Ok, here is what I was thinking, For tuning my car I thought, Gee, it would be nce to have a dyno. Then I thought hey, I have a G-Tech, unfortunately it does not have any way to output information so it can be recorded.

Then I thought What would be REALLY KEWL, would be a G-TECH & a program like DIACOM or some other Engine analysis program, maybe Craig Moates software, that would take the information and give you a relationship between G's and RPM, MPH, etc, etc, etc.

Thoughts???

------------------
87 Formula Yellow/Black
Engine & Transmission
.040 over 5.0 converted to TPI, 9.5:1 Compression, SLP Cam Dur 206/212, Lift 480/487, Cent Line 112, SLP headers, SLP cat-back exhaust, K&N Airfilter, Modified Stock cold air intake, Ported & polished Stock TPI intake, Holley Adjustable Fuel Press regulator, AC Rapidfire Spark Plugs.
Gil Younger (no yo-yo) Shift Kit, Aluminum Driveshaft, 3.73 Posi gear
Suspension & Brakes:
Baer 12" brakes 4 wheels, KYB AGX Adjustable Shocks & struts, Hotchkis strut tower brace, South side machine frame connectors, Custom reinforced control arms and pan hard bar.
Interior:
Custom Leather interior, CUSTOM Yellow Gauge Faces, Kenwood Receiver, Pioneer Speakers, Kicker Substations, Kenwood 10 Disk Changer, Pioneer Amp.

87 Formula TPI (5 Speed) Yellow/Gray STOCK

1967 Buick Riviera 430hp Turns high 14's (Not bad for 4300 lbs)

83 Camaro (Parts)
83 T/A (parts)

http://www.3rdgenformula.com

 

You mean like this?

http://www.charm.net/~mchaney/homedyno/dynokit.htm

A performance measurement tool is essential for WOT testing and tuning, and it seems to me the home dyno would fit the bill.

 

That's truly slick, I hadn't seen it before. Very elegant!

 

So, has anyone actually tried it????



------------------
86 406

 

I've tried the older version and couldnt get it to work.

 

Direct Scan does exactly that, but it's 148 specific. What surprises me is no one has looked at improving lockers, and use the edge board connector on the emc's PCB.

 

I actually developed an algorithym in an Excel graph that calculated your theoretical HP based on the gas consumed via the injector pulse widths. All I had to do was do a WOT run and then graph the results based on the amount of gas consumed.

It obtained reasonable results, when compared to the "HP 1/4 trap speed calculation" but the formula is only accurate IF you are running the correct mixture (ie 128/128). If you are running too rich, your injector pulse widths will be too long and overstate your HP.

 

DataMaster, gives you an ET estimator (& HP & a bunch more, but have not used them) & the last time I was @ the track I compared time slips to the datamaster estimate. In all the runs the gains/losses were matching from the time slip to the data master run, but the absolute times were not all spot on. I had only one this past trip to the track that was way off (.5 secons) but when I went into the manual mode for the ET estimator it came out just as accurate as the rest of the runs. I think that it could very easily be used for tuning & done with accuracy.

BW

------------------
Bobalos
aka Bob W.
www.r71camaro.homestead.com
r71chevy@earthlink.net
<><

 

I'm surprised that noone has attempted to reverse-engineer the Directscan module to see how he's doing it. Surely the same concepts could be applied to the 747, 165, 730, or whatever!

 

Well I know of the formula to figure your average HP & Torque. What would be interesting about what I am proposing is you could get a relationship between HP/Torque and RPM, therefore you could really tune that sucker and see where you have the most drop off.

What would be even better is the fact you would not need to go to the full speed 1/4 mile. Let it run thru 1st & second to get an idea where you need to tune.

 

You guys should take a look at this.
I have the AP-22 and it works great. It allows you to store data and upload it into your computer where you can graph it. It seems to give very consistant numbers.
It easily shows the differences between 1000 ft altitude changes and hot vs cold intakes. Back to back runs show consistancy in the 0-2 hp range.

http://www.race-technology.com/WebPa...eterIntro.html

 

You don't have to reverse engineer anything, just look at Lockers, it's at the DIY_EFI.

