Fiat x1/9 + bike engine challenge build

I had to work 3:30a to 3:30p today (not normal, I had a software implementation go live today). But none of my users are working on sundays so any issues will start up for me tomorrow.

So, I designed (well, stole the design from some super smart person who invented the 'monostable multivibrator') a circuit to make a quickshifter for the bike transmission in the fiat challenge car. Here it is:

The idea is with bike transmissions that they can be shifted up (2 to 3, for example) without the clutch, you just need to release the torque on the trans for a moment. This is usually done by rocking your throttle hand while applying modest pressure to the shift lever. With full throttle torque on the trans, it will not shift, but as soon as you release the throttle just a bit and free up the teeth, the trans will snap into the next gear. Yes, many people are quite good at this.

But, you can make the process repeatable for even a gorilla driver like me by making a quickshifter that kills the engine for a very short amount of time (like 50 milliseconds). If you trigger the kill of the engine with the moving of the shift lever, you now have a system that will auto-upshift for you and you can keep the throttle matted. Basically you floor the throttle, and when you want to shift you start pushing on the lever. As the lever moves (right before it gets to the point where it starts moving the shift forks), it hits the microswitch that initiates the monostable multivibrator circuit that cuts the engine for 50 or so milliseconds, just enough time for the shifter to find the next gear with the unloaded transmission.

How am I going to kill the motor you ask? Well, first I'm going to try using the kickstand switch - the switch on bikes that kills the motor if you put it in gear with the kickstand down. If that doesn't work I'll think of something else.

This is a simple circuit and I could likely buy all the parts for less than $5, but for the challenge it would be cooler if I can scavenge this stuff from an old broken VCR or something. So now I'm off to find some broken junk in the trash!

p.s. if you are building or want to learn about circuits, falstad.com/circuit is about the coolest site known to man. Build, test, and watch your circuits function. It's an entire undergrad degree packaged in one simple java page.

(also posted this in the 'how did you use your day off thread' on the main page but didn't want to lose it from the build thread)

Just have it kill the spark, leave the fuel running- works great, shoots a fireball every time you shift.

¯\_(ツ)_/¯ said:

Just have it kill the spark, leave the fuel running- works great, shoots a fireball every time you shift.

Totally in for fireballs.  Also those things will be on their way soon.  

I went to a makerspace on Tuesday night since I am in NYC for work. They we're having an open house and had everything I needed including tools and such.

Unfortunately I couldn't get my circuit to work just yet, it was a combination of me being rusty at circuit assembly and testing and the fact that I had no idea where anything was in a shop where organization is clearly frowned upon.

I'm on the airplane headed home so this quickshifter will probably wait until the next time I'm here in NYC and not able to work on the major effort of getting the car to run and drive.

That said, instead of using the kickstand switch input, should I have the relay just kill the ground to the coils? 

In reply to Robbie :

I'd probably do something like that, yes.  Worked fine on high strung FSAE engines, as far as I can remember- that way spark comes right back, timed like it's supposed to, afterwards, rather than the timing/fuel/whatever else shutting off and having to find its' place again.  It might not work very well at really low rpm, but that's not where you intend to be flat shifting, is it?  

In reply to ¯\_(ツ)_/¯ :

Nope! It would also be sweet to integrate an rpm input, so then you could just lean on the shifter and the engine would automatically shift up at the desired rpm. That might even be better than a lever driven switch for drags.

I got back to the makerspace Tuesday night, and got my circuit working!

When you push the button, the LED stays on for anywhere from .05ish to 1.5ish seconds. This is adjustable with the potentiometer.

Unfortunately, the circuit does not filter out a long single input, which I will need it to do. Currently, if you hold the button down longer than the LED on cycle, the circuit will just reset and trigger again, meaning the LED stays on as long as you hold the button. This is bad, because it means the motor would be off for the whole shift. I want a 50-100 ms off time no matter how long the button is pressed.

I did find some devices that are basically very low content arduinos (attiny) mounted to a board with a voltage regulator and usb input. For a whopping $2.48. This circuit will be simpler to program than to make with analog discrete components, so I ordered a few to play around with. The attiny board means I could do stuff like rpm-based shifting and stuff, and it gives easier control of engine kill time (just type the number of ms you want the time to be, rather than trying to mix and match resistors and capacitors to get the right rc time).

On the car itself I haven't done much (I got home from Thanksgiving trip and had to do a ton of yardwork to prep the leaves before the first winter snow), but I did look at a couple gas tank locations. I think both the frunk and right above the motor (basically the stock motorcycle location) would work.

