GT380 ground up build

themotoworks

Been Around the Block
Starting a new project with a new idea, I picked up a gt380 for free with a valid registration and want to use it as a design experiment to see if I can put some ideas I've had into one single project. The idea is really to conceive the entire bike before really building anything, my hope is that it both makes a really functional bike, speeds up the mechanical side of the build process, and makes for a really nice looking build. I pulled the motor already and have been measuring it to put it into solidworks and integrate it with the other parts. The front end is a gsxr-750 with custom made triples and a custom machined spoked hub, rear end is all custom fab, going to run an ohlins shock for a 900ss ducati because I'm familiar with them, planning on a large diameter tube backbone frame and machined aluminum swingarm mounts that combine the engine mount, frame to engine mounting and subframe/seat mount... came up with some sketches, and talked to jeremy lacy about him basically designing the bodywork and look of the bike, should be interesting.

IMG_20180525_153012159_zps3spjhcee.jpg

IMG_20180524_155600_238_zpswerp6och.jpg

gt380_zpsilzwcsty.jpg
 
This looks like a sweet project!
Just wondering how you ended up with the frame style you chose.
I presume by large diameter backbone you mean something like an Egli frame,
c9100c0c0664e870fb60c275ff3ac88c.jpg

That one is for a tz350 i believe.
What did you plan for material?
 
chromoly, I'm having trouble finding large diameter thin wall tubing in that material though, I may need to have a sheet rolled.
 
Figured it would be chromo, was hoping maybe you would have had better luck than me at finding something in a proper size.

Did you model it yet? What sort of weight are you aiming for for the frame?

I modelled several frame designs this design seemed to give the best stiffness to weight ratio, it was about 18lbs
 
YZ frames were pretty light. When the shock tubed into the frame, the backbone was stamped.

csteel124_3.png


They later went to this in 80.

6082133407_47f94a75dc.jpg
 
How large do you want the backbone?

http://www.onlinemetals.com/merchant.cfm?pid=10434&step=4&showunits=inches&id=250&top_cat=197

That's 3.0'' but it's pretty thick. You may have to go to rolling a sheet into tube as you suggested.

in DOM you can get up to 4.0" http://www.onlinemetals.com/merchant.cfm?pid=12690&step=4&showunits=inches&id=283&top_cat=197
 
I talked to Fritz Egli and he actually told me what I had suspected for a while, he doesn't use chromoly tubing, the modulus of all steel is basically the same, and there's not going to be a difference in the stiffness from a513 to 4130 unless it's being bent to yield, which it shouldn't be. 4" .083" is available, I'll model a frame from that and see how well it does, from my experience, changing the diameter or wall thickness of the main backbone section very predictably changes the tortional stiffness of the front end
 
A friend of mine is working on building a tube chassis autocross truck, and we've been researching the use of docol r8 tubing vs chromoly. Similar weight, yet there are far less brittleness issues in the heat affected zone from welding. It just became available in the u.s. market fairly recently, though it's been used in Europe for a bit. Just another possible option. When i build a frame (someday) ill likely use it.


http://www.aedmotorsport.com/catalog/product/docol-tube-r8

Sent from my SM-G920V using Tapatalk
 
Don’t they all weigh nearly the same? Like less than 1% density difference?

I was under the impression that you have more elastic deformation from 1020 vs 4130
But that they both reach plastic deformation at nearly the same points.
 
yeah, the density of all steel is basically the same, the modulus is basically the same, the things that vary with different alloys and heat treatments are the tensile strength and the yield, which really only determines where they "break". so if I were designing a roll cage, chromoly would be a good choice, because for a given weight roll cage, the chromoly one would take more of a beating before it gave way, but in a design built to be predictable under known loading, with an adequate safety factor in the design, there wouldn't be a difference in performance between 4130 and a513
 
gt3801_zpsp9dirbzd.jpg

trying to finish designing the swingarm and rear frame mount, which is a big p.i.t.a. because the rear engine mounts are kicked way off to the right of the engine centerline, not ideal but I can make it work
 
gt380%20timing_zpslz5ozhte.png


I'm planning on slightly modifying the engine, cutting the piston skirt and adding piston ports as well as reed valves and tuned pipes, anyone on here know much about 2 strokes? Will this timing be alright? It's a bit modified on the intake side to allow for the reed valve to stop reversion back into the carbs instead of the piston blocking the back flow.

also made a lot of progress on designing the rear suspension, need to finalize the shock choice so I can finish designing all the linkages etc.
 
