BW 350 PSRU Failure, please share the word!

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
Prop torque is proportional to rpm^2, prop power is proportional to rpm^3. Same with all turbomachinery, such as centrifugal and axial flow pumps and turbines, torque converters (19 years a torque converter engineer), turbocharger turbines and compressors. There are some non-linearities, but they are small effects compared to the rpm^2 to torque relationship.

The big point is that when rpm running a prop is small, torque is quite small, and so energy to drive amplification is also quite small. Sometimes you can get away with operating at resonance if it occurs at low rpm. But I still would not recommend depending upon that ...

Billski
 
Last edited:

Himat

Well-Known Member
Joined
May 5, 2011
Messages
2,868
Location
Norway
The more cylinders, the more firing pulses per engine turn. A 4 cylinder at 1200 rpm is firing at 40 Hz, while a V8 at 1200 rpm is 80 Hz.

Making the thread drift complete, would it be better to have an engine where the crank throw and then firing order spacing where odd and uneven? Like the old saying on the bridge sign: “Hundred men I can carry, but not if they are marching”.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
Seems to me that anyone reading this thread might come to the conclusion that direct drive is the best way to go.
Thread drift Alert! This thread is about a high power Chevy with a PSRU that broke.

If you want to build with a direct drive engine, we are all fine with that. If you want to talk about the pros and cons, of direct drive vs geared systems, it might be better done on a thread with that topic...

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
Making the thread drift complete, would it be better to have an engine where the crank throw and then firing order spacing where odd and uneven? Like the old saying on the bridge sign: “Hundred men I can carry, but not if they are marching”.
Making the thread drift complete, would it be better to have an engine where the crank throw and then firing order spacing where odd and uneven? Like the old saying on the bridge sign: “Hundred men I can carry, but not if they are marching”.
Define "better". Remember that you have to put all of your natural frequencies outside of the entire range of firing frequencies, and not just by a little either. Half at bottom, twice at top is customary. Let's look at examples:

Even fire engines have the firing evenly spread around the crankshaft rotation and give one firing order. If min rpm is 900 rpm and max is 4500 rpm:

Even fire 8's have four pulses per rev, one every 90 degrees, firing frequency is rpm*4/60 = rpm/15. Firing frequencies run from about 60 Hz to 300 Hz;
Even fire 4's have two pulses per rev, one every 180 degrees, firing frequency is rpm*2/60 = rpm/30. Firing frequencies run from about 30 Hz to 150 Hz;
Even fire 2's have one pulse per rev, one every 360 degrees, firing frequency is rpm*1/60 = rpm/60. Firing frequencies run from about 15 Hz to 75 Hz;
Singles have one pulse per rev, one every 720 degrees, firing frequency is rpm*1/2/60 = rpm/120. Firing frequencies run from about 7.5 Hz to 37.5 Hz;

Odd fire 2's abound. Just for an example, a parallel twin with a two throw crank (Honda 350 and 450 from the 1970's and many many others were done this way) gives you a 180/540 spacing on firing pulses, so you get 2/rev and 0.667/rev. Firing frequencies for the same rpm range would be firing frequencies at both, so your max would be 150 Hz, but your min would be 10 Hz. Yeah, you have to clear a much bigger range of frequencies, but you only get half as many pulses per second so amplification builds up more slowly. In an airplane, where we go steady state, slower amplification is not a help.

Then think Harley-Davidson, where they have one crank pin, 30 degrees between the cylinders, and both fire on the same rev. 30 degrees, then 690 degrees. Some will argue that this acts like a single (7.5 to 37.5 Hz for our example reve range), and it is close to that, but the 30 degree spacing does show up, so you actually have to deal with about 7.8 Hz to 900 Hz.

Odd fire engines are used for a bunch of reasons, but they make the vibration management picture harder to get to success.

