Bishop Rotary Valve...

[HotRod Todd]

Has anyone here considered abandoning the trusty ol' poppet??
Who can argue with controlled burn durations approaching 20CAD (10-90%),
No pesky piston-valve cutouts,
Therefore huge CR if you so desire?
Port flow profile at your command....
'dual tumble' charge filling and resultantly expedient flame propagation!!!
Dreams???
Yep.... mostly wet.
But.... Applicability to FSAE??

Discuss!

[AxelRipper]

Well they don't use them in Formula 1 so they can't be good/work... http://fsae.com/groupee_common/emoticons/icon_razz.gif


(yes, I know they're banned in the Concorde Agreement)


I've always been interested in them, especially the Coates setup ( http://www.coatesengine.com/csrv-system.html ). Had actually contacted them seeing if they'd sell a set for my truck (since one of their test motors was a small block Ford also) but they're focused on industrial generator applications.

Would be interesting if you could make it work.

[Dunk Mckay]

Not sure it would be of as much benefit on a 600, but on a 450 it could work. Making a custom head for a single would also be a lot easier than for a 4cyl. i suspect this is probably one of those things that goes into the category of "I wish it was a little easier, a little cheaper, and a little more worthwhile doing this," along with custom made engines, CVT/IVTs, fully active suspension, etc. Do-able but the ROI would be very low in FSAE.

A team that can combine all the things in that category with total effectiveness, no reliability issues (nigh on impossible) and well trained drivers, would probably win all dynamic events, although not by as much as some might think. But pay the price in cost. Design judges may or may not be enthusiastic about the amount of effort involved relative to the performance gains.

[jlangholzj]

Originally posted by Dunk Mckay:
Not sure it would be of as much benefit on a 600, but on a 450 it could work. Making a custom head for a single would also be a lot easier than for a 4cyl. i suspect this is probably one of those things that goes into the category of "I wish it was a little easier, a little cheaper, and a little more worthwhile doing this," along with custom made engines, CVT/IVTs, fully active suspension, etc. Do-able but the ROI would be very low in FSAE.

A team that can combine all the things in that category with total effectiveness, no reliability issues (nigh on impossible) and well trained drivers, would probably win all dynamic events, although not by as much as some might think. But pay the price in cost. Design judges may or may not be enthusiastic about the amount of effort involved relative to the performance gains.

i think it could be more of a benefit on a 4 cyl than you chalk it up to be. Granted you're probably correct in saying that the most net gain would be in a single but I can think of some ways this could help out a 4cyl as well. I imagine that if they can hit 18K (on what I'm assuming is a SBF from their 5k limit) then a sport bike 4cyl would have no problem hitting upwards of 20k. Of course at that point it would probably choke out more but the added valve flow would help it make broader power at the higher RPM range vs. the single I'm assuming would make more peak power due to the RPM and flow advantages.

[Dunk Mckay]

Man I'd love to see someone push a R6 or CBR600 to 20k. i would imagine the greater benefits would be the ability to increase the CR, maybe up to 18:1? There's probably 10 or 20 ponies to be found there combining both, but with the amount of issues you're going to get with those condition that aren't valve related, you'll be changing your block once every 22k (hopefully not less) just to avoid catastrophic failure.

[tgman2]

There is a 450 dirt bike modified to use rotary valves same people that did the original bishop rotary valves. From memory it produced a fair amount more power pretty much across the rpm range. Admmitedly it was using almost the same valves as they had on their f1 engine. I would love to see someone drop this engine into their car and see what happens.

I can't remember his name off the top of my head but the guy that now owns the rights has formed his own small company in order to sell the technology.

[rjwoods77]

"Bishop Innovations' patent for the rotary valve engine was bought out by BRV Pty Ltd, owned by Tony Wallis, one of the valves original designers. BRV has constructed several functional motors using the rotary valve technology, such as a Honda CRF 450, which had greater torque at both low(17% increase) and high (9% increase) engine speeds, and also produced more brake horsepower up to around 30% more at functional engine speeds[10]. The engine was also considerably smaller and lighter, as the cylinder head assembly was not as large."

