[TrumbullHucker]

If your going to mock me, then do it straight up, no sarcasm please.

The winter months make me( and i think alot of people) loopy and itching to ride.

I have always wanted to design and build my own downhill frame.

My first frame will most likely be Mild or Chromoly Steel ( dont really care if it will weigh 55 pounds + ) and most likely a single pivot. As my knowledge and frame building skills increase, I will graduate to multi linked, Aluminum frames

I do have AUTOCAD designs( which i took 2 years of college courses of) of a really "interesting and out there" suspension design.

I have the skills and knowledge of G.T.A.W (Gas Tungsten Arc Welding). I graduated with an Associates in Welding Technology( We covered M.I.G, T.I.G, Stick, Oxy-Acet welding). I have learned how to weld those process's just mentioned on Mild Steel and Aluminum.

At my new job (which I was blessed to receive a full time, 40 hour work week) I am a full time welder/ machinist. I was STOKED to get my hands onto some bridgeport milling machines and lathes. I now know how to use a Daewoo A20 60" Lathe like the back of my hand! I also learned how to read inside/ outside micrometers and all the "thousandths" measuring devices.

OKAY, now that im done giving ridemonkey my life story, I am interested in seeing how frame builders/ frame designers figure out how to design swingarms and front triangles in accordance to shock placement and leverage ratios. now i am just talking about single pivots here, I cant even imagine how they design/ cut/ cnc mill links for multi link bikes, like M6, Liaperre exc exc.

Now i am aware i did not major in Engineering and Physics, so please dont butcher me haha.

So...how do you smarties do it?!

I would show you my "wild" 3-d frame design that i designed over the past 2 years, but my computer has a nasty virus, and I havnt had the time to invest to remove the virus.

take 'er easy

-jordan

 

[Kanye West]

If it's nothing more complicated than a 4bar or dual link setup, try the program "Linkage" to start with.

 

[kazlx]

Basically, a leverage ratio is the ratio of tire movement to shock shaft movement. If have a leverage ratio of 2:1 and your tire moves 10 inches then your shocks shaft will move 5 inches. So in your CAD drawings, you should have the amount of travel that the bike rear is supposed to have and depending on shock placement you can pick a shock that fits with the correct eye-to-eye length and stroke. No matter how many linkages you have, in general, when you look at the frame, the shock shaft will move in a proportionate ratio to the rear wheel. It gets more complicated with different wheel paths, but on a single pivot, it's pretty straightforward.

From Chumba:
http://chumbaracing.blogspot.com/2010/01/chumba-tech-article-leverage-rates_4629.html

Google has a million different sites that give explanations.

 

[tacubaya]

First you want to determine the suspension behaviour you want. Progressiveness or regressiveness (leverage ratio and leverage curves), wheel path (influences cornering - turning radius), pedal efficiency (anti-squat A2 value), chain extension range etc. Then when you have those parameters you can modify the linkage to compile with all (or most) of them.

I'm in the same situation as you are and I am still learning. There are guys out there who have been working on suspension designs for more than 10 years and they are still learning new stuff...

Cheers

 

[davec113]

Hi,

I am a ME, and worked as a welder and auto mechanic / fabricator for a while. Welding and machining is a great skill to have. I have a TIG and MIG, but never got good at aluminum, I mostly fabricated off-road vehicles. I just moved and am going to make room in the garage for a small metal shop, maybe get better at TIG. I have a high end Thermal Arc, so I can't blame the machine...

One theory is you want the leverage ratio curve to be a straight line, but because most bikes use rotating links you do not end up with a straight line. This is one of the reasons Yeti uses the rail system, the other is axlepath.

I have owned a Uzzi VPX, a VPP bike, and it makes too many compromises for the sake of a theoretically superior axlepath. I have gone back to single pivot (Trek) and couldn't be happier.

So, here it is, the theoretically perfect leverage ratio curve:

 

[davec113]

From Chumba:
http://chumbaracing.blogspot.com/2010/01/chumba-tech-article-leverage-rates_4629.html

Check out the graph of those leverage curves, and notice how f'ed up the Santa Cruz and Mojo are. Also, look... a single pivot and FSR are exactly the same. With the pivot in the right location, the axlepath is 99.9% the same too.

 

[karpi]

the problem with the Trek is that while beeing a single pivot, it doesnt act like one when the suspension compresses since it has a floating shock, where one side compresses and the other decompreses. The blur is actually real good, you can see that the maximum leverage rate is right around the SAG point. What I understand by this is it helps break through the initial part of the travel making it more supple, but then, it spikes up to much. Thats my opinion on the graph, cant comment any more.

