This started as a project about 2 and a half years ago when I started drag racing motorcycles with a friend who has a bike shop. The original project started as how to make things more efficient, easier on the motorcycle and of course more constistant. Of the bikes we race the fastest when pushed to the edge has been 7.92 seconds at 181mph if I remember the mph correctly, it consistantly runs 8.05 to 8.15 seconds in the 170's so obviously safety is paramount. Now of course this is the disclaimer that has to come with posting a project like this. Racing, Air Shifters and of course Nitrous Oxide can be very dangerous even in a well maintaned and planned system. I highly recommend that only individuals with full understanding and experience work on or operate either of these systems and claim no liability for there use. With that said, on with the project.


A short explanation of the systems in question would probably make the whole thing more understandable at this point. First the Air shifter portion of the project. An air shifter is a pneumaticaly operated shifter almost excusively used in drag racing. When the rider presses a button at any throttle position the air shifter system activates a pneumatic cylinder that pulls the shift lever and kills the motorcycles ignition at the same time. During this the forward momentum of the bike unloads the transmission and allows it to change gears even at wide open throttle without the use of the clutch. A well sequenced systems adds virtually no wear or damage to the transmission. In short, it allows gear changes at wide open throttle without the use of a clutch which in turn results in 1-4 tenths of a second on average faster 1/4 mile times. The orignal air shifter modification was to add a pressure switch to the line feeding the cylinder and activate the ignition kill from it. This resulted in not losing time with a dead ignition while the shifter built pressure and started is motion since the ignition didnt kill until the shifter was in position to make a gear change, sequencing the shifter operation like this resulted in approx a tenth faster 1/4 mile by shortening ignition kill time as much as 30 milliseconds. 30 x 4 gear changes = 120 milliseconds or 1.2 tenths. These modifications lead me into building the ignition kill unit myself since the current retail units had a minimum time to kill the ignition that we were below on a well adjusted system. The original unit was analog using a 555 timer circuit extrmely high component count and good but not excellent reliability.



The next hurdle we encountered was with Nitrous Oxide systems. I'll keep this one short, beyond a certain amount of horsepower the general rule of thumb is that you have to dissasemble the transmission and make certain modifications to allow higher horsepower applications to make the gear change. (Having the shift drum cut at minimum). In the original systems that we were using there was a complex series of relays and another 555 timer circuit tying the nitrous oxide system and the shifter system together to cut the nitrous system an adjustable amount of time in milliseconds before the actual shift was made. This allowed us to make consistant gear changes at higher levels of horsepower without the need for transmission modifications.



All of this of course was a very complex system of relays and analog circuits that was tedious to assemble and just about as tedious to install and adjust. Here in this long story is where Parralax comes in. During the early spring of this year I discovered Parallax and the BS2 OEM board, at this point with a little analog circuitry understanding and little else. However I knew enough to realize that this could make things far easier, less complex and more reliable. I purchased a couple, spent several months tinkering and eventualy converted the air shifter over to operate from the BS2, the next step was to eliminate the relays and timer tying it to the nitrous system which was very simple at the level I was at by the time I'd gotten through most of the WAM kit.



The end result was that now I've developed and succesfuly tested an SX18 based system that uses an SX18, 3 fets, and some minor support components that does all of this extremely reliably, and consistantly. The final circuit is installed on a 2004 Kawasaki ZX10 that with very little modification has run as fast as 8.71@164 mph on a totaly stock engine and transmission, and runs consistantly in the 8.70's. To most people this may not mean much but trust me, for what it is, its extremely fast.



The key features using Parallax components allowed me with the BS2 and later with the SX were, far lower cost $30 relative to $400. Simplicity, one unit rather than going with the standard 2 pieces, and with both controls being integrated it allowed me control of the NOS system relative to what the shifter was doing without interfacing the two together. Most important of all, consistancy, for my own use the BS2 and the SX have both been extremely reliable and consistant. The BS2 system that was created at the beginning of summer while less advanced is still running on another bike, slated to be replaced this winter.



My first and biggest thank you goes to Parallax for the products and support, and of course also to all the members of these forums. While I wasnt posting over the summer I was reading and many of the projects and discussions in these forums answered the questions I had and taught me what I needed to know to make the control portion of this project succesful. The next stages are to install it on the rest of the race bikes that we run from Daves shop (currently versions are running on two) and wait for spring. Needless to say, thanks goes out to Dave for having faith in the work I was doing, and to Eric for having enough faith turn down large drag bike teams while the system was being developed on his bike. Neither of who will probably ever read this but it had to be said anyway.



The schematic is attached and the SX/B source code will be posted tonight after Ive had a chance to go over the new code one more time. I dont care who builds it, uses it, I just dont want it manufactured, without going into details the nitrous oxide control is progressive and this project replaces about $400.00 worth of retail components and has several large advantages. Side note, portions of the schematic are for safety reasons, like the activation inputs also supplying solenoid power etc. Again, racing is dangerous and parts fail, this setup has worked well for me but I can take no responsibility for its use, modfication or failure.



Thanks again everyone

Ok, there's the code, sloppy and horrible but it works. As I get time I'll clean it up

Chris

Post Edited (ChrisP) : 6/30/2007 3:13:37 AM GMT

Image Attachment : ECS12.jpeg   212KB (image/pjpeg) This image has been viewed 1397 time(s).

File Attachment :
SX Progressive 2.SXB   4KB (text/plain)

 

 

 

Chris

Image Attachment : ZX10Control.jpg   33KB (image/pjpeg) This image has been viewed 1223 time(s).

