As has been the case with so many other aspects of my RC learning experience, I came to the appropriate thread too late. Last week while flying my 3D HobbyShop Extra 300 in an harrier about 8 feet off the ground, she quit flying quite suddenly and mushed into the ground. Thankfully only the gear broke. Today while practicing a "wall" followed by a lame attempt to hover about 50 ft infront of me and about 10 to 15 feet up she suddenly died again. After the first accident I assumed it was operator error of some sort, but today the shut down was quite dramatic and the power came back on an instant before she hit. After reading through this thread I am convinced I fell victim to the "cone of silence". I admit I was guilty of having the antenna laying parallel to the ground and was pointing it directly at my model in both cases, and given that the batteries were less than 1/3 discharged and the wind was calm during both crashes I feel confident in my assesment. My reasons for posting are 2 fold. First I'd like to share my experience with the great big world of RC Groups just so I can do my part to improve the body of common knowledge, but second:


Has any person or group documented this phenomenon in the real world, in repeatable trials with a DX7? And, if so how big has this cone been proven to be? Or, how big is the cone in theory in terms of degrees?



Thanks for any and all input, and of course I promise to bend my Spektrum antenna 90 degrees and lay it to the right from now on


Chagrined in NY

Joe

This is not unique to the DX7 or to Spektrum systems. You can do this on 72 MHz systems too, but the 2.4 GHz systems seem to be more suseptable to it. Probably something to do with the wave length on 2.4 GHz.

I speak from reading, not from personal experience.

I'd be looking for a bad connection in the electrical system somewhere, especially the switch system. Flying that close should have no antenna affect. Been using both the Spektrum radios since they both came out, and always fly with the ant. straight out with the planes out to the limit of visibility at times and never had a radio glitch. Had other problems unrelated such as the ESC quiting from overload that would be easy to blame on the radio, and I did suspect it.
Gord.

Gord (or anyone), I'd be interested to know how much range you get with the binding button held in (ground range check procedure) and the antenna straight out, pointing at the model. I ask this because I get less than 20 paces with it straight out, but 40 or more with it left or right (this is with the AR6100 receiver). So my system meets the specified ground range of 30 paces with the antenna to the side, but not with it straight. In that sense, my experience has been similar to 2Sunny's "cone of silence" one.

I've only had my DX7 for a couple of weeks and I don't know if what I'm seeing is normal or not. I did have a lot of trouble initially getting my AR7000 receiver to bind properly, but for some reason this seemed to work better each time I tried it (over a period of a week or so) and it now seems to bind OK. I'm still trying to figure that one out, because as far as I know nothing changed.

Basically, I'm trying to determine whether or not my system is functioning normally or if I need to return it.

Harry

If I remember correctly, with the AR6100 (in an Electrifly S.E.5a) and the antenna 'extended' and pointed at the model, 33 paces was good but 34 was bad. With the antenna canted to the side, I could go all the way to 42 paces.

When I tried the same thing with the AR7000, it wasn't nearly so bad. I can range check to 100 paces (yes, with the bind button pressed) with that Rx.

Sting

I'd be looking for a bad connection in the electrical system somewhere, especially the switch system.

I was thinking of the same thing Fly, but the connections are so simple with no switches just motor, esc, battery, and rx all of which I went over a hundred times and ground tested forever trying to duplicate the failure. The only other thought I had is that in each instance the plane was virtually standing still thus making the chances of hitting and remaining in a dead spot far greater than a plane that is moving rapidly.

but the 2.4 GHz systems seem to be more suseptable to it.

Does anyone have links to others who believe they have suffered the same problem with their Spektrums?

Thanks again for the input.

