vernw

iTrader: (2)

I'm trying to field test the manual tranny S_AUJP v4 for one of our illustrious developers (not sure he wants to be identified yet).

He's implemented the knock sensor detected light in this BIN on pin F5.

My question is this: Does anyone know if this pin will supply a ground or a voltage when it is "turned on"? I don't want to burn something up in the ECM by trying the wrong one.... LOL

JP86SS

iTrader: (1)

Glad to hear someone is able to try that out.
It was my idea so i'm on the hook for that.
Sorry to keep you hanging, didn't see the thread till now.
From my investigation it should provide a ground path.
this particular pin has 2 quad drivers in parralel so current should not be much of a problem.
In any case try to use a small light that won't draw too much, LED would be best if you have one.
If you do fry something let me know, i'll fix you up with my spare if needed. Not that I'd rather someone else be the tester, my motor is on the stand right now (looking real nice BTW )
Not sure on the actual amp rating of one of those. Can't find info on the chip directly to confirm it.
anybody have a better guess on that?
(single and parallel paths)

JP86SS

iTrader: (1)

From Ludis's schematic, The bottom 2 are F5.

junkcltr

iTrader: (1)

About a month ago I came across a thread were RBob said what the drive strength was and it is something like a National Semiconductor XX chip. I think the drive was up to 600mA from memory. Do a search for RBob and National.
I think a write to the F5 PWM reg with 0xDFFF turns the driver on, correct?

EDIT: If it was me, I would use an LED with a 1.4k ohm resistor to V_ignition or a 330 ohm to 5 volts. You could measure a relay coil resistance. I think they are usually 30-100 ohms. Then measure the light bulb you are using to find the resistance. You probably don't want the light bulb resistance higher than a relay coil resistance. Using an LED with a resistor (some have internal resistors) would be best.

vernw

iTrader: (2)

Thanks for the info guys (although honestly I don't understand all of it - LOL)

Anyway, I've got an 12v LED with built in resistor from Radio Shack already installed next to the trunk popper switch. I just ran the wiring over to the ECM and stopped until I heard the right way to hook it up. So I'll proceed (probably Thursday night or Saturday) with the idea that it is a ground supplied by the ECM. Guess I'll know for sure if I do some data logging and see knock but no light.

Or I see or smell smoke

Thanks for the offer of the spare ECM if something fries, hopefully that won't be an issue! LOL

junkcltr

iTrader: (1)

Nahh, you won't smell smoke. You could connect a light bulb like the SES light bulb and it would still be fine or you can use the LED.
The knock light will work just like the SES light circuit. It uses the same type of electronics to make the light turn on. So you could connect it up exactly like the SES light is wired (one side to V_ign and the other to the ECM).

I like LEDs because they are not as bright as a light bulb and are usually colored.

Could have used pin F8 also. It is straight forward to toggle. It is a discrete I/O, not PWM.

vernw

iTrader: (2)

GREAT! Smelling electrical smoke was never one of my favorites anyway LOL.

I like LED's, too, where possible. They're small, easy to put in, seldom burn out, and I'm looking forward to my orange "knock detected" indicator!

vernw

iTrader: (2)

One more S_AUJP_MT v4 item, although this may be the wrong place to bring it up. The BIN Scott sent me to test has the shift light stuff all built in and it's working fine as well. One thing about the shift light is a little bothersome....

I'm trying to use it like the original GM shift light, i.e., indicating a good economical time to shift (as opposed to a WOT/red line usage). The GM light from the factory, as long as you were using the factory rear gears, would not come on in 5th gear. At least I never saw it do so altho I've not had the car over about 115mph

Anyway, I suspect they originally used N/V ratio or some MPH check in the code to try and have it only indicate it's time to shift up thru 4th gear. Since I went to my 4.11 Ford 9 inch, it would come on trying to get me to shift above 5th on my T-5. So it had to be tied into the gearing somehow. The new V4 code doesn't seem to have anything to avoid this event either.

