|
-
POSTING RULES

-
Donate yearly (please).
-
Advertise in here!
-
Today's Posts
|
Insert Pics
|

01-18-2013, 11:25 PM
|
|
|
|
Join Date: Dec 2012
Location: Mukilteo
Posts: 20
|
|
Quote:
Regarding cowl pressures, the important thing is to actually measure them and not assume anything about how air is going to flow. A simple gauge that measures inches of water can be had for under $50 and save many headaches if it just gets used. From my experience, radiators, oil coolers, and inter coolers will perform pretty well once the differential pressure across them gets up around 6 or 7 inches. It only takes a little leakage to allow the differential pressure to equalize and destroy the flow. Air scoops and NACA inlets cannot do their job if they are having to exit air into a high pressure area, like the inside of the cowling. The pressure in my cowling behind the radiators will have around 7" water column pressure, so to get good flow to another exchanger requires the air inlet or scoop to overcome this pressure before any flow can take place. It is easy to waste a bunch of time putting things together that looks like it will allow good air flow and cooling...
|
Indeed. I helped one of the sport class reno racers and got him to instrument the plane briefly. I ran some calcs based on the measurements and made some recommendations. He picked up about 15 kts in the first iteration! Unfortunately, I think he got impatient with the process and I had a hard time getting any more readings.
I also got to compare notes with Dave Anders for a couple of hours at the race and concluded that both in the race plane and in most of the planes I've seen, there is likely quite a bit of low hanging fruit in the cooling system, but as you say, you can't assume, you have to measure, unless you are a lot better than me.  At least theoretically, I also see some apparent low hanging fruit in water cooled engines, but I've not actually had opportunity to touch one.
__________________
Scott Hogan
KAWO
Former PA22/20 owner
RV-4, 0-360 Catto prop
Last edited by shogan50 : 01-18-2013 at 11:26 PM.
Reason: quote problem
|

01-19-2013, 04:47 AM
|
|
|
|
Join Date: Apr 2007
Location: Norway, Stj?rdal
Posts: 598
|
|
Quote:
Originally Posted by shogan50
Anyone have an exploded assembly view of one of these gearboxes or photos? I've not followed, but as a mechanical engineer I'm quite curious.
|
Here is exploded view with parts for the Verner 133 (80 HP, twin). Verner used a belt in their earlier versions, but switched to this gearbox. The exact same principle is used on the newer Rotax 912iS, but different design (the ULS also has a free play for proper idle operation, but that is removed on the iS).
I'm not an expert on this either, but as an engineer working with dynamic analysis of turbines (and stuff) I'm curios how things work. The "solution" is rather straight forward and obvious once you think about it; spring and dry friction (Coulomb) damper. The dangers of TV is at low RPM and low propeller load, and during start/stop. This is due to the rotating inertia of the propeller vs torque peeks from the engine and changing direction of forces during no load conditions that also could cause high peek values. At those conditions you want to protect the gears from overload and fatigue. At all other conditions, you want the gear to act as a solid body. The spring will soften the forces, and the clutch will dissipate energy, a shock absorber. When the dynamic forces are below a pre-set value determined by the pre-set spring force and clutch, the whole assembly acts as one unit. So you can build a light weight gear that lasts "forever". To use rubber instead of spring/dry friction will also work I guess since rubber acts both as spring and damping, but the gear will never act as one solid unit, so energy is constantly dissipated causing reduced life time and maybe even a heat problem, maybe also higher order TV. That is my theory on this (right or wrong).
http://www.vernermotor.com/pdf/iomvm133m_s.pdf
Regarding cooling. ULPower has included in their manual how this setup MUST be for the engine to operate without overheating (or without excess drag depending how you look at it). Look at page 30 http://www.ulpower.com/engines/manua...ion-manual.pdf
If you stick to the manual regarding those pressure drops, the engine will run in any airframe without any heating problems.
Rotax also have stuff like this available, but their site is a bit more confusing.
When looking at the Viking engine site, none of this basic engineering stuff is available, and that should ring some bells here and there. I believe the Honda engine is perfect for the job, much more so than any Subaru, because it also runs without problems in outboard marine application. But if Egg's setup is perfect is another story altogether.
|

