5

In an answer to another questions @Ceshion introduced a fascinating kind of valve train that I have never heard of, desmodromic valves, as introduced by Ducatti. That answer is here:
What are the various kinds of tappets / lifters and the advantages / disadvantages of each

and here is the image she provided as an example:

Desmodromic valves example

Here are the specific questions:

  • Are these only used in motorcycles, if not, where else is this used?
  • Given the complete lack of valve float, just how high can these rev? What is the next limiting factor in how high they can rev (ignoring the rest of the engine for now. Unless the only other thing that would rev limit is something in the bottom end.)
  • How much more or less reliable are these than the usual modern valve train setups?
  • Are these restricted to racing or are there street applications for these?
  • Does anyone but Ducatti use these or are they patented by Ducatti?
  • Given the lack of valve float, how much bigger and heavier can the valves get to allow more flow and more power?
  • From the looks of it, the position of the valve is totally dependent on the cam followers for both opening and closing cam shafts putting it exactly where it needs to be. There are no springs or hydraulics to take out any slop in the system. How does a system like this account for wear on the mating surfaces? Is there any adjustment in a system like this? Do they leak more than the usual setup because it's hard to get the closing and opening cams to position the valve for a good seal?
  • Does a system like this require higher oil pressure to prevent premature wear as the revs go higher?

The high revs means so much more power, there have to be down sides to this for them not to be in use in a lot of places.

Thanks in advance!

2

I can actually answer some of this question as well. There are other companies that use them, while Ducati is the best known. Ducati uses them in their racing as well as streetbike engines. Mercedes Benz is another company that uses them, namely in F1 engines. The theoretical limiting factor, the best I understand it, is heat and friction. There's no practical reason to work the engine so hard as to reach those limits. They are somewhat less reliable, due to the complexity and number of parts that need to stay in sync, and power delivery is similarly less reliable. There are, in fact, springs in most desmodromic valve setups, they just aren't used for cam following. They are, however, used to reduce slack in the system, as can be seen in the following:

enter image description here

This also shows an alternative setup to what I described in the other post, in that this example uses the same camshaft for both opening and closing the valves. They are adjusted with shims, labelled in the diagram above. This is what makes them such a nightmare to service, since there are two clearances on each valve that interact with each other.

Now for what I don't know: I'm not sure how the oil pressure needs compare to conventional valves, but keep in mind that in most cases as engine RPMs go up, so does oil pressure. So it may be necessary, yes, but it may also be produced somewhat automatically as a byproduct of such high speed. Also, I'm not really sure how potential valve sizes compare, however I haven't personally seen a huge difference in valve sizes between desmodromic and conventional setups.

  • +1 more great info thanks! I was thinking about the power as RPM increases. Just to feed the engine as RPM goes up means either the velocity of the air from intake to exhaust increases for a constant sized path. And I think the velocity is free just because the RPM goes up. At least to a point. Or, the diameter of the path would have to increase to allow more air and fuel at a lower velocity. I know from nothing about which would be better for either street or racing applications. – cdunn Mar 19 '16 at 19:43
  • You would likely damage the engine from overheating or bearing friction before you reached the maximum breathing capability of the engine, however that's also assuming a fuel delivery system that can keep up with the airflow. Basically, there are several potential limiting factors, but the only ones that you can't really work around are heat and friction becoming excessive for the materials used. – Ceshion Mar 19 '16 at 20:25
  • That is, besides using progressively stronger materials, I'm sure with an infinite amount of money you could build a reciprocal engine that spins as fast as a turbine, but you'd need the specialized bearings used in such high speed applications (fluid or electromagnetic), as well as a very efficient cooling system, in order to avoid damage. – Ceshion Mar 19 '16 at 20:28
  • You didn't really answer the questions though :-( – Ppoggio Mar 20 '16 at 4:39
4

You asked

Are these only used in motorcycles, if not, where else is this used?

No other manufacturers of motorcycles currently leverage the desmodromic system.

Given the complete lack of valve float, just how high can these rev?

The Ducati GP7 MotoGP platform of 2007 revved to as high as 19,000 RPM.

What is the next limiting factor in how high they can rev?

Perhaps a shorter stroke with larger valves and less reciprocating throw in the valve will allow the next increment in RPM. Currently the intake tract is reaching the speed of sound which imposes issues of back pressure and stalled air pockets within the intake tract. The current GP15.X and GP16 RPM's numbers are, as of yet, unpublished by Ducati but speculation has them at 20K + RPM's.

How much more or less reliable are these than the usual modern valve train setups?

The have more moving parts, we can surmise that reliability is tied to simplicity. My own experience with Ducati Desmodromic valve trains professionally has been very good. Statistics for failure rates are not available but I believe we would see that they have an overall lower tolerance to long term stress cycles than a standard under the bucket shim valve train that is used in most high performance modern motorcycles.

Are these restricted to racing or are there street applications for these?

Ducati uses this valve train in all of it's production and racing vehicles.

Given the lack of valve float, how much bigger and heavier can the valves get to allow more flow and more power?

I would defer a response from a mechanical engineer as the mathamagic get's pretty complex surrounding the stresses at such high RPM's. Additionally, all of the weight and angle information would need to be published to answer this question.

Does anyone but Ducatti use these or are they patented by Ducatti?

No other manufacturer currently uses this system. Mercedes used them in F1 in 1954 and 1955. Ducati has patents on various components but this design goes back to steam engines in late 19th Century.

Here is a fairly comprehensive list of of true uses of a desmodromic systems in internal combustion engines.

How does a system like this account for wear on the mating surfaces? Is there any adjustment in a system like this?

Older two valve Ducati SR2 engines suffered from increased wear with particular poppet valve angles. Current and legacy platforms use a helical spring on the negative rocker to prevent bounce and seal the valves properly upon closure. Here is the modeling on a Ducati design.

Does a system like this require higher oil pressure to prevent premature wear as the revs go higher?

Oil pressure tolerances on modern platforms are within normal smaller rocker arm designs of the past.

Downsides

  • More mass

  • More moving parts

  • Increased inertial mass to the overall ICE

  • Complexity of maintance/valve adjustment

  • Manufacturing costs are higher

  • More weight above the center of gravity of the motorcycle

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.