Crank / Cam Signal Emulator

kidturbo

kidturbo

Piston Tester
Jul 21, 2010
2,332
1,103
113
Somewhere On The Ohio
www.marinemods.us
Built this cool little Arduino project I found online a while back to generate E38 ECU data.

https://www.facebook.com/ardustim/

It basically allows you to generate a matching 5v crank and cam PWM signal of any common wheel types. It has a couple 60-2 crank wheel options like modern LS built in, but didn't have the correct Duramax cam wheel. Which isn't a common to any other GM.

So while my Internet was down last night, I decided to tackle the cam pulse pattern using a signal capture from an eninge on Marks dyno. After a couple hours, couple drinks, and some major trial and error, I was finally able to make a stock LMM ecm think it was actual controlling an engine.

Nothing to crazy yet. Didn't have a pots on fuel pressure input, no resistor in place of injectors, so it was in major limp mode with 20+ codes. But little sucker was still computing fueling and timing according to EFIlive. Which opens up some possible uses in bench tuning and testing.

Anyways handy $25 tool, or $35 if you buy a nice case to go with..
Source code available below. And I have the def file with good Dmax patterns.

https://gitlab.com/libreems-suite/ardu-stim

Cheers,

-K
 

Attachments

  • Dmax-Stem-Test-1.jpg
    Dmax-Stem-Test-1.jpg
    127.3 KB · Views: 62
  • Dmax-Pattern.jpg
    Dmax-Pattern.jpg
    50.9 KB · Views: 59
  • 20190225_042215.jpg
    20190225_042215.jpg
    207.5 KB · Views: 63
  • 20190225_232508.jpg
    20190225_232508.jpg
    137.4 KB · Views: 63
Last edited:
2004LB7

2004LB7

Well-known member
Dec 15, 2010
6,394
1,755
113
Norcal
That's pretty sweet.

In the past with my LB7, I've found that a resistor in place of an injector did not trick the ecm. Maybe I never got the value correct but I was successful with an inductor. I think a resistor doesn't have the proper back emf or inductance and the ecm/fcim could detect this. A bunch of trial and error until I got one to work. Drove around with one injector turned off / bypassed for about two weeks until the new one showed up. Kept that inductor and a fuel line cap in the glove compartment for roadside emergencies
 
2004LB7

2004LB7

Well-known member
Dec 15, 2010
6,394
1,755
113
Norcal
PS: once you get the ecm out of limp mode it would be interesting to see how fast the "engine" revs up

I also bet that ecm has never seen such a clean cam and crank signal. Must be confused :rofl:
 
DAVe3283

DAVe3283

Heavy & Slow
Sep 3, 2009
3,682
236
63
Boise, ID, USA
2004LB7 said:
PS: once you get the ecm out of limp mode it would be interesting to see how fast the "engine" revs up

I also bet that ecm has never seen such a clean cam and crank signal. Must be confused :rofl:
Click to expand...
I don't think there is any internal restriction on how fast the ECU will allow an engine to rev. I know there isn't on the LB7. I have a log file from my brother's old truck where I got on it, it lost traction and picked up RPM so fast it spun the balancer on the crank. It picked up 2000+ RPM in like 1/4 second.

I remember a thread a while back (not sure it was this site or not) where Mr. Jack Stand Racer was complaining about ECU limitations on increasing RPMs, but could never provide any data to back it up. I can't find that thread at the moment, but I stand by what I said then: I don't believe there is any ECU limitation on RPM increase. It is all the physics of the motor, drivetrain, etc.

K, what I'd like to see is how the ECU responds to intentionally bad crank/cam data. If the cam signal goes out of sync with the crank, how far off does it have to be before the ECU sets a code? A couple degrees? A whole crank pulse? Multiple crank pulses?

How does it handle missed pulses? Do they get logged correctly on the scan tool? (I have some suspicions they do not get properly logged.) What about really noisy signals? (I suspect that your mis-adventure with the previous engine harness on the twins with twins might have led you down this road to begin with...)
 
kidturbo

kidturbo

Piston Tester
Jul 21, 2010
2,332
1,103
113
Somewhere On The Ohio
www.marinemods.us
It's been said that they will run with no cam signal. I known this to be true, cause I first tested this tool with a 4x cam pattern and it registered rpm signal. I'll get it spinning then pull the cam wire and we'll see.

What happens with interference, we found it will just cut fuel till corrected... On the dyno we were picking up injector 2 pulse on crank signal wite, and it just stop firing for a rotation until it got a clean signal again. Then when it fired, interference returned. Pretty nasty sounding loaded..

