Gyro Warning System

More testing to tune GWS to AR-1 in Beta testing phase.
160 rotor RPM pre-rotation, stick jammed back and gunned the throttle immediately on a 914 creates a flapping risk amber alarm but then rotor recovers back and you fly off.
Behind the curve warning speed calibration done.
Full power (Rotax 914 115 HP) very nose high attitude with single pilot and low fuel (light loading) results in rotor RPM going from 320/330 to 280 and stick starts to shake a little. DO NOT TRY THIS AT HOME. The recovery is done very gently reducing power and get the bunting risk warning (amber alarm). We didn't have the guts to get it to the point of getting bunting red alarm level which is closer to disaster.
Bunch of parabolic flight curves to induce the bunting risk warning. It actually gives the behind the curve warning first because the G loading went down to only 0.72 G and that does not indicate a bunt over is going to happen. Doing a high rate of descent at idle power and then shoving the stick forward getting nose down 45 to 50 degrees quickly got 0.67 G but no bunt warning came on as it didn't stay there much and recovered loading. A pull back to level from that attitude gave a 1.8 G peak.
More testing tomorrow.
 
Last edited:
Keep safe.

Hi Phil:
Thanks. Yes we are analyzing data and doing it stepwise based on what we see in previous steps/tests so we have a good idea how close or far we are. We are doing this so when pilots stuff it up, we know how far they must have gone to get to the point of blade sailing on takeoff roll and/or God forbid a fatal bunting over. So there is data proving how far things must have gotten for something to happen. We did pitch PIOs as well one after the other. starting 1G pitching over for 3 to 4 seconds getting to 0.66 G, losing rotor RPM quickly then pulling back, getting to 1.4 G and rotor RPM comes back just as quickly as it went away. The whole sinusoidal parabolic form was around 6 to 7 seconds on these tests. We could sit there and do this PIO for 10 minutes and rotor RPM decays and comes back and repeat and nothing happens. Reduce the engine power, leave the stick alone in neutral and it sorts itself out. Power before Pitch ... the old school mantra is correct.

Rotor RPM to 160, stick all the way back and full throttle immediately on a 914 and the flap angle was only 4 degrees. Our limit is 10 degrees. I believe ELA is 8 degrees according to Mike.

Edit - 1:45 pm 03/26/2022
1) Latest test for bunting - Got down to 0.47 G for just 0.5 seconds. No problem for rotor to recover and got the bunting risk alarm. If we had remained there a bit longer we would have gotten bunting red alarm. This is some indication of an idea how long you have to sustain low G before even getting the red alarm which we did not even get, we got amber alarm in 0.5 seconds of 0.47 G and then it will take a bit more to actually have your rotor hit something in the air (your tail usually) and start to torque roll. 0.5 seconds of below 0.5G did not even come close to that. The derivative or rotor RPM decay rate at this 0.47 G was -12 rotor RPM per second. Interestingly enough right now from limited data points the rotor RPM decay rate at 0.6x G is -10 rotor RPM per second.
2) 150 rotor RPM - aggressive stick full back and power full on on a 914 AR-1 - Flapping risk alarm but flapping angle only 4 degrees and it took off but it could get close if I was one up because the acceleration would be way quicker than 2 up. Also a more powerful engine like 915iS would give a higher acceleration. This shows how badly you really have to stuff things up to chop your tail. We list minimum rotor RPM to start ground roll at 160 in the POH. Better number to reach is 180 or higher

This setup is a great tool for certification testing I think. It gets rid of all the "opinions" and just gives the data graph of all these parameters to analyze. That is if you are a test pilot and got the balls to get this data on the machine you think is good.
 
Last edited:
Mike and Abid. Kudos to you for taking the time, trouble, expense and RISK associated with the development of this instrument. If nothing else, it promotes the discussion/understanding of the mechanics of the ever-changing flight dynamics the gyro experiences through the various stages of flight.
 
Mike and Abid. Kudos to you for taking the time, trouble, expense and RISK associated with the development of this instrument. If nothing else, it promotes the discussion/understanding of the mechanics of the ever-changing flight dynamics the gyro experiences through the various stages of flight.

Thanks for the encouraging words. They mean a lot to me and I am sure Mike and Chris Buchanan (SC). They have been right there with me and my other pilot David in at least half the flights executing these scripts for testing. Engineering with just theories coming from smart minds without experimentation and careful measurement (not subjective feel) is no engineering at all. Half the theories generally go right out the window when data is properly collected
 
It is a very very interesting project / piece of work and as you say it is nice to get some quantitative data upon something that has almost evaded such a thing even though we are in 2022!

This is very interesting:-

2) 150 rotor RPM - aggressive stick full back and power full on on a 914 AR-1 - Flapping risk alarm but flapping angle only 4 degrees and it took off but it could get close if I was one up because the acceleration would be way quicker than 2 up. Also a more powerful engine like 915iS would give a higher acceleration. This shows how badly you really have to stuff things up to chop your tail. We list minimum rotor RPM to start ground roll at 160 in the POH. Better number to reach is 180 or higher
On that flapping angle of 4 deg verse 10 deg full back stick - does that have much influence upon the rotor RPM acceleration? Also what is the environmental wind at that? i.e. if we did this same test into a 30knt head wind what speed would flap the blades at 150RRPM?
 
