Spychalla Aircraft

My Feedback: (34)

About a month ago my son lost a giant scale Ziroli Stuka in a very dangerous crash. Luckily nobody was hurt. Total airplane loss was around $3,000. After a detailed study of the wreckage I have determined that this plane could have been saved by the application of 65 cent fuses on the servo leads. What happened was that a servo failed in a dead short. It got so hot in side it melted the surface mounted components right off the PCB. I have had 1/2 dozen servos fail on me in the past 30 years but they always failed open. In this case it failed closed. Consequently the servo pulled over 30 amp continuous draw on the servo leads and through the Receiver. Standard Receivers are only rated at about 30 amp so in less than 10 seconds it melted the wire leads and burned out the Receiver. The following crash was at near full power with absolutely no control of the airplane. Failsafe didn't kick in because the RX was gone. From my study of this situation I believe this could happen on any R/C airplane flying. I think we as modelers need to ask ourselves is what will happen in my airplane set-up if I have a dead short on a servo or a wire lead. Will the system sustain the amp draw? Will I still be able to shutdown and control the airplane? If you can't you better change things!!! Consequently I am now going to the following minimum setup on all my giant scale WWII airplanes:

1. Fuse all servos.
2. Run dual Receivers with completely separate batteries and switch harnesses. Separate redundant servos to separate receivers as best as possible.
3. Mandatory Engine Cut-off capability if and when power is lost.

So Why Fuses? My answer is that it takes any shorted component out of the system in less than 2 seconds. What is left will allow you to either land the plane or conduct a controlled crash in the worst case. Think about it. Standard practice is to fuse nearly every electro-mechanical machine made except our airplanes. They fuse cars, motorcycles, snowmobiles, boats, and even houses. Why in the world wouldn't we fuse our multi-thousand dollar airplanes?

So here is how I am applying fuses. I chose the 65 cent 5 amp bel Type MS Slow Blow Micro Fuse P/N 507-1036-ND from Digi-Key. (Call 1-800-344-4539) You can follow the picture sequence. 1. Measure about 1.5" from the end of the lead. 2. Separate and cut the black wire. 3. Cut the black wire shorter to provide space for the fuse. 4. Tin the wire leads and trim the fuse to fit. 5. Apply 1/8" shrink tube over the black wires on each end. 6. Solder the fuse in place. I use a handle little holding fixture I made form clothes pins. 7. Shrink tube the leads. 8. Slide 3/8" shrink tube over the connector and shrink the whole works together to isolate and vibration from the solder joints.

This is now what I am doing. This set-up would have saved my family 2 giant scale warbirds over the past 2 years. I wish I could go back with the knowledge I have now, but the best I can do is implement this moving forward and hopefully save what I have or will have some day. You can too. Leo Spychalla

Spychalla Aircraft

My Feedback: (34)

Here is what the application of fuses and dual receivers looks like in 2 different airplanes. The first is a 1/5th scale Yellow Aircraft Spitfire that weighs about 27 lbs. I used a JR R921 and a Spectrum AR9000. Make sure to separate functions for ailerons, elevator, engine cuttoff, and engine throttle to separate RX. Flap, gear, and rudder will likely have to go all to one or the other RX depending on your set-up.

The second picture is a Ziroli Stuka with dual AR7000 receivers. You can see the fuses on all the leads. You will notice that I just glued my receivers to block of foam for mounting. It was quick and effective. Hope I don't ever have to remove them. Leo

Spychalla Aircraft

My Feedback: (34)

And if all this wasn't enough to think about, you need to know this. All of us for years have set-up our ailerons or elevators on a Y-Harness. It is redundant.. right? Not so. Only if you have one servo fail open. If the servo fails closed you lose both servos. Try it. Take a Y-Harness setup and ground the signal wire using a jumper plugged in a lead instead of a servo. Make the jumper as in the picture (White or Yellow Wire) to either the ground (black or brown) or positive (red or orange). You won't burn anything up. You will just eliminate the voltage signal. What happens is you kill the whole channel and the other servo will not move. NOTICE! DO NOT CONNECT THE POSITIVE (RED OR WHITE) WIRES AND THE NEGATIVE (BLACK OR BROWN) WIRES WHEN DOING THIS TEST. That would get a little exciting and you will smoke some things.