[Dunk Mckay]

30% extra. But that's unrestricted. Although the smaller and lighter part sounds much more interesting, and that'll be saved at the head, the highest part of the engine, so some nice CoG lowering there too. If you make your own you could easily play with the packaging a bit an get some favorable inlet and outlet opening so you can get the engine right up behind the driver, reduce your yaw moment. This is starting to sound more attractive. But I still stand by the fact that the ROI is gonna be too low for any real value in Formula.

[DougMilliken]

... But I still stand by the fact that the ROI is gonna be too low for any real value in Formula.
Another non-poppet valve that showed promise but was never fully developed is the Butterworth swing valve. Here's one link, there are many more including patents,
http://carfixguru.com/the-history-of-valves
Basically the idea is to swing the intake valve head out of the way of the incoming charge...

[Jan_Dressler]

Was about time for a rotary valve discussion again http://fsae.com/groupee_common/emoticons/icon_smile.gif
Maybe combine it with this (http://www.ilmor.co.uk/concept_5-stroke_2.php)...

[rjwoods77]

The whole point of this "valve" system is to eliminate the poppet to increase airflow. Sounds good but take it one step further and get rid of the piston as well and you have a rotary engine with even less intake restriction. I still wish they allowed rotary engines in FSAE like they did in the old days. If the old FIA x1.5 displacement factor was used this would be awesome to use especially with a cvt...

http://www.woelfle-engineering...rotaryengine_en.html (http://www.woelfle-engineering.com/Produkte/produkte_XR50rotaryengine_en.html)

Who ever designed this kart CVT setup is pretty clever because the secondary cvt actually houses a planetary reduction gear that has a 40mm axle (i.e big enough for a tulip snout) that goes through the whole thing. Connect this to the new Taylor Race or FSAE Drexel diff and it would be a real nice tight drivetrain package.

http://www.karting1.co.uk/cont...n-for-karts-test.htm (http://www.karting1.co.uk/continuously-variable-transmission-for-karts-test.htm)

http://www.youtube.com/watch?v=PWkpLrnAjEU

[Z]

Rotary valves have potential, especially for mass-produced low cost engines, but I don't think the development effort is worth it for FSAE.

Here (http://fsae.com/eve/forums/a/tpc/f/125607348/m/824105905?r=54520089051#54520089051) is an earlier post on a 40 year old, 20kg, air-cooled, turboed, 2-poppet-valved, 450cc single, that made ~125hp at ~8,000rpm.

Given that the maximum power through the 20mm restrictor is also ~125hp, why bother with anything else?

In fact, why not just copy that engine (well, 1/12 of it http://fsae.com/groupee_common/emoticons/icon_smile.gif)?

Z

[jlangholzj]

truthfully the BRV would even be a "safer" platform as well for a high-po engine. couple things come to mind:

-don't have to worry about valve float (YES i know F1 engines already can rev up to 20k but imagine how far THOSE would be pushed with a brv http://fsae.com/groupee_common/emoticons/icon_eek.gif)

-Assuming that a higher flow profile would have a smoother idle than its poppet counterpart. ie you don't need a big lopey cam to make power

-more convenient intake/exhaust port placement. Since the darn thing spins in a circle you could pretty much put your ports wherever you wanted to.

-tighter combustion chamber. there's no valves extending into the chamber so it would be very possible to make some MASSIVE compression bumps. Even if you stuck with something around 15:1, having the smaller chamber would allow for some tighter packaging in the motor.

-less source for failure, less moving parts

however I do agree in that the ROI would be hard to get over. It would take a couple years of Sr. design students to get the package put together and then I'm guessing another 2 years to properly implement it in a FSAE vehicle....but good lord that would go like stink

[Z]

As noted in the past, I am fully in favour of FSAEers pushing the envelope on "creativity and imagination". But I just don't see the benefits in rotary valves.

