AndyKunz

Do you understand how spread spectrum works?

Andy

djmp69

Do you understand how physics works?

djmp69

You ask that, though I have, in multiple posts on this thread, explained that I understand how it's SUPPOSED to work. In THIS case, it did not work as designed and as you explained. What you keep missing is that for your diagnosis to be correct, certain things had to have happened. And those things DID NOT HAPPEN. Yet the anomaly was manifested. You seem to keep forgetting or at least ignoring the fact that what you descibe, i have tried with other models, and the anomaly was not present. If it happened as you say, then it should happen for all the models. Except it only happened with this situation. Why can't you just accept for a moment that there might actually be a problem? Spektrum for me has proven to be a VERY good system. Which is why I am so invested in the system. But I am not so invested that I am brainwashed into thinking it is infallible. That's what research and development is all about. It's ongoing. There is no end of the road. There is ALWAYS something that can come up that was not anticipated that will spur a new direction of development. That's where updates and upgrades come from.

rgburrill

Andy, you aren't listening. And you are being very condescending. No one is questioning your design capabilities. But it sounds like his receiver is responding to a signal model for which IT HAS NOT BEEN PROGRAMMED. Something is wrong with his receiver.

AndyKunz

Yes, I do. That's why I asked about spread spectrum. It was an honest question for you, not a smart*** one. I'll take your non-answer as a no.

Spread spectrum works by having, essentially, a bunch of different notes playing at once. You (a receiver) hear lots of notes, and they don't make sense. It's no different than white noise, or tape hiss (if you're old enough to know what that is). Just like lots of screaming kids at Disneyland, you will hear your child's voice (or in our case, my wife will) even though there are lots of similar, indistinct sounds. Your ears are tuned for those you love.

Spread spectrum uses a similar concept. The transmitter is putting out a distinct sound, but to make it so others can talk at the same time, it does it in a unique voice which is made by pseudo-random changes using a coding scheme shared with the receiver during bind. That coding scheme is based on both the transmitter's unique ID and the model ID, same as referenced above. So even it a transmitter could talk to two models at once, they would be encoded differently (these are called PN codes, if you care to dig more) and each receiver would hear only the voice that it was tuned to hear. That voice would have a message of the channel data, but it also includes the transmitter ID and model ID. That way the receiver is doubly sure that the message it got is intended for it. It's not a million kids hollering "MOM! DAD!" it's hearing "ANDY KUNZ! THIS IS IAN!" I know it's for me because I hear a voice I'm tuned for, plus it's addressed in a way that is uniquely going to identify it. If I heard my son Ian's voice calling for "JACOB SMITH" it wouldn't be something I'd listen to. Maybe later I'd ask him who Jacob Smith is, but it wouldn't be a message I'd pay attention to.

The transmitter is always going to be saying "This is transmitter 1234563223a1 model 223 with channel data 1, 2,3 ,4 ,5 ,6 7" and it will always be encoded using its unique PN codes for "transmitter 1234563223a1 model 223" - a very unique voice, with a very unique name (1 in 4 billion). The only way a receiver is going to hear that message and it not be complete noise is if it knows to listen for transmitter 1234563223a1 model 223, and if it knows the PN codes that transmitter 1234563223a1 will be using for model 223.

For a receiver to think the sound it heard might be for itself, it would need to have a section of its memory set up to listen for the voice of a particular transmitter/model. By that, I mean it would need to know the name to listen for ("1234563223a1 model 223") as well as be using the correct encryption (PN codes) for provided by 1234563223a1 model 223. Those two keys would have to match.

So, by saying that the receiver was never bound, you are saying that your receiver thought it was bound (it was, at the factory, to a different transmitter), that the model in that factory transmitter was the same as the model in your radio, and that the unique ID of your transmitter matches the unique ID of the factory radio (which is based on the RF chip's serial number, guaranteed unique by TI or Cypress). The unique IDs are a 1 in 4 billion thing, and there aren't 4 billion of these chips made yet, so no repeated serial numbers are possible.

Does that help clarify why I say that it's virtually impossible?

Andy

AndyKunz

Nothing is wrong with the receiver, nor with the transmitter.

Perhaps the explanation in my previous post will help explain why a receiver will not hear a transmitter unless it knows both the serial number and model number of a transmitter in order to get the "voice" right, and also those same items (1 in 4 billion chance) for the messages coming in. In short, if the receiver hasn't been bound to this tx/model, it simply won't listen to anything the tx has to say (if it could even hear it).

