...that can ruin your whole day.

While replumbing the tanks in my Super Stinker, I wanted to relocate my main power switch to the opposite fuselage side.

This switch is one of the more popular, "heavy duty" models with a charge recepticle incorporated into the unit.

A flying friend saw mine last weekend at the field and commented that he had lost an airplane because of a faulty switch just like mine. Evidently, one if the wires had a bad solder and popped off the tiny board, causing loss of current and the inevitable crackup.

Since I had the switch in my hand, I decided to take a quick look to insure my switch was ok. Afterall, I had already flown the device with no issues.

I popped the black plastic back off and carefully removed the switch from the housing. I immediately noticed that the foil that is adhered to the epoxy board was peeled up on the one edge and that the black wire had a rather oversized blob of solder.

Using a jewelers loupe, I examined the wire and the board and discovered a flaw. Evidently, the wire had been soldered, removed and re-soldered at the factory. Problem being, when the 1st solder was removed, it peeled a small portion of the foil off with it. In order to bridge this "gap" an inordinate amount of resolder was needed to make the contact. However, solder does NOT adhere to epoxy board and since only a very slight edge of foil actually was making the contact, as soon as I slightly touched it with a pik-probe, it popped right off the board, large solder blob and all. The board was bare, exposing that 2mm gap in the foil.

I decided to remove all the wiring, clean the board, reflux and resolder the wiring harnesses to insure reliable service.

I unsoldered all 6 contact points using a braided wick, cleaned it up with tuner cleaner(contact cleaner for you kids that never saw a manual tuner in your life) and fluxed the contacts for the redo.

I used a jumper to bridge the 2mm gap in the foil, resoldered all the leads and reinstalled the switch into my fuselage.

Man, it sends chills down my spine to think of what the vibration from the 41cc gasser could have done to that crappy solder job from the manufacturer!

It pays to pay attention to the little things that can ruin your whole day, (and investment).

PM

Good tip, and explanation of the fix. Any chance of getting you to share the make of the switch as a heads up to others that may face the same problem? .....and perhaps some pictures of your work?

Chad

That could really ruin a guys day. I think I'll check mine as well.

I really did consider posting the make of the switch but hesitate to do so.

Since I've already solved my issue, posting photos would show the "cure" but not the disease, as it were.

However, I guess you're right...any visual is usually better than a written explanation!

I'll fire up my cheap digital camera tonight and pull the switch back out of the Pitts to post a few tonight.

The model of switch will be self-evident once I post the photos...

PM

Hi,

I always run dual switches and dual batteries on every plane I fly to prevent this exact problem. On the larger planes, I also run dual rxs. You may want to consider the dual switch option.

Elson

I would not even fly with two of those same brand switches.....

Hi!
Why not use electronic switches?There are several on hand. Or the new Spektrum Ar 9100 receiver with a soft switch.

Ok, I pulled the switch, unsoldered my previous work and started over so y'all could "see" what I'm rambling on about...

The photos tell the story. The brand name of the switch should be obvious.

PM

I had a look at a couple of my switches and they are OK. Mine are a different brand as well. I did have a brand new Futaba servo I had to rewire a while back. I found the wire was cut about a 1/4" from where it goes into the servo case. It was a very thin cut like a razor blade would make. The wires weren't completely severed, but very close to it. It would have failed in the air for sure.

I'm curious why you are so afraid to mention the manufacturer by name. If a product is made with poor quality there is nothing wrong with saying who is responsible. You just might help save someone else some problems.

I realize that we can figure it out from your pictures but still, you can say the name. They won't send someone to your house to break your legs

It's about time we begin to demand better quality products.

We demand high quality stuff and want it to cost next to nothing. Can't have it all I guess.

Hello Point, I would like to know if the switch in your post is a knife type or the rub two contacts together type, sorry for this explanation, maybe , "the other type" would be better, if it is the "other type", then I would not use it at all, vibration will cause one problems, also the bat handle types are not acceptable for the same reason, they are ok for transmitters, but they are not subject to vibration. This is from actual experience.

The product should at least be of reasonable quality. If it's not then I don't want to buy it even if it does cost next to nothing.

Hey All
In electronics, the higher the sophistication of a part, the tighter the packaging of that part, and the less access we have for that part. So we are now at the mercy of the assembler/manufacturer.

We are all in the same boat. A higher cost for a part SHOULD mean a higher degree of design features, better components in the unit, higher quality control and assurance. The all in one switches made ALL our lives easier. They use heavier gauge wire, the plastic box is more substantial than the "older" flimsy looking designs. The perceived added value costs were also at least $10US per switch over the "older design".

The perception is (was) that there is $10US more of value. Unfortunately, that is not really the case here. The design is good... It integrated a need for us, the additional wire gauge sizing also is a plus too. The biggest plus was that instead of wires being soldered directly to a little switch tab, the wires were soldered onto a pad (larger area) on a piece of circuit board. This is also its weak point.

