SMEC to LM adapter discussion.

[Aries_Turbo]

Back on moparchem.com geoff allan attempted to make an adapter to go between the logic module red and blue connectors to a SMEC logic board. he didnt finish it and id like to. I have quite a few SMEC boards but i dont want to rewire my k car and i have a LM in it.

Rob, you can post that moparchem stuff here.

Brian

[boost geek]

That would be cool, I like it in my glovebox.:)

[turbodaytona87]

This will be a cool project, I'll be keeping on eye on this thread to see how things come along.

[bakes]

Someone here has done it already i just cant remember the name.

[black86glhs]

Hey Brian, is this an adapter the goes between the harness and the SMEC itself?

[bakes]

Ah found the link
http://www.turbo-mopar.com/forums/showthread.php?t=41973&highlight=smec+lm

[boost geek]

I'll just grab a harness out of a turbo mini the next time I find one. Easier and the pm stays by the battery anyways. I was hoping to move it to make room for a cold air intake.

[black86glhs]

Ah found the link
http://www.turbo-mopar.com/forums/showthread.php?t=41973&highlight=smec+lm

yep, thats it.

[ShelGame]

Here's what I managed to save from CHeM. It's not much, unfortunately. As I recall, Geoff never got it to work right.

[risen]

I was going to work on one of these, and I have a way of doing it without hacking up a a LM to pull the plugs off. I have a test LM-> megasquirt plug-n-play adapter at home, but haven't had a chance to test it out.

I have the LM male plug side figured out, I can post a pic of it later. It uses all new parts and plugs right into the LM connectors. I also believe I have a source for the new smec 60 way connector, but I haven't picked one up yet.

However there's at least 2 items that I discovered during the making of the smecstim:
It appears that the injector signals from the LM are ground switched while they appear to be power switched from the smec. I'm not sure if I'm just not doing something right but this is how it appears at the moment.

It seems that the CEL is also reversed, the smec supplies power while it seems that the LM switches it to ground.

Also, the coil signal from the LM is much weaker than the SMEC, but this may not matter if still using the LM power module.

I haven't looked at the voltage regulation signal, but I doubt that's changed.

I was going to put some transistors on my converter to convert these signals so all you had to do would be plug the 2 LM connectors into the board and have the Smec plug into the other end. Is there a lot of demand for this sort of thing? I can put the coil on plug setup on the back-burner if there's a lot of interest.

[1FastCSX289]

Excuse my ignorance......but whats the point? What is the advantage of running an SMEC over a LM?

[risen]

Excuse my ignorance......but whats the point? What is the advantage of running an SMEC over a LM?

The 89 T1 code (used in Turbonator SMEC) is much more advanced than the LM code. I believe the MCU is also a generation newer and faster in the SMEC.

Then there's always the 'just cause I/we can' argument... :)

[ShelGame]

Division! That's the big reason. The 6803 used in the LM does not have any built-in divide instructions. That makes it difficult to do a lot of calculations. It's also the main reason our tables are stored in the point-slope format. If they had a 6811 available in '84, they probably would have used a more traditional interpolation routine. But, without an efficient means to calculate the divisor, interpolation becomes cumbersome.

I've looked at finding or developing a drop in processor replacement for the LM that would use a 6811-based core. At least then, some of the code could be 100% compatible.

[1FastCSX289]

Division! That's the big reason. The 6803 used in the LM does not have any built-in divide instructions. That makes it difficult to do a lot of calculations. It's also the main reason our tables are stored in the point-slope format. If they had a 6811 available in '84, they probably would have used a more traditional interpolation routine. But, without an efficient means to calculate the divisor, interpolation becomes cumbersome.

I've looked at finding or developing a drop in processor replacement for the LM that would use a 6811-based core. At least then, some of the code could be 100% compatible.

Are you speaking Greek? :confused: :lol:


The 89 T1 code (used in Turbonator SMEC) is much more advanced than the LM code. I believe the MCU is also a generation newer and faster in the SMEC.

Then there's always the 'just cause I/we can' argument... :)

I see now. :lol: Gotcha. ;)

[Aries_Turbo]

Is there a lot of demand for this sort of thing? I can put the coil on plug setup on the back-burner if there's a lot of interest.

im interested. :)

AJ's adaptation looks a little easier though. i may do that instead. lemme get more info from him to see which is the way to go.

Brian

[risen]

im interested. :)

AJ's adaptation looks a little easier though. i may do that instead. lemme get more info from him to see which is the way to go.

Brian

Ok. Well, I'm still going to get to it eventually, but if you want to see it sooner rather than later LMK.

[zin]

Just to clarify, this mod would put a SMEC's logic board inside the car? or just the "logic" board part? With the point being to improve the electronics... If I'm clear on this part, having never dissected a SMEC, would the logic board fit into the LM case?

Mike

[risen]

Just to clarify, this mod would put a SMEC's logic board inside the car? or just the "logic" board part? With the point being to improve the electronics... If I'm clear on this part, having never dissected a SMEC, would the logic board fit into the LM case?

