Old airplane driver's view of Part 107 test

[spudster]

So the real solution is multiple classes of drone license?

Sounds like a better solution, but whether there's resources for it is another matter.

In my humble uniformed opinion the current system makes no sense. I can fly around all day at 400' AGL with no problem other then registering the UAV if it >249g. But as soon as I fly 100' AGL above my house and take a photo to sell it, I then need a part 107. It has no real life logic.

Again, IMHO it would make much more sense to require a license with the relevant knowledge that is actually needed rather then the intention of the flight is to license by UAV weight. One such example could be:
<250g - Trust
250g to <1000g - Amateur License
1000g to <5000g - Professional License
>5000g - Industrial License.

With computer on-line testing most of this would take no more then creating the tests, put them on-line and put the system into place to automatically issue the license for at least the Amateur License level which would cover 99% of all pilots.

 

[BigAl07]

I'm not sure many would call the 107 test "silly easy". When I took mine, I knew the stuff cold and fully expected to score 100%. When I complained to the monitor that some of the questions were deliberately vague resulting in a score only in the mid-90s, he quipped, "If somebody got a 100 they'd probably change the test!"

Well for the record, that "monitor" doesn't know what they are talking about. There are "100's" made on that test. Greg over at the Pilot Institute even has an announcement system for each of their students who actually score 100.

 

[Zbip57]

I would hazard a guess that the sort of people who diligently study all the material they can get their hands on, fully intending to score 100% on the exams, are not the ones who would then go out to create mayhem with their drones. They're the type that would likely thoroughly research the necessary knowledge requirements beforehand for any project they choose to tackle, regardless of whether an exam first needs to be passed.

The people who, without studying at all, think they can harvest enough information off Google to scrape a passing score on the exam are already demonstrating questionable morals.

Then there are people who are scared away altogether and give up entirely when they see the absurd types of questions being asked on these exams. They'll just go out and fly their drones anyway hoping nobody will catch them in the act.

Which of these categories of people is most likely to be causing problems, and how is the current exam structure aimed at solving this situation?

 

[adcimagery]

So the real solution is multiple classes of drone license?

Sounds like a better solution, but whether there's resources for it is another matter.

I don’t think it’d be particularly difficult. Take the sections covering safety and airspace, and make that the “basic” test. I’d argue that basic test should even apply to rec flyers, as TRUST is currently insufficient, imo.

Then add endorsements or certifications for things like night ops, near airports, near people, etc. Many of those questions already exist, but could just be expanded into a test.

 

[Robert Prior]

it would make much more sense to require a license with the relevant knowledge that is actually needed rather then the intention of the flight

That is, in theory, the Canadian system.

Microdrones — no license needed (at least a minimal test would be better).
Basic certificate (250g-25kg) — must meet all of:

• You fly it in uncontrolled airspace
• You fly it more than 30 metres (100 feet) horizontally from bystanders
• You never fly it over bystanders
• You fly it more than 3 nautical miles from a certified airport or a military aerodrome
• You fly it more than 1 nautical mile from a certified heliport

Advanced certificate (250g-25kg) — if you want to do any of:

• You want to fly in controlled airspace
• You want to fly over bystanders
• You want to fly within 30 metres (100 feet) of bystanders (measured horizontally)
• You want to fly less than 3 nautical miles from a certified airport or a military aerodrome
• You want to fly less than 1 nautical mile from a certified heliport

One could argue about details (and I have) as the licenses are for all remotely-piloted aircraft systems in the weight class, not just quadcopter drones with GPS, so writing the exam is rather like having to know about driving any vehicle from a moped to a moving van for a drivers license. But the basic idea is, I think, sound: licensing based on the type of flight, not on the reason for it.

 

[Robert Prior]

I don’t think it’d be particularly difficult. Take the sections covering safety and airspace, and make that the “basic” test. I’d argue that basic test should even apply to rec flyers, as TRUST is currently insufficient, imo.

Then add endorsements or certifications for things like night ops, near airports, near people, etc. Many of those questions already exist, but could just be expanded into a test.