Billski
 

Winginit

Well-Known Member
Joined
Jun 12, 2016
Messages
811
Location
x
Can the firing order affect the resonance ? I know when GM went from the smallblock Chevy to the LS motor, they changed the firing order from 18436572 to 18726543. Supposedly this was done to correct vibration or resonance. Many engines over the years have employed different firing orders. If you notice, the Ford 351 and Mod motors have the same mechanical firing order as the LS1, but the cylinder numbering is different, so in effect they both have the same firing order. (Ford was first)

msd-firing-orders.jpg
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
The included cylinder angle on all Harley-Davidson engines, except the V-Rod, is 45 degrees. The V-Rod is 60 degrees.
I remember having this discussion a LONG time ago with fellow Ford Tech Specialist who is also a motorcycle enthusiast, and we talked 30-690. So, instead of relying upon my faulty memory on these details, I looked things up. Classic H-D V twins are 45 degree V's and normally use 405-315 firing. For dirt track racing, the 45-675 firing scheme was done by racers and the factory offered a couple 750 cc homologation models to support their use in AMA racing that also did this. They were not concerned with vibration in that arena, and the big single firing characteristic supposedly makes control while sideways on dirt easier than with even firing engines. The 405-315 firing is easier for torsional vibration issues, but still huge for anyone who has ever sat on or ridden a Harley.

While I stand corrected on the detail, it leaves the conclusion that odd firing is more difficult to isolate intact.

Billski
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
Can the firing order affect the resonance ? I know when GM went from the smallblock Chevy to the LS motor, they changed the firing order from 18436572 to 18726543. Supposedly this was done to correct vibration or resonance. Many engines over the years have employed different firing orders. If you notice, the Ford 351 and Mod motors have the same mechanical firing order as the LS1, but the cylinder numbering is different, so in effect they both have the same firing order. (Ford was first)

View attachment 55396
Torsional vibration downstream of the engine only knows what the spacing of firing pulses are - if the engine stays even firing, our downstream issue stays the same. The various V-8 firing sequences influence crankshaft stress and rocking modes.

You really do not want the engine firing the cylinders in sequence from one end to the other, that can drive more massive cranks and blocks.

Rocking modes pertain to passenger comfort. The thing not being shown in the MSD diagram is that the right and left cylinder banks are offset from each other, and so influence block acceleration in the rotational axes as well as the linear axes. Using airplane frame of reference, they are vibration of the engine block in vertical and horizontal axes, and pitch, yaw, and roll axes. In modern cars and trucks, the linear axes are not really an issue, roll axis vibration is fairly easily isolated, yaw less so but still fairly well able to be isolated, and pitch is the toughest to isolate well. Changing the firing orders can help with isolating the engine from the chassis...

With regard to torsional vibration of the rotating parts, once you scatter the firing pulses around on the crank and provide even firing, the internal torsional problem was set.

And as you pointed out, the firing sequences were well worked out a long time ago, and they don't confront us anyway. I suspect that some of the variations in firing sequences was of specific design teams fancying themselves as artists needing to do something different from their predecessors, much the same as they numbered their cylinders differently than their competition did.

Billski
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
Lots of stuff on this thread. I am feeling a need to clear a few things. I have written a primer on how torsional vibration works, natural frequencies, how to adjust them, and so on. Feel free to read it at https://www.homebuiltairplanes.com/forums/showthread.php?t=14215

On to specifics from these threads:

Direct drive is being advocated. It is thread drift here, but I will comment that if you want an auto engine for direct drive, you had better learn what the VW and Corvair people have learned. Our classic direct drive engines have a very large diameter crankshaft with a very long bearing at the prop end to support the prop against the various bending moments a real prop has on it in use. The VW and Corvair guys have been adding prop bearings of similar proportions to support their props. If you just hang a prop on the back end of a V8 crankshaft, you might have durability problems, particularly if you like to move the rudder pedals or stick rapidly or do spins and snap rolls.

The Bud Warren drive in question does appear to have been run at higher torque and power than anyone recommended, but when you compare numbers, it does not appear to have been by a huge overload. When you get over the edge, you do get into fatigue, and looks like the Bud Warren drive had to have been close to that edge with their recommended engines.

Stuart Davis thinks his AutoPSRU BW350 is a solid product. One guy ran a HUGE prop and much higher power than recommended and broke the drive – and Stuart claims it is not his fault. I think I can buy that argument.

Andreas exceeded torque and power limits and had a number of other issues, and broke a drive. Well, Andreas did exceed current recommendations, but not by a huge amount. If the other transgressions contributed to the broken prop shaft, and they can normally be avoided, it might turn out that AutoPSRU’s BW350 is sturdy for our uses. If the other issues did not contribute to the broken prop shaft, it might well be they are treading close to limits with some of the recommended engines too.

