The use of mixed engines isn't unique, but I would have concerns that any natural frequencies that could cause problems have a far greater chance of happening. Synchronizing changes the available modes for vibration, but you'll need to better characterize the airplane and engines to determine if that's a problem. Are the new props heavier or with a greater moment of inertia? Getting the best performance will take more work, but if some inefficiency is acceptable manual controls should work. If the engines are higher thrust it will mean the pilot needs to be ready to react more quickly if one of them fails than would otherwise be expected.

It's my observation that every combination of engines and placements has been tried without necessarily generating failure, but just limited by cost of operations. I would expect a lot of ground testing to check as many contributions as possible.

OK,
With 4 props on an existing aircraft there's only a few examples to choose from. I'll let that guide the way I picture what you're doing.
The starting point is that out-of-sync propellers are very annoying. Wahh-wahh-wahh from the old World War II movies, droning on for hours and hours.
If it's an existing aircraft like the ones I'm thinking of, it already has a synchrophase system and you should look up the functions in the AMM and POH to understand how it works. Then dig into the prop maintenance manual. Props have their own TCDS and maintenance manuals. Since the plane already has 4 engines, the current synchro system is designed to regulate 1 master and 3 slaves. It won't be able to cope with 2 systems at different speeds without alteration, but there are plenty of ways the synchro could be altered to suit the two different ranges on board. You have to be willing to void the warranty to do this, of course... the planes you're dealing with are terribly old, so digging into the pulse generators should give away the obvious secrets pretty quickly to a practiced eye.

Please remember: we're not all rednecks!
www.sparweb.ca

Random thoughts

4-engines??? All Turboprop? Weight/size? Lockheed Electra class? or smaller [UAV]?

Seems that 4-engine jets might be a better testbed match if [2] are replaced with turboprops. Also... Why mod the OB engines [#1, #4]? IF a testbed... why not place the test engines inboard [#2, #3] so the 'known quantities' [reliable engines] are in the most critical engine-out location?

Regards, Wil Taylor
o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

Is the issue that the the different engine types will need to run at different speeds? If the blade passing frequency is different, then synching between the two types is not possible. Human beings can't detect phase between different frequencies, unless they are very similar in frequency (within a few Hz).

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

thx guys, great responses.

full size plane (not UAV)
why outbd ? replacement engines slightly less power than OEM, and inbd nacelles are more problematic.

Greg, thx. you feel that this may not be the problem I think/fear it could be ? That with the very different prop speeds then the interference may be acceptable. As proof of concept I guess we'd be willing to "suck it and see" ...

another day in paradise, or is paradise one day closer ?

rb…

Are You pursuing limited flight at relatively low speeds/Gs [initial proof of concept, engine shut-down/restart/feather... initial fuel-consumption/operating-parameters, etc, etc]... or are You looking at flight up to the edge of the envelop? Why [2] engines... not just [1] to 'tickle the dragon'.

Even-though 'beat-frequency' phenomena [whaa-whaa-whaa-whaa-whaa…] is highly irritating to the crew even in small doses in a MIL turbine cargo bird [ever ridden in the back of an ancient/high-time C-130E or a CH-47C for hours?]... sudden deep/interacting harmonics and/or 'prop whirl' are very dangerous/destructive phenomena.

WARNING. I learned an early/valuable lesson from an under-performing 'dog' OV-10A. It scared the operators when another OV-10 was lost during an over-mountain flight and had to shut-down engine due to oil loss... crew ejected.

The particular 'dog' was brought to SA-ALC for evaluation by us OV engineers and mechanics. I inherited it after a few weeks of frustration and getting no where. There were no obvious structural or mechanical issues for its terrible performance numbers. WORSE... Local USAF mechanics were useless when it came to the engines/props... which I suspected were the true culprits. When managers finally invited North American mechanics/engineers into the problem [they angrily dismissed me from the effort... p*sssed me off] the NA folks discovered major issues with engine operation... both had become detuned and operating sub-par [~90% power]… AND worse-yet both of the props were in horrible shape... blades not matched, pitch angles miss-rigged and [significantly] were NOT dynamically balanced. One prop was replaced. Both props were re-rigged and dynamically re-balanced and the engines tuned for 'peek performance'.

