A little background first...

Purchased Ms2 v3.0 back in April or so, and spent a few months on and off constructing the board. Read all I could, which I guess wasn't enough , and went ahead with a basic install. By basic I mean that I am not running anything with the ms yet. I tapped into TPS, Crank Angle, stock o2(did I even need to?), installed both GM sensors, and wired into power and established ground. I am eventually going to go turbo, as I have already rebuilt for it, and wanted to get the ms installed before doing so. Now the issue...

Everything ms seems to be fine as far as the readings that I am getting while in MT, except the damn tach. My first mistake was when I built the ms, I had read the posts by Corbin, Wicked and Gil about the tach spikes that they experienced with the 3.0 board, so I wired my board with this in mind and followed Corbin's fix for the situation found here:

MSnS V3 Board Tach Spikes Fixed!

No go. I still saw jumps in the tach reading. So I continued searching and came across Wicked/Gil's fix:

Running MSnS v3 Board as FULL Standalone (No Spikes)!

Now with this setup, I was able to get a steady tach signal, however, it seemed too far under what my stock tach was reporting( stock tach was at 950 or so and ms was in the high 600's/low 700's and as I would rev the gap would grow until it was around 1k apart up in the 4000's), so I continued to adjust the pots, and when The tach reading on the ms was VERY close to the stock reading, the spikes were evident again. It would spike from @950 up to @1300. I tried even finer tuning the pots, however to no avail. After searching more indepth on the ms2 site, I stumbled across what I thought was going to be my answer:

MS2-Extra Neon/420A Manual

I followed this setup and now my tach is up in the high 3000's and low 4000's. Also, it is still spiking. I was hoping that some of the ms guru's would have some suggestions?




One quick question about the wiring schematic. In this picture, are D1 and D2 supposed to be open? That is how I have them now. I had them both jumped before. At this point, the only thing that I can think of doing is increasing the value of the c12 capacitor. Thanks for any responses in advance.

DirtMegirt
Wiki Contributor

A few guys tried to increase the value of the ED capacitor but claimed it didnt help. I never went any further into this problem. I chose the easy way and stayed with the older board. In fact the spare 3.0 i have still isnt built.

Another thing to try and make sure the MS case isnt touching the vehicle chassis at all.

DirtMegirt
Wiki Contributor

How stable is your battery voltage at the Megasquirt? Can you post a datalog of it with the car running? Did you clean all the solder flux off real good?

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

DirtMegirt
Wiki Contributor

If those are your min/max volts from you datalogs than that is good, I was just looking for spikes there. Are you running 8v regulated power to the crank sensor as suggested in my writeup? Does the crank ground go to the Megasquirt? Is the signal wire shielded? If all is true, then have you noticed my comment about soldering in those extra capacitors to greatly stiffen up the overly sensitive MS1 processor maybe the MS2 needs the same capacitors.

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

No Joel, i'm really not doing anything that is suggested in your write-up except what Gil suggests to do about halfway down in your post. I noticed that you run MS standalone, and now that I think about it, Gil does too, doesn't he? For some reason, I thought he was piggyback like Corbin. Looks like i'm going back to your post. I just tapped into the blue/white crank wire at the ECU for my connection. The wire in my db37 harness is shielded. Also, maybe you could let me know how to get those datalogs you are talking about?

DirtMegirt
Wiki Contributor

I was just talking about a datalog from MegaTune, then looking at it with MegaLogViewer, people in the Megasquirt forums usually just post screenshots of the areas of logs they are concerned with, and then attach the complete datalog.

I don't know of anyone who's been able to run the v3 board with spark without taking the stock ECU out of the picture. It's just too sensitive to all the noise that gets induced with all that extra wiring everywhere.

I'm pretty sure you're going to have to run a dedicated power, ground, and signal line from your MS to the crank sensor.

This was what made my car drivable with the MS1 processor v3 board, although, I did have to add some additional capacitors on top of that to eliminate some seldom spikes thereafter.

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

DirtMegirt
Wiki Contributor

Those are the caps, but they are soldered onto the processor pins, not the db37.

Battery voltage shows up as battvolts I think. If it's not in the list, you need may be using an older version, and will have to add it as a logged parameter.