Stu Davis’ failure analysis bothers me. How does he get to these conclusions without seeing the parts? My experience with gearbox design and troubleshooting is that failure analysis requires some quality time with the parts. There is a lot of information available with a hand lens and the sun on the parts. A lot more if you have a decent lab available.

Stu postulated that the clearance in the gears was taken up by the elastic deformation of the case in the big turn that Andreas did, and that overloaded the shaft in question, leading to the fatigue failure. If true, Stuart is saying that gearbox is not really up to handling Utility Category flight. I sure hope not. I would personally hope that we have a safety factor above the g-limits of the airplanes. Now maybe Stu is saying that with overload plus the G’s, Andreas used up the clearance. Let’s look into that. How can he use up clearance? Well, if the case is highly loaded over large areas by these G’s, elastic deformation of the case will be measurable, if still pretty small. If the case deformation is greater than the radial clearance (tooth flank lash or root to tip) between one or more pairs of gears, they could have bottomed and greatly increased the load on the shafts and bearings… These clearances would have to be pretty small as gears usually go. If 4 g’s uses up all of the clearance, even with the overloads, there is a design issue to be resolved.

I believe that the prop shaft could easily get its biggest bending moments from pitch/yaw rates that might be seen with aggressive stick and rudder use. Not high g’s, just high pitch/yaw rotation rates. The prop is a big gyroscope, and the moment required to change the plane of the prop rotation is a function of Mass Moment of Inertia of the prop, rotation speed of the prop and the combined pitch and yaw rate of the airplane. This issue will never be found on a conventional run stand, but we get it in airplanes…

On to the myriad of PSRU comments.

Geared PSRU’s were used in almost all of the combat aircraft of WWII. V-12’s, radials, they all had them, big props turning slowly were needed while piston engines like to turn much faster to make good power at min weight. Solutions to torsional vibration ran all over the place. Two types of solutions dominated: Soft systems, where the first primary resonant mode is safely below the operating speed and the prop is isolated from the firing pulses, and; Stiff systems, where the lowest resonant mode is above the operating speed and the entire system sees nearly full value of the firing pulses.

Soft systems - We know that the Allied powers’ V-12’s had quill shafts between the crankshaft and the gearbox for the purpose of driving the 1st resonance order frequency (and maybe some others) down below firing frequency from idle speeds. This is a simple torsion spring put in the system to drive 1st resonance below the lowest firing frequency. Nice when you can do it. This scheme also requires that all other resonance frequencies must be above max firing frequencies and usually above 2x of firing frequencies. That is a lot of analysis and testing. It is done in cars/trucks/gensets/ships/etc with springs, sometimes several different spring rates, sometimes elastomeric elements are included, etc. It is also done in Rotax and several of the PSRU’s using rubber springs (which are not linear, but have rising rates).
Soft systems using a quill shaft have a single spring rate and very little damping, but they worked in the big V-12’s. Soft systems using a set of identical coil springs work almost the same way, except that they can have some friction and they can wear. They are used extensively in automatic transmissions (in the torque converter clutch) and in manual transmission clutch discs. To do this sort of thing more compactly, there are clutch discs with multiple spring rates arranged to give rising spring rates with rising torque. Rising spring rates are kind of neat in that they do not amplify much. And they last lifetimes when done right.

Let’s also remember something else about “soft” systems. The springs are not there to allow a resonating system some travel to prevent damage. The springs are there to reduce the natural frequency, generally significantly below firing frequency at idle.

Some of the PSRU’s out there do similar things with rubber springs, which have a low spring rate at low torque, and go up in rate as the load goes up. They do not really have a single resonance point. It is wider range, but not as strong, which reduces the amplification when operating near the resonant range, and makes them behave better as when they go through resonant range during startup and shut down.

Some things to remember when you have a soft spring in the system – In the operating region, the spring does not see the full value of the firing pulses. With the system operating significantly above the 1st resonant frequency, the stuff downstream of the springs vibrate at a small fraction of the engine side torque variation. Travel of the spring (rubber isolators count as springs) is small, and firing pulses transmitted to the prop side are thus also small. The only times the system will see large strokes are:

Engine start – as the engine accelerates through the resonant mode on its way to idle speed;

Engine shutdown – as the engine coasts down from idle to stop. The idea is to place the resonant mode safely above cranking speed but significantly below idle, so that neither one will amplify much, and then you count on going through the resonant mode quickly.

