Originally Posted by 2walla

AR 7000 in the cub , ar6200 in the flying king, and an ar6000 in the quad. Xmitter is in us mode..

Originally Posted by AndyKunz

Hi Jim,

In the Spektrum system, here's how binding works.

The receiver enters "bind mode." It is waiting for a transmitter to attempt to bind to it. It will do this as long as it is powered. If power is removed, if the bind plug is removed and you power it up, it will look for the last bound transmitter. If the bind plug is in, it will return to looking for a transmitter wanting to bind. When the receiver is in bind mode, it flashes the LED rapidly.

When the receiver hears a transmitter that wants to bind, it will tell the transmitter about itself. During this time, the LED will blink slowly. The tx and rx will negotiate the protocol to use, and the transmitter will tell it the GUID to listen for. The receiver will then save the new bind information (transmitter GUID and bind type) in EEPROM or Flash for later use. Then it executes the connect, and the LED turns on solid.

Note that the bind information in the receiver is not overwritten until it has replacement information.

On the transmitter side, when put into bind mode the RF deck will immediately erase the previous bind information (which is bind type and GUID to use). This will allow the transmitter to meet the timing requirements for transmissions consistently - a necessary item for the receiver to see. (It takes a "long" time to erase Flash/EEPROM; this erase-before-transmit is necessary.)

It will then begin issuing the bind message that the receiver is waiting for. It hears the receiver and works its side of the protocol negotiations, then saves the bind information where the erased bind used to be and then continues its transmissions at the negotiated frame rate.

If power is lost by the receiver, upon restoration it will immediately listen for the saved GUID using the saved protocol, using a special "re-sync" algorithm to find the required frequency at the proper timing. This will restore the connection.

If power is lost by the transmitter, upon restoration it will transmit according to the GUID/protocol that was saved during the bind process. It doesn't have to do anything special to restore the connection; that is all handled in the receiver.

I've explained this several times, both here and on RCG. With a little googling you may be able to find the references, if you feel they'd further clarify your understanding of the bind process.

As for non-power-related failure modes, there are only a few ways for a bind to be lost. One is with extremely high temperature for a sustained period of time, causing the non-volatile memory to be erased. I doubt this would be possible without other obvious major physical damage to the receiver or transmitter. An extreme EMP condition (atmospheric nuclear detonation? cosmic particles or rays?) could also cause bit errors in the memory which would prevent a reconnect. Over-voltage is another possibility, but it would probably require power insertion on the load side of the internal voltage regulation. I suppose extreme over-voltage on the servo bus could cause that, but I'm not about to try it myself. Some things that WON'T cause loss of bind data are a crash or power loss.

A connection can be lost by a host of means. As others have noted, a momentary power loss can cause the receiver to execute the reconnect mode. In the very first DSM2 receivers this could take up to 3 seconds; version 1.6 corrected this and reduced it to about 1/2 second. With DSMX the reconnection is established within just a few ms (less than 100). A connection can be lost due to an extremely low signal level - this can be demonstrated by doing a range check and then attenuating the transmitter by grasping the antenna with your hand or putting your body between the devices. Connections can be lost by having the antennas shadowed behind large conductive surfaces such as battery packs, engines, or carbon fiber (wings, fuselage, etc.) - this is why the physical diversity of antennas is important to all brands.

I hope this clarifies the difference between a bind and a connection. If you have more questions about how the stuff works at this level, just ask.

Andy

Originally Posted by Jim Branaum

Thanks Andy.

I am pretty sure I knew, know, and agree about bind but am not sure I agree totally about all the non-power related bind failures. Everything you have said is software related and inside the chip, but if the electronics supporting that chip fail there is no way for that bind to be effective. I suspect that is where the difficulty in comprehension for many comes from as the term 'bind' is a software related issue but may (and I am sure does) have a hardware (electronics) component.

I have designed, built, tested, and sold inventory control systems starting with the software all the way to the completed product that included multiple cash drawers that were discretely controlled and could be opened based on the data acquired (which product, what funding, and which user = which drawer). A systems failure could have been either a software failure (only happened once in beta when the users network crashed) or a hardware (electronics) failure. That is all I have been saying.

Originally Posted by AndyKunz

Actually, most of what I said was electronically related, so I'll just continue here with a little more electronic detail.

Another electronic component of "bind" is the storage of a binary value in EEPROM or Flash non-volatile memory (NVM). As you may know, these sorts of devices are guaranteed erase/write cycles with a minimum of 10K, typically 100K or more cycles. Reading a value does not degrade the life storage, only erasing and writing.