Mike
Yes, it would allow you to run the logic part of the SMEC instead of a LM. It won't directly fit in the LM case. I can post a picture later if you want of the 2 side-by-side, I have a few laying on my desk at home. If you wanted to hide it behind a kick panel, you might be able to do that, but the connector + the SMEC logic board is probably 3-4 inches high.

[1984rampage]

Ok. Well, I'm still going to get to it eventually, but if you want to see it sooner rather than later LMK.

Id be very interested

[risen]

Id be very interested

Ok, well what I'll say is this: I think I've got the final redesign of the COP setup (which will now handle t3 style coilpacks too :) ) done. I'm going to finish the schematic and setup a testbed tonight. I'll put it to the side after that, while I wait for test PCBs to come in.

Later in the week, I'm going to start with the LM -> megasquirt adapter I have in-hand. Once I get that working (give me 2 months, min) the smec adapter should be easy to knock out. I'll have already figured out all the signaling needed for the LM's PM and it should be easy to translate the SMEC version.

So, reasonably, I think I'll have *something* by mid-late June.

[1984rampage]

Sounds pretty sweet! Whats the LM -> megasquirt adapter your talking about?

[risen]

Sounds pretty sweet! Whats the LM -> megasquirt adapter your talking about?

Since megasquirt natively supports our hep, I figured it would be easy to make a plug n play adapter for it. The adapter takes the lm plugs on one side and has a db 37 on the other. So you pull the lm, put the adapter in it's place with a megasquirt. This makes going to megasquirt a plug and play affair. I have to handle the voltage regulation and do something with the knock signal, but it appears pretty straightforward otherwise.

[zin]

I can post a picture later if you want of the 2 side-by-side, I have a few laying on my desk at home.

I'd like to see them!

Thanks!

Mike

[Aries_Turbo]

nice. lemme know if you need any of the factory LM knock interface circuitry stuff. I ID'ed all the components and recreated it on a breadboard hooked to an arduino. i just havent had the time to do the testing. im going to eventually finish my knock interface box (perfect go-between for your MS to LM adapter to retain precise knock indication/retard with a MS) but i can share schematics and stuff. especially if it get a DIY kit out there in use faster.

Brian

[risen]

nice. lemme know if you need any of the factory LM knock interface circuitry stuff. I ID'ed all the components and recreated it on a breadboard hooked to an arduino. i just havent had the time to do the testing. im going to eventually finish my knock interface box (perfect go-between for your MS to LM adapter to retain precise knock indication/retard with a MS) but i can share schematics and stuff. especially if it get a DIY kit out there in use faster.

Brian

I did read your page on Frank's site about the knock stuff. I have some TI DSP chips that I was actually planning on using in the future with the MS board as the knock circuit is on pin headers for easy expandability. I was actually going to ask where you got the information on the bore size determining frequency of the knock (which I saw in another post), because I'd like to read a little further if you have a good source of knock info :). The TI chip requires that you select a resonant frequency in order for it to function and that's part of what's holding me back from using it right now.

The current board has a attiny24 for the voltage regulation, it's only used to do PID and switch the control circuit to gnd at the moment. So the code for the TI chip could be used with a arduino afterwards with minimal effort, if I ever get to it. As you can see, my project list is pretty long :)

[risen]

I'd like to see them!

Thanks!

Mike
Ok here they are. The 1st pic has the smec (brownish color), the LM (with red and blue connectors) and the lm2squirt board (as I'm calling it). The 2nd pic has a LM connector (body module actually) against the pin header with the same pin offset. 3rd pic is just LM and SMEC logic board.

[Aries_Turbo]

I was actually going to ask where you got the information on the bore size determining frequency of the knock (which I saw in another post), because I'd like to read a little further if you have a good source of knock info :). The TI chip requires that you select a resonant frequency in order for it to function and that's part of what's holding me back from using it right now.

Detonation Frequency

The reverberation resonance generated by detonation generally lies in the range between 2 and 12KHz. The following equation can be used to estimate the detonation resonant frequency for a specific engine.
Detonation resonant frequency = 900/(3.14 x cylinder radius)

Where the resonant frequency is measured in Hertz and the cylinder radius in metres.

[risen]

Detonation Frequency

The reverberation resonance generated by detonation generally lies in the range between 2 and 12KHz. The following equation can be used to estimate the detonation resonant frequency for a specific engine.
Detonation resonant frequency = 900/(3.14 x cylinder radius)

Where the resonant frequency is measured in Hertz and the cylinder radius in metres.

I got that much from the other thread (thanks for posting it again, too), but I'm more curious about the original source of that info. I understand your knowledge is encyclopedic, but you don't normally post like an encyclopedia :).

[Aries_Turbo]

i saved it in a txt file on my work laptop an just pasted it up. as far as the source... i dunno but ive seen that formula all over the place.

i wonder if its discussed in john b heywoods book...

Brian