Rex4Life

Now I'm trying to get running with Full MS control--got good news and bad.

Good news, everything seems to be functioning--bad news it's not all working correct.

1) spark is working, engine can run--but the timing marks aren't even visible. I get good rpm on MS and a tach signal in the car. So this means the leading and trailing are firing off. Right now the timing in MS is fixed to -5 deg. I get minor kickback while cranking--most likely an indication of bad timing. If I don't fix the timing to -5, it seems to use the ign map. I must have the idle advance set up wrong.

2) fuel pump comes on--but turns off after running about 3 sec. If I jumper the fuel pump to be on the car will run.

3) BAC seems to be operating correctly.

4) Using stock temp sensors, before I was operating in piggy back mode. The readings appear to be reasonable.

5) The MS responds to hyperterm inputs and reports back OK:

T --> MegasquirtnSpark Extra 026i
S --> MSnS-extra format 026_mod
V --> cdKMOIRRRR

I do not get any MS startup errors in hyperterm.

Seems like I'm really close but these minor things need to be fixed before I can start driving. If I don't get it running by dinner time, I'll have to switch back to the stock ECU till I have more time to work with it.

A few questions:

There is an option for wheel decoder routine, should it be 024s9 or 025?

What is turning off the fuel pump?

What can I do to fix the MS timing--any ideas?

I'm really hoping for this to be a success story--we've certainly had enough bad ones lately.

Scott

custom13B

Scott I had a lot of trouble setting my CAS too. I originally was using the yellow pulley mark. I decided to try the red mark and it worked great.I then set the timing to-10 in mt,you know what I mean. That gave me the best and smoothest run yet..even if it was wrong it worked good for me..so its worth a shot,Shawn

muythaibxr

err, the red mark is 20 deg ATDC, you can't use that to set your timing. If you're using that your timing is retarded about 15 degrees from where it should be.

I have new settings as of today if you're not restabbing the CAS from the stock settings.

I opened up the CAS with the CAS properly set for the stock computer and the engine at the yellow mark... and the teeth are off from where they are when you stab the CAS... Mazda expects the CAS position to move about half a tooth... To get the timing right, I set the triggers/returns to (11,1) and (5,7), and the trigger angle to 65... which gives me a cranking advance of 5 deg btdc... that's a little too much advance, and it caused a few backfires when cranking... You can set the trigger angle to 60 and turn the CAS a bit to get good cranking behavior and such (assuming you're using the new (11,1) and (5,7) settings.

Do you have an S4 TII? How do you have the fuel pump set up? How do you have it wired? S4's normally use the AFM to turn on/off the fuel pump, so I'm just wondering how you have it wired now.

Rex4Life

Right, I don't want to re-stab. I thought the settings in the faq didn't need to change the cas setting.

Basically it's wired like an S4 T2. I connected the pin 37 fuel pump to the ground side of the circuit opening relay. I does run for a few seconds while cranking but then drops out. Funny the pump doesn't start immediately with the ignition switch going on--I thought it was supposed to start for the priming.

Scott

muythaibxr

Hrmm, I'm not sure the way you have it wired will work... the S4 uses the AFM to turn on the fuel pump... I'd suggest taking the wire you have running to the MS, and grounding it to the chassis while cranking, and see if you still have the same problem you did with the MS hooked to that.

It should turn on the pump for 2 seconds when you first turn the ignition key, so my guess is that the MS isn't really controlling your fuel pump right now.

The settings in the FAQ were based on the alignment of the CAS before you stab it using the line on the CAS housing and the dot on the gear. I found over the weekend that when you stab it, the CAS moves over about 1 tooth on the 12 tooth wheel... (maybe half a tooth, or somewhere between) and mazda expects this to occur... So after looking at how an un-tampered-with S4 that runs perfectly on the stock ECU is set up, I came up with the new settings. Basically until yesterday, I'd never actually seen how the CAS is aligned on a completely stock rx7.

You'll want to turn the CAS a bit with a timing light hooked up after you get it running to get good cranking behavior out of this setup though.... you'll be cranking at 5 deg btdc if you don't, and that can cause backfires like I said.... Essentially with a trigger angle of 60, the stock alignment is 5 degrees off from what the MS wants... You could set it to 65 deg trigger angle, and time based cranking with an advance of 0 and hold ignition of 3 or more, but as I've said in other threads, using time-based cranking can be dangerous if your starter or battery are weak. It's much better to simply turn the CAS a little and continue using trigger-return.

EDIT: I updated the FAQ settings for the wheel decoder to take into account the way the stock CAS is aligned once it's stabbed.

Rex4Life

I disconnected the AFM when I had the MS installed. Basically I took the same wire that goes thru the AFM flow switch and hooked it up to the MS pin 37--trust me I think that part is wired ok. The fuel pump ran OK during cranking, it's a few seconds after that it shuts off.

Right now two things are bugging me: 1st the reqfuel not letting me change it, and second, when I did run the engine with the fuel pump jumpered, the timing marks were not even visible. I'm thinking it may be running 180 deg out. Is it possible the timing signals being so far off are causing the pump to drop offline?

