# Direct Drive Honda/B&S limits

Discussion in 'Firewall Forward / Props / Fuel system' started by LHH, Jan 13, 2020 at 2:30 AM.

### Help Support HomeBuiltAirplanes Forum by donating:

1. Jan 13, 2020 at 2:30 AM

### LHH

#### Active Member

Joined:
Aug 1, 2013
Messages:
30
3
Location:
IL
I have been reading the Briggs conversion messages.
I understand moment of inertia issues, twisting and bending to some extent, but what is the best way to arrive at a factual answer?

For example, X industrial engine can be used as a direct drive if the propeller weight is less than x and or RPM is less than X. If propeller weight is greater than X or RPM exceeds X, then a reduction drive must be used to handle the load.
Can this be calculated accurately using data from the engine manufacturers?
I am sure I am missing a fact or two, but you get the idea.

2. Jan 13, 2020 at 6:33 AM

### TiPi

#### Well-Known Member

Joined:
Aug 25, 2019
Messages:
160
155
Location:
Mackay (AUS)
There is no simple answer as there is no available engineering data for these engines.
Honda, Kohler (and possibly Subaru & Kawasaki) have published some data for max belt loads on the flywheel side and max overhang on the PTO for the rated loads.
The fitting of a propeller is way out of the scope from the OEM for these engines. Unfortunately, you will have to do the calculations or testing yourself, or copy someone else’ work.

regarding direct/re-drive: this is primarily determined by the propeller (ideal diameter and pitch for your airframe) and the required power from your engine. If they happen to be close to each other with regards to engine rpm, the best compromise is direct drive. If they are more than maybe 30% apart, a re-drive might be the better option.
If you are weight-constrained, you can go 2 ways as well:
Direct-drive with a heavier prop and ditch the flywheel (see Columban Luciole)
High-revving small engine like PPC

3. Jan 13, 2020 at 6:53 AM

### Vigilant1

Joined:
Jan 24, 2011
Messages:
4,240
1,973
Location:
US
Yes, I'll echo TiPi. There are no published rules/formuli. There are some things that can be readily calculated (e.g. the side-loads on the "prop bearing" given a particular prop weight and distribution, RPM, and angular rate of change, etc), but that still leaves a lot of unknowns about the important stuff--are there torsional vibration issues (direct drive >or< PSRU), etc.
Generally, I have the impression that some brave pioneers do the best math they can, apply basic principles to reduce the likelihood of failure, test their setup on the ground and eventually in flight, and gradually build flight time and confidence. The rest of us make a LOT of use of the knowledge gained by the pioneers.
In your boots, I think the first step might be to get some idea if a PSRU is likely to be beneficial for your envisioned application (this usually boils down to how fast you'll be flying--a PSRU lets you turn a larger prop at lower RPMs, and that's more useful at lower airspeeds). After that, maybe look at what others have done with the same engine and layout, and look at the success/problems they are having.

Last edited: Jan 13, 2020 at 6:58 AM
BoKu likes this.
4. Jan 13, 2020 at 10:43 AM

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
7,183
2,044
Location:
North Carolina
Prop diameter and RPM are constrained by keeping the prop tips subsonic. Allowable bending loads depend on prop RPM, MOI and gyration rate. Figures for these loads are not published, so you can either guesstimate, or test. Plagiarism of known good setups will save a lot of time...
Even a lightweight prop wil have a higher MOI than flywheel.

5. Jan 13, 2020 at 3:31 PM

### LHH

#### Active Member

Joined:
Aug 1, 2013
Messages:
30
3
Location:
IL
Thanks for the input.
Am I correct that calculating numbers for the flywheel is the only potentially close number?
For example, if flywheel weighs 15 lbs and is rated for 3,600 rpm, a number can be estimated or calculated. Then prop can be estimated, if prop forces are smaller than flywheel numbers, then a potential exists for direct drive?
This would mean very small composite propellers like the FES system for electric gliders. Each blade weighs 9 ounces( 260 grams) for a one-meter prop diameter.
A 1-meter prop weighing 18 ounces at 3,600 rpm could be calculated and could be compared to a 15 lb flywheel at the same rpm right?
I understand prop forces will be different compared to the flywheel, but looking for some grain of factual data vs 100% guessing.

6. Jan 13, 2020 at 3:41 PM

### BBerson

#### Well-Known MemberHBA Supporter

Joined:
Dec 16, 2007
Messages:
12,450
2,511
Location:
Port Townsend WA
The flywheel stub on my GX 670 is about the same diameter as a VW conversion.

