Im having a strange problem with my V3 based megasquirt. I just firuged out spark issues with muys help. I believe the wires from the CAS 2 pin wheel briefly touched together and may have fried something. It was working and the car alomst running then stopped. The megatune screen now shows 2 RPM even when cranking and the TPS guage reads 87% all the time and even the MAP sensor stops giving readings when cranking. According to megatune to ECU is talking to the laptop. All my wiring is fine I double checked it or should I say checked it for the 20th time. A piece of electrical tape came off my trailing VR sensor wires and I believe they touched togethor. Should I **** can this megasquirt and just put the factory ecu back in? At least it would run again. Every time I seem to get close I have a set back,every wiring joint is soldered and heat shrinked nothing is done half fast. Thanks in advance but I doubt anyone can actually help.

 

Did you try connecting up to the stim tester?

I wouldn't think shorting the CAS wires would fry the whole thing. Maybe the fuse blew, can you check that.

Sounds like something else is wrong.

 

touching the VR sensor wires together shouldn't cause a problem after you pull them back apart unless somehow you had other current going through one of those wires and you fried the lm1815 chip. That'd be pretty hard to do as well though.

My suggestion is just reflash the firmware with whatever version you were using, reload the settings, and try again. Something you did may have corrupted the firmware (power surges can do that for example). Also, if you have a stimulator, it'd be good to hook up to that to see if the TPS works... Also, try taking the CAS out of the engine and spinning it, and see if that gives you anything on the RPM gauge. If it does, I'd say you have ground problems or a short circuit somewhere.

Reflash your firmware and reload your settings and let me know if that helps, if it doesn't I have some other ideas, but they involve more troubleshooting.

 

Ok I calmed down a bit . I didnt go postal worker ..almost but contained myself. I agree the stim is a good idea I'll try that tomorrow. No voltage was anywhere near the VR wires. I hate the idea of reflashing the firmware was a PITA the first time but maybe its necessary. Just dont understand what happened this thing is as quirky as my X wife! A ground problem is very unlikely I have several grounds and made sure they are making good contact with bare metal,soldered connections and soldered ring terminals. A short is more likely. A wiring problem is so hard to accept I mean this harness is very high quality spent tons of time building it and used grommets where ever it passes thru metal. I am a electrician by trade so took my time and made it right. Thanks for the tips I get on it asap and post the results.

 

I tested the MS on the Stim and it does the exact same thing. So the problem is inside the MS board itself...somewhere. The RPM reads 2 all the time ,the TPS reads 87% all the time and the MAP sensor does not do anything. Theres no obvious burned components or dark spots on the board. Im wondering how to reflash the firmware? I had a heck of a time the first time so is there a different procedure for re doing it? Thanks back to work,ttyl.

 

It's very easy, just recompile your firmware with the proper inc files generated by easytherm, and then in the firmware directory, just doubleclick the "download" batch file, which will ask you a couple of questions, and then upload the firmware to the MS.

You should select "N" when it asks if you're upgrading from the standard MS firmware.

Ken

 

Copy your source .s19 file to the "megasquirt/car1" folder.
Run EasyTherm, browse to the .s19 file in the car1 folder.
Grab the rx-7 defaults from the EasyTherm menu, and verify com port setting.
Adjust the air temp bias resistor to match the one you used during MS assembly.
Hit 'write', then 'download', and ET will create new .inc files for you, embed them in the source .s19 file, copy the .inc files to the mtCfg folder automatically, and download the newly created .s19 file to your MS box. The entire process should take 5 minutes at most.

 

you have to make sure you have start address right when you burn, later versions of msns-extra have moved them I believe (could be wrong though). This is why I suggested using the download batch file.

 

All commonly used code versions use the default (F500) start address. The steps I posted prior are all that are needed. I put that start address option in ET to suit Eric Fahlgren's custom DualTable code version, which had the thermistor tables at F300.

