LocostUSA.com

Update: 3/14/08

Today's pi day!

I've been busy ordering stuff and finishing finals, which are generally mutually exclusive. Oh well.

I bought some 10ft control cables from surplus supply as well as a few other doodads (including a 35amp bridge rectifier for my coilgun project). I also ordered the cheapest intake availiable off ebay, a radiator tank, emergency brake handle, and the final missing knuckle/brake.

I looked into brake fittings. It seems the cheapest way is to do 10mm to AN3, some stainless hose, and then hard lines. However, I feel sufficiently nervous enough about working with hard tubing that I can make a pretty good case to my parents to pay for full braided stainless lines. The Ar-i-el At-om uses them, so I would assume that they're good.

I just got my box from pitstopusa...which included aurora 5/8 joints and a Howe steering clamp. I highly recommend the clamp - it's really well built and a great time saver. It should allow me to have a bit of tilt-steering.

For the steering, I ordered some cheap 5/8 heim joints. However, these have a play of about .1-.2 mm. In terms of steering stability, do you guys think that this will be a problem?

The seatbelts I ordered from speedway also got here. I would highly recommend their cheapest offering ($60), which seems to be bulletproof with close to 3" webbing and 3/16 steel hardware. Plus, they're inspected and approved unlike similar offerings on ebay. Still, one of the most ridiculous parts of the registration process will be putting normal 3 point belts on for inspection.

Over the next week, I'm going to design the suspension. I'm going to aim for: around 7-10 degrees of steering axis, 10 degrees of caster angle, and the front track of 57.9". The camber gain will come out as it will when I model the a-arms. I do have a bit of choice in this matter, but it seems to me that significant movement of the a-arm mounting points will be required to create significant changes in camber gain.

The other main task will be creating a cad drawing for waterjet cutting. I'm going to have the majority (60+) of the parts I would normally need made cut out of 1/8 and 3/16 steel. I believe waterjet services are reasonably priced and well worth the hundreds of hours saved.

Awhile ago I bought an immobilizer/key assembly. JonW seems to be running into the same problem. I believe that the ecu won't allow the car to start unless the correct code is read by the immobilizer. There is no way to bypass this, and i've researched it pretty extensively

Since I'm running four front knuckles/brakes, i'm going to have to fit my own parking brakes. I considered the official version, but I'm leaning towards Go-kart cable brakes modified to fit the rear wheels. They would attach on the ABS sensor mount (which is part of the knuckle casting). And would save probably $150.


Some links to my most recent pictures:


http://www.flickr.com/photos/25435483@N00/2333850238/
http://www.flickr.com/photos/25435483@N00/2333015849/
http://www.flickr.com/photos/25435483@N00/2333017329/

_________________
"I've just been paid a visit by Q. He wants to do something nice for me."

"I'll alert the crew." (Picard and Riker)

Good idea on the cart ebrake, just need to make a bridge spacer to adjust for rotor thickness. I would add a spring to maintain tension after the rotor cools.

Why so much kpi? Do you need that much for the wheel offset and tires your using to eliminate the scrub radius? I was suggesting how you could reduce the kpi in a previous post.

The more kpi you have, the more the camber will change when turning. It may be fine for a vehicle like a suv with a high cg, a lot of body roll, and power steering, but less is more for a locost.

_________________
Miata UBJ: ES-2074R('70s maz pickup)
Ford IFS viewtopic.php?f=5&t=13225&p=134742
Simple Spring select viewtopic.php?f=5&t=11815
LxWxHt
360LA 442E: 134.5x46x15
Lotus7:115x39x7.25
Tiger Avon:114x40x13.3-12.6
Champion/Book:114x42x11
Gibbs/Haynes:122x42x14
VoDou:113x44x14
McSorley 442:122x46x14
Collins 241:127x46x12

Are you saying using braided line for the entire run? It is not as stiff as metal tubing and will give you a spongy pedal. You should minimize braided line use and also not use too large a diameter. You don't need to be drastic - just don't run extra feet of the stuff.

A hard brake pedal feels a lot more responsive and is one of the pleasures of a light weight car with a good brake system. It should feel like there is a brick under the pedal when you use it. You'll like that. Just one of the ways modern cars have become crappy... IMHO...

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.

Can a mod duplicate and move some of this recent material out to it's own appropriate thread? I would like to see discussion continue without burying the build log...

And, what about that rail gun!

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.

Update: 3/21/08

I got the intake, e-brake handle, and radiator reservoir today. All seem to be in good condition. The intake, which is a cold air/ram style from ebay, will require some modification but should work great.