Yes, but someone has to write regulations to make that happen, vet the regulations for loopholes and unintended consequences, create the tests, set up a portal for the tests and issuing certificates, figure out how to handle the transition period and create regulations to implement that, create guidelines for decide penalties and exceptions…

Even apparently simple things can take a fair amount of effort. And given that this is a federal agency, the people doing that need to be paid, and presumably would be pulled away from other work so other projects will need to be rescheduled…

 

[spudster]

Since it is hard to do we should just not do anything? Every month wasted doing nothing means the end results is one month further away. At some point of doing nothing people just says screw it and give up

 

[Robert Prior]

Since it is hard to do we should just not do anything? Every month wasted doing nothing means the end results is one month further away. At some point of doing nothing people just says screw it and give up

No. But it isn't a simple matter of pointing out how logical and better a different system would be and then expecting the FAA to magically implement it. It will require real and sustained effort to persuade someone (more likely several someones) at the FAA to make changes and allocate the internal resources necessary to make those changes.

I think it's worth doing. But it's also important that anyone attempting it realize that they will need to devote serious time and effort to it. Success is more likely when people understand that they may not see any apparent return for their effort for quite a while, so they don't get discouraged and give up prematurely.

 

[DoomMeister]

In all seriousness I believe most tests have psychologists behind them. If a quality, seasoned airline pilot had to take the current ATP exam they hopefully would not fail but may get lower than expected. As a manned pilot I earned my 107 easily in 2016. That said I did not purchase and fly a drone until Spring 2021.

Trust like other FAASafety.gov courses, including TSA training are exactly as you properly state. I would hope flyers of all kinds want to have knowledge to be safe and competent. I said "I would hope..."

As a former course developer in the US Air Force I can tell you there is a lot of work put into test development and validation. With modern day technology tests could be constructed on the fly from a vast pool of questions for each knowledge area tested in the 107 certification. Multiple choice tests are the easiest to grade as they are completely objective in nature and not open to interpretation by the test scorer. Many people are under the impression that multiple choice tests are quite easy to pass, and in many instances that is true, but if the test is designed properly a person that knows the material will be able to pass with high marks where those with a lesser knowledge will have great difficulty.

The TRUST was a bust in my opinion. Especially since it was not tied to your registration (for those operating in the 250g+ categories).

 

[sar104]

That is, in theory, the Canadian system.

Microdrones — no license needed (at least a minimal test would be better).
Basic certificate (250g-25kg) — must meet all of:

• You fly it in uncontrolled airspace
• You fly it more than 30 metres (100 feet) horizontally from bystanders
• You never fly it over bystanders
• You fly it more than 3 nautical miles from a certified airport or a military aerodrome
• You fly it more than 1 nautical mile from a certified heliport

Advanced certificate (250g-25kg) — if you want to do any of:

• You want to fly in controlled airspace
• You want to fly over bystanders
• You want to fly within 30 metres (100 feet) of bystanders (measured horizontally)
• You want to fly less than 3 nautical miles from a certified airport or a military aerodrome
• You want to fly less than 1 nautical mile from a certified heliport

One could argue about details (and I have) as the licenses are for all remotely-piloted aircraft systems in the weight class, not just quadcopter drones with GPS, so writing the exam is rather like having to know about driving any vehicle from a moped to a moving van for a drivers license. But the basic idea is, I think, sound: licensing based on the type of flight, not on the reason for it.

The only reason that the purpose of the flight was ever a factor is because Congress was successfully lobbied to require it to protect the model aircraft industry. There is no point complaining about the FAA implementation - their hands were tied from the start.

 

[Mike Drop]

As a former course developer in the US Air Force I can tell you there is a lot of work put into test development and validation. With modern day technology tests could be constructed on the fly from a vast pool of questions for each knowledge area tested in the 107 certification. Multiple choice tests are the easiest to grade as they are completely objective in nature and not open to interpretation by the test scorer. Many people are under the impression that multiple choice tests are quite easy to pass, and in many instances that is true, but if the test is designed properly a person that knows the material will be able to pass with high marks where those with a lesser knowledge will have great difficulty.

The TRUST was a bust in my opinion. Especially since it was not tied to your registration (for those operating in the 250g+ categories).

I think I get it. There could be a large pool of questions. No two tests need be alike. With a large enough pool, the odds against two pilots getting the same tests are vanishingly small.
And yes, I agree. The TRUST exam is a real failure.