01-19-2013, 07:51 AM
|
 |
|
|
Join Date: Oct 2005
Location: 08A
Posts: 9,500
|
|
Quote:
Originally Posted by SvingenB
Here is exploded view with parts for the Verner 133 (80 HP, twin).
|
Interesting design.
http://www.vernermotor.com/pdf/iomvm133m_s.pdf
I suspect more than a few readers are confused by the use of the word "clutch" in this application. It is not a clutch in the conventional sense of the word, examples being:
Any of various devices for engaging and disengaging two working parts of a shaft or of a shaft and a driving mechanism.
or
a coupling used to connect and disconnect a driving and a driven part (as an engine and a transmission) of a mechanism
I suspect the word was borrowed with the parts. The diagrams do in fact illustrate clutch plates as used for power transmission and disengagement, in a style typical for motorcycles and small diameter, multiplate applications. However, in this case the clutch plates cannot disengage the drive train. The plates merely serve as a friction damper/torque limiter in parallel with a soft element (the 13K-05-45 springs). As noted, at low torque power is transmitted via the plates. When start up intertias, resonant oscillating torque, or even the normal torque oscillation of internal combustion exceeds the frictional capacity of the plates, they slip and power transmission is assisted by the springs. When slipping the plates serve as a true frictional damper, converting mechanical power to heat.
I have no idea how well the system really works. I would note that static friction is higher than dynamic friction. I'd also note that continuous plate slippage would add a lot of heat to the system, and friction plate lifespan would be an issue. Last, its performance would vary with wear.
I see one design detail I do not like at all, but i'll try to remain open-minded. The 13M-05-04 damper body is keyed to the crankshaft (13S-00-86 key).
Quote:
|
To use rubber instead of spring/dry friction will also work I guess since rubber acts both as spring and damping, but the gear will never act as one solid unit, so energy is constantly dissipated causing reduced life time and maybe even a heat problem, maybe also higher order TV. That is my theory on this (right or wrong).
|
Lovejoy publishes technical information for their popular line of rubber couplers, including damping coefficient. It is very small, so heating is equally small in normal operation. That said, yes, if you spec a coupler that is too soft you can damage it with both heating and mechanical forces.
It is possible to develop a viscous damper in parallel with a soft element. Dana has a patent on one in parallel with springs. I built one for a Suzuki conversion in parallel with a Centaflex rubber element. A viscous damper eliminates all the wear issues of a damper based on surface contact. This one sheared a viscous silicone fluid between large area close fitting surfaces (0.010' clearance). The large area also offered a huge capacity for heat dissipation, thus fluid viscosity could be treated as stable.
Here I'm pumping in silicone after having it apart for for inspection. Education and recreation....

__________________
Dan Horton
RV-8 SS
Barrett IO-390
|

01-19-2013, 08:41 AM
|
|
|
|
Join Date: Dec 2012
Location: Mukilteo
Posts: 20
|
|
That's a clever gearbox design. I imagine much more important on a 2 cylinder than a 4.
It is interesting that they prescribe an exit area of 500% of inlet. That will make for some cooling drag! 
__________________
Scott Hogan
KAWO
Former PA22/20 owner
RV-4, 0-360 Catto prop
|

01-19-2013, 08:55 AM
|
|
|
|
Join Date: Dec 2012
Location: Mukilteo
Posts: 20
|
|
Quote:
Originally Posted by SvingenB
Regarding cooling. ULPower has included in their manual how this setup MUST be for the engine to operate without overheating (or without excess drag depending how you look at it). Look at page 30 http://www.ulpower.com/engines/manua...ion-manual.pdf
If you stick to the manual regarding those pressure drops, the engine will run in any airframe without any heating problems.
Rotax also have stuff like this available, but their site is a bit more confusing.
|
Interesting. From memory (I'm on a flight to Dallas right now) lycomings like about 5" of water. It sounds like UL can deal with less than an inch (20mm). I wonder if I'm forgetting something. That's a big difference! Cooling this thing with minimal drag should be much easier. It is also interesting that they don't prescribe a picolo tube for the plenum. Measuring the reno racer, I found that inlet velocity was substantial enough to alter readings quite abit, but that plane was doing north of 375 indicated (don't recall if he is calibrated in mph or kias, I think the latter).
__________________
Scott Hogan
KAWO
Former PA22/20 owner
RV-4, 0-360 Catto prop
Last edited by shogan50 : 01-19-2013 at 08:59 AM.
Reason: clarify kias vs mph
|

