Automotive Transmission simulation

[UZengineer]

Hi all,

I am an undergraduate mechanical engineer, working on 'Dual-CLutch Transmission' as my final year project. I was wondering if anyone could help me with simulations with MSC MD ADAMS software, more specifically, how to use the contact joints, gear joints and step functions?
I would really appreciate the help.

Cheers!

 

[GregLocock]

STEP5 works exactly like it says in the manual.

Contacts are a bit tricky, what sort of bodies are you trying to detect contact between?

gear constraints are conceptually weird but work like the manual says, in my experience.

Incidentally are you using VIEW or one of the more specialised front ends?



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[UZengineer]

Thank you for replying mate

I am trying to detect the drive and driven gears. Also I'm trying to substitute a body for synchronizer function which can attach with one gear to rotate it on the same speed, and then move on to other for the same task.. What kind of functions can I use for this particular job?

And, can you please tell me an example command for step5 function?

I really appreciate your time! Cheers

 

[UZengineer]

And yes, I'm using ADAMS VIEW

 

[SomptingGuy]

VIEW!!!

I used Emacs when I was an ADAMS jockey. View just provided pictures for the marketing people.

Step functions are bad, really bad. They aren't continuous (unlike real life). Consider using the ATAN function, which behaves like a step, but is continuous and can can be scaled to behave just like a smooth step.

- Steve

 

[SomptingGuy]

Oh, BTW, student posting isn't allowed here. Your post may vanish.

- Steve

 

[UZengineer]

Oh okay...

Thanks for your help. Can you tell me how to use ATAN function, or is it more or less like STEP function?

Thanks once again

 

[SomptingGuy]

Digging through my notes from '94...

ATAN(nx)/PI+0.5 <=> STEP(x,-del,0,del,1)

Where n is constant and large values of n correspond to small values of del.

This is dead simple, but very integrator-friendly.

You can simulate a backlash function by combining a couple of these functions. If I remember correctly, ADAMS had/has a "BISTOP" function, whhich was just a couple of discontinuous STEPs flipped and added.

- Steve

 

[UZengineer]

Cheers mate

 

[GregLocock]

STEP5 is a smoothed step function. I do not know if it has any advantage over SG's approach

http://ti.mb.fh-osnabrueck.de/adamshelp/mergedProjects/solver/f_hlp/functions/func_step5.htm

The exact same arguments work for STEP, which provides true step functionality, at the cost of killing your model, or upsetting it very badly in the best case.

People do seem to be using VIEW a bit more than I'd expect for big models, personally I enjoy using it to rough things out, but often end up exporting the adm code and using that as the basis for the final model/DOE or whatever.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[SomptingGuy]

The discussion digresses...

I found that any piece-wise function in a stiff system caused the ADAMS integrators to lock up. The ADAMS people always suggested using a different integrator (never worked). The only solution I found was to make all functions continuous (in all differentials).

I think, although I never did any real vehicle modelling, the Pacejka "magic" tyre model uses ATAN or something similar.


- Steve

 

[UZengineer]

Guys, please assist me on 'gear constraints' if possible? I'd really appreciate you help... Well, I don't know if VIEW is better or worse than any other simulation package, but for now its a compulsion for me to use it, so I'll have to manage things within the functions and limitations of ADAMS/VIEW.

Thanks

 

[UZengineer]

And are there any online catalogues available for these functions? I mean any functions manual or something?

Cheers

 

[GregLocock]

Yes of course there is, the help manual for the SOLVER lists them all in their byzantine complexity.

So click on help, then ADAMS/help which opens a page in your browser, then in your browser click basic package, solver, c++

In the case of gear we get the following, in effect a gear pair is assumed to be a pair of frictionless slip-free wheels in contact, and the only point of interest is where they mesh:

GEAR
The GEAR statement defines a gear pair constraint. Examples include a spur, helical, planetary, bevel, and rack-and-pinion gear pairs.
Format
GEAR/id, JOINTS=id1,id2, CV=id
Arguments

CV=id Identifier of the marker that designates the point of contact and implicitly determines the ratio of the two gears. The z-axis of the CV marker must point in the direction of common velocity at the point of contact. The direction of common velocity is tangent to both pitch circles and would be normal to the tooth surfaces of the gears if the pressure angle was zero (Adams/Solver (C++) always assumes a zero pressure angle). The CV marker must belong to the carrier part that also hosts the J markers of the joints.
JOINTS=id1,id2 Specifies the two joints whose relative motion is to be related with a gear. Each of these two joints must be translational, revolute, or cylindrical. Note that both J markers of these joints belong to a carrier part that must also host the CV marker.

Extended Definition
The GEAR statement uses the location of a common velocity CV marker to determine the point of contact of the two gear pitch circles. The direction of the z-axis of the common velocity marker indicates the direction of the common velocity of points on the gear teeth in contact. This is also the direction in which gear tooth forces act. Note that the CV marker has constant position and orientation in the carrier body coordinate system (BCS). The CV marker through its z-axis only provides a direction, and there is no need to be more specific than this. In this context, there is no need to indicate things such as how the power flow occurs through the gear.
The reaction force reported back to you for a GEAR element is the reaction force measured on the J marker of the joint that is specified first in the GEAR definition. It is important to keep in mind that the J markers for both joints associated with the GEAR element must belong to the carrier part (the part that hosts the CV marker).


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[patprimmer]

Greg

Maybe these features only come on legal registered copies as clicking help seems extremely obvious if it was available.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[SomptingGuy]

ADAMS is a monolithic hunk of old Fortran with a GUI slapped on top. It's likely that the precise definitions of values in the "input deck" have been lost as on-line help has replaced the precise block definitions of old.

(We only provide the equivalent "blocks" manual on demand these days for our software.)

- Steve