 

[GregS]

When I took my written test for my private ticket many years ago, I passed with a 92 percent. I was bragging about it to my CFI when he said "Well, you'll only be wrong 8 percent of the time."
Watching YouTube videos about the Part 107 test, I see a lot of people bragging about dragging a low passing score over the goal line.
I'm not sure that's the goal at all.
For the drone drivers who scored 72 percent, how much knowledge did you retain? The point, in my opinion, is not to just squeak by on the test but to demonstrate that you thoroughly understand the material.
Well, do you?
Its in my best interest (and maybe all of ours) not just to chew up and spit out the material but to digest it.
Having said that, I went back to review the AIM/FAR. I was shocked and not a little disappointed at what I had forgotten. Before I take the part 107 test, you can be sure I will study until I have a command of the material.
I don't want to be wrong 28 percent of the time.

Mike, you are absolutely correct. As a former Military pilot and Commercial pilot, just getting by might someday get you or someone else killed...in a real aircraft.
For drones, it could still come close to that if flying where one shouldn't.

 

[Kilrah]

@adcimagery you do realize that Part 107 is a little bit bigger than just "mavics taking Real Estate pictures" don't you? Part 107 encompasses up to 55lb aircraft and operations other than just your local neighborhood. As someone who has indeed actually USED airport markings for a large project we did on the grounds of KAVL I can say some people need that information. Just because it doesn't apply to you and how you currently operate doesn't mean the material isn't relevant to others in the industry. Think BIG PICTURE not just your mavic.

That's the whole point, if there is some knowledge that 99% of people won't use but is useful for some scenarios you make 2 license classes, one that doesn't encumber the minds of those who'll never need it anyway, and one for those few others that does encompass what's needed for them. Like IFR isn't a thing every aircraft pilot needs to learn, it's a separate license that you take if you need it, or how you the car driving license doesn't include trucks.

 

[Robert Prior]

With modern day technology tests could be constructed on the fly from a vast pool of questions for each knowledge area tested in the 107 certification. Multiple choice tests are the easiest to grade as they are completely objective in nature and not open to interpretation by the test scorer.

Writing and validating that vast pool of legitimate (ie. not trivial) questions isn't an insignificant task, nor is categorizing them so that the randomly-assembled test covers all the knowledge it's supposed to.

Validating the questions is important, as what the writer intended may not be how the reader interprets the question. After my father sat his requalification exams in toxicology virtually the entire group of professionals sat down with the committee who set the exam afterwards to go through the exam and point out the ways that most questions could be legitimately interpreted in different ways so that more than one answer would be the best answer, depending on initial assumptions. These were all professionals at the top of their game.

I'm certain you know this, but as a retired instructor I know that many people think "throwing together a test" is trivially easy.

So yes, could be done. And I think should be done. But doing it properly will take time and resources.

 

[Zbip57]

That is, in theory, the Canadian system. [...]

The only reason that the purpose of the flight was ever a factor is because Congress was successfully lobbied to require it to protect the model aircraft industry. There is no point complaining about the FAA implementation - their hands were tied from the start.

The intent is worthwhile, but the execution makes no sense. The result implies that recreational flights will always somehow be conducted more safely than commercial flights, or conversely that commercial flights are always inherently more dangerous than recreational and thus requiring stricter regulation.

The AMA lobbied for leniency based on their proven safety record, but how and why should that extension apply to every other non-AMA recreational flyer who gets a new drone for Christmas?

Canada does have a similar exemption for the Model Aeronautics Association of Canada (MAAC), but there are several conditions attached.

See: MAAC Exemption

The Canadian exemption is based on you being a member in good standing of MAAC and operating under all of their strict conditions. Our exemption certainly does not apply to every other recreational flyer based solely on whether they're flying for fun or business.

 

[Mike Drop]

Mike, you are absolutely correct. As a former Military pilot and Commercial pilot, just getting by might someday get you or someone else killed...in a real aircraft.
For drones, it could still come close to that if flying where one shouldn't.

I'm thinking of the Kara Hultgren incident. It's said she barely qualified for Tomcats. Another example of someone who 'just got by' quals.

 

[Classic flyer]

When I took my written test for my private ticket many years ago, I passed with a 92 percent. I was bragging about it to my CFI when he said "Well, you'll only be wrong 8 percent of the time."
Watching YouTube videos about the Part 107 test, I see a lot of people bragging about dragging a low passing score over the goal line.
I'm not sure that's the goal at all.
For the drone drivers who scored 72 percent, how much knowledge did you retain? The point, in my opinion, is not to just squeak by on the test but to demonstrate that you thoroughly understand the material.
Well, do you?
Its in my best interest (and maybe all of ours) not just to chew up and spit out the material but to digest it.
Having said that, I went back to review the AIM/FAR. I was shocked and not a little disappointed at what I had forgotten. Before I take the part 107 test, you can be sure I will study until I have a command of the material.
I don't want to be wrong 28 percent of the time.

