11sORbust

Some of you know, I have an 87 formula with a 6.6l . I like to drag race it at the track. The torque is way too much for the 245 DRs to handle. IMO, it would be better to go WOT instantly and retard the timing just enough to prevent wheel spin (for the launch) VS gradually applying the throttle in order to prevent the hides from roasting. I think the factory knock routines could be utilized. This is what I have so far....

FYI, I run without a knock sensor. Since I don't use a KS, all the tables are zero'd out right now. First thing would be to feed the knock sensor circut with a signal that'll produce continual knock counts at the right rate. Second would be to apply knock retard only to the WOT tables. This will create the problem of timing retard past the launch.

My solution to the above problem would be to add a threshold to "disable WOT timing retard (based on mph)". I'm not fluent in ASM. Hoping someone could help me add that constant. Let me be clear, I want to do the work myself. Just need help on "how" to. This would be a great thread IF it works out! maybe add the constant to the super AUJP..

There is one final thing I would like to add. My car is a 700R4 and I'm not using the lockup feature. Maybe this could be simple as patching in some addresses.....maybe use something like the lockup prevention constant. Bottom line is the ECM will need to ignore pin GF9 after XXX rpm.

BMmonteSS

It would be neat if you could have a buffer "box" that would compare front wheel speed to rear wheel speed and then output a signal that would increase with the tire spin to feed the knock sensor pin. If you designed the box correctly you could make it completely adjustable and even have it shut off at a certain speed. Of course a better idea is to integrate this "box" into the programing and go from there. If done correctly you could even have this work with a knock sensor in place.

This isn't exactly the type of suggestion you were looking for, just something I've been thinking about lately.

11sORbust

My idea is much simpler. It's based on the fact that wheelspin from a dead stop(when instantly going WOT) is repeatable and consistant. So there is no need for "smart" traction control. Find out ho much retard the engine needs, set it and forget it. Less additional hardware, the better.

rooster433

You can probally spend way less time learning how to drive better.

In cases like my 550rwhp 6speed firebird smart traction controll can be very useful because I can manage to pull 1.91's on street tires all day long.. its not until I start ramping into that 15psi really hard until I start spinning (usually between the 60' and 100' mark)


one of my runs on 7psi last november was 1.91 8.10 @92mph on remington H-rated radials. You can see it likes to spin half track by the MPH.

Our 01 LS1 would have the same problem at this track.. I actually pulled a 1.87 on Kumho's and we recorded it on video.. its not until about the 100' mark until you can actually hear the tires wheelhop chirping. (not one chirp between gears but a bunch of little ones as the tire is loosing it)

RBob

iTrader: (1)

In this case it may be easier to use a WOT retard table. Have a table of SA retard values based on MPH. Use TPS% and maybe RPM to enable the table usage. When the qualifier(s) are met have the values subtracted from the current SA.

RBob.

11sORbust

I'm a good driver.




here is something I made really fast........

11sORbust

or if the TCC output is a ground then the relay ground turns into a 12v +..

rooster433

How are you going to generate a knock signal.. its not just looking for a ground or anything.. It needs a freq

Grumpy

Shhh, you're ruining the fun, when you tell them what works.

BTW, the above, and something just slightly more complex, do work. I thought someone would have noticed it on the last log I posted, but, nope.

Craig Moates

I was thinking about the other half of this equation. How about, assuming you get your wheelspin under control with a nice soft launch (you controlling it via gentle roll-out), you then let your ECM take control of the last part of the pass?

For instance, you go ahead and have a calculation block that uses your MPH and transition time from MPH=0 to 1 as the start time. You calculate how far you've traveled and how much time has elapsed. On down into the pass, say at the 1/8 mile point, you take another look at where you're at with respect to time and distance. You then start correcting by pulling timing out to hit a target time on the 1/4 mile point.

So this way, maybe your car is capable of like 10.5 in the 1/4 mile. Give yourself 3 seconds to roll into it, and allow yourself some room on the backend for slowing down. Dial in at maybe 12.5 or so, and let your ECM take you there in the right amount of time.

Just a thought. It might be tough to get it to work with the ECM alone, but I bet with a little PC-side help and an emulator/datalog in the mix, you could hit it spot-on or at least within 0.1 every time. Plus you won't be taking your machine to the limit every time.

Is this cheating?

poncho@home

"Is this cheating?"

Sounds like cheating to me, but it won't be until you caught. I've been knocking around ideas like that in my head for the last couple of years, unfortunately I don't have the electronics talent to pull it off.