Interesting question Phil about flapping angle having what effect on rotor acceleration. Probably best to talk to me or Mike or Jean Claude on a DM.

As to your second question about wind. We had 14 knots crosswind on this trial. Stick in the center forward pre-rotation and takeoff attempt to get the flapping risk alarm and then abort. So not much headwind factor.

And yes you would flap earlier if you had a lot of headwind. Airspeed is airspeed. I know you know that. That is why 150 is not recommended for short takeoff technique. Best is to get to 180 or more. Otherwise don’t gun the throttle

Note: I am changing my technique for pre-rotation and takeoff in crosswind. Stuck into wind pre-rotation gets you much closer to rotor flap than stick centered
 
Last edited:
Our current thinking is that the GWS will only be sold to manufacturers. We’re pretty sure that we will have to spend some time with each as we train and help them carry out the installation and select the correct parameters for each gyro model. This requires some very precise flying to calibrate the GWS internal ASI and rev counter and to establish behind the curve and other parameters. The owner/pilot would not have access to these parameters only the manufacturer to limit the risk that someone changes something he doesn’t understand.
Hi out there,

my first post in this forum on this very intriguing thread:
How about the wear and tear of the instruments, aka ASI, RRPM-gauge, etc. Don't they run off the initial calibrated values during years and may affect the decisions made in the GWS about when to alarm?

Sincerely,
Peuqui
 
GWS does not rely on other instruments. It has its own ASI and Altimeter and OAT and rotor RPM measurement logic.
The drift that can produce a delta in its calibration needs to be zeroed every few months inside a hanger wit no wind and it requires only pressing a knob for a few seconds.
 
Two videos that should be watched in the order listed.
They tell the story of what happens to your rotor RPM as you go from 1 G to 1.8 G to 0.4x G and back
It also shows that as you do a vertical descent where you already lose rotor RPM a bit and then do a sudden pushover, even without power you will drop rotor RPM close to the danger zone. With power on in this you may go beyond recovery

1)

2)
 
A definite vote of thanks to Mike, Abid, Chris, and the team who are researching and obtaining the hard data on a critical area of gyroplane safety.

An area that has remained not well understood by many, and one where many inexperienced gyro pilots have come to grief.

The figures, procedures and recovery techniques they are discovering with the tools they are developing, are important for us to learn from, in order to stay safe in one of our most basic everyday functions in gyroplane operations and general handling.

Certainly big thank you from me, and I am sure all who have been following your efforts to date.
 
Last edited:
Leigh thank you for your post, it was a pleasure meeting you at Bensen Days, I hope you can get Blue in the air soon.
Mike
 
Likewise Mike, glad to see how this is progressing.

Denis was most impressed and enthusiastic, and good luck with the development of this important safety instrument.
 
The GWS considers two main Take Off (TO) flapping scenarios:

  • Pre rotation below POH Rrpm, stick back TO.
  • Correct pre rotation Rrpm but accelerating with the stick forwards.
I treated both equally and this, sort of, assumed that flapping accidents during TO were equally distributed between the two scenarios.

During early testing of the GWS I used the flapping angle measuring device attached to the rotor hub bar to record the flapping angle during TOs with ever decreasing pre rotator Rrpm to test scenario 1. I found that I could still TO with very low Rrpms without excessive flapping angles and asked myself and Jean Claude

“how are these guys managing to flap their rotors?”

While Beta testing with Abid at Silverlight we started at his normal 180 rpm pre rotation and I kept asking him to pre rotate to lower and lower rpms and not to open the throttle gently but go to full WOT. Abid was understandably very reluctant. He, like most (perhaps all) manufacturers, had no way of knowing where the lower Rrpm limit was without actually flapping a rotor. With the flapping angle measurement fitted to his AR1 he could see after each flight that the flapping angle hadn’t increased to a dangerous level and eventually, he said

“how are these guys managing to flap their rotors?”

We came to the same conclusion that most of the flapping TO accidents we are seeing are probably due to the stick not being fully back after pre rotation.

Because I also realised just how much faster Abid’s 914 powered AR1 accelerated on a smooth runway than my tired old 912 powered ELA accelerated on grass, I’ve gone back to the GWS logic and algorithms to see what could be done to make the “TO with stick forward” alarm more reactive.

We are now looking at incorporating a new alarm using a stick position switch and engine rpm to give a very early warning of a stick forwards acceleration.

My question is what should the alarm message be?

“Stick position”

“Stick, stick, stick”

“Stick not back”

“flapping risk”

anything other suggestions?

Keep in mind that I do not want to give an instruction, as I’ve said many times, I am not a CFI and am not qualified to tell you what to do. I’m an engineer, I can only tell you what is wrong. The gyro manufacturer or the owner can record whatever message he thinks will get him to take the appropriate corrective action.

Mike G
 
The GWS considers two main Take Off (TO) flapping scenarios:

  • Pre rotation below POH Rrpm, stick back TO.
  • Correct pre rotation Rrpm but accelerating with the stick forwards.
I treated both equally and this, sort of, assumed that flapping accidents during TO were equally distributed between the two scenarios.