The solution here is that the Y-Harnesses have to go if you are going to keep true redundant servos moving on one channel. You have to use a MATCHBOX or similar device which will isolate the signal. It keeps the signal alive to each servo connected. You can do the same test with the shorted signal wire and maintain complete function on the remaining servo.

Photos are the interior of my WWI Balsa USA Fokker DVII. It was not practical to equip this airplane with dual receivers after the fact. That would be best, but second best is that both the Aileron and Elevator Y-Harnesses had to go and MATCHBOXES took their place. I also added a 42% brand engine kill for safety. You can see the new setup. Notice all the servos are fused as well.

So all along when you thought you had complete redundant servos, you did not in all cases. Food for thought. Leo

dirtybird

My Feedback: (5)

All I can say is in the 40 years I have been flying theste things that include 7 years as a Kraft & Logictrol factory rep. I saw 1 shorted servo. It was a new one that came with a Futaba set I bought. When the Hobby shop said it was good and refused to replace it. I threw it away

Campgems

It a great idea. One suggestion would be to fuse the positive line vs the ground.

Don

k_caller

My Feedback: (1)

Great idea, can really attest to the fact of what happens with a shorted servo. My servo leads for the elevator shorted out when the extention leads managed to contact the reso in the fuselage. Caused exactly what happened in your scenario- comeplete burn out and loss of receiver and components and subsequent Hollywood crash. A setup like you suggest here might have saved the plane.
Just 2 quick questions- why the ground lead instead of the positive? and is there any reason for a 5 amp fuse over - say a 3 amp fuse?
Another idea might be to make a simple fuse panel- that way you could eliminate the soldering of all the servo leads.

Spychalla Aircraft

My Feedback: (34)

Here are my thoughts on the suggestions:

1. Fuse on positive lead vs. negative - This is a close call. No doubt. If the body of our planes were metal and could be ground you would definitely want to fuse the positive but they are not. They are wood or fiberglass. Both would work but I chose the negative lead because you have 3 voltage carrying wires. Taking out the negative cuts the current flow on all 3 because there is no potential to ground to a frame. Taking out the positive lead still leaves you with voltage path to the receiver between the signal wire and negative ground. Yes this is a very small current but a current none the less, so I chose to fuse negative ground. I would like to hear a persuasive logical case for the positive lead.

2. Fusing Leads versus a fuse panel - Both would work. I like separation in order to manage potential failure modes related to heat and vibration. I just see a panel having some mass to it and vibrating. Putting the fuses in the leads does not add any connectors to the system and it is quick and easy once you start doing them. I can do one in less than 5 min. now.

3. 3 amp fuse versus 5 amp - A mid size analog servo will draw about 1.7 amp under load. The manufacturer will tell you 2.8 amp peak when reversing a heavy surface quickly. A mid size digital servo will draw about 3 amp under load. You don't want to fuse too close to these values. You want some margin. When things go wrong, like a dead short, you will get current upward of 25 amp. depending on your wire size, wire length, and battery supply. With the help of my electronics friend Pierre Deschene I studied the bel fuse blow charts. The 5 amp could handle 5 amp for about 16 min. but would blow in 1.5 seconds under a 10 amp load or higher. That is pretty nice. I considered the 7 amp fuse to increase the current spike margin, but it would take 12 seconds to blow under a 10 amp load. 12 seconds at 10 amps is enough to melt the insulation off the servo lead wires and short other component in the system. Not good. So 5 amp is my choice. People running huge 1/4 scale servos or larger would have to adjust fuse size based on their peak servo loads.

Another general comment. I am told that the standard servo connectors and leads on all our systems are only rated for something like 1 amp continuous current. Check this out yourself by running a 1.5 amp load through a lead with a connector in the middle. See how warm the wires and connector get. You will be surprised if you have never done it. They are fine for short burst of 3 amp but they cannot handle much beyond that.

Leo

Dick T.

My Feedback: (243)

Leo,

Your points are well taken as well as the solution.

I have flown a couple of 33% aerobats and isolated the installation with two receivers, batteries, switches, etc. to seperate the amperage loads. Adding the fuses as you did builds in another safety, Those not wanting to do the modifications could use power expanders, matchboxes, isolaters and other types of devices to keep high amperage loads out of the receiver.