Let me explain...The entire unit is hand produced and assembled, what all of us have is personally crafted switch. The only automation there is the screw gun that attached the screws to the body, and even that is held in a hand. The soldering of the wires is done by people applying a drop of solder to the pad. The opportunities of manufacturing a bad part are unfortunately quite high.

1. Circuit board material is single layer glass. No problem. However, the manufacturer of the board may not use the best processes to assure quality traces. After all it is only a wire connection device...hmmmm.
2. The circuit boards piece that you see in the picture has been hand cut via pneumatic or electric scissors from a "mother board" that could contain a hundred of so of these parts. An operator could easily and accidentally nick or cut a trace.
3. Have the boards been cleaned of oil and crap. My guess is as good as yours.
4. Solder - They PROBABLY use solder appropriate for the job.
5. Assembly of the board - Here is where it gets interesting.
I used to teach new employees a 2 day course on hand soldering. Yup... there is that much to learn about the process and what it takes to do a good job and have consistent quality.

The easiest way to see a good quality solder job is to look at the soldering job. Is it bright? If so the chances are high that the bond/connection is secure. Is the surface dull? If so, the bond is not there and the connection though probably has conduction, it will not have so for long. Use these as indicators. If they did not use a proper flux on the board, the solder may be bright but it would look blobby. (yeah... I know that is a lame REALLY technical term. ).

So now the little circuit board is soldered, and the person assembles it into its casing. It has to be manipulated, with wires being worked and twisted slightly for the fit. The unit is finished. Total time for all of this... all the processes from start to finish including packaging is less than 90 seconds.

Has the finished unit been powered up to test for circuit continuity? Does the manufacture do any life testing? (hahahaha - I know my guess). Are there any destructive testing aspects conducted....

The cost of buying these switches is about $15 and up (depending on the features). The cost to the manufacturer is less than a buck. (I have an engineering background in automotive electronic mfg., along with 25 years of Asian assembly experience so this cost estimate is bang on). The rest is marketing, distribution related costs and profit. So waddya want for a buck guys?

I lost my favorite plane to a switch - because I thought that the price included some quality control.(talk about my OWN experiences not being listened to0 by myself, I still kick myself in the butt over that!!!) Now I use 2 switches and 2 battery packs on any plane over 20cc.

my 2 cents worth!

Henry





[link=http://www.zrcgf.com]RCGF Engines - North America[/link]

Thanks for the headsup! I have the selfsame switch that was going into my Extra 330L!
I opened the back up and had a good looksee![:-]
This switch is a whole lot better than the one in the comments.
Maybe it's from a later production run!
As you say, it's the little things that can re-kit one of our pride and joy!
Thanks again for the pix.
Zee
AMA129570

Nothing new about defective electronics and wiring. Planes have been lost to bad solder joints, wires breaking loose from batteries and switches, crystals falling out of receivers, and the list goes on and on, since radio controlled flight began. That's not going ot change. The "lifetime perfect" affordable electronics solution has yet to be made. Same with computers and software. If it's electrical or mechanical there's a chance of failure. Way of life.

BTW this is the second instance of a defective switch that I've encountered in less than a week. Perhaps a different manufacturer from yours but the same switch housing is used under several brand names, both OEM and after market.

Good catch.

Thanks Piston for the great write-up. Taking a look at the internals of all my switches just got moved up to the top of the list.

I wounder how many "reciever failures" have been caused by switches. Especially since the latest recievers are more sensitive to power failures.

I'd like to venture a question in that regard. Are the 2.4 receivers more sensitive to "power failures"?

We've always had this problem in our radio systems, even when all we had was simple 4.8v nicads. For a long time the largest flight battery we had available was a 600or 800 mAh nicad and when the voltage fo those dropped from around 4.2v or 3.8v or so we would lose airplanes. At the time few planes used more than 4 standard analog servos so those batteries usually managed to make 4 flights or so before calling it a day. 6 flights was pushing the limit hard. That's why "field charging" came about.

So in todays aircraft we are using 5 or 6 high torque, or more, high current servos. Larger batteries as well but today's servos are far more "hungry" than those in the old days. Deflections are greater, planes are heavier, flying style is more agressive, so battery demands have increasd tremendously. Whne the battery voltages drop down to a certain point the receiver can no longer send an activation signal and some juice to the servos. Same as the old days. The difference is that the receivers in use today stop doing the job at a lower voltage than when they stopped working in "the old days".

Nothing's truly changed with the exception of the new receivers being able to hang in there longer than they had been able to do so 15 or 20 years ago. We've placed a greater load on our batteries and typically few sit down and perform even a rough calculation of how much current their servos might use for a given flight. Used to be we considered the idle current, normal load current, the stalled current, the average flight time, and did some basic math to determine what size battery we might need. Then we checked the battery condition after a flight to verify our calcs. Then we field charged to assure that we never ran out of that electrical fuel. We still lost planes to "brown outs" but the number was significantly reduced.