 

[dklopp]

here a good thread that has a TON of info on the general building of a frame (hope you have time on your hands because it is very helpful): http://www.pinkbike.com/forum/listcomments/?threadid=39840

but the best place to start with finalizing your design is using linkage. it is dirt cheap for a personal licence (the free one is only for viewing frames in the library, no editing): http://www.bikechecker.com/

i know this wont perfectly answer your questions, but it will open up another good source of information, which can never hurt.

 

[jonKranked]

its hard to mock someone with Howlin' Mad Murphy as their avatar

 

[TrumbullHucker]

its hard to mock someone with Howlin' Mad Murphy as their avatar

40 watt...woommbb

 

[TrumbullHucker]

and thanks for those links, dklopp. It is great to read about what the pinkbike people use for material ( AL and Steel series )

and Iv used bikechecker before ( only free version though ) maybe with this full time job, i can purchase it!

 

[LMC]

I would just scan through existing bikes in Linkage, find one that has a leverage ratio to suit and copy the coordinates. then you can move or rotate the coordinates in one group to suit your frame design. The cove peeler would be a good one to look at.

 

[LMC]

Check out the graph of those leverage curves, and notice how f'ed up the Santa Cruz and Mojo are. Also, look... a single pivot and FSR are exactly the same. With the pivot in the right location, the axlepath is 99.9% the same too.

The Mojo is designed around an air shock, hence the high ratio at the end of the travel to compensate for the natural progressiveness of an air shock. High leverage at the start also to overcome stiction.

 

[davec113]

The Mojo is designed around an air shock, hence the high ratio at the end of the travel to compensate for the natural progressiveness of an air shock. High leverage at the start also to overcome stiction.

So are all the bikes on that chart.

 

[EVIL JN]

I found some pagen on google from a book, cant remember the exact title but something along motorcycle chasis.... it explained the basics of suspension, single pivots and sp with likage actuated shocks. But in reality i think you could quite easily grasp what the basics are about. Think of doing a biceps curl, from the begining when yore arm is straight down until it comes up to parrallel to the floor it gets progressively harder to lift. Then when you pass 90 degrees it gets easier as your arm comes closer to your chest.
It all comes down to the shocks shaft moves straight while the rocker of the shaft moves in a circle. That means that the shaft will not just compress but also rotate which means that the rear wheel will move through its travel while the shock compresses and rotates. Which means that the leverage of the wheel against the shock will be more than that the average leverage rate is. As the rotation of the shock decreases and the rocker starts to move to 90 degress to the shaft will mean that the leverage rate decreasses.
So with short links you can get way more progressiveness since they will have to rotate more to make the shock move the same ammount distance. And with long links or swingarms its easier to keep the ratios closer to linear.

This just what i have grasped of suspension kinematics and i am by no means any kind of suspension guru just decently good at math and geometry.

 

[LMC]

So are all the bikes on that chart.

The blur's curve does strike me as a little wierd, seems like it would amplify the negative traits of an air shock.

 

[TrumbullHucker]

now hopefully Howlin' Mad Murphy will protect me again through the coming flamage.

I created a ROUGH 2-D sketch on AUTOCAD 2009 of my frame that WAY down the road i would love to weld and build up

And i have purchased the Linkage program ( awesome program, very cool and fun )

here!
(click for bigger resolution)


and here are the Linkage graphs

By joscerb55 at 2010-02-20

By joscerb55 at 2010-02-20

By joscerb55 at 2010-02-20

By joscerb55 at 2010-02-20


I am really curious to see what you guys think
about how sh*tty everything is

 

[mtg]

No flaming, just some advice:

Your leverage ratio is higher than 3, I would try for much lower. It also is fairly flat until it falls off near full bottom out. Riding wise, it would feel normal through the travel and then blow through the very end, unless the shock ramps up a lot. The Bullit you have overlaid is a falling rate design- generally a bad idea. However, the leverage rate is much lower, which is good.
Personally, I prefer a progressive wheel rate, so with a coil shock that behaves pretty linearly, the leverage rate would decrease throughout the suspension travel, and near full bottom out, decrease at a faster rate.
If you want to keep a linear wheel rate, then I'd lower your leverage rate, keep it flat as it is, but get rid of the increase near bottom out.

To change the leverage rate, I would raise the shock mount on the swingarm and lower the mount point on the frame- that should help with both leverage rate and curve shape.

The pivot point concentric with the BB makes for a forward axle path, as shown in your graph. This will make it harsh on square edge hits. Right now, your rear wheel moves 2" forward once it is 8" into the travel

It sounds like a cool project, have fun.