Stiff systems – You make the entire system really stiff so that the 1st resonant frequency is above the highest firing frequency of these engines. Our current direct drive engines are stiff systems. Stiff systems have to be stiff everywhere. One soft element and they are not a stiff system.
Stiff systems are what we are used to in little airplanes, but they can be tough to achieve in geared systems. What happens is that engine is trying to speed up and slow down while the prop with its big inertia is trying to maintain a steady speed. Any gear lash gets opened slightly as the torque eases and then banged closed again as the torque rises. If the lash is big enough, the impact is huge and the item with the lash can be damaged over time. Keep the lash small enough and the elements beefy enough, and the elements live nicely. This is what the PowerSport system had, and it appeared to be durable.

One other thing was invented in the 1930’s that made a big difference - Torsional Pendulums – Attached to crankshaft counterweights, these can be tuned to whatever order you want to absorb. In the big radials, one was usually tuned to firing order and the other to the next most obnoxious vibration order. These were variously applied and they are capable of really toning down the max prop shaft torsional vibration. These neat gadgets have been making their way into cars in the last few years. With these systems, a stiff engine and gearbox system had much smaller speed fluctuations and could then be reliable, even with significant lash in the planetary gearsets they used.

All of this stuff takes careful engineering to be assured of success. Have the PSRU on the market for auto engine conversions gotten similar engineering attention? Most, I must agree with Ross do not appear to have. The ones I know the most about, even the successful ones, well, they were “developed” rather than “designed”. It appears that PowerSport is about to show up again, and it has been gone through by folks who know what they are doing. While the analysis has now been done, Everett Hatch beefed up the design until he and Alan Tolle could not break it. Alan had quite a few dead stick landings in his RV3. Sometimes “F--- with it until it works” can succeed. Ev Hatch knew he was aiming for very low gear lash and very stiff, and when he drove for that, it appeared to work. The big W-8 engine being marketed for replacing turbine engines on Queen Airs and the like appears to have been done well too and it seems that the same guy went through this engine and the PowerSport… These engines have a real chance of being durable.

The others? I would want to see either similar reviews or a lot of successful experience. Autoflight and Marcotte appear to be there on successful experience. Maybe AutoPSRU is there, maybe not. All three of these systems are “soft” systems. Autoflight and Marcotte use rubber spring elements, while Auto PSRU has coil springs imbedded in polyurethanel, so they have both. I will watch and see.

The Ballistic Drives look promising. Some folks are putting a bunch of power through them. They have rubber drive elements – “Soft” system. A couple potential problems occur with airboats as a proving ground: The highest precession rates on an airboat I have seen are an order of magnitude lower than is possible in an airplane. Think spin entry, spins, abrupt yaw or pitch inputs, and severe turbulence. The precession moment thrown on the prop shaft is linear with the precession rates and is likely higher in airplanes than anywhere in the airboat environment. If that results in failed shafts and gears, well, that would be bad. We shall see with the P85 and a couple other guys running Ballistic Drives…
 

Winginit

Well-Known Member
Joined
Jun 12, 2016
Messages
811
Location
x
That was a heck of a post ws. I sure wish I could retain all of it. Since I'm more into direct drive I probably interpolated different points from it than most people will. If the use of a large propellor and it's incipient momentum is a serious contributor to failure issues, THEN does it also follow that a benefit of direct drive using a smaller prop at higher rpms would be smoother power and less tendency toward a vibration or harmonic problem ? Let's say that both the engines are identical LS3 engines operating at 3600 rpms. One is swinging a large metal propellor operating thru a PSRU. The other is a direct drive with a supporting external bearing also operating at 3600 but with the appropriately reduced size propellor. With the small prop a composite or wooden would probably be available, but for Apple to Apple comparison I guess a metal prop should be compared. What say ye, ole wizard. (meant as a compliment)
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
3,749
Location
Calgary, Alberta, Canada
That was a heck of a post ws. I sure wish I could retain all of it. Since I'm more into direct drive I probably interpolated different points from it than most people will. If the use of a large propellor and it's incipient momentum is a serious contributor to failure issues, THEN does it also follow that a benefit of direct drive using a smaller prop at higher rpms would be smoother power and less tendency toward a vibration or harmonic problem ? Let's say that both the engines are identical LS3 engines operating at 3600 rpms. One is swinging a large metal propellor operating thru a PSRU. The other is a direct drive with a supporting external bearing also operating at 3600 but with the appropriately reduced size propellor. With the small prop a composite or wooden would probably be available, but for Apple to Apple comparison I guess a metal prop should be compared. What say ye, ole wizard. (meant as a compliment)��
You wouldn't operate an LS engine at 3600 after adding the weight and cost of a PSRU to it, wrong engine choice then.

