Caddywompus sensors

[yellowoctupus]

So, the short story is I have a Ford 351w V8 with stock sensors hooked up to a MS1 unit running the 0294a firmware. I'm having issues with the sensors showing up completely incorrect. It's roughly 60F here, and they read -14F. I figured, ok, it must just be the Ford / GM sensor calibration issue, so I fired up Easy Therm, used the following values for the IAT and CLT sensors (33deg=94k, 123deg=11k, 208deg=2.37k), but when all was said and done, I still show bogus sensor values in Tuner Studio. Additionally (just to add confusion here...the Battery Voltage dashboard shows 0.12V, and the O2 sensor shows 1.98V (narrowband Ford...).



I will gladly attach an MSQ if that's helpful, at this point it seems like more of a basic configuration problem to me....

[Matt Cramer]

A data log and MSQ may still be helpful.

[yellowoctupus]

Hi Matt, thanks for the reply.

For what it's worth, here's the tune and a short (non running) datalog. I have some 27k resistors on order to swap out on the board, unless of course I get the easy therm settings to stick...

Thanks in advance for any help/ suggestions.

[yellowoctupus]

Could this issue have to do with the fact that I can't locate (or copy) the newly created '.inc' files to put into the TS project directory? I feel kinda stupid at the moment, as I can see the newly created .s19 files in the EasyTherm directory through the 'Write .inc and .s19 files' button, but when I navigate there through Windows Explorer, the 'new' files aren't visible.

(yes, I have Hidden Files/Folders selected in the folder options...)

Then, as a sanity check of sorts, I copied the most recent Easytherm modified .s19 file from the EasyTherm program, and tried to paste it on my desktop, and I get the error "Could not find this item. This is no longer located in C:\Program Files (x86)\EasyTherm. Verify the item's location and try again."

What in the world am I doing wrong??

[yellowoctupus]

SOME MINOR SUCCESS!!

So, my EasyTherm woes were somewhat solved! This the solution:

Figured out what was going on with Easytherm and Win 7. The files were being written to some virtual directory that was not visible to me in the normal directory. So ET thought everything was fine but Win 7 was isolating the output to keep me 'safe'.
Found them by viewing compatible files, fixed it by giving ET administrative rights (which it did not have for some reason, even though I installed it as an Admin). http://www.miataturbo.net/megasquirt-18 ... post593303

This is how you change the administrative control:

To mark an application to always run as an administrator, do the following:
1. On the Start menu, locate the program that you want to always run as an administrator.
2. Right-click the application’s shortcut, and then click Properties.
3. In the Properties dialog box, click the Compatibility tab.
4. Do one of the following:

To apply the setting to the currently logged-on user, select the Run This Program As An Administrator check box, and then click OK.
To apply the setting to all users on the computer and regardless of which shortcut is used to start the application, click Change Setting For All Users to display the Properties dialog box for the application’s .exe file, select the Run This Program As An Administrator check box, and then click OK twice. https://technet.microsoft.com/en-us/mag ... 31742.aspx

Now ET creates the files right in the Program Files directory on my computer, as I would expect! Next up is attempting this on the laptop, and see if MS is happy with the solution too.

[yellowoctupus]

To close the loop here, the admin rights thing is totally the trick to creating and seeing the s19 and inc files correctly. EasyTherm seems to be hit or miss whether it wants to Write to MS or not properly (it seems to just freeze at "Line 25" about 30% of the time, after wiping the old firmware...). In this case, I just used the download-firmware.bat loader, with the note that you have to have your new file in the same folder, with no other .s19 files. And it has to be labeled msns-extras.s19 (ET puts a "_mod" in the filename that is not automatically recognized by download-firmware.bat.

Additionally, the three .inc files created by ET need to be moved from the ET folder to your car's project folder in the 'inc' folder, not the 'projectCfg' folder.

Now my temp sensors are happy. I don't think it's 68deg in here...but they're only off by maybe 5 deg, so we'll call it good enough. Closer than -19F or whatever they were reading before.

[yellowoctupus]

Should I be concerned that the 02 sensor (not running) gauge in TS has the text displayed as ~2.02V, whereas the gauge shows ~0.8V??

2nd gauge I'm wondering about is batt voltage which the gauge shows ~16.8v, but the readout says 0V.

Seems weird to me, but I'll probably try to fire the car up sometime tomorrow regardless. Let me know if anyone has ideas on these two...

[Matt Cramer]

Check the voltage coming in through the O2 sensor circuit. For the battery voltage, measure the voltage across the H1 boot header; it should be 1/6 battery voltage.

[yellowoctupus]

Check the voltage coming in through the O2 sensor circuit. For the battery voltage, measure the voltage across the H1 boot header; it should be 1/6 battery voltage.

The voltage at the boot header was 2v, so I guess that checks out, although I'm not sure why it reads different in TS. (actual battery voltage is 12.56.

The voltage at the 02 actually reads 2.7 at the relay board (no bueno...). It's just a stock 02 Ford narrowband 02 sensor, but now that I have it unplugged, it's not reading ANY voltage, so maybe there's a sensor issue. (it should be reading ~100mV just sitting there, cool and unplugged, right?)

I had it wired like this:

1 wire goes to the injector 12V+ (heater leg) on the relay board.
1 wire goes to the 02 sensor connection on the relay board.
1 wire goes to the battery ground terminal (not on relay board, but same ground)
1 wire goes to the MAT Return.