11sORbust

Didn't see your post earlier. THat sounds good, how hard would it be to pull off?

On a sidenote, I was at a book store today looking for something on ASM. Found 3 books and don't think they could help me. It was all PC based info and seemed too general. For $40-$55 a pop, I didn't want to buy a book with information that couldn't be applied to the automotive ECM. I think the only way for me to learn it is working on little projects like this one. Just need someone to help!

Grumpy

No more so then ABS, or the computer controlling anything else, IMO.

Back when NASCAR was having problems policing them, and saying it couldn't be done with the existing inspections, just for grins, I made one. And I'm no EE. Just reading Forrest Mimms, and building all the lil silly DIY kits Radio Shack offers, can really be interesting. There's also an outfit called Rainbow Kits, that makes some DIY kits that are *fun*, to put together. While I can't begin to do what some other fellows can, it's still fun to try. BTW, the TC used just a couple simple chips, and was just slightly larger then 1/2 a #2 pencil.

Only trouble was that one night the temps went under freezing, and it went nuts. The car wouldn't go over 3 MPH, LOL. It was almost a long walk home that night.

Grumpy

For bracket racing, maybe.
But, I never understood the premise of 1/2 throttle racing.

Yes, just use a series of timers. Might just run it off the brake lights to make it easy..... Launch Control, then traction Control, then HP timer. Yep, it'd make a programming game out of the 1/2 throttle racing game......

rooster433

Does the 727 have a ASR built in?

The 93 corvette computer has it, not sure how exactly it works though.

dimented24x7

iTrader: (2)

I like RBobs idea. Just add initial retard in with the ammount controlled by the rate of throttle opening and have teh computer decay it out with a multiplier. Honduh seems to have a similar stratagy with their cars to stomp out the initial torque multiplication from the converter and it does a good job of killing any hint of off-idle throttle response. I can get around it with careful modulation of my right foot, however.

11sORbust

So, when you are "foot braking" at 3600rpms and launch, you can totally control the wheelspin? We are talking over 450ft-lbs of torque.

dimented24x7

iTrader: (2)

With careful control of the timing you should be able to. Since youll be powerbreaking go with the initial suggestion to use some sort of qualifier for rpms/tps to add in initial retard when you launch. The tricky part is when to add in the retard and how fast to remove it. Obviously you dont really want all the retard while your stalling the TC at the line. Maybe something like an initial ammount of retard added in past 3000 rpms at the line to provide the initial control and then additional retard added in on top based on the rate that the tps is increasing once you finally launch. After this, no additional retard is added and the current retard is decayed out as the car gets underway.

JP86SS

iTrader: (1)

i'm needed that too, can't even get a good launch unless theres 4 people in the car!
Maybe trigger the sequence with an external switch.
Just to give the command to insert the retard when staged.
Rev on the line, hit the switch indicating zero speed reference currently. Then upon a 1 MPH movement the decay can be started.
Make the decay rate adjustable and try it a few times.
that way the retard can be timed or distanced out (depending on what is easier to do)
Have to keep the impact minimal or invisible to the other routines I would think to make it effective.
Simpler is better.

JP86SS

iTrader: (1)

Doubled up on me

Grumpy

Maybe not all.......

RPM is you friend when your staging, it's one of em big word things, kinetic energy. So if you keep the rpm up without stalling the converter, guess what?.

Might actually want a launch control, and then traction control.

Just guessing.....

Craig Moates

Cool, maybe a built-in two-step sort of thing. That'd be slick! I like it the more I think about it.

Wish there was a way to tell the ECM to drop spark entirely or something, alternating, sort of like a 2-step.

rooster433

anyone know of a simple circut to compare pulses from the rear abs relcutor to the front?

I don't have a problem launching.. its half track thats doing me in.

JERRYWHO

Droping the spark will work better, Retarding the timing will not drop a lot of horsepower. I have only 1.4 degrees total timing at launch and then ramp all of the timing back in down track and still have lots of wheel spin.

This Datamaster run was 1.5 60Ft.

Jerry

11sORbust

Ahh, a T-56 car I think we are on the same page. My claim is that going wot out of the hole with timing retard would produce a faster car than a human that modulates the throttle blades. First off, the person is not going to apply the throttle exactly the same, every pass. Second thing is closing the throttle blades will reduce power MORE than keeping them 100% open with spark control. FYI, you have more than 1.4* of timing. Look at the value you are displaying. It's the reference spark advance. That is not your TOTAL advance number.