I've spent an ungodly amount of hours working on suspension stuff. I've come up with the following numbers for the front suspension:
tire diameter: 24.8
KPI: 15.6 degrees
Caster: 5 degrees
Static Camber: -1 degree
Static toe: none for the moment
Scrub radius: 17 mm

Now for the fun stuff:
Roll center: 3.5" above ground, movement of around 5 inches in 3 degree roll
Bump: with 2" bump, one wheel toes out .37 degrees or around .04 inches.
Ackerman: 43.7%

Suspension junkies? How do these numbers look? What should be improved.

_________________
"I've just been paid a visit by Q. He wants to do something nice for me."

"I'll alert the crew." (Picard and Riker)

Update: 4/3

I've been on an ebay rampage lately, having bought the instrument cluster with some ripped wires, a throttle pedal, axles, shift cables, and several other small parts. I'm still scouting for a pedal assembly and some coilovers. I ended up buying three steering racks, all made by Sweet, two with a 2.5 ratio, and one with a 3.0 ratio, for around 80 on ebay. Somehow it was cheaper to buy three than one. Once I've chosen the one I want, I'd be happy to sell the other two. They are circle track racks with widths between 18-19".

I test fit the engine, and am about 1/2 done with the motor mounts. The passenger side mount is done, complete with improvised rubber bushings. See my post in the suspension section for more info on this.
On the honda k20 engine, there are four mounts:
1. front: a tripod and a rubber moulding. For me, one of these threads is destroyed, and the other is on a cracked part of the bell housing.
2. Passenger: three bolts in the side of the engine, facing towards the wheel.
3. Driver: three bolts facing up.
4. rear: three bolts on an arc along the engine where the bolts are parallel to the axles.

I'm rebuilding all of these motor mounts, except that I'm not using the front mount. I hope that the three will be enough. Since the engine weighs only around 300 lbs and only generates around 130 lbft of torque, these three mounts should hopefully be enough.

I spent a day driving around Salt Lake, and found a laser shop which will cut a 2' by 3' 1/8 steel plate to a cad drawing for 75$. This should save me 30-40 hours of grinding not to mention tens of grinding wheels and blades.

I grabbed some PVC bends and a 3" sewage connector to allow me to twist the intake pretty much any way I want. So far, i've spent around $40 on the intake = very locost.

I'm trying to find a throttle cable that will work. It needs to be at least 12 ft long with a ball end. I can find all sorts of cables, but none with ball ends that aren't bike brake cables.

Over my spring break, I read and worked in solidworks and Wishbone.exe to figure out suspension. While I don't have a final design, I know where I want to end up. My intention for the final front suspension adjustments are as follows:

KPI/SAI: 16 degrees
Caster: 9 degrees
Front weight: 200 lb/wheel
Rear: 300 lb/wheel
Scrub Radius: negative, less than 1 cm
Ackerman: 50%
Roll center Height: 0 inches above ground
Bump Steer: Less than 1/4 degree for 1 inch droop, 2 inches bump.
Bump Steer: Less than 1 degree for 2 inch droop, 3.5 inches bump.
Static Camber: -1 degree
Static Toe: 1 mm toe out

I test-fit the shifter assembly. As I thought, the easiest solution for a middy was to order some long shift cables. The 10ft cables are pretty thick and the shift effort is much heavier than the stock assembly, but the feel is improved with a satisfying clunk.

I bought a Pilot steering wheel with two buttons built in for $20. This was a great buy - the wheel feels really good and fits the contour of the hands well. The only negative is that the attachment is a 5 point, where as the hub is 3 point. I'd highly recommend pilot to fellow locosters.

Also, my dad may buy a small lathe in the near future. Yay! no more fake-turning aluminum on the drill press.

I found some different-shaped threaded rod ends in order to fit the knuckle better. This should allow me to fit the upper control arm further into the knuckle to decrease the KPI from the current 18 degrees and decrease the angle of the upper control arm to a more reasonable angle.

Well that's enough for now. Check my flickr account for more pictures. See earlier posts for the URL.

_________________
"I've just been paid a visit by Q. He wants to do something nice for me."

"I'll alert the crew." (Picard and Riker)

You could probably use a throttle cable from a fiero or MR2, I have seen a cylindrical end at the tb, but not a ball. I don't have any Honda cars though.