 

[davec113]

No flaming, just some advice:

Your leverage ratio is higher than 3, I would try for much lower. It also is fairly flat until it falls off near full bottom out. Riding wise, it would feel normal through the travel and then blow through the very end, unless the shock ramps up a lot. The Bullit you have overlaid is a falling rate design- generally a bad idea. However, the leverage rate is much lower, which is good.
Personally, I prefer a progressive wheel rate, so with a coil shock that behaves pretty linearly, the leverage rate would decrease throughout the suspension travel, and near full bottom out, decrease at a faster rate.
If you want to keep a linear wheel rate, then I'd lower your leverage rate, keep it flat as it is, but get rid of the increase near bottom out.

To change the leverage rate, I would raise the shock mount on the swingarm and lower the mount point on the frame- that should help with both leverage rate and curve shape.

The pivot point concentric with the BB makes for a forward axle path, as shown in your graph. This will make it harsh on square edge hits. Right now, your rear wheel moves 2" forward once it is 8" into the travel

It sounds like a cool project, have fun.

I agree with everything mtg said.

Check out other popular sp designs, like the Session 88, 303R, Commencal, etc in your program to get an idea of how you can improve on your design. I am biased because I own a Session 88, but I think their execution of a sp is the best on the market. The 303R, with it's rail to control leverage rate, is probably the closest to ideal in that area, and the pivot point is well chosen too.

 

[TrumbullHucker]

alright, got some very good advice in this thread, with peoples thoughts and these helpful links. like woah.
I also am getting a very firm grip on the program Linkage. what a great program! i tried trial versions years ago and i thought it was just a program that had pre-loaded frames that you can look at. but changing polygons around and adding and all the graphs. its great! super helpful!

once i get a solid single pivot frame design, I will do a quick 2-D sketch on AUTOCAD, then really put the time in and make her into 3-D, with correct dimensions and all

then invest about 2 grand into stepping up my parents' house to a 200 amp service( they still have only 100AMP, SUCKS! ) then another 2 grand on a welder, then another 1500 on a decent little milling/lathe combo
PSHH

p.s- not including the time making the frame jig!

 

[Slater]

Welding is honestly one of the easiest parts of building your own frame. I would spend a lot more time considering how the pivot hardware is going to go together, how the derailleur hanger is going to integrate into the dropout, how the rocker is going to be made, how you are going to bend the chainstay tubing, etc.

 

[jeremy_2640]

alright, got some very good advice in this thread, with peoples thoughts and these helpful links. like woah.
I also am getting a very firm grip on the program Linkage. what a great program! i tried trial versions years ago and i thought it was just a program that had pre-loaded frames that you can look at. but changing polygons around and adding and all the graphs. its great! super helpful!

once i get a solid single pivot frame design, I will do a quick 2-D sketch on AUTOCAD, then really put the time in and make her into 3-D, with correct dimensions and all

then invest about 2 grand into stepping up my parents' house to a 200 amp service( they still have only 100AMP, SUCKS! ) then another 2 grand on a welder, then another 1500 on a decent little milling/lathe combo
PSHH

p.s- not including the time making the frame jig!

Nice one mate, you're on the right track now. I suggest you scrap the Autocad idea tho, it's a waste of time. The reason I say this is it isn't parametric i.e. it's a pain in the ass to modify stuff. Get yourself a evaluation or student copy of Solidworks or Inventor (but preferably Solidworks, I've worked with both and Solidworks rips Inventor) and do all your sketching in that. That way you can then use this sketches to drive your final model. It will be far quicker and easy to modify.

Have fun with the build! If you can, invest in a small CNC mill - it will open up a world of possibility.

Jeremy

 

[TrumbullHucker]

UGGHH i would jizz if i had solidworks.
i did classes for autocad so i know how to do it extremely well

i would love to learn how to use the solidworks program. how difficult is it compared to autocad??

cnc would be too expensive for me buddy

 

[tacubaya]

Solidworks is VERY easy to learn, but it's a mid level CAD/CAE program and Autocad is low end so it is going to be a bit more difficult.

I know how to use Pro/Engineer and Solidworks and although the ProE WF5 is pretty fast and good, I like to work in Solidworks.

 

[mtg]

Yeah, another vote for Solidworks. You should be able to come across a student version for free. Don't even bother with 2D CAD, it's almost a complete waste of time. Solidworks is 3D, parametric, easy to use and very powerful.

If you get really ambitious, COMSOL gives out free trial versions of their multi-physics software for FEA analysis, plus more capabilities than you could ever imagine.