Ken--could you double check my drawings in the FAQ and verify they are correct? I'm going to re-check my CAS and ignition wiring just to make sure something isn't backwards. Then I'll double check the CAS install and make sure it's all lined up correctly.

I had to remove the MS yesterday since I need to drive so right now I can't troubleshoot, but I'll start some testing with the stim. Maybe this weekend I can hook it back up. I should have taken a datalog, but forgot--oh well.

Scott

jayroc

I KNEW IT! I had heard that mazda did some kind of stock compensation for CAS installation, but now I know for sure.. Should I try these instead of the 1,3,7,9 settings you recommended earlier, muy?


Glad you got your car going, man. Good luck tuning her up!

muythaibxr

Have you done the wheel decoder changes I told you about yet? Those should get the timing mark back where it should be....

If you're using 12,2 6,8 with a trigger angle of 60 degrees and the CAS stabbed like stock, you're going to end up with timing that's 25-30 degrees retarded... possibly more... which could make it so you can't see the timing marks.

I just checked your S4 diagram against what I'm doing for my plug 'n play unit, and it looks like you do have the G wires backwards....

1N is G+, and 1P is G-

You also have the Ne wires backwards... 1T is Ne+ and 1Q is Ne-

Can you update that graphic so I can fix it in the FAQ? I wonder if this is what has been causing people problems, I hadn't megasquirted an S4 when you made the diagram, and never thought to look at your diagram after I determined the proper wiring, so hopefully this helps people with the problems they have been having.

Rex4Life

That sucks. How bout for an S5, are they correct or do they need to be changed also?

And there is also the coil wiring--does that one check out OK?

Let me know soon--TIA,

Scott

muythaibxr

Actually it's not my car, it's my friend Julian's... It wasn't any trouble at all getting it to start... it started the first time I had the MS hooked up to it (this was my plug 'n play test unit).

My car is still waiting for its working stock harness and new leading plugs (stock ECU flooded the engine really bad because the sensor ground wasn't working, which made the ECU freak an inject a crapload of fuel, destroying my plugs... it won't even act like it's trying to fire now, even with the CAS stabbed perfect and the same settings that started my friend's car on the plug 'n play unit, so the plugs are shot).

you can use whatever wheel decoder settings you want as long as you have the lid off the CAS and make sure that whatever you set for return A is lined up with the VR sensor after you have it stabbed.

muythaibxr

I haven't checked those yet, but the S5 FSM actually tells you which ones are + and -, so I thnk it's more likely than not that you got those right.

I'll check the S5 and the coil wiring later... but I doubt anything else is wrong in the diagrams since the FSMs actually tell you what all the other wires do.

muythaibxr

As far as wire colors go, I have no way of checking right now without going and looking at a car, but according to the FSM for S4 and S5, you got everything else right, so it's just the S4 CAS wiring is backwards.

Rex4Life

Installed the MS ECU today--no more piggy back.

Ken, the new wheel settings worked great. Put the timing marks about 1 cm from where the stock setting was. I adjusted the mark in and checked trailing and it also is running perfect.

For the two problems I had, here are the solutions:

1) Fuel pump turning off after 3 sec: I had the car side wiring wrong.

2) Ignition timing: The CAS S4 drawing has the wires backwards--like Ken noted already. I reversed the ne and g wires and all is good.

Two things left for me to work on:

1) Timing map.

2) A slight stumble coming off the idle advance on to the spark map.

Overall I'm very happy with how it all went. Now to the fun part <Tuning>

Special thanks to Ken for all the hard work he has put in getting the 2nd gen ignition working and the staged injection mods.

Scott

p.s. Gotta change my sig!

muythaibxr

the coming off idle advance part I already told you about in a PM, if it works for you, I'll post the message here. Basically I think your #1 (timing map) is influencing your #2 (stumble off idle advance). Oh, if you haven't read my PM yet, I forgot to mention, be careful when you're upping the advance, do it one degree at a time and note the result. I'm confident that you'll get the results you want by tuning the advance... Tofuball and I have.

Also, you can upgrade to 029d on your firmware if you want, I've tested it and it works great, and has my second parameter code for staging.

Eagle7

The colors listed were inconsistent:
1N is G (green) - not red
1P is L (blue)
1Q is W (white)
1T is R (red) - not green


Muy, are you sure the 1N and 1T are the + signals - which should connect to the active circuitry (not ground)? I've had mine wired the opposite way all this time. Maybe that contributes to my noisy signals.

Thanks,

muythaibxr

yes I'm positive.

Having them backwards will also make it hard to get accurate timing when you stab the CAS, and will make the timing inaccurate depending on RPM.

I have to review a new diagram from rex4life... as soon as I've had a chance, I'll post it in the FAQ... most likely tomorrow.

Rex4Life

Yes, some of the colors for S4 were also wrong. The new version Muy has will correct all the mistakes. Obviously I'm running with the new wiring setup--and no problems thus far.

Sorry for the confusion, but no one had wired up an S4 at that time so we didn't know for sure about the connections. All part of the open source learning curve.

Scott