PSRUs allow you to increase the power to weight ratio of the package- main reason for using them.
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
If the use of a large propellor and it's incipient momentum is a serious contributor to failure issues, THEN does it also follow that a benefit of direct drive using a smaller prop at higher rpms would be smoother power and less tendency toward a vibration or harmonic problem ?
Thanks for the compliment.

I can not say it enough: There is no substitute for doing the engineering correctly with regard to torsional vibration.

Less or more tendency? If all you want to argue merits, well, it ain't that simple. All sorts of issues force good engineering before you can tell.

Prop size, all by itself does not make for failure issues. What makes failures in products are inadequate understanding of dutycycle and of the five noises as they relate to this product. I can give you some understanding regarding your question.

Resonance mode shapes (the shape of a resonant vibration) and their frequencies are analyzed using Eigen Analysis. There are classic lumped mass and spring tools good for when your system is comprised of elements that are either massive and not springy or springy and not massive. There are more sophisticated tools for distributed mass-spring elements, and then there is FEA... My previously mentioned primer on vibration talks about vibration modes and how each resonant frequency is a function of the square root of the stiffness/mass ratio of the involved parts. The stiffer a system is, the higher its resonant frequency, the more massive it is the lower its resonant frequency. The effect of the ratio is not linear, you need a four-fold change in the ratio to make a two-fold change in frequency. In torsional vibe, we are talking Mass Moment of Inertia for masses and torsional springs, but if it helps to look at linear springs and masses to get your arms around the concepts, go for it.

Direct drive prop systems as executed in legacy airplane engines are stiff systems with beefy support of the prop flange against torsional vibration and precession loads with high inertia metal props. That makes them pretty darned secure when folks start hanging lighter props on prop extensions. If they took the crank and prop and apportioned the connecting rods and pistons appropriately and added in the camshaft and accessories with appropriate gear ratios, then ran their Eigen solutions, they would get an estimate of the frequencies at which the first so many Eigen Modes occur in that system. My bet is that with these aforementioned legacy engines, the lowest frequencies of their Eigen solutions are safely above the max firing frequencies.

You can do similar analysis for a SBC with a prop bearing system added. With a faster turning engine and prop, the prop will be smaller diameter, but because it will have higher blade loading and disc loading, the hubs and blades might be beefier than the legacy prop, so do not count on huge reductions in MMOI. Since even a small air screw is still a big inertia in most airplane engines, the prop does tend to be the biggest inertia in the system. If it has reduced MMOI, it will tend to drive resonant frequency up compared to legacy engines. So, a smaller prop contributes to making a stiff system, potentially with less massive connecting components than with a larger prop. Yeah, your crank extension/prop bearing might be less massive with the SBC and direct drive. On the other hand, if you were trying to isolate the prop from the engine torsional vibration (a soft system), the light prop would drive how low the spring rates would have to be in your torsional damper.

Now take that same SBC engine and gear down the prop speed. Now the prop diameter will be larger, but it may be somewhat more lightly constructed because of the reduced blade and disc loading, but its increased diameter will likely result in considerably higher inertia. You also have to consider gearing. A geared prop running at a speed ratio relative to the engine speed is being accelerated and decelerated by crankshaft vibrations. The inertia through the gearing can be expressed in the frame of reference of the crankshaft as I*SR^2. If the prop is running 1/2 speed, its inertia relative to the crank is 1/4 of its measured MMOI. Yep, the prop is more massive, but that more massive prop is also acting like it is less massive through the gearing. Since prop inertia is a function of mass and swing radius squared, the bigger prop is likely to have more inertia, and the gearing brings it back down with a square of the SR too, hmm, sounds like you might be working in the same range both ways, but you would have to do your own math and design your own system to work.