 

The software end of this sounds like a candidate for my new software revision. Easily done, should let you track (and log to disk) relative horsepower/torque as a function of time (ie. RPM, etc). It would have to have some level of time-averaging, so there would be a little fudge in there, but as stated it could be used for tuning purposes. Look forward to it.

 

Craig, (I am not a programmer, so go easy on me) I have done some, so I know it is a REAL PAIN sometimes, not always. Would you be able to graph the information so we could get a visual Idea of how it looks? and where to modify variables?

Just a thought

 

I used that Home Dyno program and it truely didn't work. Most people have trouble with noise. The program can't deal with noise AT ALL. I worked on the cabling and built filters that eliminated the noise. On their web site you will find several listings for changing resistors and adding capacitors and other ineffective junk. After eliminating the noise I found the program still would only accept 1 out of 10 recordings. When it did the horsepower/torque looked right until about 2500RPM. At that point it would jump to about 5000HP and drop off to 0HP at about 4000RPM. This is with the newest version of the software (which ain't that new). In analyzing it I found the problem was that his ignition pulses were unrealistic compared to most ignition systems. All the ignition systems I played with has a strong oscillation in the spark pulse. His gently ramps up and then down. As a result his software goes insane and figures four or five ignition pulses rapidly and then a long period of no ignition pulses. This is why it reports extremely high horsepower. A low pass filter gave me believable results to about 2500RPM but still was ineffective. His algorythms need serious work, but its just $40 down the crapper. The idea is sound, especially on non-traction limited cars, but this program should be made freeware with the source code available.

I thought I might add that his general method should be applicable to a program monitoring the fuel injection computer. All that it is picking up is the RPM, which should be available inside of the ECM.

[This message has been edited by Grand Prix (edited July 02, 2001).]

 

I'm confused. Why not just use the Diacom readings directly?

I been using the Diacom RPM graph (accelerating uphill in 2nd gear on the same stretch of highway entrance ramp) to time each RPM band from 2000 to 5500 or so. Run-to-run it gives a direct comparison for evaluating modifications, and if you want to work out the math you can use the info to derive horsepower numbers for acceleration on the street (although it only has resolution down to about 0.1 seconds as I see it).

C. Jacobs' book "High Performance Ignition Systems" has a section on the math involved in performing such evaluations - that's where I got the idea.

Is there some inherent inacuracy in Diacom's timing that I haven't heard about?

DrJ

 

You are heading in the right direction. .1 seconds would be a little rough and might lead to inaccuracies. But using math you should be able to create a program that works fairly well. The problem with this method is it assumes the engine is loaded by the car. A loss of traction is interpreted as an increase in acceleration.

 

Here's another option;

http://www.geocities.com/MotorCity/D...treetDyno.html

It's freeware, and doesn't use the inductive pickup. Might not work for you guys using coil-packs, but it will for us with distributors. I already have schematics for using it with a V8/HEI setup (I have the hardware built, just need to install it.)

I also bought Chaney's Home Dyno. I never could get it to work either. I gave up on it and about a month ago, a friend tried the above program with some hardware he built.

Ken

 

It looks like it would function as a dyno. What I want is a program that would give me ALL of my variables including horsepower, torque, RPM, Speed, Spark in °, TPS, etc. Like a DIACOM with a dyno, Therefore I would be able to decifer exactly where I need to improve my programming.

 

Wow, ask and ye shall recieve, I guess. Really makes me want to convert my 4.3L TBI Bravada to 747!

Is this Locker board really as well-kept a secret as it seems? Everyone should be talking about this!

 

Lockers is at these links:

ftp://ftp.diy-efi.org/incoming/Lockers0.zip

ftp://ftp.diy-efi.org/incoming/Lockers1.zip

 

Just checked 'em out.. Pretty good stuff. It surprises me that anything this powerful isn't getting talked about here or on the GMECM list..

 

Yes lockers certainly looks fantastic. The only thing is, it requires more than avg electronics knowledge IMHO. I don't think my burner will program a PIC and I don't have an oscilliscope to sync what ever it is that needs syncing.