Today many simply post a question is a forum asking "what size battery do I need", or "how many milliamps". Then there's a disagreement about an answer and a free for all ensues with regard to what type battery they should have. Most go with the smallest, lightest, cheapest, minimum AH battery the can lay their hands on. Less weight you see. There's no consideration for wire length and voltage drop. Increases in resistance? What's that? Connector counts from numerous extensions lower voltage to servos and increase resistance, generating heat and reducing battery efficiency. Most forget that the battery is the second "gas tank" they carry in the plane. When the liquid one goes empty they can glide home. Not so when the electrical tank goes dry.

It's all the same as it was, we just aren't as vigilant as we used to be. When the inevitable happens the first thing everyone does is blame the radio manufacturer for designing a sub standard system. The manufacturers did a wonderful job, especially considering what we ask those radios to do. The receivers don't drop out until at a lower voltage than they used to.

We're the ones that are causing the failures. We pay less attention to our systems than ever, and have "dumbed down" the knowledge needed to participate in this hobby to a point where simple failures of basic components are assured.

PointMagu- I had that exact problem with the same (LOOKING) switch as you did. The first thing that went bad on mine was the charging pins, they fell out, no biggie, I just charged off of the battery. But then the ground pulled off, luckily I was using it for the ignition and it only caused a dead stick. I have since re-soldered it and filled the area with silicon before re-assembly to help with strain relief, which is what I believe caused the ground to pull off.

I have 4 of those switches. Two each on two different planes. I've noticed that when checking the voltages of my batteries with my loaded voltmeter through the charge jacks, I have to jiggle the connector around in the jack to get a true reading. Seems the female charge jack in the switch has developed some play or looseness over time. Anybody else have this problem?

Funny thing about those switches. I can name four different companies that use the exact housing. One of the four makes switches for another distributor but the "other distributor" switch is inferior to the after market switch because the "other distributor" insists on a less expensive design schematic. So externally, all four switches appear to be the same when in fact there are fourcompletely different assemblies. There may be more than four vendors using that same case but I'm only personally aware of four.

The probelm in this thread is that everyone is going to look at the external appearence of their switches and if they look the same they are going to think they have a switch with a potential defect. You have JR, Cermark, Smart Fly, and MPI all using the same switch case-that I know of. There may be more.

The manufacturer needs to be identified for people to have a heads up.

Pat,
What I was referring to is that the 2.4 receivers have a lag time between when the brown-out occurs and when control is restored. I have seen the reboot time on the order of 5 seconds (2 seconds seems to be the norm). I know the receiver manfs. are redesigning their receivers to make the reboot time negligible.

But, I agree with everything you posted, It seems people do not pay proper attention to the power systems in their birds. Batteries must be able to handle the amperage load with an acceptable level of voltage drop, wiring gauge needs to be properly sized, extension lengths need to be kept as short as possible, connector usage should be minimized, connectors should be sized for the amperage load, etc...

You should feel lucky. In the golden days of analog servos and 4.8v 600 mAh nicads, when the receiver had a voltage shut down there was no such thing as a "re-boot". The plane crashed every time. If it was extremely well trimmed and in a level attitude when it happened sometimes they just flew away....

I agree with Pat about naming the brand name of the switch. I'm new to the bigger aftermarket switches and I don't recognize which one it is. I have 2 smart flys ordered that I'm installing in a new 35%er. I'll check them out first thing but still don't know which one you are talking about.

Something came up in a conversation today with RTK that has a lot of bearing on how well switches in general hold up.

How many people provide support for switch, battery, receiver, and servo wiring reasonably close to where any given wire makes a connection or exits an electrical "appliance"? How many provide for any type of strain relief in their wiring?

Permitting wires to hang loosely from a battery, switch, or servo places that wire and connection under tremendous flight loads as a plane is flown through any type of aerobatics. Leaving extension connections laying loose inside a plane, or wires left unsecured at the receiver, are under the same loads and subject to failure at any time. Wires pulled tight in order to stretch that last little bit of distance, rather than buying and using a longer one, place constant tension on the connections at both ends. That may be a battery, regulator, switch, ignition, or any other component.

Failure to adequately secure wiring can cause solder joints to break and foil traces to be pulled from pc boards of any associated component. Over a period of time various components and wiring should be expected to fail if left unsupported during flight. This is one of the little things that's rarely mentioned in aircraft set up and should be covered from time to time so we don't set the stage for flight failures through neglect. We need to take a hard look at our installations and consider how much movement is available to all the wires. How much weight will each bundle of wires be when subjected to aerobatics and how far can they swing? What will they impact at any portion of a swing? Will any of the wiring be pulling directly on it's connected end when subjected to a flight load?

Little things one by one but big in causing crashes.