The main benefit (as mentioned by Jlangholzj above), is fewer moving parts, so cheaper and more reliable. I can see a low cost, no-user-serviceable-parts, B&S or similar lawnmower engine, with a "Ceramo-graphotex (TM)" coated rotary valve that lasts the life of the engine. When worn out (eg. the "Nikasil" coated bore gets scratched) then the whole lot is thrown away.
~~~o0o~~~

But regarding the other claimed benefits:

1. Higher RPM.
This is mainly set by mean piston speed, not valves. Any engine intended to last more than about an hour has a MPS ~25m/s or less. This figure applies to F1, Nascar, and almost every other race engine. Drag race engines go up to 40m/s (max!), but blow up after ~10 seconds.

If valve float is a problem, then either use stiffer springs, or pneumatic springs like F1, or desmodromic like Ducati. Desmo is more efficient than steel or air springs (ie. less friction, especially at low to medium RPMs)), so good for fuel economy.

2. Higher Flow.
How? Most rotary valves I have seen, or considered, have horrible flow paths most of the time, except for the brief moment when the hole in the "rotor" is aligned with the I/E pipes. Also, steady rotary motion implies that the flow path is "not optimal" for a lot of the time.

Once a poppet valve is open more than 1/4 x diameter, there is no more restriction from the valve than from the pipe-to-chamber transition alone. In fact, the valve head can help turn the flow around the corner. If the poppet is opened and closed quickly (eg. desmodromically), then it has maximum flow for a long duration.

"Variable Valve Timing", for good breathing, and hence good torque, at different RPMs, seems to be equally difficult whether using rotary or poppet valves.

3. Higher Compression Ratio.
This is really set by the fuel's octane rating. Low octane fuel needs low CR, or else detonation, and dead engine. A secondary (but still important) factor in detonation is the shape of the combustion chamber. Briefly, a good chamber is compact with no sharp corners or pockets. Poppet valves can give a very smooth "hemi-head" shape to the chamber's ceiling. Piston pockets for valves are not essential, especially with turbos and/or low B/S ratio (ie. not TOO over-square). Rotary valves do not (?) give this smooth hemi-head shape .

4. Easier Valve Drive.
This one I agree with. Just a simple chain drive from crank to rotary valves, similar to a SOHC or DOHC. Poppets then require some kind of opening buckets or rockers, and closing springs (or desmo-rockers). So those, and their attendant size/mass, are the poppets' main disadvantages.

Z

(Edit: Added more waffle... http://fsae.com/groupee_common/emoticons/icon_smile.gif)

[RenM]

Originally posted by Z:
3. Higher Compression Ratio.
This is really set by the fuel's octane rating.

Thats not the deciding factor for high reving 4 cylinder bike engines. Because of the the small stroke you get problems with valve clearance to the pistons when going for a high compression ratio. Even with normal gasoline you should not run into knocking problems.

[Z]

RenM,

"High reving 4 cylinder bike engines" have a bore/stroke ratio = ~1.5. F1 engines have B/S = ~2+, and a CR that, although undisclosed, is certainly very high (maybe = ~15+?). Diesel engines with approximately square B/S (ie. = ~1, or less) have CR = 16 - 25, and this includes their necessary bowl-in-piston design. My point here is that poppet valves are no restriction to high CRs, even with short stroke (high B/S F1), or pocketed piston (diesel) engines.

I reckon the reason sport bike engines don't detonate too badly is that they have long valve open durations to help their breathing at high revs. This means both I&E valves have to be considerably open during overlap at TDC, which thus forces either a lower piston top, or deep piston pockets, both of which make really high CRs a little more difficult, though not impossible.

Furthermore, the long intake duration means the intake valve closes a long time after BDC (ie. during the compression stroke). This means that at low revs, when there is no "ramming" effect, some of the air breathed in gets pushed back up the intake pipe, and this lowers the effective, or "real", CR. Hence no detonation at low revs. At high revs there is little time or distance in the small chamber for severe detonation to occur.

So does anybody know the I&E valve opening and closing angles for these engines?