Andy

djmp69

I'll try this one last time. I never said the receiver wasnt bound. I said i hadnt yet binded it to the other model. The receiver was bound to model x in the TX. Model x is the ONLY model it has EVER been bound to. Model x was a fresh setup, and no other models had been deleted from the TX. When I put it into model z, it was STILL BOUND TO MODEL X. So when I turned it and the transmitter on, the only model it should have responded to is model x. Yet it was responding to model y. I had NEVER BOUND THAT RECEIVER TO MODDEL Y, ONLY MODEL X. That has never happened before. And THAT is the way I understand it is NOT supposed to work. And I certainly would hope it is not designed that one TX can control two different receivers AT THE SAME TIME. You keep mostly explaning spread spektrum in a way that has nothing to do with what I experienced. If you look at your above response, you said EXACTLY what I'm saying. "In short, if the receiver hasn't been bound to this tx/model, it simply won't listen to anything the tx has to say (if it could even hear it)." That's why I'm thinking trouble, because it was never bound to the model to which it was responding.

Yes, I do understand how spread spektrum works, by the way, in answer to your question.

djmp69

Afghhjn

djmp69

At least YOU got it, and at least I know I'm making sense, thank you!

AndyKunz

Good that you understand. Now please explain to me how it knew the secret information, both the IDs and the PN code. It had never been given the data, and there is no "universal" voice it could hear. I told you at least one way how it could hear it. Does that possibility make sense?

Andy

djmp69

And the Light goes on. Now you understand my confusion. No, i cant explain how it knew the secret codes, which is why i reached out here. I notice you quoted the easy stuff. Why didnt you quote yourself saying the receiver wouldnt respond to a model it was never bound to? And this is exactly what happened. Can YOU explain how the receiver was responding to a model to which it was never bound? You yourself said this isn't supposed to happen. You tried to explain it, but the only things you said could cause that NEVER HAPPENED. Its pretty simple. I binded a receiver one time, and it responded to a completely different model that it was never bound to, while that model was contolling the receiver it WAS bound to. Could it be, oh, I don't know, there's some sort of problem? Radio is radio. Make no mistake, this happened. Whether you want to believe it or not, it happened. It's OK to say you don't know. This may be one of those flukes to which no one will ever know the answer. It might not even happen again. All I know is, I have a new receiver, and it works the way it's supposed to, not like the bad one. It responds, just like all my other 18 Spektrum receivers, only to the model to which it was bound, whether i deleted models or not, whether I binded it to 1 or 10 different models. That's the way I have always known this system to work from dsm, dsm2, dsmx. When something like this happens, that tells me there is an unforseen issue. Maybe try being open to the possibility that there is a way something could be wrong instead of treating someone like they have no clue what they're doing because you think your system is perfect and there's no way there could be a flaw. This is how research and development works. You find out about certain things, research the issues and develop fixes or better systems. Not treat your customers like theyre idiots.

AndyKunz

I did. The receiver is not bound to a model. It is bound to the transmitter + model number ID. The model number ID can be changed according to the ways I listed.

"The receiver was bound to model x in the TX. Model x is the ONLY model it has EVER been bound to."

I tried to explain to you that it's not the MODEL, it's the ID of the model. The ID can be re-used multiple ways. The system is working correctly, as designed. End of story.

Andy

djmp69

OK cool. So the system is correctly allowing for a rx to be controlled by a model id to which it wasn't bound. Even though you said it couldn't happen that way. Got it. So then by that I don't even have to change models in the tx,, I can just turn on and go for it.

Thanks for the clarification!

AndyKunz

No, it's ONLY able to control a model with the same ID. Same as I've been telling you since the beginning.

Model is not the same as Model ID.

Andy

djmp69

regardless. All I know is that a receiver was responding in a way it was not supposed to. Ifti had gone to fly that plane it would be crashed because even though i gad control, it was accepting the controls set up for a different plane.

bottom line, everything is working the way it should now, the bad receiver is gone.

I'm done.

rgburrill

Obviously, Spektrum can never have any bad parts. This is part of the reason I never buy Spectrum and rarely buy Horizon. I used to buy Tower all the time but not since horizon bought them.

And Andy, I got my master's degree in digital systems engineering in 1974. And prior to that I was a product engineer over X and C band transponder systems. And prior to that I had my first RC aircraft. All of this under government procurement regulations and specs. And I've seen my share of bad parts. It happens, even to Spektrum.

AndyKunz

Yes, bad parts do happen. They are electrical failures caused by static, poor handling, bad soldering.

They don't magically make mathematical functions decrypt data, not when you're looking at large primes.

Andy