Bit of my pm to Z69, this is my latest idea....

11sORbust

Craig, there is electronic controlled throttle bodies on newer vettes. I'm fairly sure(can check) that it uses a seperate control box. So with your idea, one could hack the box and program the throttle openings. Then it would be a computer program that's controlling throttle application throught the run. I wouldn't use it for racing though, where's the fun in that? Besides, there will be a new rule banning your device in no time.

MrDude_1

if it helps, i recall reading that the the early C5s use a seperate controller... but the later ones and C6s, along with the GTO, use the PCM for throttle modulation.

dimented24x7

iTrader: (2)

Dropping and reintroducing the spark is hard on the motor, though. The SAE team at Rutgers where I went to college had that problem. One of my friends was taking the car around the track for yucks for a few minutes and he kept hitting the rev limiter. Good bye, motor. The added stress of having the spark come in and out on top of the rotating stresses caused it to puke a rod.

Although, maybe a soft control that pulls all the timing and then cuts the spark if the loss of timing alone didnt work would be better. It would do the oposite when coming back on.

Come to think of it, you could use one of the ecms quad drivers (with external hardware or something) to control power to the spark coil to cut the spark.

11sORbust

IMO, it's much better to drop some spark advance than to kill power to the coil. That's just asking for problems.

RBob

iTrader: (1)

Drop individual injectors?

RBob.

JPrevost

Different car/engine combos are going to respond differently to torque limiting routines.
The hardest torque limiter to the least;
All cylinder spark cut
All cylinder fuel cut
Bank spark cut (rather hard to implement)
Bank fuel cut (this depends on TBI single vs dual plane and TPI)
Individual spark cut
Individual fuel cut
Fuel lean
Spark retard

That is a generalized list, nothing set in stone so please don't nit pick. So where as spark retard would work well for a 13 second car it might not do enough for a 11 second car.
I'd stick with spark retard as a first step, it's the easiest to implement, then fuel lean for launch control. For those crazy traction monster cars a bank fuel cut might work really well for staged launching. I've personally tested this method of unhooking one whole bank and the sbc will run, rather smoothly I might add. So using a digital out into a relay you could effectively pull more than half the horsepower out.
I'm not sure if this is my idea or not but... using an accelerometer to get the true vehicle speed you could compare it to the vss signal speed and effectively retard timing based on the discrepancy . Talk about simple, that would only require a $30 accelerometer circuit and for you p4 guys, the extra analog input.

Grumpy

Hey now there's an idea.
But, that'd mean figuring out which GM ECM (OBDI) that offers SEFI.

11sORbust

The zr1 vette is a c4 with "sefi". Then you'll need a cam sensor to trigger each injector.

My friend uses spark retard to help control wheelspin. He programmed a 01 vetter to come out of the hole on a 300 shot of nitrous. 11.0 @ 127mph is what it ran on the stock goodyears (1.7 60ft). I don't know any driver that could control that. We are talking 500+HP and 600+tq instantly. On stock hides. based on that, I know spark control will work fine for my car.

I'm not really looking for input on the thread's tittle. Was hoping someone would teach me how to transform this from an idea to something that works, just by manipulating the code. any takers? I learn fast.......

RBob

iTrader: (1)

Something like this is a start:

Just stick it into the bin prior to the max retard check. This kinda' stuff is why NASCAR doesn't allow EFI. Was funny a few years back when a BIL was ranting about how much better carbs are because everyone in NASCAR ran them. "If fuel injection was better then everyone in NASCAR would be using it." Uh, FI is not allowed by the rules. . . LOL.

RBob.

Grumpy

Might need to look a little further. The 8331, and variants, are kinda rare.
There might even be a dual sensor distributor out there.

Z69

Still at the napkin stage and I know very little of the stock SA code, but was going to look at using the stock retard table that Tim pointed out to me. zero out stuff as needed. And a mph and a/c on qualifiers.
Or something similar.
Of course this only works w/o a k/s.

If you really want to DIY this.
The source code for dummies thread is really where you should start on this stuff. You'll end up there sooner or later....Once you start to see what you can do with src....
You can use this as your motivation to learn it.

Wonder what one or two DRP's of single fire would do?

JERRYWHO

No, It has a 700R4 with a stall converter. The first RPM drop is the car hooking up.