Where did those data points come from? Is that what you measured stock timing as or where you want it to be? For a free revving motor, those numbers are not close to where you want them to be. IMHO. try EO at 90 ATDC, TO at 120 ATDC and Intake at say 1110 BTDS as a starting point and work out the time area at target peak revs and see how wide you would have to go to reach the sort of BMEP levels you want at target RPMs.

Of course you will also need to design a pipe to work with the target port shapes - assuming there's enough metal to port it to those numbers. Look at what Barton motors did with the water cooled Sparton 500 which was based on that bottom end.

Suitable sized reed boxes are a good way to add additional intake time area but that may cause clearance issues with the rear cylinder base studs.

Just as a point of reference with a more modern motor, an RS250 Aprilia lightly cleaned up makes over 60hp from 250cc, so a triple based on the same ports could theoretically make around 90hp. Your motor will not have exhaust power valves or the port area, or crankcase reed area of an RS250 so that gives you a base to consider. By comparison, a full race GT250 or T20 makes closer to 40hp with a TZ crank, later style multi port barrels etc .

You may want to look at what anyone with a T20/GT250 or even a T315/350 does to make power and start there. I recall riding a beautiful T350 with 6mm spacers under the cylinders many years ago. It looked totally stock and went remarkably fast.

How much HP at what Revs is your target?

Last time I looked at one of those, I ended up with EO at 80 ATDC, and transfers at 110 ATDC peaking at almost 10,000 with restricted intakes to keep revs down and some semblance of bottom end (ie below 6k).
 
those exhaust and transfer numbers were based on measuring the stock engine, the intake to crank timing is based on adding windows into the piston skirt and cutting the base of the skirt back, which gave longer duration, requiring a reed valve, but possibly making more power. I downloaded some long book on 2 stroke performance, looked like it was from the 70's, but it has a lot of info on port size based on rpm etc., I want to make this thing peppy but not go so overboard that I lose months to engine tweaking. The idea for this build was to really test the idea of building a bike completely in the computer, then machining and welding to the prints, then assembling, and seeing if it would work, or what would need to be changed to make it work. I hesitate to go neck deep into the engine, but at the same time I don't want to leave it as sleepy as stock, especially when the rest of the bike is really build for performance. also since my slow time at work is usually the summer months, if this goes past september, it may not see the light of day until next summer

Here's where I'm at with the rear suspension so far, plots look good for the geometry, clearances look pretty good in SW, just need to flesh out things like the linkage and swingarm, I'm also a little hesitant to finalize the design without having the shock in hand, and since ohlins doesn't give out cad info or drawings, that might force a bit of a wait...


gt380suspension_zpsqchgnuew.png

gt380%20suspension%20plots_zps3itofdk8.png
 
I dug out a previous simulation run for a more or less stock 380 and changed inlet timing to yours and lost power everywhere. The ports are way too large and lose intake velocity and therefore crankcase filling. Raise the exhaust 2.5-3.0mm, transfers by 2mm and drop the intakes 2mm to start. Check those numbers with your engine builder. There are infinite number of combinations and a few of them work better than others. You will have to machine teh head to get compression back too.
 
283 ATDC or 77 BTDC modeled in Mota 6.1 or -x beta version. 283* is 24.2mm BTDC and your 123* works out to 44.1mm BTDC (plus piston length of course).

MOTA ignores actual intake port dimension when using reeds because it assumes that the reeds will effectively modulate flow but with excessively large ports, it loses flow velocity even with reeds. The sections of the reeds and port must be matched to work effectively.

BTW, you still get harmonics in the intakes with reeds and reversion - just not quite as bad.
 
IMG_20180602_102202153_HDR_zps25rvla8v.jpg

IMG_20180602_180857_587_zpsqa3wvezy.jpg


finished up the rear hub, it's modeled in the solidworks bike so it's nice to know already that things will fit, next going to try to finish the front hub sunday before I need to do actual work at the shop haha
 
Back
Top Bottom