 

Renns and Muy thank you both ..you wre right. I reloaded the firmware and entered in all the values and it works again. So now we know if anyones v3 ms freezes up like mine did thats the issue most likely. I still havnt got the car to actually run..it sparks and stumbles and is very close. I have one small issue with the fuel pump still. From reading everyones posts it seems the fuel pump is suppose to run a short time when you first turn the ignition key to the run position. The relay for the pump is triggered from the MS terminal 37? on the db37. Mine does not work that way I have my fuel pump running all the time when the key is in the run position. I have it wired exactly like the diagram on the ms forum so I think its a setting issue in the MS itself. Everything else works great I have a GM full sweep TPS and GM coolant and IAT sensors calibrated beautifully with easy therm thanks to renns for that. Fitting a GM TPS wasnt that difficult and worth the time IMHO. Anyways Im back to moving forward again,thanks again.

 

there aren't really any settings for the fuel pump that I know of... it's a wiring issue if it's just on when the key is on... I'd suggest disconnecting the fuel pump wire and checking if it still stays on.

 

Yea I agree I'll check into it ..just not a high priority for me. Im just happy the MS is functioning and back to getting the car running.

 

Well Im confused and at ther end of the rope here. I reinstalled the firmware and it solved the MS from locking up problem. Now for the life of me I cannot get a RPM reading. The firmware setting on megatune2.25b514 and MSNS are exactly like in the FAQ Ive checked it about 8 times now. My wiring is also per the FAQ and worked before my firmware problem above. Tonight I double checked the wiring tracing each wire back to the DB37 each is where its suppose to be and connected well also insulated well. Every signal works except RPM so therefore no spark. Im wondering 2 things....first is it possible my CAS went bad? I cant see how because its a magnetic sensor but Im asking anyhow. Second I have a perfectly good electronic distributor and coils is there a straightforward way to hook that to the MS? I read on the MS site and as usual it raised more questions than it answered. Thanks.

 

Without a working lm1815 circuit, you won't get any rpms on the screen... if you did something bad enough to require a reflash of the MS, you may have killed the lm1815 chip. At this point that's the only thing I can think of... it sounds like you're not getting a signal from the lm1815 to pin 11 of the ms.

 

Sounds logical. My only question now is none of the chips on the v3 board are labeled lm1815. Remember this is leading and I have it wired to the db37 connector. All the jumpers are correct as it worked before. Im hoping its not some wacky chip and I have to special order one from digikey. I do have a spare new LM1815 chip. The leading comes from the CAS to pin 25 and 7 I believe, if those numbers are wrong pay no attention I have the correct pins just my memory sucks. Also I dont see a 14 pin chip anywhere so Im prolly SOL for now..lol. Any further tips would be great,thanks again. Shawn

 

again, leading and trailing have nothing to do with the 24 tooth wheel vs the 2 tooth wheel....

The 24 tooth wheel is used for e-shaft position, and the 2 tooth wheel is used to reset the counter on the 24 tooth wheel (so it's used in place of a missing tooth). This is how the megasquirt uses the CAS.... you need both the built-in vr sensor conditioner and the second lm1815 conditioner to both be working properly to get RPMs to show up on the screen, and for even leading ignition by itself to work. NOTHING will work right if you don't have both signals being conditioned and fed into the MS properly.

The built-in v3 conditioner also isn't an lm1815 conditioner, it's a custom circuit designed for that board... only the extra conditioner circuit on the daughterboard is an lm1815 conditioner... that's the one that's probably broken... That's the one that will cause you to not have RPMs shown because without that, the count on the 24 tooth wheel never gets reset.

It should be to pin 24 on the db37 for Ne+, the Ne- doesn't matter as long as it's one of the ground pins.

 

I think Muy is much better at this then me. I'll quit responding.
One note though, on the V3 boards, it doesn't use the lm1815 chip. The chip is U7 on the board.

 

yeah, that's what I was saying, there is no 1 chip on the v3 board that is used for vr sensor conditioning, it's a combination of a couple of transistors, an op-amp, and a few other types of components...

