What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

HR-I vs RV3

Lufthans

Well Known Member
Hi guys,

I own an RV3B that is now taken apart for some serious upgrades. I’m a multiple offender when it comes to building and modding aircraft (and cars for that matter) and am going to put a Subaru conversion on the RV.

Not here to discuss the merits of this engine. There’s other places on VAF to do so. Suffice to say that I’ve been flying one on my Jodel for the past 15 years, which after three iterations, tons of engineering, money and headaches, I got to function flawlessly and thus far 100% reliably. I have done a few more on other aircraft since and even got to do some serious R&D for an Oregon-based company (not the mothership, there are more companies in Oregon).

Anyway, on the RV3, I will be putting the cooling in a P-51 style belly pod, with proper ducting, and will turbocharge the engine. With that (and modified electrics, fuel system, custom ECU, engine mount, cowlings, instrumentation and a whole bunch of upgraded engine internals), it has the potential of putting out 300 hp.

And here’s where my questions in the Rocket forum come in:

I know that 300 hp is serious overkill on such a small aircraft, and I know that its 210 mph Vne can be reached with half that power. So I will be using the other half to make it climb like a mountain goat with ambitions (which really suits my type of flying anyway - quick dashes to the clouds, play around between them and then drop back in).

However - it would be nice to have a bit more headroom in the speed department.

I’ve sent an email to John Harmon (with much of this text), however am not sure if I can expect an answer from him, since he’s apparently no longer supporting the HR-I. So I also post the questions here:

Does anyone here know anything about the specifics of this aircraft?

What did John state as a Vne on the HR-I, and what did he do to safely achieve that?

I believe he’s switched to an RV4 tail? Just for the sake of its mass balancing, or is it stronger too? Just the horizontal or the vertical tail as well?

Did he do any other structural modifications to the airframe like thicker skins? Did he modify the wing in any way? (Mine has the B-wing). (And yes, I know about the razorback, I just like the P-51-like looks of the slider more)

Bolting on a more powerful engine is one thing, but making it safe to operate is another thing altogether….

Curious to know and any tidbit of info more than welcome!

Thanks
 
Perhaps you could find Dave Anders' discussions about his very fast RV-4. In at least one of them, he outlines the major structural changes he put in. Among other things, he built a new empennage with modifications.

A couple of builders added mass balancing to the RV-3 tail. Generally, as far as I know, those were following the concept from the later RVs. Also, as far as I know, none of them were professionally tested for a higher Vne (i.e., with proper instrumentation and analysis), although one or another of them may have been tested more casually.

Are you going to have a variable pitch propeller? With that much power, it would probably be desirable. Oxygen, of course.

How much does the engine installation add to the weight? That alone might suggest some structural changes, and of course the CG position is always a consideration.

A related alternative is a new build, of the F1 Rocket instead of an RV-3 and installing that engine of yours, and I believe I'd choose that instead. They are updating their website so you'll need to contact Vince directly. [email protected]

Good luck - and please keep us posted.

Dave
 
Last edited:
ooh boy. You'll definitely be into the design margins. There's lot to consider but I'll state what I think I know, i.e.wings. The rockets re-station the ribs 1" closer throughout the length. There's also the ~4" wider fuse this also effectively "shortens" the wings. There's the wing skin thickness increase but that's not universal and reportedly for CS versus dimpling ops to save time.

While it obviously results in a stronger wing, the stiffness is improved to decrease structural divergence/increase aero-flutter margin. This is (IMO) the most unforgiving aeronautical design alligator of them all. Proceed with caution.
 
Sounds like a fun project-----keep us informed as you move along.

Unfortunately, I have to finish two car Restomod projects first (work obligations) and build myself a new workshop. So I can only start in 8-10 months from now, at the earliest. Gathering information and keeping my mind sane from lack of aircraft projects in the meantime. But will do, in time.
 
Perhaps you could find Dave Anders' discussions about his very fast RV-4. In at least one of them, he outlines the major structural changes he put in. Among other things, he built a new empennage with modifications.

A couple of builders added mass balancing to the RV-3 tail. Generally, as far as I know, those were following the concept from the later RVs. Also, as far as I know, none of them were professionally tested for a higher Vne (i.e., with proper instrumentation and analysis), although one or another of them may have been tested more casually.

Are you going to have a variable pitch propeller? With that much power, it would probably be desirable. Oxygen, of course.

How much does the engine installation add to the weight? That alone might suggest some structural changes, and of course the CG position is always a consideration.

A related alternative is a new build, of the F1 Rocket instead of an RV-3 and installing that engine of yours, and I believe I'd choose that instead. They are updating their website so you'll need to contact Vince directly. [email protected]

Good luck - and please keep us posted.

Dave


ooh boy. You'll definitely be into the design margins. There's lot to consider but I'll state what I think I know, i.e.wings. The rockets re-station the ribs 1" closer throughout the length. There's also the ~4" wider fuse this also effectively "shortens" the wings. There's the wing skin thickness increase but that's not universal and reportedly for CS versus dimpling ops to save time.