Joe

Joe:
Make sure your ESC can handle four servos if it's running on a 3 cell batt. That was the problem I mentioned above. The radio just quit cold. When I did a bench test, I rotated the right stick and moved the left back and forth and the ESC kept getting hottor and hotter till it quit from thermal overload. A few seconds later it came back on again. This was on a bipe with a servo for each wing plus elev. and rudder. I put pushrods from the lower wings to the top to eliminate the one servo and had no further ptoblems. On my bigger 40 size electrics I use a separate UBEC also, to solve that problem. We're having our 1/4 scale event tomorrow so I'll try the range check and let you know the findings. The AR7000 is in a 1/4 scale Corby Starlet with never a twitch out of it. Once you get the bugs out of the plane you'll find the radio 100 % reliable compared to the 72 meg. systems. I put my full faith in it and it's a nice feeling.
Gord.

2Sunny,
I am with Flypaper on this one. You shouldn't get a "cone of silence" like that. I suspect something else. Although I do notice reduced range when pointing right at the model with the bind button pressed, I have never seen a dropped link in the air. Where you using the AR7000 Rx? Did you get a good range check?

As a side note, I laid my Tx down in the grass flat, and then turned on the Rx and it did not link up. (I usually stand it up so the antenna is pointing to the sky). Then I picked up the TX, and "pop" it linked up to the Rx. So you may have trouble connecting if you have your Tx flat on the grass.

Um, the 'cone of silence' is real. That demonstration in the range check proves it... the range in the air is going to reduce by the same relative amount, about 20%. But I think power is far, far more likely to be your problem... you were pumping heat into your ESC with hover power and constant near-full-travel control movements, I bet it just overheated the BEC and shut down. I no longer trust linear BECs with a Spektrum RX... switchmode or batteries (and big, quality batteries at that) only.

Looks as if the latest DX7's shipping out of Horizon are now being supplied with the new larger SPM9522 110ma TX/RX Charger. I just picked up a new DX7 Heli model and it had the 110ma charger on board. Also got one of the new Spektrum Neck Strap Adaptors. These are really nice. All metal and very easy to install and adjust.

I don't know why anyone relies on a linear BEC these days, too inefficient and heat is always going to be problem. #3S is just OK but 4S and above is a no go. A separate switch mode BEC is the way to go or a separate battery depending on the model etc. If you sit down and work out current draw and voltage to be dropped with a linear BEC you will understand why they get so hot so quickly

1. Please explain the difference between switch mode and Linear mode in laymans terms. Examples of each type by name so we know what may be the better way to go.


2. Which type are Castle Creations', PHX series, esc's? I've yet to have any troubles with CC esc's.

Thanks

Take a look here:

A Linear BEC uses a linear regulator:
http://en.wikipedia.org/wiki/Linear_regulator

A switchmode BEC is one of the other types described here:
http://en.wikipedia.org/wiki/DC_to_DC_converter

All BECs are some kind of DC to DC convertor. The basic problem with linears is that the amount of power they waste as heat is (Vin-Vout)*I, where I is the output current. That means a 3S lipo 5V BEC dissipates about 6.1W per amp of output. Given that the BEC chip is about a millimeter square, at 3A it would do OK as a soldering iron.

Switchmodes make very little heat, by comparison, which means they can be used at much higher voltages; any BEC rated for 5S or more is pretty much certainly a switchmode.

If an ESC manufacturer doesn't say the BEC is switchmode then it isn't. That's because switchmode has such big advantages.

For Castle in particular, some models have linear BECs and a few very new ones are switching. They also make a brand new 10A BEC, which is such a huge rating you'll never see it from linear regulators (except the Fromeco li-ion regulators for giant scale planes... and they need huge heatsinks and cooling fans).

Hi Andrew,

Is there actually any advantage to a linear BEC?

Originally Posted by Andrew McGregor

Um, the 'cone of silence' is real. That demonstration in the range check proves it... the range in the air is going to reduce by the same relative amount, about 20%. But I think power is far, far more likely to be your problem... you were pumping heat into your ESC with hover power and constant near-full-travel control movements, I bet it just overheated the BEC and shut down. I no longer trust linear BECs with a Spektrum RX... switchmode or batteries (and big, quality batteries at that) only.

Linear regs or BEC use fewer components and for low voltage applications they are very good. But when it comes to higher voltages and current the switch mode is king. Money seems to have something to do with it when producing ESC's. Cheaper ones will have linear and the more expensive ones switch mode.