Would it be very difficult to add a tuneable parameter in V4 so you can make sure it doesn't tell you to shift above whatever your top gear is? If it's too much of a hassle, don't worry about it. Just a thought (I'll be putting in a T56 in the next few weeks anyway - already have it and am just trying to get the $$ for the 2 pc RMS flywheel and $$$$ for the clutch, shouldn't be as much of an issue then I hope).

JP86SS

iTrader: (1)

"F8" (4004, b1 IIRC) is FAN 2 so its best to leave that one open for use. It is also a communication lag device meaning you need to send it out the CS and then turn it on. This one should be faster reaction running directly off the CPU?
F5 was a good candidate to prove how to turn on/off the actual PWM outputs. Have to see if just putting the output Duty Cycle in 3FCC would do it.
Not 100% if there is another qualify bit on the actual output "enable", have to go back and look what was done.
Also seeing if the timer value is ok (need to see if the light stays on long enough when knock is flagged)
Might be too long/short of duration. Something to tweak as needed.

Z69

Vern
You have the beta bin.
I set up the code so the knock light will come on anytime rpm is above 3k.
I didn't think you should have to go out and make your engine knock just to test a light bulb.
It was also for bench testing purposes.
Just see if the pin went to gnd above 3k.
The timer should work though IIRC.
Might have to play with the setting.
I wasn't sure what the math was to set the timer up.

vernw

iTrader: (2)

Thanks, Scott. With my tuning abilities, getting a little knock is not hard at all. I also set the shift light to come on at 2400 RPM, just as a little reminder that you ought to be thinking about shifting now with today's gas prices

So the knock light will come on at 3K RPM, that will be easy to test. Will try the timer values as well. SInce this is such an easy wiring change (adding F5), I should be able to do that and test it tonight for you. I'm all for helping out however I can on this project! It's the least I can do!!!

junkcltr

iTrader: (1)

It is good to see new code tweaks. Is the 3K lower limit adjustable with the ECU file? If so, then set it lower to 0 RPM and do the hammer knock trick on the block by the knock sensor to verify the bulb installation.

Z69

No it's hard coded.
And it is only setup to look at rpm currently.
I wanted to verify we had the light correct first.
I used source so it was simple to comment out the Knock status check and add a cmp to rpm.
The bin vern is using has already gone well beyond the original
scope of the first beta test. I just wanted to verify the IAC code change worked and no SES in a fully functioning car.
I don't really have time to work on the code project at this time.
Or time to fully bench test it.
I just wanted to do a quick check of all the stuff that was done in the source. I was pleasantly suprised that no problems
have surfaced-so far
It seems all that's left is to verify the new stuff works as intended.

I wanted to do a staged shift light as suggested by CM.
But I ran out of time.
Vern is currently using the shift light as a CARS light.

vernw

iTrader: (2)

Looks like there's a little tweaking needed with the knock detected light. The LED comes on as soon as the key is turned on, and doesn't go off until the key is turned off. This is with it wired one wire to a key on 12v and the other wire to the F5 pin on the green connector. The light also does not go off if RPM is over 3000, I checked that just in case the logic got flipped and was backwards somehow.....

Also, the WB logging still doesn't work. I reset the LC-1's programming to use 1.0-2.0v as recommended by JP and the only output I'm getting stays between 100 and 105 for the most part. Same values are showing up in both TP and DataMaster (in the Min IAC field)

JP86SS

iTrader: (1)

Hmmmm...
Checking...Maybe I can spot something in there.

vernw

iTrader: (2)

THANKS! I appreciate it, don't want to go back to my AXYC based bin if I don't have to.....

JP86SS

iTrader: (1)

Think I found it.

Nope...
Still looking at it.

Z69

Missed a couple of *'s which aren't required in ashc11 which is what I orignally used for the WB patch.
Hopefully that was the problem with the wb.
Need to find time to go over the rpm calc again.
Might have missed something there too.

RBob

iTrader: (1)

This will also allow the indicator light to be modulated in brightness according to the level of knock. . .

RBob.

vernw

iTrader: (2)

Sounds great! Let me know when you have something I can do to help out on this....

JP86SS

iTrader: (1)

exactly !
1st step is just on with timer decrement. Next phase we could look at doing variable DC by increment decrement DC with possibly a first knock burst to get your attention.