 

[davep]

UGGHH i would jizz if i had solidworks.
i did classes for autocad so i know how to do it extremely well

i would love to learn how to use the solidworks program. how difficult is it compared to autocad??

cnc would be too expensive for me buddy

They have a free 90 day trial now...they call it their stimulus package. http://www.solidworks.com/sw/esp/esp_download.htm

Last time I checked, they had no student or trial downloads so I grabbed autocad because I could get a trial. In my little experience with autocad, SW is much nicer (and easier IMO) to use...but I do know SW much better.

 

[Kanye West]

SW and AutoCAD aren't even a comparison in my opinion. The real applicability of each one is wildly different.

 

[HAB]

Autodesk Inventor does have a free student version, and it's pretty good.

 

[fluider]

I haven't tried SW yet, although my vote goes for Inventor because I've been using it to my satisfaction for the last 2 years. The only major disadvantage of it I see it's not capable of creating and/or handling NURBS surfaces (if you want to play with organic shapes...).

 

[TrumbullHucker]

awesome, thanks for that 90 day trial link, im trying it out!

going to try to do the Xpress stress analysis thing too, i doubt i will be able to learn the program in 3 months.
the reason why i am standing behind Autocad is because i took a class for 2 years, so i know the program inside and out, probably more then some people on here

 

[tacubaya]

awesome, thanks for that 90 day trial link, im trying it out!

going to try to do the Xpress stress analysis thing too, i doubt i will be able to learn the program in 3 months.
the reason why i am standing behind Autocad is because i took a class for 2 years, so i know the program inside and out, probably more then some people on here

Well yeah, but it's equivalent to knowing inside out a piece of turd... you may be a turd expert but it's not useful.

3 months to learn Solidworks is very doable...

 

[davec113]

I learned Solidworks in school, and now have to use autocad, and it sucks. I'm just doing 2-D architectural-type work, it's ok for that, but I'd much rather use SW for 3D.

 

[fluider]

I'd recommend you:
1) don't stick with one software (especially 2D CAD) for many years. While you're younger, you have better ability to learn. So learn .
2) CAD modeling, simulation and analysis moved to 3D. Became more complex and interactive. 2D CADing is history.

 

[Handy Man]

Use the walkthroughs that come with solidworks to learn it. They are easy to follow and very good. I taught myself to use it in less than a week, but I already knew how to use ProE so some carried over.

 

[TrumbullHucker]

awesome guys, yeah i will be using the walkthroughs and "help" section alot

work is cancelled due to northeast snow right now, and i got some beers, so you know what i am going to be trying today ( and no not jerking off upside down with a belt around my neck )

ill let you guys know how my solidworks progress is going

 

[mtg]

so you know what i am going to be trying today ( and no not jerking off upside down with a belt around my neck )

Hilarious

 

[LMC]

alright, got some very good advice in this thread, with peoples thoughts and these helpful links. like woah.
I also am getting a very firm grip on the program Linkage. what a great program! i tried trial versions years ago and i thought it was just a program that had pre-loaded frames that you can look at. but changing polygons around and adding and all the graphs. its great! super helpful!

once i get a solid single pivot frame design, I will do a quick 2-D sketch on AUTOCAD, then really put the time in and make her into 3-D, with correct dimensions and all

then invest about 2 grand into stepping up my parents' house to a 200 amp service( they still have only 100AMP, SUCKS! ) then another 2 grand on a welder, then another 1500 on a decent little milling/lathe combo
PSHH

p.s- not including the time making the frame jig!

i would just get someone to CNC for you for the first few times. Draw it up and then find a reasonably priced local machine shop. Or ask the engineer at your local college to do a bit of work on the side. or download emachineshop software (although i find them to be expensive)

Bear in mind when drawing parts how the machine will be able to locate edges, clamp parts in place etc.

 

[essenmeinstuff]

the problem with the Trek is that while beeing a single pivot, it doesnt act like one when the suspension compresses since it has a floating shock, where one side compresses and the other decompreses.

I'd just like to point out that this is garbage.

 

[4130biker]

Hey @TrumbullHucker, did you ever pursue this any further?

May start a thread myself as I design my first dually.

 

[TrumbullHucker]

Hey @TrumbullHucker, did you ever pursue this any further?

May start a thread myself as I design my first dually.

no i never pursued any further, i still have been in the welding field, and with about 3 years of CNC operating 2 years VMC programming; but i have no engineering background
quite insane that you quoted this thread. couple weeks ago I fired up solidworks and started sketching around ( its going to be a long winter )
my original 3d model is gone, and i started sketching some new DH frame with the same dimensions as a large DHR to just screw around with.