One other issue I love to bring up is that the crank extension/prop bearing system has to be attached rather sturdily to the crank. If it has lash or is rather flexible in its connection, the rather light crank with big oscillating accelerations will be vigorously going through the lash and/or vigorously exercising the attachment method. Sounds like you would need to design to beat these failure modes while you are also trying to isolate the prop precession moments from the crank. They do this in the VW and Corvair world. Maybe your SBC crank is up to this, maybe not. But you sure better know and design accordingly to have a successful solution.

Five noises? The world is usually different from what you initially imagine. To make your product robust, you need to think about variations in how the customer uses your product, how it is made, what the environment is, etc, and then expand your design effort to cover these things. Big topic unto itself.

I can not say it enough: There is no substitute for doing the engineering correctly with regard to torsional vibration. Less or more tendency? That sounds like hiding your head and hoping.

Billski
 

rv6ejguy

Well-Known Member
Joined
Jun 26, 2012
Messages
3,749
Location
Calgary, Alberta, Canada
Great posts Bill, nail squarely on the head of all the important points that so many folks either ignore or don't understand in this field. Too many crossed fingers being used as a substitute for proper engineering/ testing in the field IMO.
 

Winginit

Well-Known Member
Joined
Jun 12, 2016
Messages
811
Location
x
You wouldn't operate an LS engine at 3600 after adding the weight and cost of a PSRU to it, wrong engine choice then.
I believe TxFlyGuy is quoting 3600 rpm cruise with his LS3 powered Titan T-51 using a 1.9 Autoflight PSRU. That would put 2700 rpms for the prop at 5130 rpms for the engine. Don't know the prop size so not sure if thats his plan. Different people have different ideas as to what is a safe operating speed for an engine. My personal opinion is that I would not want to operate an engine at over 5000 rpms for extended periods.






PSRUs allow you to increase the power to weight ratio of the package- main reason for using them.[/
QUOTE]

Yes, thats true. Its just that most homebuilt airplanes are not capable of using 300-400 hp even if it "only" adds an additional 80 lbs to the already heavier 400 lbs of the LS engines. With other types of engines and airplanes, PSRUs can work well. Trying to fit an LS engine into most experimental airplanes is difficult because of the weight. For the majority of experimentals, sufficient reliable hp would be availble without the complexity, weight, and cost of a PSRU. Its easy to get 250 HP from an LS, and that would satisfy most peoples needs. It really becomes a matter of what an airplane can handle and how the builder wants to get there. TxFly Guy is building a very high dollar build of an airplane that is financially out of reach for most builders. The option of a somewhat lighter and less expensive version of an LS that still produces all the hp their airplane is capable of using seems to be a better choice. Rather than do that, most builders will just choose a lighter engine with a PSRU, like a Sube.

There was a fellow in Memphis a while back who built a custom one off design and used a direct drive LS. He was a very experienced Naval pilot with over 7000 hrs and upon retirement worked for American Airlines and FedX as an airplane mechanic. His total cost for the airplane including a used LS1 was less than $10K. He took off at 3400rpms (230 hp) and cruised at 3000 rpms/ 120 KIAS from Memphis to Oshkosh. ROC was 1700 FPM. He used a simple unsupported 6" prop extension and a ground adjustable warp drive propellor. Another more well known airplane is Gary Spencers Ford powered Long EZ that has been flying with a simple prop extension. He followed the same path that I reccommend with an LS engine. He took a 302 Ford and increased its capacity to 360 cu in with a stroker crankshaft. He can readily exceed the VNE of the aircraft (220 mph) with a 1500 fpm climb out.
He had over 500 hours on it in 2002 and as far as I know he is still flying it today. Both of these guys proved it can be done successfully without a PSRU, and even a support for their extension shaft. Examples like these are proof of concept. Personally I would prefer a support and bearing of some kind, and that is still lighter and cheaper than a psru. The point I'm trying to make is that there are proven examples that you can get all the horsepower that most airplanes need without expensive redrives, so builders should at least consider whether they need all the hp from a redrive or not.
 
Last edited:

TXFlyGuy

Well-Known Member
Joined
Apr 25, 2012
Messages
1,783
Location
Republic of Texas
I believe TxFlyGuy is quoting 3600 rpm cruise with his LS3 powered Titan T-51 using a 1.9 Autoflight PSRU. That would put 2700 rpms for the prop at 5130 rpms for the engine. Don't know the prop size so not sure if thats his plan. Different people have different ideas as to what is a safe operating speed for an engine. My personal opinion is that I would not want to operate an engine at over 5000 rpms for extended periods.

