Marc D

iTrader: (7)

Case Ideas:

Ok, so I've built the board, and I've taken account on the measurements. There are several ideas that I've come up with as to mounting the board.

From what it seems, it's not possible to use the stock MS case to fit both the board and the adapter board; it's simply not tall enough. For those who wish to mount the board and the MS together, there is a "taller" version of the case, available at fry's (available in the same brushed aluminum as the stock MS case, or anodized black). I was thinking about doing it this way, but I was toying with the idea that I may want to upgrade to MSIII in the near future (when it is released) so I want to leave room for the large MSIII adapter board and other attachments (i.e. mods for sequential or whatever). If you've been following with the development of MSIII, you'll find that there will be no room so spare with the new case that MS3x will be using.

MegaSquirt-3 Case Preview




I decided to mount the board externally in its own separate case, and thanks to abe's perfect sizing of the board (lucky ), the board will fit in a 5.00"x3.12"x1.85" case made by context engineering (which is available at frys for about 13 bucks, cheaper than buying from their home website at 15). Getting it to fit perfectly will require some work, but it fits right for the job. The adapter board on the harness plug side where the Abe's board label is located will need to be shaved a bit for my mounting idea (sorry Abe).

I will be working on this over the next few days and I'll let you guys know the progress. I will update this post to keep clutter to a minimum.


Current parts purchased:
5.00"x3.12"x1.85" Aluminum Case
2-56 x 3/4" Screws with matching nuts for securing the board
Size #2 washers for adjusting the board height.

These can be purchased all at fry's.

Tools required:
Screwdriver
Mini adjustable wrench
Cutting Tool (dremel)
Patience.

EDIT: Here is my post updated with the additional information on how to make the case:

Pictures!

I shaved the top of the board where the harness plug sits. Its shown here. Don't mind the extra wire there, I felt like i accidentally severed another trace while i was killing the alternator one, so I just ran another wire for insurance.


There is glue here covering the DB37. Just used hot glue. I used it for insurance that the box side cover doesnt touch and short it.


Here is another shot on where I shaved the board.


Another Shot where it was shaved


This is the bottom of the board. THere are just two screws there at the bottom, and with a bunch of washers to level the board. The board sits within the side rails, so basically, the screw posts just hold the board in place from moving. On the back side, towards the top right side, there is a hole for the terminal posts where the wires can be fed.


Another shot.


Take note here. I had to shave the "rail" area beacuse the black mount of the DB37 wouldnt fit in there of course. On top of that, I shaved down the side of the DB37 too. You can see it in the pics above.


This is the "top cover" of the box. it just has a cut hole for the harness plug.


Here is the board in place.


A shot on how the board is mounted.


The other side of the board. Yes, there will be space here. The board doesnt fit THAT perfect


Here is the other side of the box. This is the hole where the terminal posts are.


Heres a shot of the terminal post area with the cover top.


Other side, harness area with the cover.


I could have done this more prettier, but I was being lazy and i Just wanted it to be done. You can just measure this and cut it to your needs. This is the DB37 connector cover. The underside of the board where the pins are sit a little close to the cover, so i used glue to seal it for insurance, which you saw in earlier pics.


Another shot.


And there you have it. It only cost me about 17 bucks in supplies, and I have a dremel handy. I wish i had something that could cut it with better precision, but I just dont have that kind of tools or access to any tools of the like. Im sure you guys can make it look a lot better with the proper tools, but this will work for me.

AbeFM

iTrader: (3)

Nice! I'd love to see pics.

Hmm, a taller case, you could just mount vertical DB-37's on there, I think the footprint would be the same. I'd have to think about the directions of everything, it's mirror imaging that gets you most of the time in this - otherwise, hunker down and solder 37 wires board to board, not too bad either.

If you're looking for just a bit more strenth, if you can find screws which will bite right on the tip, there's ~4 more holed you can pick up - those in the OEM connector. All my screws were too long, but if you were bolting it TO your case, you might be able to use those.

Glad to hear the board went together ok.

lsc224

iTrader: (13)

I worked on my plastic ABS case and it was a tad too small for the board, so I ordered an aluminum case from Fry's. I hope I ordered the right one, there was only one for $14.99, per MarcD's notes, that's the price on their website. Part no. is 2090782. I am also going to do a relay wire like MarcD's. Thank you MarcD for your tips.

Marc D

iTrader: (7)

Too bad. The one on context engineering's website is 12.99 at fry's though. 2 bucks cheaper... unless you wanted black anodized. THe box fits just right, but its going to require sanding down the board a bit underneath the ECU harness plug. This is required so you can plug in the harness and its easy to pull out, instead of having the plug recessed within the box itself.