 

Ok now that I understand,thank you. Now I have a stoopid question. If you need the vr lm1815 ckt and the onboard ckt why is this not said anywhere? Damn thats my problem I actually had it working on just the 24 tooth wheel by dumb luck yanno. I unhooked the lm1815 ckt thinking it was to trigger trailing..now I understand. All this time I was on the wrong track. Well at least I know now. Feel free to laugh at my expense. And to think I was about to go postal worker. Im gonna reread the FAQ and throw the LM1815 ckt back in there since I spent so much time building it and making a box to enclose it et etc. Any documented cases of MS sending anyone to a nut house? And to think I couldnt even look at the car today was too upset with it. Thanks for explaining that in detail...I'll shut up for a while now.

 

I did say it in the FAQ that either you have to have the 2 conditioners working, or you have to cut 2 teeth out of the CAS, and I'm not going to laugh... everyone starts learning somewhere.

Go relax a bit... then get back to getting the MS to work!

 

It works again! Just had to solder 2 wires and the LM1815 is back in the loop and I have spark....small miracles do happen. My fuel pump is still running whenever the ignition is on and I think its causing the engine to flood. Im working on that issue tonight. Also Im using 720cc injectors 4 of them because its a turbo 13B so maybe my fuel map is a bit rich for idling. My fuel map is the standard MS map prolly for a V8 havnt been able to get any more specific setting so I left it alone. Thanks again off to the garage.

 

make sure you adjust your req_fuel value based on the fact that you have 720 cc injectors...

In the window that lets you calculate req_fuel, set 2600 cc's for your engine size, and 720cc for your injector size, and it'll recalculate req_fuel for you..

getting a good idle with those 720cc primaries could be pretty hard if the idle pulse width gets down below 2 ms...

 

Kewl thanks for that info....I will do that first thing tomorrow. If the idle is lumpy I'll deal with it. I just know that I will need all the fuel the 720's can supply so going smaller even just for primaries isnt a option. That fuel pump wiring problem is still doing the same thing. I have a harness I made using the V3 wiring diagram on msefi.com. I used a bosch relay and wired terminal 30 to constant power,term 86 gets switched with ignition power,term 87 is the output to fuel pump and 02 sensor and term 85 goes to pin 37 on the MS. Wired this way the pump will never turn on. If I connect term 85 to ground the pump runs as it should. So now I have terminal 85 grounded and the pump runs whenever the ignition is switched to run or crank positions. I was thinking this was more than needed and causing the engine to flood. Upon thinking further into it the ms should only open the injectors when the program calls for them to be open no matter how much the pump is runniing so it really shouldnt matter either way correct? Anyways tonight I double checked that wiring tracing each wire by itself and doing continuity tests on each I also replaced the relay with a new one and the problem is still there. Im thinking its minor and just leaving it grounded should be fine/. It wants to start but wont quite hope fully after adjusting the req fuel it will tomorrow. Thanks again,Shawn

 

Having the pump running will not cause flooding unless you have very leaky injectors. I'd recommend getting that fuel pump circuit fixed, as you don't want the efi pump spraying fuel out a leak onto a hot engine in a fire. With MS controlling the pump properly it'll shut off when the engine stops in an accident. If you check in Megatune, you should see the fuel pump output turn on for a couple seconds when MS boots. Then it will remain off until it sees a tach signal, indicating the engine is cranking or running. If the fuel pump indicator is working, then the code is configured properly, and you have either a wiring or output circuit (smoked transisotr) problem. If the pump output stays on or off continuously in MegaTune, then there is a configuration or code problem.

 

I recommend for people needing bigger injectors that they stay with the stock primaries and get bigger secondaries... so do something like the 550 cc primaries and 1000cc secondaries... that allows you to tune the idle and low load stuff better, while still getting enough fuel when the secondaries come on.

You and renns are correct that the engine should not flood from the fuel pump staying on...

If you're flooding your injectors are leaky (like renns said) or you have some other wiring problem... OR you just have your cranking pulse-width and req_fuel set too high...