So is there any interest out there for this sort of thing or should i just dump the empty boards on ebay or something?

[JamesM]

Thought I would through in a quick update on the MS/d-jet project. Due to the fact that I work 70 hour weeks and the added issue of all the random snow/sleet/hail/rain storms we have had here lately, I have not been able to test with the car as much as I would like. The latest board setup I made is working though, the install was pretty simple (MUCH cleaner then my install using the standard MS/Relay) and it fired right up with no fuss. The weather seems to be clearing up now so hopefully I will get some time with it out on the road. I have an autox in a couple weeks so that should be the first real world test of the new board. I have a temp probe setup to monitor the environment inside the ECU casing so I can get some idea where we are at temperature wise with the ECU mounted in the engine compartment.

With the idea of producing these things for other people now in my head I have been focusing my thoughts a lot more on simplifying the build process for both myself and anyone who might be installing it. One of the issues I was looking at is the time and effort required for me to build one of these. While I do prefer the install using the minims board rather then the standard megasquirt board, there are still some tedious assembly steps that I would like to simplify, the biggest one being soldering all the jumper wires directly to the ECU connector. This is an annoying task to say the least, and while I have been happy with the result of my work thus far, I would be a lot happier knowing the connector was soldered directly to a PCB rather then jumper wires. This line of thought has led me down the path of designing a simple PCB to mount the connector to and then jumpering from one board to the other in order to facilitate a cleaner/stronger install. For the moment this is going to increase my time and money involved in this project however in a production environment I think something along these lines is the way to go. My other thought along these lines involve integrating the d-jet connector into the actual PCB for the ECU and enlarging the PCB at the same time in order to mount the same way as the d-jet PCB did. This would greatly increase the speed of the build as I would basically be just swapping one PBC for another without a bunch of custom wiring. This however takes me out of the realm of just getting rid of my spare boards and more into a full production situation, so even though I would like to do it I am pushing it to the back burner for now. Plus I have not found any PCB design software that I like enough to take the time to do this yet.

Another exciting (at least I think it is) development. I have the sort of brain that does not like to shut down at night, so 3am comes around, lying in bed, and all of a sudden it hits me to see if I can modify a megasquirt board to send switched power to the injectors similar to the way that d-jet does rather then switched ground which is how MS normally works, thereby eliminating the need to add the return wires to the ECU with the wiring harness. The hardest part of the install right now from an end user point of view is determining which 2 of the 3 ground connections are the ones you need to use, this should eliminate that, as well as the extra wires all together. I may have just been delirious from it being 3am but I went and looked at one of the boards and I think it will be a pretty easy mod. I am not sure why I, or anyone else, never thought to do this before.

I am going to build another board tomorrow with that and a couple other mods I have thought up and see if it works out.



As much fun as this is I wish my brain would shut off once in a while; maybe I could get some sleep at night. (IMG:style_emoticons/default/yellowsleep[1].gif)

[JamesM]

A couple other smaller issues i need to investigate , that maybe someone has some input on.

1. With the new board setup i am pulling all the power for the megasquirt through the stock harness, while this does make everything look nice an clean it has had the strange side effect of keeping power to the MS system for about 5 seconds after the key is shut off. not a deal breaker at this point but it is high on my priority list to sort out as it is annoying. I am not sure if this is due to a sticky main relay on the stock relay board, or perhaps something else. I have not had time to investigate this yet but maybe someone out there has some ideas?

2. The throttle position sensor setup. I have been thinking a lot about this as it is probably the biggest area I would like to find a development of some sort. Right now the options are
A. Eliminate it all together, which would allow a stock look but i do not like mainly due to losing the flood clear feature but also due to losing throttle position in datalogs.
B. Chop up a d-jet TPS to make an adaptor. While this enables everything to work the way it is supposed to i am not to happy with it because it involves sourcing a d-jet TPS, destroying a d-jet TPS, and on 2L motors does not look stock due to the fact that i have not found a decent TPS that fits completely inside the old one. Not really an issue on 1.7s as it is hidden under the throttle body. Also it is another PITA tedious thing to construct. It works fine, but i have a feeling a better soultion is out there. Looking for input here as well.

[charliew]

A couple other smaller issues i need to investigate , that maybe someone has some input on.

1. With the new board setup i am pulling all the power for the megasquirt through the stock harness, while this does make everything look nice an clean it has had the strange side effect of keeping power to the MS system for about 5 seconds after the key is shut off. not a deal breaker at this point but it is high on my priority list to sort out as it is annoying. I am not sure if this is due to a sticky main relay on the stock relay board, or perhaps something else. I have not had time to investigate this yet but maybe someone out there has some ideas?

2. The throttle position sensor setup. I have been thinking a lot about this as it is probably the biggest area I would like to find a development of some sort. Right now the options are
A. Eliminate it all together, which would allow a stock look but i do not like mainly due to losing the flood clear feature but also due to losing throttle position in datalogs.
B. Chop up a d-jet TPS to make an adaptor. While this enables everything to work the way it is supposed to i am not to happy with it because it involves sourcing a d-jet TPS, destroying a d-jet TPS, and on 2L motors does not look stock due to the fact that i have not found a decent TPS that fits completely inside the old one. Not really an issue on 1.7s as it is hidden under the throttle body. Also it is another PITA tedious thing to construct. It works fine, but i have a feeling a better soultion is out there. Looking for input here as well.

I am not up to speed on ms yet but I was reading the ms info on the version 3 board and it mentioned a run on condition in the fuel pump flyback clamp circuit and changing it to a zener diode to correct it if I remember correctly. This may or may not help.

Also I found this tps on ebay and it looks pretty adaptable. http://cgi.ebay.com/ebaymotors/ws/eBayISAP...e=STRK:MEWAX:IT
Charlie