I only used the two center mounting points to secure the board in the case, but the board underneath the harness plug sits in the slots, so its very well stuck in there and has no chance of moving.

Good luck, and when you do sand down the board, becareful not to damage the traces.

AbeFM

iTrader: (3)

FYI everyone:
I was originally planning on being out of town this week, but I won't be - so hopefully I can help with any questions!

AbeFM

iTrader: (3)

Looks like the baseline MSQ is going to be the next tall pole.

I'll start another thread, not because we couldn't post one here, but as MS-II firmware continues to evolve, just posting a file won't do it. I should go through and list all the settings to set.

Baseline MSQ for '99-00, including Abe's Board - Miata Turbo Forum - Home of the turbo Mazda Miata.

I'll take notes here, and try to keep this at least mostly hardware.

AbeFM

iTrader: (3)

I'm sure I've said this before, BUT...

PLEASE remember that because it's "PnP" it's not "plug and race around like an *******".
1) This is pretty experimental
2) It's easy to blow up your motor even with something that is perfect.

So go slowly!

The real advice is this, it's how you start:

Load up megatune. Make sure the MS and MT can talk. That means on the stim you can see what's going on. Turn the *****, make sure it all works.

When you plug it in, be aware of burning smells. Put your hang on your coil pack within the first minute and again a minute later to make sure they aren't cooking. Ideally have someone check them 30 seconds after powering everything. After you're sure nothing is burning, turn it on, do it again.

Do this with the vents/fans off, the stereo off. Listen for funny clicking, buzzing, cooling fans coming on. All this is settings you'll have to play with

You may choose to start with the coils (IGN A and IGN B) unplugged, test the rest, then plug them in and check for heat.

After that, see if all the sensors work. Open up "real time display" on the MT. Make sure the TPS, baro, MAP, CLT, MAT (AIT), are all sensible (does the MAT and CLT roughly agree? If one says 105* and the other 57 or -108,835 there is something wrong)... When you move the gas, does TPS go smoothly from 0% to 100%? If not, recal it.

You can mess with output test modes to make sure the fuel injectors and ignition and idle all work, but they are probably fine so don't bother. Same with the fuel pump.

Odds are the idle valve, cranking and warmup settings are a bit funny, so it might not start easy. With any luck, it will. MS-II takes two full engine revolutions (most of the time) to even notice the motor is there. So, it will take 1-2 solid seconds of cranking even on a good day (which, once things are tuned, it normally is). Don't run out and get the starter fluid right off the bat.

Pulling the jumper on the board for the fuel pump should let you manually verify the timing. You shouldn't have to, but it's a VERY good idea. Before it's running I like to do this. Yes this means you'll need a timing light. :-)

Remember, after you're running, see if the fans come on when it gets hot. You can experiment by setting them very low initially to make sure they work (perhaps 100* instead of 190*). VICS you can set to switch at 1500 or 2000 RPMS, blip the throttle and see if it moves.

Make sure the battery voltage is sensible (~12V without running, about 14.3-15.0 running)

The basic idea is to quickly test each thing, so you don't get 15 miles from home, overheat and warp your head, etc. I *know* it'll add a frustrating hour or two to driving away, but it'll save you an hour waiting for a tow truck and three weeks waiting for parts.

I know you are all smart guys, but I just wanted to re-emphasize this. :-)

AbeFM

iTrader: (3)

In case someone is watching this thread, it turns out there is an issue on the alternator, please make this change to your board. See post 4, end of that entry.

AbeFM

iTrader: (3)

Another message for thread watchers:

Turns out I poorly speced a part I bought, the transistors for idle and "spare" aren't *really* up to the job. I'll be ordering some new ones. So the "big" transistors are going to be replaced. Sit tight, I'll mail them out. The ones you have may work (I had one running on my car just fine) but they are borderline. If you've yet to populate your board, don't do those yet.

Marc D

iTrader: (7)

I updated my post about how to make the case for this board. Its on the top of this page.

AbeFM

iTrader: (3)

Captain Douchebag, reporting for duty. Sorry everyone, I have the transistors and utterly forgot to send them out. Raj, I think I owe you something as well (was it jumper blocks?)

Let me know where you are all at, I'll get them in the mail today or tomorrow. I delivered some locally, and that leaves:

Leo C.
Marc
Raj
Todd
Paul

which I need to mail still. I'll dig up the addresses and try to get them out today, if there's anything else I owe anyone, try to let me know now (Raj, your's will have to wait one more day so I can get the blocks)
-Abe.

redrider706

iTrader: (4)

Yep, I was missing the jumper block.