After the VERY first post-maintenance test flight, the pilot was shaking his head in disbelief that it was the same aircraft that he'd flow just weeks ago: climb, cruise, throttle-response, fuel consumption, systems... were all at/above 'the book' making it literally the best OV he had ever flown. NOW, what he was-MOST-impressed-with was how uniquely QUIET and SMOOTH and responsive it was over-all... the noises and low level vibrations he assumed were normal... simply weren't there; and the flight controls [stick/rudder-pedals] 'felt' butter-smooth. His smile of disbelief said it all. Optimum engine tuning and prop balancing can make a huge difference in any aircraft.

NOTE. This was NOT the first-time that props repaired/overhauled/managed at WR-ALC were problem children... I've investigated a terrible O-2A prop-hub-blade failure mishap [1984?], several C-130 prop governing/lubrication issues [over-seas]... and, just a few years ago... the USMC suffered a C-135E prop-blade failure, that led to a catastrophic in-flight break-up.

Wind-tunnel testing might prove useful.

I am sure the OEM for Your test-bed might have some ideas.

Regards, Wil Taylor
o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

OK, what are the number of props, rpms, for the two engine types? I'll put a simulation together.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

4 props, 2 (1,4) at 1900rpm and 2 (2,3) at 1200rpm. maybe the fuselage isolates the 2 wings (so modelling 1 and 2 is sufficient) ?
maybe the issue is synchro-phasing the two pairs (1,4) and (2,3) ?

cheers,
greg
(you're in victoria right?)

another day in paradise, or is paradise one day closer ?

Yeah in the Otways, you aren't involved with that Liberator in Werribee are you?

That's how I'd play it, assuming 4 blades per prop. That is, synch 1 and 4, and regard 2 and 3 as a separate synch problem.

Musically 1900/1200 is not a neat relationship B2 to F#3, but I don't think that's a big deal, it's not 'nice' but it isn't too close as it is well outside the subjective bandwidth of the ear so beating (heterodyning) is not an issue.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Greg, slight tangent, it seems to me there are further components to the frequency problem besides the prop rpm; i.e. the number of blades in each case, and, if the prime mover happens to be a recip, the firing frequency. According to your expertise, what role if any do these play?

"Schiefgehen wird, was schiefgehen kann" - das Murphygesetz

yes, I also figured you needed more info.
1200 rpm props have 4 blades, 3.4m diameter, 3.6m off CL
1900 rpm have 3 blades (yes, I know this makes the problem worse, as 1200*4 is close to 1900*3), not sure of diameter (maybe 2m), 7.2m off CL.

thx.
"Liberator in Werribee" ... nah, not me mate. Did my uni in Sydney (in the 70s), worked at CAC for a bit, then got a job in UK, rest is history.

another day in paradise, or is paradise one day closer ?

Lou- Blade passing frequency (and its 1,3,5,7... harmonics) is probably the dominant noise in the cabin . Howveer, yes the engine firing frequency and its harmonics will add to the general fun, I just don't think they'll be dominant.

rb1957, ok, I'll take a look at that. Fundamentals of 95 vs 80 hz don't ring alarm bells.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

thx very much

another day in paradise, or is paradise one day closer ?

The three files above had a very low sample rate, and although they sounded good they were wrong

here's engines 1 and 4 only

%User input
N_blades=[3,4,4,3];
nominal_rpm=[1900,1200,1200,1900];
harmonic_structure=[1,0,.3,0,.1,0];%integer multiples of BPF, use complex for phasing
rpm_error=[0,0,0,0]; %usually LHS will be 0
wingpos=[-7.2,-3.6,3.6,7.2];%location in m. + and - don't matter
amplitude=[0,1.5,1.5,0];%arbitrary
amplitude=[1,0,0,1];%arbitrary
duration=10;%seconds
dt=1/4000.;%seconds
wavfileout='waaWaa14only_0_0_0_0.wav';

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

• https://files.engineering.com/getfile.aspx?folder=3cec431a-5e5f-47bc-8695-6

2 and 3 only
%User input
N_blades=[3,4,4,3];
nominal_rpm=[1900,1200,1200,1900];
harmonic_structure=[1,0,.3,0,.1,0];%integer multiples of BPF, use complex for phasing
rpm_error=[0,0,0,0]; %usually LHS will be 0
wingpos=[-7.2,-3.6,3.6,7.2];%location in m. + and - don't matter
amplitude=[0,1.5,1.5,0];%arbitrary
duration=10;%seconds
dt=1/4000.;%seconds
wavfileout='waaWaa23only_0_0_0_0.wav';

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

• https://files.engineering.com/getfile.aspx?folder=32e7d338-b8f5-465f-9b82-4

All synched- pretty horrid already!