During early testing of the GWS I used the flapping angle measuring device attached to the rotor hub bar to record the flapping angle during TOs with ever decreasing pre rotator Rrpm to test scenario 1. I found that I could still TO with very low Rrpms without excessive flapping angles and asked myself and Jean Claude

“how are these guys managing to flap their rotors?”

While Beta testing with Abid at Silverlight we started at his normal 180 rpm pre rotation and I kept asking him to pre rotate to lower and lower rpms and not to open the throttle gently but go to full WOT. Abid was understandably very reluctant. He, like most (perhaps all) manufacturers, had no way of knowing where the lower Rrpm limit was without actually flapping a rotor. With the flapping angle measurement fitted to his AR1 he could see after each flight that the flapping angle hadn’t increased to a dangerous level and eventually, he said

“how are these guys managing to flap their rotors?”

We came to the same conclusion that most of the flapping TO accidents we are seeing are probably due to the stick not being fully back after pre rotation.

Because I also realised just how much faster Abid’s 914 powered AR1 accelerated on a smooth runway than my tired old 912 powered ELA accelerated on grass, I’ve gone back to the GWS logic and algorithms to see what could be done to make the “TO with stick forward” alarm more reactive.

We are now looking at incorporating a new alarm using a stick position switch and engine rpm to give a very early warning of a stick forwards acceleration.

My question is what should the alarm message be?

“Stick position”

“Stick, stick, stick”

“Stick not back”

“flapping risk”

anything other suggestions?

Keep in mind that I do not want to give an instruction, as I’ve said many times, I am not a CFI and am not qualified to tell you what to do. I’m an engineer, I can only tell you what is wrong. The gyro manufacturer or the owner can record whatever message he thinks will get him to take the appropriate corrective action.

Mike G

Yes 100%
Its very very difficult to flap the rotor even if you only go to 140 RRPM but pull the stick all the way back and then even gun the throttle with a 914 powered AR-1
The only possible conclusion is that either there are huge gusting wind (which I know in tail accidents there wasn't in my customers) or that they are not pulling the stick back all the way and keeping it somewhere in the middle and then pull it back quickly. Even a small movement there quickly from higher speed will cause a flap. Extremely important to pull the stick all the way as you start to move forward.

I do think a stick position sensor should warn the pilot that they have not pulled the stick back all the way if airspeed is increasing. The problem is that it could be dangerous to at that point try and pull the stick back because that likely will cause a flap. They simply need to abort
1) Cut power
2) Stick forward
3) Apply wheel brake

So the warning needs to tell them to abort and aborting needs to be trained into muscle memory by practicing with a CFI. Talking about it is not enough. When you go to do it with your student, you will quickly find they don't usually do one of those 3 things no matter how much you talked about it. Specially the students from older age group.
 
Abid has a good point. Have to be warning that would not cause further issues. I was going to say “stick back” until I read Abid post. I don’t think the others would tell the pilot what to do.

Doesn’t some of the other warning systems like low altitude tell pilot what to do such as “pull up” which of course is not relevant for this scenario. Maybe “abort abort”… Tough questions…
 
It also shows that as you do a vertical descent where you already lose rotor RPM a bit and then do a sudden pushover, even without power you will drop rotor RPM close to the danger zone.
I don't agree with you on this point, Abid.
In a vertical descent with the engine at idle, the low asymmetry takes away from the flapping divergence. The "lower" Rrpm is only due from a smaller stall area, and is not an indication of a closer danger.
A stick pushed abruptly to the front stop at this instant will not reduce the Rrpm any more, since the angle of attack of the disc going from 90° to 80° will only change the load of the disc very slightly (-2%).
 
Last edited:
I don't agree with you on this point, Abid.
In a vertical descent with the engine at idle, the low asymmetry takes away from the flapping divergence. The "lower" Rrpm only comes from a smaller stall area, and is not an indication of a closer danger.
A stick pushed abruptly to the front stop at this instant will not reduce the Rrpm any more, since the angle of attack of the disc going from 90° to 80° will only change the load of the disc very slightly (-2%).

Hi Jean
Thanks for the comment. I understand what you are saying in theory but applied the data shows opposite. Mike has this data and you can see where rotor RPM dropped to I think 270 and G’s went down as well. The red line for rotor RPM was set at 260 which I believe is too conservative and comes from your modeling. The ELA French dealer at Bensen days was doing something quite dangerous in his show off routine where he did something like this but introduced banking spiral to keep positive Gs in a corkscrew at nose facing the ground from 3000 feet to 150 feet and would recover at 150 feet or so. Quite dangerous and unbelievable to me that a manufacturer would associate themselves to such a maneuver shown in their showcase. But anyway, he was taking rotor RPM down to 260 on every such dive and corkscrew from what I am told.

So if the red line is set a bit too conservative we know the modeling and theory are a bit off. Now should we move the red line down? I don’t know Not without an explanation because there may be only razor thin margins remaining. May be ELA guy has no clue how close he was coming to disaster.
 
Top