I am in no hurry, take your time.
-Raj

Marc D

iTrader: (7)

All good abe, just send it out when you can. When I get those transistors, I'll go ahead and solder them along as well as modify my MS board for bluetooth.

Did you see my post on it? Its al updated now with full instructions, but sadly no pictures yet. When I do get pictures after the install, ill post it up.

AbeFM

iTrader: (3)

I was aware of it, but haven't checked it out in detail yet. I DO like the idea, although, MS's serial is bad enough as is... The last thing it needs is MORE errors. :-) I wouldn't flash code or set tables with it, but I imagine for logging it would be awesome. Who knows, maybe it'll rock.

Found little protective shields for the transistors, so out they go tomorrow.

AbeFM

iTrader: (3)

Some notes I worked up for another user. Probably as useful here as in the baseline thread.


From the Install Notes thread, there's some info:
D14 - IGN with pullup (You should have done this already)
D16 - IAC2B with pullup (You should have done this already)
You'll want to do up the middle LED the same way as you would the normal "spark mod", with a pull up, and run that wire to IAC1B
D15 - IAC1B with pullup
JS11 - SPR1
JS6 - SPR2
JS4 - SPR3
JS0 - SPR4
JS10 - IAC1A
JS2 - IAC2A

and from the Baseline MSQ thread:
Output Port Settings:
PM5(warmup):
coolant > 190 5.0
no second condition
0
0

PT7 IAC2 (VICS, I believe)
rpm > 5250 500
no second
0
1

Knock Sensor:
Safe Mode
MS2 AD7/JS4

Tacho:
on
JS11

So:
IAC1B is running the fans
IAC2 is running VICS

Spare1 is TACH OUT
Spare2 is the SPARE output. Interesting, I didn't remember hooking it up. :-P
Spare3 looks like KnockIn
Spare4 is EBC out - That one I guess you just have to know. :-P



edit: William - It looks like to run VVT in the most simply way, you've got two options. First, recycle the VICS transistor. I wouldn't recommend this, however, I used a smaller transistor for it. So, for you, that's the spare, and you'll deactivate that output in output settings (PT7).
But you'll want to add in the SPARE/VVT transistor. That's JS6. Let me check the MSV3 schematic... Hmmm, not SURE if the software will let you control that how I'd like... If not, you can swap two of the jumpers I said to add in the MS case, that from SPR2 and IAC2, then set PT7 to run VVT how you'd like (i.e. "on" when 3000<RPM<6000 or whatever your tune desires).

Yeah. It looks like that's going to be your best bet, the software doesn't support running it the other way. So, in your case, run the wire from the CPU at pin JS6 on the CPU to the IAC2A, set it as above, and you should be running VICS in a digital fashion.

If you want to get fancy, later, I've been kicking around an idea with Joe P. to use the EBC output to PWM the VVT in a variable fashion. It'd take more work to develop, and be sensitive to oil pressure (and hence oil temperature) but... In theory, it'll work in an open loop mode, even now, based on a table of RPM and TPS. If you want to mess with software, you could likely make it RPM and oil pressure, which would be awesome but invovled.

Hope this helps everyone!

redrider706

iTrader: (4)

Abe, why are you connecting the ebc out to SPR4 on the DB37?

Isn't SPR4 connected to the base of the power transistor?

-Raj

AbeFM

iTrader: (3)

Ah, yes, I was using the idle function of the MS-II... It has a chip on it for driving stepper motor coils... I was hoping to use the output of the MS-II to drive a bigger chip for more maring in driving EBC, but the truth is, the chip on the MS-II can drive coils. So, I'm letting it - remember, you're not supposed to populate that transistor, nor the one for VICS. The reason is that the chip puts out 12V, and there's no easy way around it.

redrider706

iTrader: (4)

Oh, OK. I put the transistor in for the EBC. I guess I will have to remove it. Will try to de-solder it. If I can't I will cut the transistor out.

AbeFM

iTrader: (3)

Ok, guys, Jason was kind enough to make a suggestion on how to fix the dim low battery light when the light should be off. I don't have a way to check it, but I'll post it here and you guys can try it and see. I'm not 100% sure I'm interpreting him right, but he says:
Ok. That's here:


...between pins R and C on U2. That's the transistor looking thing which is installed closest to the proto-area and the jumper block, and it goes between pins 1 and 2 I believe.

Give it a shot, see what happens.