Hi Megasquirt builders,

How many of you have actually completed a working MS1 system with RX7 ignition control using the dual LM1815 VR conditioner circuit as shown in the schematic below:



Here is my second attempt, which only has a working 24 tooth circuit (half it it doesn't work):



I tried it both ways, one with reading the diagram as there being a node at the intersection pointed out in red, the other without. Both situations don't work.

It has been bench tested using a 5v source, FC CAS, and multimeter. Only the left half works, and this is exactly the same situation I had with my first LM1815 VR conditioner circuit. It seems unlikely that the bad half's LM1815 was a victim of ESD in both attempts.

So...... Has anyone actually made one of these and had it work?

 

I've had several work..... Actually, I've never had one fail to work. I'm not sure what you're doing wrong from looking at your circuit. The spot you have pointed out just means that both should be connected to +5v. The only time I've seen one of these circuits have problems is when there are potential ground loops and such.

How did you have it wired to the MS? The output of the conditioner of the right should go through a 1k resistor to pin 11 on the MS. Also, how do you have your software set up??

 

The wiring to the MS is inconsequential at this point as I'm only trying to verify that the circuit works as is, before putting it into a larger system.

With that in mind, I should be seeing a 0v / 5v squarewave from both circuits right?

By the way, I gave up. I just shaved two teeth off my 24 tooth CAS wheel. It's single circuit time. If the 24 tooth and only the 24 tooth outputs the correct square wave (unless my understanding of the required output from VR conditioner circuit is wrong), I'll use the one that works.

 

you're not going to be able to verify the 2 tooth wheel without an oscilloscope or at least without hooking it up to the MS... Trying to verify with a multimeter, you'd have to spin the CAS VERY fast before you see anything.

I guess it's too late now that you cut teeth out of your CAS, but your second trigger was probably working fine.

 

I just checked this on my testbench, and a measurable voltage is seen at the 2-tooth output even at cranking speeds. I measured around 150mV on DC, 30mV on AC. It's certainly not ideal, but any small voltage that increases with rpm will give some comfort anyways. It wouldn't hurt to put some notes in the FAQ on this, as those building these twin LM1815 circuits without a full array of test equipment will need some way of testing.

Those building there own can avoid the hassles of these circuits/daughterboards by using the V3.0 board and knocking opposing teeth off the CAS wheel like 88IntegraLS ended up doing. That'd be the simplest way I know of to get up and running with FC fuel & spark.

 

Hmmm. With the circuit built as shown in the schematic I saw .15ish vdc coming from the 24 tooth circuit and constant 5v coming from the 2 tooth circuit. The .15 would move up and down by about .03v when the Cas was spun by hand.

When I removed the intersection shown in red, the 24 tooth side fluctuates from 0 to 5v in accordance with the teeth swiping past the VR sensor.

Now that I have it configured for 12, -1, missing tooth in wheel decoder settings with the 24 tooth circuit outputting 0-5v pulse waves to pin 6 on the opto isolator, the rpm goes insane, 0-20000 basically. This was checked in megatune. Even touching a screwdriver to pin 6 and nothing else touching it produces the same effect. The megasquirt has been flashed and reflashed and works correctly until I try to touch pin 6. Pin 6 is being taken as the pin to the right of the white dot, with the dot and mark facing up.

Is there anywhere I can search for specific instructions for wiring a single VR conditioner circuit to the MS to work as a missing tooth setup? I've searched the rotary ignition thread over at msefi.com last night, and in here.


thanks

 

My Lm1815 ckt has yet to work as well. My ckt is for trailing only as I have the v3 board. I know its built correctly and wired correctly yet it still dosnt work. This isnt my only issue so Im just concentrating on getting leading spark for now. I wish you luck MS isnt easy..its cheap but not easy.

 

the lm1815 has nothing to do with trailing vs leading on the megasquirt... All it does is tell the ECU when it can reset the count on the 24 tooth wheel... so it'll count up to 12 on the 12 tooth wheel, then a tooth from the 2 tooth wheel passes, and it resets back to 0, and counts up again.

If you're having trouble getting it to work, 1) make sure it's hooked up the way I detail in the FAQ, and 2) make sure the v3 conditioner potentiometers are both set all the way counter clockwise... and 3) make sure your ECU has proper grounds... don't ground it to the chassis as that'll cause problems.

 

So you're saying you get good RPMs until you touch pin6? The output of the lm1815 that is on the 24 tooth wheel should go to pin 6, and it sounds like that's what you've done, so I'd say touching that pin is introducing a lot of noise... so don't touch it

 

You know, I've seen this problem before. It could be a number of things, but I fixed one of these circuits that had this EXACT problem....