QUOTE]

Yes, thats true. Its just that most homebuilt airplanes are not capable of using 300-400 hp even if it "only" adds an additional 80 lbs to the already heavier 400 lbs of the LS engines. With other types of engines and airplanes, PSRUs can work well. Trying to fit an LS engine into most experimental airplanes is difficult because of the weight. For the majority of experimentals, sufficient reliable hp would be availble without the complexity, weight, and cost of a PSRU. Its easy to get 250 HP from an LS, and that would satisfy most peoples needs. It really becomes a matter of what an airplane can handle and how the builder wants to get there. TxFly Guy is building a very high dollar build of an airplane that is financially out of reach for most builders. The option of a somewhat lighter and less expensive version of an LS that still produces all the hp their airplane is capable of using seems to be a better choice. Rather than do that, most builders will just choose a lighter engine with a PSRU, like a Sube.
At 3600 engine rpm, the LS3 is loafing. This will produce slightly over 300hp at sea level, standard day. The max potential of our engine, with the GM LS Hot Cam, is beyond 500hp.
With our standard 84" prop, this will equate to 1890 prop rpm. This rpm setting (3600) would be considered max cruise power.
Honda owners will cruise their engines in the 3400-3600rpm range. This is with a 2.21-1 gear reduction.

The LS3 is the least expensive, high power option available. Speaking of power, we rarely (if ever) will exceed 300hp. 4000-4200rpm also will be the most we would ever see, depending on density altitude conditions. Reliability? the SBC is legendary.

Weight? The complete drivetrain package with the LS3 is about 10lbs more than the Honda V6. The 84" carbon fiber 4 blade (hollow) Whirlwind Aviation propeller weighs in at 34lbs.
 
Last edited:

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
Seems that the SBC crankshaft end is sturdy enough for some level of flying with the prop bolted to the crank flange.

I would like to do an estimate of the maximum safe precession rates for this sort of installation.

If anybody on the forums knows the diameter of the SBC crankshaft is at the rear main bearing and material of the crankshaft. Also, the make and model of the props commonly used with direct drive SBC and their MMOI will help too.

As long as we are at it, anybody have the prop bearing diameter of a six cylinder Lycoming or Continental (500 cubic inch range, not from the Conti O-300 through O-360) and common prop inertia for them as well?

Billski
 
Last edited:

Toobuilder

Well-Known Member
Log Member
Joined
Jan 19, 2010
Messages
4,781
Location
Mojave, Ca
There was a guy flying behind a DD SBC in a replica biplane for a few years. Prop bolted directly to the crank via a short aluminum adapter. I spoke with him some about it and while he had no real issues with the powerplant, it just didnt perform well enough to recommend doing it again. He opined that he should have done a 500 inch Cadillac V-8 instead. Almost the same weight, but a LOT more low end grunt.

There was also the "Junkyard Dog" which was a LS-1 DD. It did fly, but I believe it was involved in a fatal crash.
 

Winginit

Well-Known Member
Joined
Jun 12, 2016
Messages
811
Location
x
Seems that the SBC crankshaft end is sturdy enough for some level of flying with the prop bolted to the crank flange.

I would like to do an estimate of the maximum safe precession rates for this sort of installation.

If anybody on the forums knows the diameter of the SBC crankshaft is at the rear main bearing and material of the crankshaft. Also, the make and model of the props commonly used with direct drive SBC and their MMOI will help too.

As long as we are at it, anybody have the prop bearing diameter of a six cylinder Lycoming or Continental (500 cubic inch range, not from the Conti O-300 through O-360) and common prop inertia for them as well?