For the cable:
http://www.pegasusautoracing.com/produc ... oduct=1270

For the terminal, a round wire stop:
http://www.gokartsupply.com/thcable.htm

_________________
Miata UBJ: ES-2074R('70s maz pickup)
Ford IFS viewtopic.php?f=5&t=13225&p=134742
Simple Spring select viewtopic.php?f=5&t=11815
LxWxHt
360LA 442E: 134.5x46x15
Lotus7:115x39x7.25
Tiger Avon:114x40x13.3-12.6
Champion/Book:114x42x11
Gibbs/Haynes:122x42x14
VoDou:113x44x14
McSorley 442:122x46x14
Collins 241:127x46x12

Here is a Honda cable from Lokar. Note the tb end:

http://www.lokarmotorsports.com/throttlecables_big.html

_________________
Miata UBJ: ES-2074R('70s maz pickup)
Ford IFS viewtopic.php?f=5&t=13225&p=134742
Simple Spring select viewtopic.php?f=5&t=11815
LxWxHt
360LA 442E: 134.5x46x15
Lotus7:115x39x7.25
Tiger Avon:114x40x13.3-12.6
Champion/Book:114x42x11
Gibbs/Haynes:122x42x14
VoDou:113x44x14
McSorley 442:122x46x14
Collins 241:127x46x12

Update: 6/29/08

Three months later, and things are happening. I was pretty bogged down with school for the last two months, but now that I'm out, things are happening at a crazy rate.

I've got till September 10th to finish my car, at which point I'm going to Stanford University. I'm working five hours a day...the rest I am free to spend working. I'm sure that I can get it drivable by that date, bodywork may be next summer. However, I'm planning on doing heavy fabric bodywork to save weight, money, and speed things up.

What's working now:
Engine is mounted and turns over.
Seat Sliders Work
Rear suspension is completely done with the exception of a couple tack welds that need to be welded over.
Front Suspension....in progress.

Stuff that happened during school:
I went on a shopping spree. Total money spent so far is around $7500. quite a bit of that is in tools, but most is in the engine and component

I was trying to fit the engine in the frame, but realized that it wouldn't fit without some pretty nasty grinding into critical structural pieces. After much thought, I decided to move the engine over 5 inches and increase the track of the car from 57.9(stock rsx) to 62.5. I avoided a big expense by only getting one axle lengthened. One machine shop I took it too wouldn't touch the axle...an axle shop lengthened it 4 5/8 inches for $100. Now, my car is wider than most, and the engine fits.



Rear Suspension:
I learned that solidworks is possibly the greatest program ever. I modeled the rear suspension using guidance from wishbone, printed out pictures, cut, and welded. I was pleased to find that the honda ball joints didn't get pried out when met with the full weight of the engine. I am using a cam system similar to the At-om's. However, I improved on it with fully adjustable bellcranks and tapered roller bearings. In combination with the yamaha R1 shocks, with 200 lbs of adjustable preload and 12 settings of bump and rebound.

Rear: 1x9x4x10x12x12 = 51840 different setups per tire
Front: 6*7*10*12*12 = 60480 different setups per tire

I've got some setup work to do.

Anyway, the bellcranks work great. Attached to the frame with 3/4 grade 8 bolts and timken roller bearings, there is no stiction in the suspension. The At-om uses bronze bearings for the bellcranks - I've heard that the valving of the shocks has to be adjusted for the friction created by the bellcranks.

I now know the length of the various pushrods/tierods. My rear pushrods came out to 9 inches, front at 12, and tie-rod somewhere near 13.

I'll let the pictures speak for themselves. Once again, check out my flickr account at http://www.flickr.com/photos/25435483@N00/

_________________
"I've just been paid a visit by Q. He wants to do something nice for me."

"I'll alert the crew." (Picard and Riker)

I know its kind of late, but I was reading that you were having problems with the immobilizer system of the civic. I don't know if you solved this problem yet, but wouldn't a remote car starter system bypass the immobilizer system. As an added benifit you would have a functional remote start system when you are finished. Just a thought.

On the remote start, I looked at that. On ebay, you can by a remote start immobilizer bypass thing for 25. I read the manual, and it reads the immobilizer code, stores it, and pretends that it is the immobilizer. You still have to have the ECU reflashed and synced with the immobilizer/key. I figured that it'd just be easier to go ahead and use the key, and the added security that comes with it.

_________________
"I've just been paid a visit by Q. He wants to do something nice for me."

"I'll alert the crew." (Picard and Riker)

A remote starter wouldn't help. The immobilizer prevents the fuel pump relay from grounding and operating the fuel pump. Just substitue a "hot with ignition on" wire to defeat it. That's what I did.

