2 speed transmision evolution----

[liveforphysics]

who let the dogs out?

fruit of tonights labor.
The deeper i get into this project the more I remember why I hate production........the up side is: repitition builds speed.
progress continues.

Hell yeah! I LOVE dogs! Tough to tell from the pics, but are you running any back-cut?

 

[12p3phPMDC]

Does back-cut make the profile of the dog look like a dovetail?

 

[liveforphysics]

Does back-cut make the profile of the dog look like a dovetail?

Yep.

This is part of the gearset I run in super-civic. It uses 8deg back-cut dogs, and you see see the dovetail-ish look in this picture.

Back-cuts are not a critical thing at all with dogs, and I think Thud's design uses a spring on the shift fork arm to ensure it stays deeply engaged, which is just fine and will work perfect. I just like back-cuts because as soon as you start to shift, and the tips of the dog's touch, it slams the dogs together and instantly causes full engagement so you never have the chance of just tip-loading the dogs, which can break them. It also ensures they never slip out under load. Thud's design here is excellent, and back-cut or no back-cut, it's never going to be failing, and with the spring on the shift arm it's never going to be falling out of gear either, so he's all ready got those things solved by skinning the cat a different way.

 

[Thud]

I'll have 1 deg of back cut in them when finished.
That'l maintain engagment.
They will get a case hardening treatment before assembly.
These things are turning into a watch makers nightmare!
you guys will enjoy the details when i get them all posted...

(hmmnn thud thinks "Instead of manufacturing, I will post the drawings & you can get them made till your hearts content") LOL.

I am certain any one who trys a 2 speed tranny like this will never want to be without one.
I will have all the components for the 1st batch of 5 this week. Another week or so & I will have eye candy for the regulars to fight over.
any one curious on pricing, I'll say..... it is not quite set yet..... but the initial run is all hand built by me.
They are going to be in the range of a big Astro motor though for certain. 4 times as many parts & a butt load of labor. (but it nearly doubles the operational range of that big money motor 8) )

Nice gears there Luke, Most of my moto trannys are only 2Deg. much more & you really have to back off to get the next gear.

 

[liveforphysics]

Nice gears there Luke, Most of my moto trannys are only 2Deg. much more & you really have to back off to get the next gear.

The shifter has a micro strain-gauge bonded to it. The moment it feels 5lbs of force tugging on the shifter, fuel and spark are cut for a 0.05second interrupt window. This is just enough to cause a brief moment of no torque loading to let it shift effortlessly and consistently, then hits it with full power again so quickly you hardly new it ever had an interruption. I take full advantage of every electronic advantage I can get in super-civic.

This guys runs my same gears (8deg back-cut) with the same interrupt setup. You can see it lets you shift well, you can even go from 10,000rpm in 2nd gear back to first... lol Can you say total engine destruction? lol
[youtube]MWUbgGarhdc[/youtube]


But I don't want to get your awesome thread OT here.

 

[12p3phPMDC]

Cool, thanks for the info and pics!!

We ended up with a 50ms ignition cut when we built a solenoid based shifter for an R6 sport bike engine.
That was the lowest cut we could get with that setup and have it work reliably.

As you have already said Luke, folks that wanted to shift that R6 engine lightly ended up rounding the dogs....we
rebuilt that sucker it stopped popping out of gear under load. It would have been cool
to try and regrind them possibly with a little back cut... They may have been too hard though.
But, when you do it quick it, it just runs great.

Are the dog "pockets" also back cut?

Thud,
do you send out all the parts to a hardening service or do you DIY too?

 

[enoob]

(hmmnn thud thinks "Instead of manufacturing, I will post the drawings & you can get them made till your hearts content") LOL.

yaa hoo . you open source diy guru you. you and frankg might be brothers you know ? adopted by any chance?

I am certain any one who trys a 2 speed tranny like this will never want to be without one.
any one curious on pricing, I'll say..... it is not quite set yet..... but the initial run is all hand built by me.
They are going to be in the range of a big Astro motor though for certain. 4 times as many parts & a butt load of labor. (but it nearly doubles the operational range of that big money motor 8) )

Chuck em in the for sale section and let the open bidding begin. cancel my order for the light sabre i want one of these and will aggressively bid accordingly :wink:

 

[liveforphysics]

I lust for your 2spd as well.

What are the ratio options available?

 

[Thud]

12p,
on small parts like these I will surface harden them in a small forge using "Kaseinte" it builds a very thin surface & won't require any extravagent tempering schedual.
Yes, the mating surfaces will be back cut to match the dogs.
(my left toe has .04 interupt for max acceleration! :lol: ) I am allways amazed at the length the straight line racers will go for .05 seconds of advantage. We motocrossers just put on sharp tires & train for a 30 min. fight (Warning! old fat guy spewing nostalgia)

Luke, Right now the size is optimised for a 1.87-1 reduction in low gear & 1-1 in high. Trying to keep it as small as possible & still handle 6 kw peak. For a bicycle or light Kart/trike, that should satisfy 99% of the applications.
(The road racer build will be a carbon fiber one off!)