My base timing is set at 0 (tdc) so it is the real running timing ( Total less any knock retard)

Jerry

11sORbust

You are totally right. I need to be able to dis/assemble my bin before going any further. That is one thing I've been slacking on. Once that is done, then maybe I can move on to the launch control. YOur idea of the qualifiers seems really simple AND effective. The existing PE retard table is set up perfect......

11sORbust

Not sure if that would be the best. I'm thinking about power output vs wheelspin (in first gear). It seems that RPM would be a much better ..uhh reference than MPH when applying retard, on this app.

RPM/retard*
800/0
1600/0
2400/0
3200/6
4000/15
4800/7
5600/0
6400/0

Look at the above table. It already exist in the code. I programmed it to reduce torque output without killing total HP production. Just need the ecm to apply that table until a certain mph. What do you think?

EDIT:Then again, one could argue MPH would work better. Maybe controlling rear wheel torque output would be more effective?

Grumpy

It's Delta MPH.
I've already played both games, the VSS works.

11sORbust

That statement implies that timing retard was less effective. IF that is an acurate assessment, could you expand on why it didn't work so well?

Grumpy

????????????????
The options are retarding the timing, as a function of RPM, or VSS.

Traction Control by definition is about traction, which is linked to VSS more then RPM. Clutch slippage, and converter slippage differ at times, and are less repeatible then tire slippage.

Oh, and BTW, here's what T/C just might look like based on VSS

It was an early data log, but you can clearly see the timing change as it goes into T/C and out of T/C mode.

Mph Rpm KPa Tps CtsF MatF Spkf
2 2850 152 100 152 57 21.5
4 2925 158 100 152 57 21.5
4 3025 170 100 152 57 21.1
5 3150 173 95 152 57 21.5
7 3225 184 82 152 57 21.1
7 3375 194 76 152 56 21.5
****T/C starts
9 3450 203 76 152 56 19.0
11 3575 211 76 152 56 12.0
Mph Rpm KPa Tps CtsF MatF Spkf
11 3650 215 69 152 56 9.9
14 3725 223 68 152 56 12.3
16 3950 224 66 152 56 12.3
18 4050 229 64 152 55 16.9
21 4150 223 53 152 55 16.9
23 4050 222 53 152 55 19.0
26 4025 222 53 152 55 19.0
26 4075 219 50 152 54 19.0
31 4100 217 49 152 54 20.1
35 4100 211 45 152 54 20.4
35 4100 209 45 152 53 20.4
38 4100 206 45 152 53 21.1
**** T/C ends
40 4100 205 45 152 53 21.5

11sORbust

I typed the wrong words, you answered my question though.

I don't know what I was thinking. What Rbob posted was really good. Guess I need to learn more about ASM before trying out the LC idea.

11sORbust

could you explain in detail what everything means just below the table? I understand about 30% of it.....

Thanks alot and I'll understand if you don't have time to go into detail.

-Tim

rooster433

has anyone thought about doing somethign with the knock sensor at the rear axle that would make it go nuts as the rear vibrates (wheel hops)

Just looking for a easier way of doing this.. and also one that would help us that spin at 40-50mph

Grumpy

Yes, it's more complex, ie in the sensoring end of things. But probably more later this spring.

rooster433

Grumpy,

I looked into the 92 vette service manuals I have (I own a 93LT1 so it seemed appropiate when researchign factory ASR stuff)

There seems to be a B+ signal sent to a pinout on the ECM that retards timing as a rear-wheel torque management.

I also noticed that when the wheels are spining the factory abs reluctor wheels are creating a AC voltage (from 100mv up)

It would be simple to have a simple circut that compares front and rear AC voltage then if V_front < V_rear than apply voltage on the ASR signal to ECM.

Does any electronics guru's know of any simple circuts that compare AC sinual voltage?

3.8TransAM

Hmmm possible application to use a yaw sensor or the like for rearend control since the oscialltions from wheel hop tend to be quite violent?

That paired with some kind of actual wheel speed reference would seem to be the easiest way out.

Would be nice for say a 383 tpi app with a converter and tons of torque.

Or is this something u whipped up already Bruce?

later
Jeremy

Grumpy

I'm sure there's an easier way.

It's the sensors, that are challenging for hanging on a non-ABS car, IMO, that's challenging.

Grumpy

Yes, and no.

I want to do something a little more sophisticated. IMO, it's about the simpliest setup that gets the job done, that's most reliable.

Figuring out what hardware for Stage II, is what's taken so much time. It's just a matter of finding/ installing the misc bits, now.