Frunk is better for weight distribution, but means I make my own fuel lines and extend wiring harness. Above motor means less dicking with lines, but weight higher up. Thoughts?

Could you please use an actual motorcycle gas tank? 

Monostable multivibrator

Use the stock gas tank and location. Two gallons of fuel is roughly 16 pounds. Plenty for a couple of auto cross runs.

Focus your time and effort on getting the car up and running for some testing time. Then you can relocate the fuel tank later.

Andy Neuman said:

Could you please use an actual motorcycle gas tank? 

Oh, we are. I was on a plane so uploading pics was super slow. Here you go:

Frunk option:

Stock location:

You can see the "stock bike" location almost perfectly lines up with the existing engine mount bracket.

Both locations would be easy to mount the tank. Probably about a wash in fab work.

Indy-Guy said:

Use the stock gas tank and location. Two gallons of fuel is roughly 16 pounds. Plenty for a couple of auto cross runs.

Focus your time and effort on getting the car up and running for some testing time. Then you can relocate the fuel tank later.

We're using the stock bike tank and pump. Not the car tank and pump. Maybe that wasn't super clear.

Both locations will involve some fab work, as mentioned.

Above the engine all the way because it looks cool. That's about head level and 2 lbs of fuel is about as heavy as a big head. Just pretend you have a big headed passenger with you (or at least their head )

Hello Robbie - you don't need me to tell you this is an exceptional and excellent project. I live in Dorset, England and at least 2 significant things happened today -

1. I went to Chard, Somerset this afternoon to meet a fabricator of grasstrack racers, and agreed a contract to install a Hayabusa bike engine in my X1/9, starting May 2019.

2. I went online to see if anyone has done this already, and ended up here.

Thus proving there are at least three people who think this is the way to go.

I expect you already did your calculations - according to mine, the power to weight ratio comes out very close to the F40 Ferrari.

I'm saying this in case anyone on this board actually owns an F40 - which is unlikely - but for people like us who don't have that sort of money, well it's all about bangs for bucks innit.

You guys have a massive headstart on us but I'm sure we have some kind of common understanding about what's involved.

You have the advantage in terms of corner weights thru choosing a central driving position and I assume you are building yours to race whereas we're building a road car which will look very stock from the outside.

It should be possible to achieve a very low c of g, perfect or near perfect balance, and with more mass further within the wheelbase, than virtually any other production rear-mid engined car I can think of.

I have some concerns about spring rates and damping but I'm sure you'll be getting in to that in due course.

I spent about 5 months this year doing a nut and bolt suspension rebuild on mine in anticipation of more power - 2-way coilovers, front ARB, Avon ZZS tyres, Billet Ultralite 13" wheels w custom offsets bla bla, spent a lot of time on the geometry - this car now utterly spankable = hoonible, hoonable er.. it really handles a treat is what I'm saying. 

See pics.

Good Luck y'all - I'm watching with interest !

MarkyMark.

That thing looks incredible!!!

In reply to MarkyMark77 :

GREAT to hear from you. Let me know if you have any questions, but I'm sure you and your shop will be able to figure this out no problem (I'm just a guy with a few tools and too much confidence!)

There is another guy I know - one of my cousin's friends - who is doing basically the same build as well. He's quite a bit ahead of me when I met him and saw the car about 18 months ago, but he lives down in louisville. Not sure he has anything on the web about his build though. 

And, your car looks awesome!

MarkyMark, beautiful X! What are the specs on the ZZSes and wheels? How do you like the Avons?

MarkyMark, I like the cut of your jib, sir.

I've got an X1/9 too, but it's still Lampredi powered. I also lust after a Hyabusa powerplant.

In reply to Hasbro : Hi there Hasbro - 

http://www.imagewheels.co.uk/three-piece-billet-alloy-wheels/

Quality range of wheels made to order, any offset, any PCD, excellent service, reliable company.

http://www.avonmotorsport.com/road-legal/performance/zzs

The control tyre for the Caterham Supersport Race Championship. Enough said. Reliable company, great tech support.