Enoob,
FrankG & I would have a good time out on the bass pond for certain.(you 2 I suspect) Can't think of much we couldn't do if we set out too.

Miles,
How are the .dxf drawings imbeded in the repository section?
Can pdf's be zipped so anyone can have a measured drawing?
thanks in advance. T

 

[Miles]

Todd,

PDFs can be uploaded uncompressed. Other files need to be zipped, I think. Just attach to a post in the normal way.

 

[12p3phPMDC]

Kasenite eh?

Cool stuff....just use it in a well ventilated area it seems due to cyanide content

We built the shifter for the fun of it and for learning...which we did learn alot. Plus, with a bunch of newbees shifting, the dogs
kept getting rounded, and we wanted to make that go away via ignition cut and high speed shift.

We need a solenoid shifter for the Thud 2 speed !!!

 

[amberwolf]

I(hmmnn thud thinks "Instead of manufacturing, I will post the drawings & you can get them made till your hearts content") LOL.

I would not complain if you did that. I don't think I could make it myself yet, but eventually, I might. Probably could do it myself before I could afford one, based on the estimate.

The shifter has a micro strain-gauge bonded to it. The moment it feels 5lbs of force tugging on the shifter, fuel and spark are cut for a 0.05second interrupt window. This is just enough to cause a brief moment of no torque loading to let it shift effortlessly and consistently, then hits it with full power again so quickly you hardly new it ever had an interruption.

I think that's what Naeem needs. I don't know if he can do it but I'm going to crosspost this quote over in his thread.

 

[liveforphysics]

The shifter has a micro strain-gauge bonded to it. The moment it feels 5lbs of force tugging on the shifter, fuel and spark are cut for a 0.05second interrupt window. This is just enough to cause a brief moment of no torque loading to let it shift effortlessly and consistently, then hits it with full power again so quickly you hardly new it ever had an interruption.

I think that's what Naeem needs. I don't know if he can do it but I'm going to crosspost this quote over in his thread.

The car version of this setup wouldn't fit the "low-cost" model for Naeem's bike.

$7,200 for custom dog-shift gearset.
$2,000 for stand-alone with shift interupt capability.
$1,800 + 1,300 for axles and clutch that can last a few events.
$1,000 in misc parts, sensors, and tuning.

$13,300 to shave 0.4 seconds off your quarter mile (and most important, the added consistancy and reliability) is an acceptable price for grass-roots budget drag/street racing setups. I think an E-bike option for pull-n-pray strain-sensor WOT shifting using Thud's tranny as a base platform could be put together for maybe $1,000. It would require Naeem to change to dog-shifting, and I don't see him springing for that quite yet. Maybe he can put together a low budget solution to make it happen.

Here is some in-car footage of how fast this sort of setup shifts, and how smooth the intrupt goes. This was finals in Sacramento CA. He is stab clutching with an interupt rather than full pull-n-pray. He is around 1,100hp for the 1/8mile here, and it's awesome when you see the English racing DSM try to creep ahead, he hits the scramble-boost, putting him around 1,400hp, and walks that DSM effortlessly. Classic FWD back-halfing the snot out of AWD.
9.2seconds at 173mph. This is a street legal Honda civic hatchback that could race a $1,800,000usd Bugatti Veyron and be finished with the 1/4mile before the Veyron reaches the 1/8th... Street racing with this car against anything could be compared to sand blasting a soda cracker.

[youtube]wqdygPrgfkc[/youtube]

 

[amberwolf]

I wasn't really meaning the shifting itself, but rather the automatic shutoff of the motor or a power cutback during the shift, to keep from breaking a chain or fouling the derailers.

I thought of a simple way to do it that fits his resources and should work in that environment, which I posted in that thread:
http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=15256&p=227734#p227734

 

[etard]

Thud,
I thought this was interesting, they use Delrin for that middle piece which is designed to shear if something jams. Might make it easy to CNC that delrin for your dog clutches, would also make for more silent drivetrain too.

http://www.oepcouplings.com/content/oldham_couplings/oldham_couplings.asp

 

[Thud]

Progress report,
Its fun to find new limitations & work around them( cheap & flexable machinery) I have worked around the obsticals & can now machine the dog engagments properly. (to my satisfaction at least)
here is what they look like:

Before they are done, each driven sprocket will have been operated on with 4-set ups, & no less than 50 passes of the varius tools to make the cuts to depth. All hand cranked on an old smithy 3 in one mill.

Here is the back side, I will machine the back off to get a net thickness of .375"

These get a bushing to free spin on the shaft when not engaged.

Here is the handle I have worn all the chrome off over the last 15 years.You cant see all the copper that is showing in the photo.