A single bit failure in the NVM would not cause an immediate loss of RF connection; it would simply not be able to reconnect after power cycle.

A single bit failure in the RAM would cause an immediate loss of RF connection, but would be restored immediately on power cycle.

As you may also know, the likelihood of either event occurring is about as close to 0 as anybody can figure. If we were in outer space, it would be measurable. Down here, the atmosphere does a really good job of preventing the sort of radiation that could cause either event.

I've been doing embedded stuff (both hardware and software) since the early 1980's. Systems I have worked on have been deployed internationally with everybody from modelers and other consumer-type good right up to the State Department and DoD. I have developed multiple products which have been manufactured in the hundreds of thousands (and sometimes just one or two very expensive things), and am constantly working on new products for similar high or low volumes. My jobs have, quite simply, required a very deep knowledge of the entire SYSTEM, not just individual modules - my value is that I am both deep and wide on the systems, the modules, and the electronic components in them as well as the firmware and software. So I'd say you and I could probably share a number of horror stories and be able to relate well over a dinner.

Andy

Originally Posted by Lifer

I've been using Airtronics for many years, and their 2.4 systems since 7 or 8 years ago. Not one single issue of loss of bind or in-flight failure.

Originally Posted by Lifer

I've been using Airtronics for many years, and their 2.4 systems since 7 or 8 years ago. Not one single issue of loss of bind or in-flight failure.

Originally Posted by AndyKunz

Your experience parallels mine with released Spektrum products.

My only radio-related crash was bad code in my radio while testing new code, stuff that never saw production. That's the cost of being the guy writing the code, I suppose.

Andy

Originally Posted by AndyKunz

I think you've already seen some very good answers to your question here, from myself and from others, so there's not much sense in repeating them.

My question for you is, why do you find it so hard to realize that somebody selling about 10X the number of products as everybody else would have more reported problems (and successes) as everybody else. And why is it that folks don't understand that the way larger number of entry-level users will have skew the results in one direction due to both their inexperience and the low-cost accessories that they use, while mid- to high-end users report the same success as everybody else?

Andy

Originally Posted by AndyKunz

I think you've already seen some very good answers to your question here, from myself and from others, so there's not much sense in repeating them.

My question for you is, why do you find it so hard to realize that somebody selling about 10X the number of products as everybody else would have more reported problems (and successes) as everybody else. And why is it that folks don't understand that the way larger number of entry-level users will have skew the results in one direction due to both their inexperience and the low-cost accessories that they use, while mid- to high-end users report the same success as everybody else?

Andy

Originally Posted by init4fun

This should be posted in every thread where a fan of some other system takes a cheap shot at Spektrum .

Originally Posted by init4fun

It's all good HoundDog , no ruffled feathers here . I was agreeing with Andy that those of us who use quality batteries (or ESC/BECs) and also good quality servos and switches seem to never have these supposed "spektrum issues" that some like to go on about . I have owned radios from all the major players since Kraft and have yet to experience a radio failure related loss of a model airplane on 2.4 GHZ . I don't think for one minute that this is "dumb luck" or "divine intervention" or any other whimsical mystery , it can be found in the use of best materials and the proper placement of the components during the build . I can't count how many times I've seen improper radio setups in others' planes , things like the main and satellite antennas both being in the same orientation (supposed to be 90 degrees apart) and then folks wonder why they lost signal ? Couple bad installation with the pig headed refusal to own up to augering it in by their own dumb thumbs rather than blame the radio , and I'd say Spektrum is getting a bad rap that it don't deserve .

Originally Posted by HoundDog

Originally Posted by Jim Branaum

Your statement, sir, is a very broad brush of indictment towards all others who may or may not have had a problem and clearly suggests you have 'superiority' "issues".

I can't speak for all, but when dealing with a plane weighing over 20 pounds and traveling at speeds well over 30 MPH, *I* tend to go with the more expensive 'solutions' to avoid issues that may or may not endanger other lives. That is why I am so displeased when I do have something go wrong. Clearly (THEY SAID) my bird lost all power to the RX even though I used 2 high capacity batteries on 2 different switches feeding 2 different points on the RX. The cheapest thing in that set up was the RX. After the crash, the only thing untestable was both switches - they got crushed. The batteries tested fine, the techs say the TX tested fine. I am not aware of sunspots and I was high enough to recover any potential dumb thumb I might have done. The plane had been flown as it was for about two years with no evidence of a problem of any nature and the batteries were cycled once a quarter (because I do try to use belt and suspenders - which means looking for potential problems before they bite me) all of which pretty strongly suggests the installation was good. So, sir, exactly HOW do YOU want to paint my result?