Thanks for bringing this to light. If it helps someone save a project and deny injury, it is worth the effort.

dirtybird

My Feedback: (5)

This is a very elegant solution to what I consider to be a non problem.
All servos made today employ a solid state integrated circuit. A lot of them have the motor soldered direct to the circuit board. The chances of shorting are very remote. Even if one was to short you will find out about it the first time you turned it on. Your fuse would save your wireing but that's about it. In fact a fuse is installed to protect the wiring in most circuit design.
As far as the wires chaffing and shorting out, decent attention to the installation should take care of that.
As far as reliability is concered, that fuse is a source of failure itself

slarty

Dirtybird,

Burnt out electronics is a fact of life especially with users mishandling servos. For example, JR servos (UNLESS SPECIFICALLY SPECIFIED) are rated for 4.8V usage. Check the manufacturers website and not the Distributors. Running a servo at higher voltage will reduce its life expectancy and should the servo bind for any reason, the current drawn by that servo increases past its operational limit and can cause sever damage to the pcb & motor. I have seen so many burnt out servo's due to excessive supply levels and the owners are usually a bit upset when I told them that the cost of repairing one of these exceeds the cost of a new servo.

The addition of a fuse should minimise the damage.

dirtybird

My Feedback: (5)

Again in circuit design, fuses are for protection of the wiring, not the components.
As for the JR 8611A, I burned out three of them attempting to torque test them at a current draw of 4A.(stalled) That would have burned out despite a 5A fuse. In those cases the motor burned out and quit. They did not melt down.
I also burned out a Hitec 5955 attempting to test it on 7.5V. In this case the amplifier let go. It also did not melt down. The amplifier just quit and it would not take anymore current.
In almost all servos the cost of repair will exceed the cost of a new one. I quit the repair of radios and servos in the late '70s when that became apparent to me.

chribrian

I dont think you understand why he is doing this. Its not to save a servo it is done to save the rest of his electronics! In most cases if the servo shorted and he had it fused, the fuse should blow before the receiver gets hurt !

Spychalla Aircraft

My Feedback: (34)

I was puzzled when I first read this commentary regarding the "non problem" but it has been clarified for me by Chribrian's comment. Dirtybird, I would agree that your comments are true from the perspective you are taking. Yet that is not the basis off this whole discussion. Turn the tables and look at this from the other side. We are not trying to save any one servo or wiring. Let them fry. For that matter let the fuse be the source of failure as you propose. It won't matter. You can still fly the plane. The worst case is that you lose movement of a surface or a surface with redundant servos will get a little slower from the loss of a servo, but you can still safely fly the plane. I don't know of any other set-up that will allow you to do that in the face of a dead short. The power expansion boards will not even sustain this. If I'm wrong on this tell me the make and model of one that does.

I do agree that the components on our airplanes will fail open in most cases. That's great. I also know that if any one of them fails closed that it will kill the receiver and/or anything else it is connected to. When you lose your receiver (on a single RX set-up) you not only lose flight control but you also lose the failsafe function on the throttle. Bad things happen. Nobody can tell me otherwise.

Now I wonder how many pilots have fallen victim to this and don't even know it. A tell-tale sign is that the airplane will never go into failsafe. I have seen several crashes where the airplane never went into failsafe and they bore-in wide open. The pilot has no clue what happened. They just say I lost it.

Most people in this hobby do all they can to insure the safety and reliability of there equipment. I have not met one yet that has not lost an airplane. Inevitably 100 different things could happen to bring our airplane down. I have just identified one of them that isn't going to get me anymore. Leo

rcairflr

My Feedback: (44)

This same thing happened to me 4 years ago. I had a HS-5645 fail as you described, the bottom of the servo case was melted. The servo drew so much current that the wires also fried. Needless to say, the airplane went in the dirt, total loss of control.

dirtybird

My Feedback: (5)

I understand what he is doing. I made that post in response to slarty's post.

As I said this is an elegant solution to a problem. I admire the engineering the man has done.
Its just that in my opinion it is a solution to a problem that has a very remote chance of happening.
Sometimes trying to prevent one a problem from happening we add so many parts we reduce the overall reliability of the system we are trying to protect.

coupar

This seems like a "better safe than sorry" approach especially for retract servos. Good idea.

I design various electronic circuits in my day job. We always include Resettable fuses on power in voltages. These devices act like fuses but when the load current drops (power is switched off), the fuse resets after a few seconds. They trip pretty fast too. This R/C issue might be a good application for these devices. You could cut the servo positive lead and insert the resettable in line. If it ever trips, you dont need to replace it, just power off and it will reset. Saves you needing to access the fuse or replace it. You can bury it away wherever you like. They are not as accurate in trip current as a real fuse but they would seem to be adequate for this application.