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

Alright, just a little update. Nothing good however. I have tried every setup that I have come across, and continued to read as to hopefully not miss anything that might help. Just to clear something up, I do not have the ms installed and it is not running anything at this point. I have made the connections and am still in the process of wiring a harness for the injectors. The issue that I am having is that I cannot find a tach signal that is both accurate and stable, it's either one or the other. After reading Corbin's post, I realized that he says that he only got tach spikes at a certain rpm range, and only after adding spark duties to the megasquirt. At this point, I must have done something wrong. Terry, in your experimentation with the ms2 v3 board, were you able to get it to run fuel successfully? That is all I am trying to do. It seems to me that I will need a stable tach signal nomatter what setup I run, but am I looking at this wrong. I have even tried to connect the ms wire to the white ecu tach wire, but the megatune rpm's seem to be about 1/2 of what the stock gauge reads. Anymore info you guys can give me will be very appreciated!

DirtMegirt
Wiki Contributor

Did you test the setup on the stimulator? Also, how are your grounds. I had some issues with coolant, egt, and IAT spikes before I added more grounds to the MS.

AEM cam gears, .030 over, .020 milled head, Unorthodox UDP, port match intake/head, polished exhaust manifold, e-bay CAI, Magnaflow muffler

1995 GS A/T: Crushed, what started it all
1999 RS M/T: Back to stock
Mazdaspeed6: sold and quazi missed
1999 Eclipse GSX: HX35, waiting for Shep STG3

After trying alot of different things, I finally went back to the VR setup that Gil laid out in Joel's post. I now have a tach signal that is stable, however only up to @ 1300rpm, at which point the rpm's drop to 0. I am happy just to be getting a tach signal that looks accurate. Also, the issues with trying to burn settings to the ecu appears to be solved. I was able to input the settings for the decoder, and all of them burnt fine. I think my map sensor is off a little bit, as I don't think the reading is correct. Just to repeat, I don't have the ms controlling anything yet, just reading values. Also, to answer the above question, I grounded everything to the point where the negative battery terminal is grounded. Anybody have any idea's about the rpm issue?


Thanks for any info ahead of time.

DirtMegirt
Wiki Contributor

Ahh, so you're piggybacked into the stock tach signal. I don't know anyone successfully doing this on the v3 board because of the noise you get.

Before I converted mine to a standalone and had the crank sensor's 3 wires go straight to the Megasquirt I couldn't drive the car over 2200 rpms, without loosing the rpm signal to noise.

You think there's anyway you could get your hands on a scope to actually take a look at the crank signal at the megasquirt?

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

Joel, for the tach signal, i'm tapped into the blue/white crank sensor wire at the ecu. I wired it that way for ease of installation, however do you think it would make a differnece if I tapped that wire at the sensor itself. On a side note, I had to replace the stock 95 crank sensor with a 97+ sensor. When I wired it in, I kept the layout of the wires the same. I shouldn't have assumed, but I thought the sensor would not function properly if it were installed wrong. Does anyone know if that is the case? It's interesting hearing you say that you couldn't get the car past 2200rpm. I thought that i've read successful posts about getting the v3 board to run fuel fine, but when you go to spark control, that is when the issues pop up. I do not have access to an o-scope. Again, thanks for the help.

DirtMegirt
Wiki Contributor

Joel, in one of your posts you say that you hooked up the crank sensor signal using the inner of a shielded wire and grounded the sensor with the shielding. Am I understanding that correctly? Would there be any negative, or positive for that matter, with me doing the same thing. Would there be any less noise tapping into the wire at the sensor, rather than at the ecu, and grounding it using the shielding or another length of shielded wire to ground that sensor to the same spot as the ms? I would like to try this, but was hoping that someone could chime in on any possible drawbacks.

DirtMegirt
Wiki Contributor

See, if you have your crank sensor still hooked up to the stock ECU than all that wiring potentially becomes an antenna for noise.

I didn't have a scope at the time to see exactly what was going on, but when I ran the 3 wires from the crank sensor to just the MS using a shielded twisted pair (because that's what I had available). I ran 8v from the MS to the crank sensor, and used the shield to supply ground (not ideal, but better than no shield at all).

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

DirtMegirt
Wiki Contributor

It's something we've seen over and over that the v3 board is sensitive to on our cars, and the v2.2 does quite well.