I also attached a longer running MSL file, and I'm learning a little bit how to use the log viewer. Looks like an awesome tool so far.

2nd file attached is a screenshot showing my battery voltage meter's needle being different than the value shown, neither of which actually match the 12.5v the battery actually has.

[Matt Cramer]

If you reload the firmware and do not make any changes to the default tune, does the battery voltage read correctly then?

[yellowoctupus]

does the battery voltage read correctly then?

Unfortunately, not. Dealing with the Win7 vs EasyTherm fiasco, I probably reloaded the firmware...10 times? And during those, it never showed correctly. One thing I may resort to is unhooking EVERYTHING from the relay board connectors, and then just plug them in one by one to see if I'm getting a stray signal from the TFI or....????

The log file I attached in my last post shows the battery voltage signal going from 0-29v, and it's all over the place getting there.

[Matt Cramer]

The voltage at H1 means something's wrong with either the processor or its socket. If you're lucky it is just firmware or a fault in the soldering on the 40 pin socket. If you're not, you may need a new processor.

[yellowoctupus]

For the battery voltage, measure the voltage across the H1 boot header; it should be 1/6 battery voltage.

The voltage at H1 means something's wrong with either the processor or its socket.

Ok...which one? Is there supposed to be 1/6th battery voltage at the boot header or not? These two comments seem to contradict one another.

Additionally, my processor is soldered directly to the board and all of the joints look good....

[yellowoctupus]

As a mini-progress update, I checked my other MS1 box (I bought both from a friend who had collected them over the years from different sources....).

1. Loaded the standard 029y4a firmware on the '#2' Megasquirt.
2. My old computer power supply voltage is a measured 10.93V.
3. TunerStudio says 10.71
4. Made a EasyTherm modified .s19 file and loaded it onto the Megasquirt unit
5. TunerStudio still says 10.71V.
6. Both times the boot header voltage was 1.79 (1.79*6=10.74)

7. Loaded the same EasyTherm created firmware (from step 4 above) to #1 Megasquirt
8. Tunerstudio Battery Voltage Gauge needle shows 13v, number readout is 29.88V.
9. NOW, boot header shows 5.75V (What??)

The only difference wiring wise between the two is that the #1 is a 'factory assembled' unit, and #2 is home built. #1 also received the Ford TFI mods as well.

In the long term, I'd like to have both units working (#2 was slated for my 280z+4.6L Ford setup which is still on the stock ECU currently...)

So, give it to me straight doc.

Is she dead? If the description above points directly at the processor being dead, I'll get a replacement and solder in a socket onto the board while I'm at it.

[yellowoctupus]

tps dying.jpg

So, now that I've changed MS boxes (modified MS#2 for Ford TFI and have that one in the car currently) I seem to have a few issues here to work through too!

TPS sweeps nice and clean. Repeatable, no jumps etc. Calibrates as expected. Once I start it up, I get a few sweeps before it basically nosedives into oblivion. See attached pic. Then it seems to run a bit rougher, although it's hard to tell if it was running better, as the TPS signal goes into the negatives so quickly upon startup.

O2 is also not working properly. It shows good voltages when it's detached from the relay board (measured a rising 0.1-0.6 with a voltmeter with the engine running) but nothing changes that I can see in TunerStudio.

Thanks again for the help/suggestions. Can't wait to start doing some actual tuning!

[yellowoctupus]

ok, here's the latest on the o2 sensor.

Roughly 0.2-0.24V on the tunerstudio dashboard for the O2 sensor readout, with no sensors plugged in. (literally, just power to MS box)

I have a AA battery with an inline 1Mohm resistor which shows 0.787v by itself, then 0.122V when it's hooked up to the 02 and sensor return lines on the MS box.

Tunerstudio now reads 0.27-0.31V.

What's the deal with this? I guess there's a floating ~0.2v on my 02 sensor input pin, is there a component that should be adjusted, or is there a setting somewhere I'm unaware of that I need to change?

The attached MSL shows the battery+resistor being plugged in, then a few quick blips of the resistor jumpered, then it gets unplugged from the circuit entirely.



**Ok, based on the MSL, it looks like the voltages MS is seeing are correct, however they're not showing up as expected in Tunerstudio. Why's that?**

[Matt Cramer]

The voltage should be 1/6 battery voltage. If it is, and the battery voltage is reading wrong, something is wrong in the processor.

[yellowoctupus]

The voltage should be 1/6 battery voltage. If it is, and the battery voltage is reading wrong, something is wrong in the processor.

Oh, I understand now, thank you. I must've misread/misinterpreted your last post.

Any ideas on the Throttle position sensor changing upon fire-up, or the O2 sensor not reading properly in TS, but looking ok in the log viewer?

[Matt Cramer]

For the O2 sensor, check how you have it configured in Project Properties.

Where exactly is the TPS grounded?

[yellowoctupus]

TPS ground is shared with the CLT and MAT coming back to my relay board at the CLT RET. The TPS /harness is right next to the distributor and a spark plug line, but that shouldn't affect it since it's just a DC signal, right? I did verify VREF stays constant at 5v ignition on and running.

O2 sensor is set as Narrowband Sebsor (Default) under project properties. I have an LC-1 I will probably use up to do my tuning, at least, but had the narrowband there to check to make sure the channel was working properly first.