N_blades=[3,4,4,3];
nominal_rpm=[1900,1200,1200,1900];
harmonic_structure=[1,0,.3,0,.1,0];%integer multiples of BPF, use complex for phasing
rpm_error=[0,0,0,0]; %usually LHS will be 0
wingpos=[-7.2,-3.6,3.6,7.2];%location in m. + and - don't matter
amplitude=[1,1.5,1.5,1];%arbitrary
duration=10;%seconds
dt=1/4000.;%seconds
wavfileout='waaWaa_0_0_0_0.wav';

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

• https://files.engineering.com/getfile.aspx?folder=3e7f05d9-c734-4bd2-b1c8-d

5 rpm error on 3, 10 on 4

N_blades=[3,4,4,3];
nominal_rpm=[1900,1200,1200,1900];
harmonic_structure=[1,0,.3,0,.1,0];%integer multiples of BPF, use complex for phasing
rpm_error=[0,0,5,10]; %usually LHS will be 0
wingpos=[-7.2,-3.6,3.6,7.2];%location in m. + and - don't matter
amplitude=[1,1.5,1.5,1];%arbitrary
duration=10;%seconds
dt=1/4000.;%seconds
wavfileout='waaWaa_0_0_5_10.wav';

Nasty

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

• https://files.engineering.com/getfile.aspx?folder=3b7939af-87b9-4c1c-a60f-f

I should add that I haven't used the distance data, so the contribution of 1 and 4 may be excessive, depending on how much is structure borne vs airborne.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Excellent, thx very much.

I should add no (or manual only) synchro for 1 and 4, but OEM (possibly !?) for 2 and 3 (depending on how the system masters).

another day in paradise, or is paradise one day closer ?

That's the easiest way, I don't see any particular virtue in trying to synch between the pairs. Incidentally you'll need a much better audio setup than I have here to really hear the fundamentals, 85 Hz is well into the roll-off of my decent speakers, and I ain't playing crude electronic crap through them. I'll dig a gaming subwoofer out of the pile of discarded toys and see if it makes more sense.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Spend some money on a real sub-woofer, one which approximates the circumstance: https://www.soundandvision.com/content/eminent-tec... . Pricey? Sure. Once in a lifetime spend. You may need to dedicate a room as the speaker enclosure.

what I meant was synch-ing 1 and 4 ... right now there's none (2 and 3 would be synch-ed by a kludge of the OEM system) ... not much thought has gone into this detail (yet) ...

another day in paradise, or is paradise one day closer ?

I would look to see if one engine is the master. Considering 1&4 are more influential it may be one of them.

Please ignore all the audio above. I have been finding out some undocumented assumptions that the wav writer makes.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

Here you are in glorious 32 bit floating point at 44.1 kHz (undocumented feature being you can write wav files in all sorts of formats, but only that or 48 kHz is actually recognized by the player. Harrumph).

so the 4 numbers in the title are the rpm errors, you have a baseline, then engine 3 out by 5, then engine 4 out by 7 , then both out. I have reduced the amplitude of the outer engines as they will be attenuated by up to 6 db compared with the inners.

Here's what i used

%User input
N_blades=[3,4,4,3];
nominal_rpm=[1900,1200,1200,1900];
disp(N_blades.*nominal_rpm/60)
harmonic_structure=[1,0,.3,0,.1,0];%integer multiples of BPF, use complex for phasing
rpm_error=[0,0,5,7]; %usually LHS will be 0
wingpos=[-7.2,-3.6,3.6,7.2];%location in m. + and - don't matter. Not used
amplitude=[1,2,2,1];%arbitrary
duration=10;%seconds
dt=1/44100.;%seconds
wavfileout='waaWaa_sin_0_0_5_7.wav';

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?

• https://files.engineering.com/getfile.aspx?folder=92e4147c-9126-468a-b966-8