7. Jan 13, 2020 at 5:23 PM

### Vigilant1

Joined:
Jan 24, 2011
Messages:
4,240
1,973
Location:
US
As you may know, you'd want to compare the moment of inertia (MOI) of a prop to a flywheel, not compare their mass. Mass that is way out on an 18" lever arm has more inertia than the same mass located 4" from the shaft.
But, that is only a small part of the issue. When the engine is running and is jostled or turned, it generates loads on the crankshaft and the bearings. How much of this loading was the engine designed to take/does it take in lawnmower service? Maybe a lot different from an airplane.
And, you can go >too< small on the inertia, too. The engine needs that flywheel effect to run smoothly, especially near idle.
And in some cases thrust loads are important (pushing or pulling on the crankshaft by a prop). There's not much of that in a lawnmower. (FWIW, these loads are relatively small and these engines seem to do okay with thrust loads in actual service, but it needs to be checked).
And, if the flywheel is retained but the prop is put on the other end (i.e. the PTO end of a typical B&S, which would be the natural inclination of someone who hasn't looked at the issue), now you have a significant rotating mass on either end of the crankshaft. The resulting torsion on the crank results in breakage of the crank--sooner or later. A PSRU pulley might be okay on the PTO end even with a heavy flywheel on the other end, because the PSRU pulley has a lower MOI than a prop.

There's a lot to this. If you want to responsibly design your own system from scratch, be prepared to read a lot and do a lot of math, there are no rules of thumb that can keep you 100% clear of problems. OTOH, if you faithfully replicate the design of something that has hundreds or thousands of hours of successful use in the same type of flying you'll be doing, your chances of success go up. If you vary from a known design, understand the implications fully.

Last edited: Jan 13, 2020 at 5:32 PM
8. Jan 13, 2020 at 9:27 PM

### LHH

#### Active Member

Joined:
Aug 1, 2013
Messages:
30
3
Location:
IL
I was rationalizing this approach due to the Onan use for years in the Quickie and seeing the DA-11 at an airshow in Texas in the 1990's. Onan was underpowered but seemed to work without major issues.

9. Jan 13, 2020 at 9:48 PM

### Vigilant1

Joined:
Jan 24, 2011
Messages:
4,240
1,973
Location:
US
Yes, and there are now several designs flying with small industrial twins in direct drive mode that are accumulating many trouble-free hours in (almost) standardized configurations, you'd be unlikely to go wrong by replicating those. If you just use the same key principles they adhere to, you'd still be safer than just eyeball engineering everything. As far as PSRUs, there are a lot of them mated to industrial twins in the trike community (where high thrust at low airspeed is paramount). I don't know if there are standardized setups with a lot of hours.
What type of aircraft are you wanting to power with one of these engines?

10. Jan 13, 2020 at 10:48 PM

### Dana

#### Super ModeratorStaff Member

Joined:
Apr 4, 2007
Messages:
8,767
3,124
Location:
CT, USA
The prop loads are one thing, remember you have gyroscopic forces, which will be much larger than a heavier flywheel of smaller diameter, and loads from propeller imbalance, the best prop balance probably won't be as good as the flywheel's balance.

The other aspect is will a small faster turning prop be adequate for the aircraft's performance envelope? A small fast turning prop, i.e. direct drive, can work reasonably efficiently on a low drag fast ship, but will be much less efficient at lower airspeeds.

wsimpso1 likes this.
11. Jan 13, 2020 at 11:19 PM

### LHH

#### Active Member

Joined:
Aug 1, 2013
Messages:
30
3
Location:
IL
Motor glider with tractor engine.
Trying to work with the Honda iGX 800 at 28 hp. 28 hp is a pretty common hp in small motor gliders. Prop dimensions are be limited due to ground clearance. 42 to 44 inches maximum.
Trying to find an EFI four stroke that will not require major reworking with a small frontal area.
Honda seems to have the smallest

12. Jan 13, 2020 at 11:39 PM

### Vigilant1

Joined:
Jan 24, 2011
Messages:
4,240
1,973
Location:
US
Honda makes good engines. There are some reasons they aren't more popular for conversions, but I can't recall what they are (induction? Unitary jug/head?). Others here know about them, and Bberson has firsthand experience.
If you are limited to a 44 inch prop and using 28 hp, you'll probably want a direct drive.

13. Jan 13, 2020 at 11:48 PM

### Vigilant1

Joined:
Jan 24, 2011
Messages:
4,240
1,973
Location:
US
Oh, yeah:. Topaz is working on a motorglider design. At this point his plan is to use a 22hp engine from Harbor Freight, direct drive.