Billski[
/QUOTE]

First, a clarification. The term "small block" is generally accepted as a reference to a V8 series of cast iron engines produced by GM from about 1955 until about 2000. There were some aluminum versions available for racing and lots of $. Although GM did produce some other V8 engines, the main other engine they produced is physically larger and heavier, so it was commonly referred to as a "Big Block". The LS engine is close in size to a "small block" and is the replacement for the "small block", but its not commonly referred to as "small block". Generally its referred to as an LS or LSX with the X representing the fact that multiple versions of the basic LS are available. All are made from aluminum. There are some truck versions with cast iron blocks that are sometimes loosely referred to as LS engines because they are basically the same engine with a cast iron block. They have odd names like LY6, so while they share the same architecture, the factory doesn't call them by the LS designation. For the purposes of our discussion, generally we should be talking about the aluminum LS engines from a 346 cu in LS1 to a 427 cu in LS7. If that is what you would like to crunch some numbers on, they have a 2.560 diameter main bearing. The factory units are nodular cast iron for the most part with a 3.622 stroke. There is a 4.125 crank available made from 4340 steel. The LS7 (427 cu in) had a 4" stroke forged crank but I don't know the material composition offhand. I know it wasn't 4340. I think it was a 5xxx series. Lycoming has a 2.375 main bearing and a forged crank, but as for it composition, I don't know. I believe the bearing size is common for the O-320 and larger Lycomings. Someone correct me if thats wrong. Hope this helps.:)
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
6,792
Location
Saline Michigan
/QUOTE]

First, a clarification. The term "small block" is generally accepted as a reference to a V8 series of cast iron engines produced by GM from about 1955 until about 2000. There were some aluminum versions available for racing and lots of $. Although GM did produce some other V8 engines, the main other engine they produced is physically larger and heavier, so it was commonly referred to as a "Big Block". The LS engine is close in size to a "small block" and is the replacement for the "small block", but its not commonly referred to as "small block". Generally its referred to as an LS or LSX with the X representing the fact that multiple versions of the basic LS are available. All are made from aluminum. There are some truck versions with cast iron blocks that are sometimes loosely referred to as LS engines because they are basically the same engine with a cast iron block. They have odd names like LY6, so while they share the same architecture, the factory doesn't call them by the LS designation. For the purposes of our discussion, generally we should be talking about the aluminum LS engines from a 346 cu in LS1 to a 427 cu in LS7. If that is what you would like to crunch some numbers on, they have a 2.560 diameter main bearing. The factory units are nodular cast iron for the most part with a 3.622 stroke. There is a 4.125 crank available made from 4340 steel. The LS7 (427 cu in) had a 4" stroke forged crank but I don't know the material composition offhand. I know it wasn't 4340. I think it was a 5xxx series. Lycoming has a 2.375 main bearing and a forged crank, but as for it composition, I don't know. I believe the bearing size is common for the O-320 and larger Lycomings. Someone correct me if thats wrong. Hope this helps.:)
OK everybody. You heard it here. More recent version of the SBC is not an SBC, it is an LSx!


A bunch of guys over at GM Powertrain were tasked with replacing the venerable SBC with a modern engine, but they had this terrific legacy and needed to not change very much or these new engines would disappoint customers and be expensive to tool up and hard to fit in the vehicles. So they held the same package by keeping the same number of cylinders and pushrod two-valve designs, asked themselves what really did need to change within the engine and made those changes and called it the LS series. Seems pretty successful too. These engines that have the same basic architecture as a small block Chevy, same material sets and look except for the centered spark plugs as a small block Chevy, fit in the same space as the small block Chevy they replace, have the same displacement range as a small block Chevy, same output torque and power range as a small block Chevy, and are produced by GM and sold by Chevy, are not supposed to be called small block Chevies! They even fill the same spot in racing with crate short blocks and all sorts of go fast parts like the small block Chevy. Looks like a duck to me... but, LSx it is.


LSx engine rear main bearing diameter is 2.560". Everything from nodular cast iron to 5000 series to giant killer 4340 in the crank. That gives a big range in fatigue strength! I had heard 100 kpsi yield strength, but that sure does not apply to all of them, and estimations of fatigue strength are usually made off of ultimate strength. Anybody got any idea what the state of heat treat is used in CI, 4340, and 5260 cranks for these guys?


Lycomings prop bearings are 2.375". Smaller than the SBC, uh LS1! Anybody know the alloy and heat treat state on them?


Why do I keep asking these questions? Well, it matters to a calculation of max yaw/pitch rate we can carry at what torque and rpm, but probably not as much as the fact that the LS1 crank end is bigger than the adequate Lycoming...


This last fact does make me wonder if the SBC and its derivatives need worry much over prop precession moments, except maybe if you are running a CI crank. Better to look at it with the numbers though.


Billski
 
Top