01-19-2013, 10:31 AM
|
 |
|
|
Join Date: Oct 2005
Location: 08A
Posts: 9,500
|
|
Quote:
Originally Posted by shogan50
Interesting. From memory (I'm on a flight to Dallas right now) lycomings like about 5" of water. It sounds like UL can deal with less than an inch (20mm). I wonder if I'm forgetting something. That's a big difference!
|
Scott, your memory is working fine. Using the IO-360 chart, 5"H2O drop at 5000 PA (measured at the engine face) is enough to cool 140 hp at <400 CHT and 0.5 BSFC, standard day OAT.
I suspect the little UL motor has a higher fin area/HP ratio.
__________________
Dan Horton
RV-8 SS
Barrett IO-390
|

01-19-2013, 10:32 AM
|
 |
|
|
Join Date: Jan 2005
Location: Sherwood, Oregon
Posts: 981
|
|
HP&Torque
http://www.vernermotor.com/pdf/iomvm133m_s.pdf
This is maybe the best and most informative engine manual I've seen by far...
However the graph on p.13 seems to violate all that is holy with engines. That is that torque and HP curves cross at 5252 RPM...
Here is ECI's article on same.
http://www.epi-eng.com/piston_engine...and_torque.htm
Any engine guys have a clue? Ross?
Jerry
Last edited by Jerry Cochran : 01-19-2013 at 10:39 AM.
|

01-19-2013, 10:58 AM
|
 |
Senior Curmudgeon
|
|
Join Date: Sep 2005
Location: Dayton Airpark, NV A34
Posts: 15,420
|
|
Makes me wonder if the rpm should be labeled as prop, not crank??
And, where did they take the torque reading from?? Crank or prop flange??
I agree, 5252 rpm is the standard crossover point for HP/TQ charts------at least for the 50 or so years I have been paying attention to such things.
By the way, anybody remember what this thread is supposed to be about 
__________________
Mike Starkey
VAF 909
Rv-10, N210LM.
Flying as of 12/4/2010
Phase 1 done, 2/4/2011 
Sold after 240+ wonderful hours of flight.
"Flying the airplane is more important than radioing your plight to a person on the ground incapable of understanding or doing anything about it."
Last edited by Mike S : 01-19-2013 at 11:01 AM.
|

01-19-2013, 11:31 AM
|
 |
|
|
Join Date: Oct 2005
Location: 08A
Posts: 9,500
|
|
Quote:
Originally Posted by Jerry Cochran
However the graph on p.13 seems to violate all that is holy with engines. That is that torque and HP curves cross at 5252 RPM...
|
....in English units. The 5252 figure comes from 33,000/2pi, 33,000 pounds feet per minute being the rough equivalent of Watt's example draft horse....180 lbs pulled 181 feet in one minute.
Anyway, no newton-meters, just pounds and feet.
Look at the dyno chart. Torque at 5200 RPM is 109 Nm, which converts to 80 lbs-ft. Using the HP scale values on the left margin, find 80 on the graph. It's right there at the intersection of HP and 5252.
__________________
Dan Horton
RV-8 SS
Barrett IO-390
|

01-20-2013, 12:09 AM
|
 |
|
|
Join Date: Mar 2005
Location: Calgary, Canada
Posts: 5,766
|
|
Some thread drift here but an interesting discussion. Dan always posts thoughtful insight.
UL outlet to inlet area at 500% is nonsense as already observed. There is no science in that design, just guaranteed high drag. Great cooling at the expense of high cruise drag does not impress me. In good liquid cooled designs we are down to 50-60% of inlet area in cruise for exit area using variable exit geometry and about 125% in climb. We can cool 200hp with as little as 28 square inches of inlet area even in hot climates with a proper rad and duct design.
I am not fan of slipping metal clutches with regard to limiting TV in gearbox designs. Make sure we are talking about clutches and not dampers. The Rotax clutch does not slip in normal flight operation.
We know it is pretty essential with aero gearboxes to have some TV absorber mechanism inline if the gearset is to last decently. Many ways to accomplish this, some ways are better than others.
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -6. The time now is 12:40 AM.
|