While it obviously results in a stronger wing, the stiffness is improved to decrease structural divergence/increase aero-flutter margin. This is (IMO) the most unforgiving aeronautical design alligator of them all. Proceed with caution.

I KNOW that this is not something to take on lightheartedly. Hence my questions about the HR-I. What was its Vne, and what was done (and how was it proven) to make this possible? Without this, mine will keep its 210 mph Vne and the power will then only be for climb performance (which would be fine with me as well, quite frankly).

I know of 4 RV-3's with counterbalanced empennages, RV-4 style. Wondering what was done to the HR-I, exactly.

Constant speed MTV-12 prop is on its way here as we speak. Same one I have on my Jodel, and a large part of its awesome performance. That thing just hauls, especially with the low RPMs that I am running it. (Cruise on the Jodel is 4200 engine rpm, which translates to only 1800 rpm on the prop).

Normally aspirated, the EJ-25 setup, including radiators and the like is comparable to IO-320 weight. Maybe a few lbs heavier. Adding a turbocharger will bring it to parallel-valve IO-360 weights.

CG-wise I have it easy. For one, the radiator setup is behing the CG, offsetting the weight penalty up front. Plus the CG of the engine/gearbox combo itself sits closer to the firewall than that of a Lycoming. And then there is the battery to play around with. All in all, the CG is not all that hard to get just right.

Oxygen is useless for me. I can't go higher than 9500 ft anywhere in this country, and I need to fly out of my way to get even that. Typically, I am limited to 6500 ft.It's all class A above that.

And building an F1 rocket now would kind of defeat the purpose of this build. I've got an RV3 that I want to upgrade, and I'll work within the limits of what that airframe will allow me to do.

And Scott, are you sure about the modified wing? They are on the HR-II. The mods on those are well-documented and well-known. But these are based on the RV4. But how about the RV3/ HR-I?? As far as I know, the only mods are RV4 empennage, fastback, wider firewall and angle-valve IO-360. But I need to be sure before I feel safe to raise the Vne limits on mine.....

Anyway. thank you for the insights!!!!
 
I actually just got some (brief) answers from John Harmon himself:

Tail of the HR-I was a regular RV-3 tail, albeit with .020 skins on the elevator and mass-balanced in the same way the RV-4 is balanced.

Vne of 210 knots, cruise 190 knots (241 mph / 218 mph)

No other changes to the airframe....
 
Looks like John's first HR1 is for sale on BS. There are some logbook entries shown in the pics. One indicates a top speed of 218 kts in level flight was obtained and flutter tested up to 260mph.
 
.....And Scott, are you sure about the modified wing? They are on the HR-II...

Thank you for catching me on that. I tend to skim especially when viewing on the phone. I believe my info to be correct though not relevant to your question; story of my life. That said, sounds like my warning was already in your memory banks but I'll reinforce it. Can't speak directly for this design but the approach is common. Flutter, both aero surface and that related to structural divergence, for the (vast?) majority of EAB world was determined empirically. Dive the aircraft and smack the stick to try and force a function. Go not too much further back in aircraft development history and many of these limits were written in blood. There may be a few more but IIRC, the Glasairs were one of the few that got shaker table testing and some surprises were uncovered. I believe they were ex-Boeing engineers.

Again, sorry about the mistake. Keep us up to date on your progress. Wouldn't your project be much more fun with a direct drive diesel (he said hopingly)?
 
Looks like John's first HR1 is for sale on BS. There are some logbook entries shown in the pics. One indicates a top speed of 218 kts in level flight was obtained and flutter tested up to 260mph.

VERY interesting. Thanks!!
 
Thank you for catching me on that. I tend to skim especially when viewing on the phone. I believe my info to be correct though not relevant to your question; story of my life. That said, sounds like my warning was already in your memory banks but I'll reinforce it. Can't speak directly for this design but the approach is common. Flutter, both aero surface and that related to structural divergence, for the (vast?) majority of EAB world was determined empirically. Dive the aircraft and smack the stick to try and force a function. Go not too much further back in aircraft development history and many of these limits were written in blood. There may be a few more but IIRC, the Glasairs were one of the few that got shaker table testing and some surprises were uncovered. I believe they were ex-Boeing engineers.

Again, sorry about the mistake. Keep us up to date on your progress. Wouldn't your project be much more fun with a direct drive diesel (he said hopingly)?

Will keep you guys posted, once I finally find the time to work on the project.

And standing on the shoulders of giants (John Harmon in this case) makes things a bit easier. Still, I really need to be extremely cautious here, I know. Thank you.

And sorry, no direct drive Diesel for me. I've experimented with Diesel powered aircraft engines in the past. Pretty hard to get rid of the torsional vibration issues. Mainly because they are inherently vibrating. Modern Diesel engines tame that vibration by injecting up to five pulses of fuel into the engine per explosion. And they measure shake and then vary the amount of fuel that goes into each cylinder to compensate. You can see how mounting a prop to that might (or might not) confuse an ECU there.