As for injector resistance, good info. Since Andrew is reworking my injectors, I might see if he can spare 8 old solenoids for this bench setup. That would be ultimate way to do it.

Later tonight I'll hook up my tuned ecm and see if can crank it 10k. Be interesting to see how it handles a 2700us pulse up there.

Sent from my SAMSUNG-SM-G890A using Tapatalk
 
2004LB7

2004LB7

Well-known member
Dec 15, 2010
6,394
1,755
113
Norcal
The one that worked for me was out of a dell server power supply. It was about a one inch od toroidal inductor with maybe 50 wraps of 20awg wire

Solenoids don't like to be run very long without a plunger or iron core inserted for very long as they tend to run hot. Not enough back EMF to keep the current in check. Dont recall if the injector solenoids have a plunger in the center or if it is filled/non-removable

For the cam (maybe crank?), I thought if it was lost it would continue to run but if you shut it off then it wouldn't restart until the signal was restored.

Looking forward to the 10k rpm test. Wondering if the ecm can handle the injection overlap
 
kidturbo

kidturbo

Piston Tester
Jul 21, 2010
2,332
1,103
113
Somewhere On The Ohio
www.marinemods.us
Ok 12888R seems to be the max RPM signal it can calculate.

However when I enter a number above 12800, ecm locks up and requires a hard reboot regain control. It kept logging 12888 even after I disconnected the emulator. Key reset didn't even clear it.. So now we have a new rpm range to shoot for . Lol

On disconnecting a crank or cam signal, it will continue to read rpm change with either sensor missing. But only continues fueling when cam is disconnected. Doesn't seem to miss a beat either.. Needs more research.

fa83215e03085c1aaff56c83ef4cea53.jpg


Sent from my SAMSUNG-SM-G890A using Tapatalk
 
2004LB7

2004LB7

Well-known member
Dec 15, 2010
6,394
1,755
113
Norcal
Does the ecm "think" the engine is just coasting? Or is it injecting fuel still? At what RPM does it completely cut fuel? What about the tune? Is it full fuel all the way out to max RPM?

Have you run any cax files to gain more control over 4k RPMs?
 
kidturbo

kidturbo

Piston Tester
Jul 21, 2010
2,332
1,103
113
Somewhere On The Ohio
www.marinemods.us
I haven't faked the fuel pressure sensor yet, so it's max at 29000. Can't get any decent readings yet on related fueling. Only pilot on/off switch and such so far.
But if ya don't give it some throttle, it cuts fuel off at 850 rpm. How I know it's alive..


Sent from my SAMSUNG-SM-G890A using Tapatalk
 
kidturbo

kidturbo

Piston Tester
Jul 21, 2010
2,332
1,103
113
Somewhere On The Ohio
www.marinemods.us
Ok switched to another ECM with one of the twins "Speed Density" tunes.

Check out how FPDes, MainIQ, MainPW and estimated load all rise in direct correlation to MAP2 aka Boost Pres. I only have a pedal, some pots, and the emulator connected. The other ECM was a MAF tune and I couldn't get it to go over 60mm3 on MainIQ. Likely cause was limped.

So first I peg throttle, set RPM to 3k, FP 24k, then upped MAP. When I added some boost, she came alive across the board. Even though RPM and throttle never changes. :woott:
 

Attachments

  • Ardu-stem-dmax-3000r.jpg
    Ardu-stem-dmax-3000r.jpg
    183.3 KB · Views: 30
2004LB7

2004LB7

Well-known member
Dec 15, 2010
6,394
1,755
113
Norcal
You need to build this into some kind of console chair thingy so you can sit and drive it all while watching the scan tool log it :woott:
 
kidturbo

kidturbo

Piston Tester
Jul 21, 2010
2,332
1,103
113
Somewhere On The Ohio
www.marinemods.us
2004LB7 said:
You need to build this into some kind of console chair thingy so you can sit and drive it all while watching the scan tool log it :woott:
Click to expand...
Got a throttle pedal and some canbus gauges hooked up on my hot rod chair racer...

Played with this emulated ecm for 4 more hours last night, scan tool linked to tune file. Watched table by table across range of input value changes.

IMO, this is a terrible tool, worthless and nobody should ever use it. . .
3971f4d5e466776e19965c6801b195c9.jpg


Sent from my SAMSUNG-SM-G890A using Tapatalk
 
You must log in or register to reply here.
Top Bottom