Just an awesome tyre - the 185/55/13's as fitted to my X are effectively a 195 with an extremely robust sidewall and outstanding grip. For testing I use a particular, tightening bend with safe runouts, repeatedly. With stock power it's hard to get enough entry speed to come unstuck, but by tightening the approach angle the car does eventually let go. The first time it happened it was a bit like "whoooa wtf, the steering wheel's come off, I'm going to die" but I tried it again, knowing what to expect. The momentary lightness in the steering, feels a little like understeer but actually it's the sensation of all 4 tyres letting go at once - then you get into a sublime 4-wheel drift which runs you out wide. I'm always on full power and you end up carving an arc, in effect. What's absolutely incredible about the X is the way it remains planted, with no discernable shift in attitude - neither the front or the rear deviate from the arc. At no point does it feel like the car's going to rotate on its axis. With more power, I'm expecting to be more careful on the throttle, but I reckon you can tighten the arc, narrowing the runout. Hope that makes sense.

On the front I've got some negative camber and zero toe, 7" 200lb springs with helpers, 26 psi, rebound damping on about two thirds, and an Addco ARB. On the rear I've got more negative camber, some toe-in, 7" 250lb springs with helpers, 28 psi, rebound damping on half, no ARB. Ride heights are about 2" below stock (front) and 2 1/4" below stock (rear) on the struts. Everything's polybushed. It seems to work ok but I'll carry on tweaking in the spring. Biggest problem is those damn wiggle bolts won't hold camber no matter how tight they are. It's a documented problem - 

https://www.youtube.com/watch?v=3p4RahE0olk

https://www.youtube.com/watch?v=Xxb8DLxMGOE&t=30s

Incidentally I'm not using these struts from VickAuto, I'm using Gaz (British) - but I'm not happy and intend to go over idc. The design of the VickAuto strut is superior imo - particularly the "pillow-ball" top mount and the wide roller bearing pivot bush. Looks like John is still using the Plaia bigfoot turret plate (same here, good idea, available from Chris Obert).

Bye for now chaps.

In reply to Mezzanine :

It's the way to go, you're in the right place !

In reply to Robbie :

Hi Robbie, good to meet you too.

I've been looking at your pics showing the stock (Suzuki) headers and the way they curve round under the sump.

We will most likely fabricate a new header that goes either side of the pan, or even back up over the head, because ground clearance will most likely be tight, a knock-on effect of lowering the car on the struts and (tyre) sidewalls.

You'll probably need an oil recovery system as an anti-surge precaution (racecar style).

I'm curious as to whether you intend to run 13" or 15" wheels. You can uprate the discs (ventilated, slotted, drilled) and calipers (bigger, floating) with the Prima brake kit from VickAuto, which fit 13" wheels, available here - 

http://www.vickauto.com/newstore/index.php?main_page=product_info&cPath=5_7_51&products_id=3936

In my opinion, the 13" wheels are an essential part of the recipe - they look correct for a car of this size, very original, very retro, very low inertial mass. The 6J Billet 93 and Billet Ultralite wheels from Image (see details in my reply to Hasbro) are about 25% lighter than the OE 5 1/2" Speedlines.

Most of the really powerful X's I've seen, do require 15" wheels to accommodate bigger brakes eg Robert Frick's supercharged K20 -

https://www.youtube.com/watch?v=mm6lPvA1hyM

.... but for a naturally aspirated 200 bhp Hayabusa I think 13" adequate providing you max out the available space and possibly use custom-built calipers.

Don't give up.

May your enthusiasm, and your shadow, never diminish.

Mark.

In reply to pimpm3 :

Thank you.

A bloke drove past in a 458 Italia, actually stopped to give it the thumbs. 

I said he should get a Stratos !

In reply to MarkyMark77 :

I fully agree with you on the 13s (looks, weight, etc). This is a budget build, so right now I'm planning on running the stock 13s (13x4.5?) or the widened stock steel wheels I picked up (13x8). I'd really love a 13x6 or something in the middle, but budget will dictate the final way I go. For brakes we are using the same methodology. Of course the pads and rotors and calipers and lines will be in good working order (new), but we will likely use stock parts. I may even get the rotors turned down to minimum thickness or something to shave as much weight as possible. This will be initially an autox car so braking will need to be effective but only for a very limited amount of time. 

For headers, we are also using the stock pieces because of the budget and the time. These are titanium headers from the factory! They weight almost nothing and cost almost nothing because the race bike guys replace the stock parts. That does mean our motor will have to sit about 2 inches higher than it otherwise would but we think it is a fair tradeoff. 

Finally, we do have an oil pan with baffles designed for a car. I'm crossing my fingers that is all we have to do for oil control!

  RX7 rear hawk blue brake pads will fit the fiat calipers, that will work really well.

In reply to Robbie :

You know, it will be interesting to see the post Challenge development.  If the car turns out so well that you continue to develop it into whatever you want it to be...  That should be fascinating.