And then direct drive automotive engines are not a good idea. Your prop wants to turn slowly. That 2700 is already a hefty compromise towards the omission of a gearbox. If possible, you want it much slower. However, you then need loads of displacement to compensate. After all, an engine is nothing but an air pump. Turn slowly, need to displace more air per revolution.

And big displacement automotive Diesel engines are HEAVY.

So no, no direct drive Diesel for me, sorry... I'll stick to my Subarus....

Will keep you guys posted! And thanks for the input
 
Hans, let me know if you want to buy my Subaru STI. You know where to pick it up!
 
..............And sorry, no direct drive Diesel for me. I've experimented with Diesel powered aircraft engines in the past. Pretty hard to get rid of the torsional vibration issues. Mainly because they are inherently vibrating. Modern Diesel engines tame that vibration by injecting up to five pulses of fuel into the engine per explosion. And they measure shake and then vary the amount of fuel that goes into each cylinder to compensate. You can see how mounting a prop to that might (or might not) confuse an ECU there.

And then direct drive automotive engines are not a good idea. Your prop wants to turn slowly. That 2700 is already a hefty compromise towards the omission of a gearbox. If possible, you want it much slower. However, you then need loads of displacement to compensate. After all, an engine is nothing but an air pump. Turn slowly, need to displace more air per revolution.

And big displacement automotive Diesel engines are HEAVY.

So no, no direct drive Diesel for me, sorry... I'll stick to my Subarus....

I'm aware of most of these issues but these performance numbers, no PRSU, diesel/jet fuel still make me fantasize.

https://vansairforce.net/community/showthread.php?t=198566
 
To fly with confidence, three years ago I conducted some calculations and measurements on the empennage of my Rocket F1 EVO ( AIEO540 cold air, 10: 1, 300hp ) including a ground vibration test and subsequent numerical analysis.
If anyone is interested in the result, click here for the link to the relevant documents.
https://www.dropbox.com/sh/bgy8ycrj423s1ml/AACqZpD-CPajaEKAOh-ik2M-a?dl=0
A spreadsheet prepared using report 45 allows a first precautionary approach to the calculation of the rudder counterweight, then a compromise iteration between the response to the flexion and the torsion of the fuselage allows to improve the speed margin.
Maximum calm air speed tested is 274 kn.
All the data are only referred to F1 Rocket I-TERA.
Claudio
 
To fly with confidence, three years ago I conducted some calculations and measurements on the empennage of my Rocket F1 EVO ( AIEO540 cold air, 10: 1, 300hp ) including a ground vibration test and subsequent numerical analysis.
If anyone is interested in the result, click here for the link to the relevant documents.
https://www.dropbox.com/sh/bgy8ycrj423s1ml/AACqZpD-CPajaEKAOh-ik2M-a?dl=0
A spreadsheet prepared using report 45 allows a first precautionary approach to the calculation of the rudder counterweight, then a compromise iteration between the response to the flexion and the torsion of the fuselage allows to improve the speed margin.
Maximum calm air speed tested is 274 kn.
All the data are only referred to F1 Rocket I-TERA.
Claudio

Very interesting Claudio, although not entirely easy to follow for someone who is not a structural engineer.

Bottom line: you've had some (mild) flutter issues on your rudder and now managed to solve these and ended up with a Vne of 250 knots?

I really would like to understand all of this and apply it to the RV3. I'll do some more studying.... Thanks!
 
Yes, that's exactly what happened.

I am unable to say if the rudder I initially made was inadequate due to some assembly or I asked more than permitted.

The flutter of a control surface does not depend only on the static equilibrium ( counterweight ) but on a series of other factors, some geometric others structural (e.g. elasticity of the rudder and elasticity of the fuselage, all according to the 3axes of movement and the axis of rotation of the surface ).
To evaluate the flutter tendency you need to know these factors.
It is possible to simulate everything with a multibody program but it is necessary to know and describe in detail every element of the plane for example all the rivets and the way in which they couple the sheets ...plus I didn't have the whole 3D project of the F1Rocket EVO.
The way currently used to overcome this difficulty is to derive the elastic modulus and the moment of inertia of the parts examined from their behavior when they are made to vibrate at the frequencies of the possible flutter, in our case from few Hz to 60-70.
Knowing the typical oscillation frequencies of the tail (but also of the wings and ailerons if needed) and of the fuselage, the old and not error-free method explained in report 45 can be applied.
It is already a good step in the direction of improving the safety.

It seemed to me, from what you wrote, that you want to make a fast plane
by modifying projects already tested .
It means approaching the design area with the obvious consequences.
It is not at all sure that by stiffening the rudder you move away from the flutter, if you do not know the resonant frequencies of the fuselage you could, by stiffening it, make a rudder that oscillates in phase with the fuselage at the same frequency, getting closer to the flutter than before.
Be careful.
Feel fee to write me privately, perhaps the technical details and calculations are a little too verbose for the forum.

Claudio
 
Back
Top