Well said sir....I got my part 61 in 1977....

 

[DoomMeister]

Writing and validating that vast pool of legitimate (ie. not trivial) questions isn't an insignificant task, nor is categorizing them so that the randomly-assembled test covers all the knowledge it's supposed to.

Validating the questions is important, as what the writer intended may not be how the reader interprets the question. After my father sat his requalification exams in toxicology virtually the entire group of professionals sat down with the committee who set the exam afterwards to go through the exam and point out the ways that most questions could be legitimately interpreted in different ways so that more than one answer would be the best answer, depending on initial assumptions. These were all professionals at the top of their game.

I'm certain you know this, but as a retired instructor I know that many people think "throwing together a test" is trivially easy.

So yes, could be done. And I think should be done. But doing it properly will take time and resources.

You are right about the validation of the test questions. All of the test material would have to be validated before being used. The test already covers different knowledge areas with a specified percentage for each area. Each area would have its own pool of validated questions to be drawn from.

 

[Tomk_]

Interesting discussion.

I get the distinct impression that many here have limited experience taking government test for various certifications. I can assure you that ones administered by the FAA are some of the best. It's my opinion that those from ATF+E are the poorest. The "best" example is the explosives handlers license. While it's stated purpose is safetly, there is actually nothing tested about the actual safe handling or use of explosives. They say it tests your knowledge of the regs, but because you actually have access to the regs during testing, it really tests your ability to understand how the regs are put together. You are given 2 hours for the test but it can be completed in 20 to 25 minutes. With an hour of coaching, I could show anyone reading this how to pass this test, and most would get 100% without ever reading the regs from start to finish. I think it's a joke...yet people fail it all the time. Some people are better at taking tests. Some people are better at retaining what they've learned. Some people are good at guessing.

When I got my private pilot certificate, my DE told me "Congratulations! you have demonstrated the minimum required aptitude to be issued a license to learn."

I'm guessing some of you have CDL's. Do you have to be a "good" or even a "safe" driver to acquire one. Absolutely not. Yet there are 100's of professional drivers out there with, perhaps, millions of miles of flat land interstate driving than would not be able to safely haul cattle to summer range in the mountains or even pull a D8 on a lowboy across town.

Most tests are ostensibly about safety. That will always be subjective as that is really not definable because "safe" doesn't exist. All we can do is merely attempt to be safe. The test is a feeble attempt to quantify the unattainable for the purpose of documentation.

 

[sar104]

Interesting discussion.

I get the distinct impression that many here have limited experience taking government test for various certifications. I can assure you that ones administered by the FAA are some of the best. It's my opinion that those from ATF+E are the poorest. The "best" example is the explosives handlers license. While it's stated purpose is safetly, there is actually nothing tested about the actual safe handling or use of explosives. They say it tests your knowledge of the regs, but because you actually have access to the regs during testing, it really tests your ability to understand how the regs are put together. You are given 2 hours for the test but it can be completed in 20 to 25 minutes. With an hour of coaching, I could show anyone reading this how to pass this test, and most would get 100% without ever reading the regs from start to finish. I think it's a joke...yet people fail it all the time. Some people are better at taking tests. Some people are better at retaining what they've learned. Some people are good at guessing.

When I got my private pilot certificate, my DE told me "Congratulations! you have demonstrated the minimum required aptitude to be issued a license to learn."

I'm guessing some of you have CDL's. Do you have to be a "good" or even a "safe" driver to acquire one. Absolutely not. Yet there are 100's of professional drivers out there with, perhaps, millions of miles of flat land interstate driving than would not be able to safely haul cattle to summer range in the mountains or even pull a D8 on a lowboy across town.

Most tests are ostensibly about safety. That will always be subjective as that is really not definable because "safe" doesn't exist. All we can do is merely attempt to be safe. The test is a feeble attempt to quantify the unattainable for the purpose of documentation.

I agree with your assessment, but I was unable to figure out whether you think there is a better approach or not.

The explosives safety example is a good one, although somewhat unique, in that commercial and military explosives behave very predictably when used in design mode, but exhibit a vast range of complex behaviors when subjected to other (generally unintentional) stimuli. And the latter category is where most explosives accidents occur. I think I can count on one hand the number of people worldwide who have a comprehensive understanding of that subject.