Criticism of the above speculation is welcome. http://fsae.com/groupee_common/emoticons/icon_smile.gif

Z

[Jay Lawrence]

Z,

From what I have read, the increased CR possible for the rotary valve engine is primarily due to the improved chamber filling and swirling created by the rotary valve (a. because it has a much larger valve area and b. because swirl is imparted on it during intake (to a greater degree than poppets)). The figures I have seen for CR are up to 18:1.

My main concern with these engines (and I believe the main stumbling block in their development) is effective sealing. I imagine that sealing against 18:1 would require some hefty seal load on the valve and inherent frictional losses. Maybe (probably) I'm missing something.

[Z]

Originally posted by Jay Lawrence:
The figures I have seen for CR are up to 18:1.
Jay,

There have been countless millions of poppet-valved diesels built with CR = ~25:1 (mainly indirect injection diesels, DI have CR = ~16+). Swirl is common on diesels, and easily achieved with poppets via port angling. Likewise tumble.

My main concern is the excessive hype that is put forward in support of these things. It makes my BS antenae tingle, which is irritating!

An example is the Coates system linked in Axel's post above. It gives an example of a 5 litre V8 that can only get "5,700 RPMs" with its poppets, but the Coates rotary valves are good for "14,850 RPMs". I have only one thing to say to anyone trying to get a 5 litre V8 up to 14,850 rpm. DUCK!!! http://fsae.com/groupee_common/emoticons/icon_eek.gif

The Coates site has this animation (http://www.coatesengine.com/files/theme/csrv_vs_poppet.html) of rotary vs poppet valves. I see,
1. Two systems of very similar overall size.
2. A better shaped combustion chamber under the poppets. The RVs have two deep pockets/ports between the chamber and RV/seals.
3. A torturous flow path through the RVs.
4. The RVs seem to be "fully open" for only a very brief moment.

Anyway, I don't want to sound too negative (http://fsae.com/groupee_common/emoticons/icon_smile.gif), but I wish they would just sell these things on their genuine merits. For example, "simple, cheap", etc...

Z

[Simon Dingle]

There's been a lot of talk about how great rotary valves are because they allow much higher engine speeds and higher compression ratio.

But I have to ask why you would want to run higher engine speed or CR in an FSAE context?

Firstly, the higher engine speed. The 20mm restrictor chokes somewhere around 11,000-12,000rpm (depending on how well it's designed), this limits your maximum power. As you increase the engine speed your frictional losses (fmep) increase exponentially. Not what you want for a good fuel economy/efficiency score. I'm not convinced that a well designed FSAE engine needs to rev above 8,000rpm, let alone the 18,000+ that would require rotary valves.

Secondly, the higher CR. As Z has already pointed out, if you want a high CR you can do this already with poppet valves very easily. But why do you even want to go to a CR of 18+? As your CR increases, your thermal efficiency increases but your pumping losses also increase. There is a peak in overall efficiency at around 14 - 16 (depending on your precise engine design). As far as I'm aware, the only reason that diesel engines go to higher CRs is in order to improve the combustion of the diesel.

Both of these effects are outlined with test data in Heywood's book. I'd reference the page numbers but I don't have it with me at the moment.

[Jay Lawrence]

Z,

As I have not seen these engines in a diesel application, my CR information is for typical petrol power. I was also sceptical about the high RPM limit that Coates claim, 14850 seems far to good to be true, regardless of the removal of float issues. However, the BRV is quite different to the Coates engine. I believe the BRV is much more applicable to FSAE (ROI notwithstanding).

Simon, I had wondered the same thing (i.e. where is the trade off between thermal eff. and pumping losses at high CR's?). Thanks for the clarification.

At the end of the day, teams who switch to a single seem to have enough problems just getting the thing running (and starting) nicely, let alone adding the complexity (from a development point of view) of an innovative retrofitted valve train. You can't win by making your engine the focus of the car (been there done that...)