14. Jan 14, 2020 at 12:08 AM

### TiPi

#### Well-Known Member

Joined:
Aug 25, 2019
Messages:
160
155
Location:
Mackay (AUS)
That new Honda looks like a nice engine. BUT it is heavy at 48.2kg. The other unknown is the suitability of the fly-by-wire electronic system for the aircraft application.

15. Jan 14, 2020 at 12:18 AM

### BBerson

#### Well-Known MemberHBA Supporter

Joined:
Dec 16, 2007
Messages:
12,450
2,511
Location:
Port Townsend WA
The GX800 is 779cc. I wonder what the price is?

16. Jan 14, 2020 at 2:51 AM

### LHH

Joined:
Aug 1, 2013
Messages:
30
3
Location:
IL
$1999 according to local dealer 17. Jan 14, 2020 at 3:36 AM ### wsimpso1 ### wsimpso1 #### Super ModeratorStaff Member Joined: Oct 19, 2003 Messages: 6,063 Likes Received: 3,349 Location: Saline Michigan You can get Mass Moment of Inertia by modelling up the flywheel in SolidWorrks or build and use a bifilar torsional pendulum. Then you can do a similar thing with an appropriate propellor. But you are not done yet. You will have to estimate the max pitch-yaw rotation speed of the base engine in its intended use to estimate max intended gyroscopic moment on the shaft, then add the moment due to belt loads. I suspect that these pitch and yaw rotation speeds are pretty small compared to what they can be in even non-aerobatic airplanes. The other way is to find out the crankshaft material and heat treat and then estimate the max bending moment under fully reversed bending fatigue. Billski 18. Jan 14, 2020 at 4:21 AM ### Vigilant1 ### Vigilant1 #### Well-Known MemberLifetime Supporter Joined: Jan 24, 2011 Messages: 4,240 Likes Received: 1,973 Location: US Some thoughts on frontal area: 1) Given the basics of all these engines (same cylinder V angle, very similar stroke length, etc), the differences in total frontal area between manufacturers will be small and other practical differences that affect the ease of conversion, ultimate expected reliability, weight, etc will likely prove more important than any minor differences in frontal area. 2) All of these engines are a bit "blocky" compared to a horizontally opposed engine. The prop extension used by Microsport on the SD-1 goes a long way in allowing for a more streamlined cowling. The extension undoubtedly increases bending load on the crankshaft, but these planes have been flying with these extensions with no problem. 3) Similarly, a large spinner can help a lot in reducing the abruptness of the nose. 4) TiPi is working on a "heads down" mounting of the Vanguard 810cc engine (see his build thread and the accompanying discussion thread on it). For your application, mounting the engine in this way might reduce drag a bit. See his cowling illustrations in his build thread for details. 19. Jan 14, 2020 at 5:14 AM ### litespeed ### litespeed #### Well-Known Member Joined: May 21, 2008 Messages: 1,519 Likes Received: 307 Location: Sydney You might like the Honda brand, but I would not put one in a aircraft. They are hugely overpriced and under engineered for the price. Given you want a smallish prop, You should look at getting a three bladed prop to absorb the power and give a good takeoff- which small props are bad at. Power only is used when needed but a real problem if you dont have enough. Get a good clone for under a$1000 and spend the extra on quality internals to make make power in the rev range you want and it will be much more reliable.

A set of quality carbs is probably a lot better than a cheap simple efi for reliability. And far cheaper.
That extra $1000 for the Honda buys a lot of high quality parts to make a much more powerful but more reliable engine. The Honda is just a overpriced version of the same motors and will be made in China most likely anyway. A nice version with forged rods, quality valves springs and retainers, a billet cam, some nice forged pistons with higher compression, header for you rev range, a proper set up not the loud but useless one standard. A quality set of carbs on properly designed intake tubes and a high output ignition set up. Add some clean up porting to make it all flow right for your revs and bingo a far more powerful version for any given rev and inherently much more reliable. These are mods that will be done for any serious air motor so just start cheap and get a great motor for the same price as a heavy Honda. Now you make sure its all balanced nice then add a prop flange extension and a suitable flywheel as needed. No out of the box engine is unmodified to fly, so spend for the best bang for buck. If you want EFI, a good solution are motor bike ones. Lots of reliable cheap complete systems on modern twin bikes. Add a cheap efi controller thats programmable and you are set. This would be a lot better than any mower efi system. And you can double up the boxes for redundant ignition is need be. You are not trying to get a very high power engine, but one that makes more power, more reliable and smoother without going pasts its heat budget. The Twin can be a great engine if you are careful and invest wisely. The extra$1000 for a Honda sticker is whats known as a fools tax.

20. Jan 14, 2020 at 12:20 PM

Joined:
Nov 5, 2010
Messages:
115