I must say because of the modifications I've had to do, I do not recommend the v3 board to someone who isn't eliminating the stock ECU completely.

Joel Baldridge, ASE Certified Master Technician, Audi Certified Expert Technician

If it matters at all, there is at least one person running MSI V3.0 with spark in parallel with the stock ECU. I am still using the input circuit mentioned in that old thread. My tach wire is sheilded in the MS harness. I tapped into the crank signal at the stock ECU. The stock coil pack is powered and fired by either the stock ECU or MS (selectable for emissions). Power for the crank sensor (and all other sensors) is provided by either the stock ECU or MS (selectable for emissions). Injectors are powered and fired by either the stock ECU or MS (selectable for emissions). It probably isn't perfect, but I don't have any trouble rev'ing to 8k RPM.

Corbin

'95 ESI-T
Now with more power and fewer leaks



Gimme fuel...Gimme fire...Gimme that which I desire

There were some known issues with the MS2 Extra neon decoder. Get the latest beta (18 right now), and it will work fine. Also, i'm still unclear why so many want to use the VR input instead of the normal opto isolator tach input circuits. The crank sensor is a Hall effect sensor, which is NOT a VR sensor. The Hall sensor outputs a digital 0-5V waveform, that works fine with the basic digital logic circuits like the MS chip. The opto circuits have a capacitor on the input side, and also one on the output side to the chip..shouldnt be any issues with noise. If anything, make sure you have good ground signal at the crank sensor's ground wire, and also good voltage at the sensor's power supply wire. The sensor just switches ground, so this is very important. You can try this updated circuit if you wish:



Instead of the 150 ohm resistor, use a 470 ohm. Remove R12. Put the one leg of the 470 ohm resistor on the RIGHT side of R12(you can also put the 470 ohm resistor IN PLACE of R12, just make sure to get the 5 volts to the top side(left side in the diagram) of the resistor. Increase the .001uf cap(C12) with a .1uf cap. You dont need those diodes on the input side, you are basically going right around them(leave them out). If you STILL have noise, it may be worth feeding the crank sensor's power with a separate 12 volt switched line(again, from the battery), using a 10k ohm resistor in line to it. It's possible there is some noise on the 8 volt supply from the stock ECU. Also make sure you have the MS grounds coming straight from the battery, and also switched 12 volts through a relay, right from the battery(battery acts as a noise filter). Make sure you have good chassis to battery grounds, and also a good battery to block ground. You should also run a ground from the battery to the crank sensor's ground wire( you are basically running new power and ground to the crank sensor, so be sure to cut and insulate the car side of the sensor wires, and connect your power and grounds to the sensor side of the cut wires).

13.5 @108 MPH-2.2 60ft(stupid FWD!)
S16G @ 18 PSI/FMIC/Running on MegaSquirt II (Now with sequential
fuel injection)
My webpage: http://eclipsed4evr.home.comcast.net
-1998 Mitsubishi Eclipse RS-T- "Toy"
-1992 Plymouth Laser Turbo AWD(SOLD)
-2000 Honda CR-V(daily)

X, I tried the above input mod, with the ms2 chip, as it is pictured with no luck. I still have my v3 board, so I may try the different resistor and cap as you explained. What I ended up doing was getting a v2.2 board, did the 420a mod, and got it running with the ms1 chip. Then I swapped in the ms2 chip, downloaded the latest beta(18, I believe) and it works as it should. I put the ms1 chip in the v3 board and that runs fuel fine. I think I may stick with the v2.2 due to all the issues, however I may try your suggestions to see if they work. Corbin, I have the db37 cable from diyautotune and it has a shielded #24 wire, so I have the that part covered, however I never tried grounding straight to the batt. Thanks for all the input from everyone.

DirtMegirt
Wiki Contributor

13.5 @108 MPH-2.2 60ft(stupid FWD!)
S16G @ 18 PSI/FMIC/Running on MegaSquirt II (Now with sequential
fuel injection)
My webpage: http://eclipsed4evr.home.comcast.net
-1998 Mitsubishi Eclipse RS-T- "Toy"
-1992 Plymouth Laser Turbo AWD(SOLD)
-2000 Honda CR-V(daily)