Image Attachment : ZX10.jpg   83KB (image/pjpeg) This image has been viewed 955 time(s).

Image Attachment : Control.jpg   53KB (image/pjpeg) This image has been viewed 1268 time(s).

Remember...

Provas, GReece

---

 

 

 

 

Chris

 

2 weeks, 10 weeks, such is the life of projects..... Version 2.0 SX28 Circuit board, programing, schematics, and code. Its a drastic improvement but its not quite there yet. There are provisions for a 20mhz oscilator which were not implemented in the hardware at this time.

 

A few things that are still being researched are the relationship to solenoid duty cycle and horsepower, which looks like it may be nonlinear or at least not a 1:1 ratio, and better ways to put horsepower down to the road.

So far testing is showing

1. 100% duty cycle may not be the best or most stable for large horsepower increases (there are program changes under testing for that to lock at 80-90% instead of 100) 80% appears to be a good number for now.
2. pulse rate has an effect on horsepower and how smooth the curve is. (currently testing at 15 and 20 hz)
3. linear horsepower isnt the best way to get a bike down the track at large horsepower increases. So were looking into a curve, and/or ramp rate change at 60-%70 duty cycle.

 

Results from first race Valdosta Georgia were promising, progressive/ramp and shift functions worked perfectly but at 60-90 horsepower increases a linear ramp proved to be nearly uncontrollable.

 

Conclusions:
1. Current testing indicates that it functions as originaly designed and works well at lower horsepower levels
2. Meets our goals of drasticaly cutting racing/performance costs while maintaining reliability.
3. Function/software design  requires changes to meet our goals of large overall performance increases
4. At higher horsepower levels a single linear ramp rate seems to result in either less performance than desired, or an almost uncontrolable pass down the track with no happy medium.

 

Current goals:
1. Quantify the ratio of nitrous solenoid duty cycle to horsepower increase.
2. Quanitfy the ratio of time:horsepower increase that is useable in a/our racing application
3. Possibly change program to achieve a dual ramp system where one ramp leades into the other for more precise control to gain the maximum value from larger increases in horsepower, or currently a more specialized ramp rate/curve for our application.

 

Current function:
1. Complete sequencing of air shifter
2. 20ms delay between shifter activation and engine kill to allow system pressure buildup and decrease engine off time.
3. Interaction between nitrous and shifter systems allowing automatic nitrous shut off during gear changes.
4. Precise engine kill time control

 

It looks like I was way ambitious in that 2 week estimate but its working well, and cleaned up quite a bit. Thanks to Geunther and PJ for the tip over in the SX forums about the ramp over time increase.

 

 

And Steve is right... some videos won't hurt....

---

 

Nitrous solnoids are only designed to be turned on/off not pulsed like I think you are trying to do. Their are special Solnoids/Valves called "Dithering Valves" ( I hope i spelled that right) that can be plulsed at a higher frequency.

 

Basically you are trying to build a traction control system, and it will need to know the relationship between the tire speed and ground speed. And the only way to do that is a wheel speed sensor and some way to measure ground speed. (Their are ultrasonic ground  speed sensor out there but i neve priced them) Since you are building a system from scrach consider puting some data logging capabilities on it. The additional information will help you tune the bike better.

 

MY race pak system has.

Data logging.

4 thermocouple inputs for EGT.

input shaft and output shaft sensors for clutch slipage.

RPM control for shifting and NOS activation. 

Ignition kill.

Oil Pressure and water temp.

 

P.S. Any form of traction control is illegal in any form of class racing.

 

 

 

 

 

 

 

Doc Evil, We've been racing the Pro Star for several years now so I'm fully aware that traction control is illegal, however I'm not trying to control traction I'm controling the amount of power output to keep it manageable on a 1/4 mile pass.

 

As far as not pulsing nitrous solenoids.... Well you may want to talk to NOS (Nitrous Oxide Systems), Schnitz racing, NX (Next Generation Nitrous systems), Edlebrock, and most of the racers running serious nitrous power. However as far as solenoid performance, NOS is at the bottom of the list, an NX works much better in progressive systems, and the new NX is supposed to be far superior to anything else out there. Its actually the fuel solenoids that give you more problems.

 

None of that was meant as a rant, I truly appreciate your attempt to enlighten me on nitrous use but it sounds like your not that familiar with high horsepower street bike setups. Your data systems sounds awsome.

 

Chris

 

P.S. We've already sprayed a busa into the low 8's high 7's a couple of years ago. This is an attempt to do something new and different without spending a fortune. The busa owner had the 3k to drop in a nitrous system with off the shelf parts.

 

 

 

 

The shift is sequenced differently than your average commerial setup allowing shorter shift times.

 

The shifter and nitrous interact, yes the nitrous is cut completely during shift, it only dithers as it initially comes on. Bikes with nitrous and shifters generally dont, they just shift through, dead ignition is a dead ignition. You usually get a pretty interesting bang through the exhaust though when the ignition comes back in.

 

Chris

 

I've found the downloads for the SX-Key/Blitz Development System Manual and am reading that now.  I've downloaded the Beginning Asembly Language for the SX Microcontrollers.  Any other recommendations?

 

 

Post Edited (Chris_W) : 11/14/2007 7:26:45 PM GMT

 

Personaly, I would probably suggest looking into a solid state method of pulling the coils down for a 2 step, timing is fairly important and most relays capable are in the 3-9 millisecond closure range, also add a cap in there somewhere or the relays releasing the ground is actually going to fire the coils when they let go.......