digital proportional remote controls

Ouch!

Not sure if it is used on RC helicopters but many Chinese toys use this chip:

Made by this company:

But I am afraid you'll either have to understand Chinese or find an engineer who does.

Aluminim foil can sometimes work miracles :-)

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

There should be an RC to set-up the decoding timer.

How many pins on this chip? I may have designed it :-) ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

       The Ground Zero Mosque IS Appropriate When Renamed...
            The Obama Monument to American Impotence

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Is't the first pulse or two a preamble to set the timing for the rest that follows? Making the RC network drift part of the decoders job.

Cheers

You need to fly it in China to work properly.

It sounds more like an RF issue that the receiver doesn't like whatever noise is being picked up. Your ceramic 'cap' is more likely a resonator. They don't perform as well as crystals and that may be the issue.

I seriously doubt there is any PCM involved as you saw the pulse widths vary by control settings. That 'sync' pulse is the marker to start counting off pulses and route them to the proper servo channel. The servo has its own monostable adjusted by a pot on the output shaft. When the local pulse natches the input pulse it stops moving.

Or maybe I'm way off base.

G=B2

Nope. You've got it right. To debug it, observe matching pulse widths at the plane and at the transmitter. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |

       The Ground Zero Mosque IS Appropriate When Renamed...
            The Obama Monument to American Impotence

If you've bought a complete unit, my first choice if it's misbehaving would be to take it back to the store and have it fixed, replaced, or refunded.

I once talked to an actual Futaba tech, at the Futaba factory in Orange County, CA, to find their PWM spec. It was surprisingly simple, just PWM. (I had an assignment to simulate that signal, with an 8031 (8051 sans ROM).

But I wouldn't start hacking into a piece of bought equipment, unless I'd exhausted all the other available options, especially if I wasn't sure what the signals were _supposed to_ look like, and that isn't the kind of experiment I'd like to undertake. =:-O

Good Luck! Rich

Hey thanks, so far you're the only one who actually answered with useful information. I'll look into it. But my chip is a 14 pin guy, this one is 16.

Bah, it's not a very complex circuit, I'm going to kill the power to the RF section. Even my digital scope's knobs were misbehaving. I guess I could fire up my trusty Tektronix 547, it won't mind the extra RF. And anyways, it's cold today in Montreal! ;)

Thanks!

I'm pretty sure it's a cap, it's marked 120nF, with a smaller, brown leaded ceramic cap poorly soldered on. My instinct is to think this was tweaked at the factory to get to whatever frequency the unmarked IC is supposed to run at.

Please read my message and don't 'doubt' what I saw. It's kind of condescending. I'm not a moron. There are about 30 pulses between syncs. It's a 4 channel remote. A handful of pulses change for the throttle, a handful for the steering, and handful for forwards/backwards, and a handful for trim. Do you want me to upload a movie of my scope screen to youTube? The only place where you can see old school analog varying of pulse widths is inside the remote, where all 4 pots go to the mysterious unmarked chip. Before the 3ms sync pulse, the chip samples all 4 pots once and you can see the RC waveform change with the stick movements.

Besides, what's to doubt? It's 2010, electronics is cheap. It's either a microcontroller, although its power pins don't correspond to any uC I've looked at, or it's an ASIC for RC toys. Like the OKI L9362 (?) in my old Futaba stuff.

Insanity. No remote sends out the raw servo pulses since about 40 years. That's nonsense. There are no servos inside the toy helicopter, only high frequency PWM motor drivers. The only moving parts are the two rotor motors and the tail rotor, all big burly brushed motors driven by the 100A power transistors (all the discrete parts are marked).

You are. IT'S A CHINESE TOY HELICOPTER. It's a CAP, and it's PCM.Probably bi-phase or some such to keep the DC level about constant to simplify the receiver.

Thanks, but it was a final sale.

).

Bah, it was really cheap. As I later found out, this model has a design flaw, I'm not the only one with throttle/control problems on this model. Of course, no one told me when I bought it. It's DOA by design, and it's just for fun I'm hacking in this thing. Because I believe it's a design flaw, I'm looking at global, general things wrong. Slapping a cap instead of a crystal is pretty bad.

Since the chips are unmarked, I'm going at it from many angles.

So far, my best bet is to remove the receiver from the helicopter to troubleshoot the tx/rx combo. But first, to get an idea of what the pulses from the remote mean.

It's hard to do much of anything with a 2 foot long helicopter going full throttle at random. It's quite impressive to see this "toy" go full blast though.

I think killing the 40MHz transmitter and just hooking my logic analyzer and doing a little pulse counting session will solve that.

Thanks!

The standard older digital proportional systems used something like a 4.5ms sync pulse and a number of 1-2ms (1.5ms for centered servo) pulses for each channel. The frame rate varies with stick positions. As far as noise sometimes there capacitors on the motors, I've had problems with a glow fuel helicopter giving radio interference from metal gears running together. Loose metal parts can cause radio interference too.

RogerN

ms

ch

e

Yeah, I know old stuff uses that method. The first Kraft systems about

40 years ago were based on that IIRC. All previous designs were reed controlled. Honestly, that's what I expected to see as well since it's a toy. I was a bit taken aback when I saw the quantity of pulses on the scope. "This isn't gonna be easy" was my first thought.

uel

Thanks, that's actually very helpful. As I was examining the helicopter, I noticed a circular PCB built on top of each motor with a delta (I guess) capacitor network, one across the terminals and one across each terminal and the motor's case. However the quality of the soldering is terrible. It doesn't even look like solder, it looks like lead. It's not shiny at all. Maybe it's some kind of RoHS stuff but all the RoHS stuff I've used doesn't have that flat gray look of pure lead. But the main blob of solder that goes to the case looks cold.

I wonder if a diode snubber network is required for when the motor becomes a generator in the PWM dead time? One of the things I want to do is twist the wires together to reduce inductance.

Anyways this is keeping me busier than I wanted.

It is fun though to see the kind of stuff you get in toys these days.

Dynamite-stick LiPo batteries, huge current switching circuits, powerful motors, etc...

Seriously, the battery is a 11.1V 1100mAh 3S battery, impressive, and the helicopter kills it in 5 minutes. That's about a 20A current drain. All on cheesy wiring and questionable PCBs and solder joints.

Camera photoflash circuits are impressive. ~150A peak through a surface mount IGBT, multilayer circuit board and wires the consistency of angel hair pasta. At least it's only pulsed.

That battery must get pretty damn got going at, what, 12C discharge? Amazing indeed that nothing burns/melts/explodes.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

e

Don't forget there's an equally impressive airflow, even with the cowling. It's a coaxial rotor chopper, like 99% of the toy helicopters and the blades sweep 18 inches in diameter...

When I started it in the kitchen, it was hilarious. My roll of paper towels unravelled, stuff went flying, etc