The solution was to replace the LM1815 chip as the person who soldered it on originally let it heat up WAY too much. I replaced the LM1815 chip in question (which was the 12 tooth circuit in my situation) and everything worked as planned.

Just a note, with the LM1815's I use sockets so I can swap around the chips, but more importantly, keep heat out of the chip.

Just something to look out for, not saying it is the exact problem.

 

Exactly, with one clarification: the good rpms come when the lm1815 circuit is NOT CONNECTED to the megasquirt, the ms unit works perfectly in fuel only with the stim running it, it goes crazy if ANYTHING, LM1815 CIRCUIT INCLUDED touches pin 6 of the opto isolator.

I really appreciate your efforts in developing RX7 ignition for us, but it appears to be out of reach to those of us who don't understand how it works (results of this survey). I'm about ready to go EDIS; I've done everything you said to do and didn't touch the LM1815 pins very long when soldering. I soldered the CPU onto the megasquirt board (no socket) and didn't have it overheat, same with all the ms chips. Unless the LM1815's are unusually sensitive to solder heat compared to other chips, that isn't the problem. I'm not saying it isn't, but it seems very unlikely.

 

Has anyone thought about switching over to a distributor type cas?

seems like it would be more accurate for the type of signal we're looking for, ... 0, 1?... just my thoughts.

 

I've got my V3 board with one added LM1815 (in the proto area) working on the bench. Plan to try it in the car this weekend.

 

Anyway you could post a picture of this? I'm curious to see how other people have built it.

 

If you haven't actually soldered the output of your lm1815 circuit to pin 6, and you're just holding it there with your hand, then I'm not surprised it goes crazy.

any kind of brushing or scraping movement for any of the VR sensor contacts or any of the connections anywhere in the lm1815 circuit will cause the rpms to go crazy. I've tested this by just scraping the G+ wire with the allegator clips I use to make test connections on the bench... just doing that will cause the RPMS to bounce all over the place without me even spinning the CAS. Everything works fine as long as you create a good connection everywhere and hook it up right.

also, everyone else in this thread who has tried to use a dual vr sensor conditioner has it working... so I fail to see how the result of this thread is that it's out of reach unless you know how it works... custom 13b even has his working now... Not trying to be confrontational, but just saying that I'm not the only one who has it working... almost everyone I've helped has it working now, and several people who have it working got no help from me at all aside from the FAQ.

Ken

 

I found the problem. I was thinking back to my electrical fundamental class this morning, half asleep and it occurred to me that I was not powering my lm1815 circuit with the same power as the megasquirt, and the did not share a common ground. The voltage coming out of my vr conditioner was useless without a return path for the current, which I didn't give it. No wonder the results with a screwdriver touching the pin 6 were the same as the circuit.

I wired it up correctly today, cas in hand, and saw the same 20000 rpm until I spun the cas: instant 120 rpm with 2 turns / sec. Thank God.

So yeah, you were right, I had the circuit right the first time and didn't have to shave off two teeth from the cas. What would have helped me would have been to know that the output from the 24 tooth circuit should be .14-.16 vdc oscillating when the cas is spun by hand, and the 2 tooth circuit should output a straight constant 5v.

thanks again for all the help.

 

Oh one more thing if you don't mind me asking, should the wheel decoder settings be 1 3 7 and 9 like with the dual circuit? Same trigger return, etc? Thanks

 

heh, I never even stopped to think that you might be powering the lm1815 circuit from a different place than the megasquirt... Keep in mind that the lm1815 circuit should be powered with a +5v source... the lm1815 itself will work fine with +12v, but the processor might be damaged by anything over +5v.

Those wheel decoder settings will be fine as long as you align the CAS right (when you insert it, have the 24 tooth sensor in line with tooth 3).

 

Here's the best closeup I could get of the proto area where the second VR circuit is.

More pictures here https://www.rx7club.com/showthread.p...=1#post5113688

 

You're build is certainly different then mine.

 

I think I like yours better. BTW, I've also got a couple components there on the bottom for a second O2 input that I'll never use.

 

I just used the msns-extra daughterboard, that way people who buy my setup can use the proto area for other things as they feel like adding them.

 

LoL, maybe! but i'm not certain if my circuit works yet.

 

Well...i'm not selling my setup...yet
If someone is interested making something different, they can buy a Radioshack board just as easily. If my build works, it'll be a single board build plug-n-play.

AGAIN...my setup is not verified yet, so we'll see

 

yeah, the other reason I don't use the proto area is because I need to enable several extra outputs for controlling things like the fuel pressure solonoid, VDI, Aux ports, etc...

There's not a whole lot of room in the proto area for more than just the lm1815 circuit