Resettable fuses were made by Raychem. I believe Tyco makes them now. The devices are shaped like small disc capacitors.

Just a thought.

Mr. Mugen

I see they ofer them up to 3a @6v. Might be worthwhile in a smaller bird.


http://findarticles.com/p/articles/m...45/ai_54717752


Kevin

coupar

Lots of Resetables at Digikey. More manufacturers make 'um than I thought.......You'd think retract servo manufacturers might include a resetable fuse in the power line for "just incase" protection..................maybe some of them do?

Spychalla Aircraft

My Feedback: (34)

Coupar, The Resettable Fuses are an option. One that we considered as we worked on addressing this failure mode. We discounted it based on the fact that something severe has to happen to the servo (or attached electronic devise) to push it over 2 or 3 amp. When that event occurs we wanted to take it out of the system permanently (Not reset it). The standard fuse was a better choice in our opinion. If that fuse ever burns you have got worse problems on your hands. Likely a failed component or melted wires that need to be discarded and replaced.

I would be interested in hearing your theory on what condition could occur that would drive a servo over say 3 to 5 amps that then you would want to reset it for and bring it back into the system. The only thing I can think of is a dead short of some type and why would you want to reconnect that to your batteries or receiver at any time? Leo

JohnW

My Feedback: (6)

I understand the argument better safe than sorry, but isn’t the most common failure mode by far an open circuit? Shouldn’t we be running redundant wires to each servo from multiple RX’s and power sources before we worry about fusing servos for shorts? This all just seems a bit over the top as a reaction an unfortunate, but extremely rare failure mode.

dirtybird

My Feedback: (5)

Exactly.

I see you are from Nebraska. I guess cornhuskers just have more common sense than the rest.

coupar

Leo:
I'm not disagreeing that the fuse will work. The Reset-able just saves you replacing fuses. Seems like an easier recovery. Fuses have been known to weaken after a while so when the fuse fails, if you are like me, the first thing you do is replace the fuse and watch it pop again! That resettable will try again and open again till you correct the failure. Makes it easier to diagnose without replacing fuses.

[ I would be interested in hearing your theory on what condition could occur that would drive a servo over say 3 to 5 amps that then you would want to reset it for and bring it back into the system. The only thing I can think of is a dead short of some type and why would you want to reconnect that to your batteries or receiver at any time? ]

Actually there is a relatively common one (it’s not a theory) for retract servo users. If you look at any thread regarding retract servos, you'll see somebody who had a retract servo pull excess current because the gear got jammed. Many of these result in drained batteries and crashes. One thread describes an 30 amp drain from the gear not going up all the way (seems hard to believe but..). The common solution is to have two batteries, one powering the retract servo and the other powering the rest of the system. Personally I think this is a pain and may be overkill (and extra weight). You should always double check retract operation. BUT it has also never happened to me ….yet! I have to admit, there are situations where a retract jams, it is not obvious and it pulls a very healthy current. Your solution is perfect for a concerned retract user. In this case the reset-able fuse is handy because the retract jam can be corrected and the fuse will reset and you are back flying right away. You don’t need the additional battery and there are no consumeables.

If fuses work for you stick with um.

dirtybird

My Feedback: (5)

[/quote]
In testing servos for maximum torque the maximum I found (Stalled)was 4 amps. That was the JR 8611A. I would like to know what servo is there, that would draw 30 amps when stalled. 30 X 4 = 120 watts. Is he using a Sirus? If not that servo would be a molten mass in a few seconds. You would need a fuse there. That's for sure.

Spychalla Aircraft

My Feedback: (34)

John from Lincoln, If you must rank failure modes then Yes you should run dual receivers and dual power sources before you fuse servos. (All key actions identified above) Plus don't forget that engine kill module. Very important. Then fuse your servos. Sure the failure mode is rare and a few of you guys will not be concerned about it until it happens to you. So we can debate that till we expire. However it is a failure mode that I will never worry about anymore while others do. Leo

JohnW

My Feedback: (6)

I have some high end servos a bit larger than 8611a that pull around 6-7A stalled. Never tested a Sirus, but even that big dog I doubt pulls 30A.

Leo, I understand, by all means do what makes you comfortable. I just don't understand why you don't run redundant lines to the servos at the same time you are fusing them to protect yourself against the more common open circuit failure mode.