The infomation to do this is in the Honda wiring manual. It would take a thief an extra 10 seconds to do this if he knew which wire it was. The relay is under the dash, not under the hood.

Clarfying edit:

The remote described has to read and copy the code fro the key to work. The immobilizer usually causes problems when you either don't have the original key (my case) or you change the PCM (Honda-speak for ECU). If you have the PCM-matched key there should be no problem. The immobilizer function can be separated from the ignition if you want to start the car without a loose key or with a button. Also, the key doesn't have to inserted in the key lock, just be located in proximity to be sensed electronically.

Alternately, the Honda dealer can reprogram the new key to the PCM. But I am still completing wiring and trying to prove complete system function before that could be successful.

would like to contact you regarding info and advice on my build. looking to build an At-om or slr. please call 917 587 6742

I'm back on locostusa, with things starting to get under control.

I've been working around 10 hours per day on this car, and it's all coming together.

The physical parts that are not yet mounted are: dashboard/switch panel, exhaust, mirrors, lights. Other than that, most everything mechanical is done. wort wort wort.

Suspension works, shifter works, fuel pump works, steering works, brakes work, clutch works, ecu sort of works.

On a side note, the brakes are incredibly strong. I can easily lock all four wheels with a light tap on the brake pedal. I'm not sure why this is, but it's a bunch of fun. I've been pushing the car up the driveway, jumping in, and rolling down.

At the moment, I'm reverse engineering the Honda wiring harness. I got the ecu and gauge panel sufficiently set up that I could crank the engine and read OBDII codes. The list of errors on the OBD reader was hilarious:

Oxygen system....failed
Fuel Emissions.....failed
A/F Sensor......failed
fuel tank....failed

and so on. However, the engine was happy to tell me that my coolant temperature was 26 degrees C. Awesome.

Which leads me to the critical part here: I've done a bunch of research, torn my hair out, tried every trick on every forum, and there is no way to bypass the immobilizer. You can bypass the fuel pump relay (it does truly look like that's the only relay that the ECU will ground once the correct key is inserted. However, the honda/acura ecu will NOT fire the ignition until the correct code is received. Since I have an immobilizer from a different car as the engine/ecu, I can't get it to work.

I called Hondata, I posted on k20a, and no-one seems to be able to help.
That is, until I tried several places around town. I found an Acura dealership that is willing to let me set up my ecu/gauge cluster/wiring harness on the floor of their shop with my trusty 12V power supply, using the data port on the bare harness to reprogram the ecu. Incredible. We'll see if they're good on their word tomorrow.

The only other issue I have is with ackerman/caster. When I turn the wheel about 75% lock, and push the car forward, the wheels continue to want to turn, and automatically move to full lock. Remember I have a huge amount of kpi (16 deg) and a caster of around 5 degrees. The strange part is that the steering corrects itself if you are moving in reverse. However, there seems to be zero tracking moving forward (mind you this is only at like 3 mph).

Many people here seem to have an issue with suspension components hitting each other. I think that my suspension doesn't hit itself at any position. The only glitch at the moment is that my R1 shocks (which work really well, btw) together with the cams (which are really smooth/no stiction) are kindof a bi-positive switch. This means that if the front suspension gets compressed enough, the cam will rotate too far and get stuck on the other side of its normal rotation. I think I will correct this with an aluminum spacer on the shock. Yes, it decreases the front travel, but I have plenty to begin with (5 inches)

I believe that this is being caused by a torque created around the steering axis by the scrub of the unloaded wheel. At root cause, I guess is my lower ackerman (40-50% or so) combined with the large amount of kpi. (The inner wheel has a wicked angle on it at full lock)

Other than this, I'm on track to finish the car by the end of the summer.

Body panels should be pretty easy. I've settled on a combination of formed acrylic and aluminum. The body pan will be aluminum. Two of the body panels, the banana peel and the engine cover, will be from two domestic items: a kayak and a doggy dome. Fun stuff.

I am this close to finishing.....I can't give up now. Yet what keeps you going is that sepia-toned image of yourself blasting through a mountain road in your creation. I'm sure you guys know what I'm talking about. :p

_________________
"I've just been paid a visit by Q. He wants to do something nice for me."

"I'll alert the crew." (Picard and Riker)

Awesome! Try more caster. 8-10 degrees may be needed.

_________________
He is a wise man who does not grieve for the things which he has not, but rejoices for those which he has.