 

[AussieJester]

Cheers for the progress report Thud, how have things progressed with the casting...or is that sore subject still LoL...

That is a lathe with milling attachment i'm guess? i would like something like that myself,
if i could find a good oldie at auction for an affordable price i would be as happy as a pig in shit
they dont make em like they used to do they

Best of luck mate lookin the bizz thus far...

KiM

 

[12p3phPMDC]

Looks good Thud!!

Thanks for posting those pictures... I need to pick up a rotary table for my 3in1 !!

How do you pull off the back cut? Custom ground tool or do you angle the table and not worry about
the 1 degree dig in the bottom of the pocket.

I like how you squared up the ends of the pockets
with the smaller cutter. I guess you could do the back cut at an angle and then at the end, square it up and make the
finish passes.

 

[Thud]

Thanks 12,

In the set up pictured I have the indexer shimed to give me the back cut angle. (its a 2nd set up for the angle on the other end of the pocket) the dig at the bottom is not an issue as the back will be gone when finished.

My rotary table is also a budget item. The tiny bit of backlash raises cain when cutting into the 1/8" grooves on the ends. I have to pre-load the table with a tommy bar to keep from snapping bits on entry into the previous cut area. It works & is a great work out LOL.

AJ, I have the casting issues worked out. just need to budget the time to make a few more patterns & fire up the foundry again. I need to cast my dynomometer parts, a couple parts for the motor build. tranny cases, hmnnn i must be forgetting something.....

 

[Miles]

My rotary table is also a budget item. The tiny bit of backlash raises cain when cutting into the 1/8" grooves on the ends. I have to pre-load the table with a tommy bar to keep from snapping bits on entry into the previous cut area. It works & is a great work out LOL.

My rotary table doesn't even have a screw - I do all the control with a tommy bar...

 

[12p3phPMDC]

I'm guessing that a tommy bar is a big lever that you bolt to the rotary table to help load it?
The rotary table doesn't have a clamping lever on the perimeter to keep it still? or is that not enough.

Do you guys have the indexing plates?

I've been shopping around for a bit, but haven't pulled the trigger on anything yet..
Grizzly imports and Enco seem to have a decent little 4 or 6".

So, is it the side load on the cutter that snaps them ?

Please post a pic of the dogs when you get there.

If I remember correctly, you basically used cylindrical drill rod dogs on the proto?
Whereas this time, they will be more tooth like?

I believe that you mentioned before that you will hone the bushing to the shaft as well?
Was that a bronze bush?

Sounds like fun!!! thanks for dealing with all my questions!

 

[Thud]

oh Miles,
A fully manual rotory table, same circa as your Myford? I love the well kept antique machinery, not that there is much "new" in the world of lathes or manual mills. I'll bet you are easyer on tooling with that than the "chatter box China special" I am playing with. :lol:
I now apreciate(& really miss) the Brown & Sharp 12" I had access too. The guy traded off that machine (bridgeport & accesories)for something goofy IIRC

 

[Miles]

It's older than the Myford, I think......

It's a Taylor Hobson indexing table with integrated tilting. Full manual control via the allen key wedged into the T slot

 

[Thud]

Here you go 12:

I'm guessing that a tommy bar is a big lever that you bolt to the rotary table to help load it?
The rotary table doesn't have a clamping lever on the perimeter to keep it still? or is that not enough.

Do you guys have the indexing plates? For what little real indexing I do, I have a couple plates I customize as required. 99% I get done with the cheap spin indexer you saw in the photo

I've been shopping around for a bit, but haven't pulled the trigger on anything yet..
Grizzly imports and Enco seem to have a decent little 4 or 6". A true dividing head is a really nice pc of equipment & will come with a tail stock & indexing plates. like anything It costs money.

So, is it the side load on the cutter that snaps them ? The cutter will "grab" in the open ends of the slots,& climb into the groove using up all the "slop" in any fixture, putting all kinds of load on the tiny flutes I am using. I call it "sudden catostrophic acceleration" & is a factor with any kind of back lash in your lead screws also(interupted cuts are a pain compaired to having the cutter buried in material)

Please post a pic of the dogs when you get there.Will do

If I remember correctly, you basically used cylindrical drill rod dogs on the proto?
Whereas this time, they will be more tooth like? The dog hub is machined from a solid block for a couple reasons. Mainly to keep integraty of the dogs & still have the deep groove for the shifting fork to ride into. May make more sense when I post the pdf drawings of the thing.

I believe that you mentioned before that you will hone the bushing to the shaft as well?
Was that a bronze bush? Yes Bronze on the free wheeling sprockets.

Sounds like fun!!! thanks for dealing with all my questions! no prob.

 

[AussieJester]

AJ, I have the casting issues worked out.

Good to hear mate eeeeexcelent...

, hmnnn i must be forgetting something.....

Stocking up on blue dye maybe

Look forward to next updat emate