Unstable control with serial interface - Using the serial interface in the instructions is glitchy

[ebourlet]

I built the interface in the instructions. The problem is that when I calibrate it never gets smoothe. Flying the planes act like one would with a glitchy radio. I have a Futaba 5UAP in PPM. I used shielded leads from the radio to my interface . I am using it on a 400 MHZ PII. Any help or suggestions would be great.

[IanC]

Keep the lead as short as possible from the interface to the port. The lead from the transmitter is ok (better if screened) but found that screening connections to the PC inproved things. Still not 100% but alot more flyable.

[Stroker]

Hi
Try turning off the sound in the sim. that worked for my computer.

[ebourlet]

Wow thanks Stroker that worked.
So is this a problem with my computer, sound card, or program? I have an older Soundblaster AWE64

[RookieOne]

Mine's jittering too. Both sound and image.

I tryed to turn off sound but still jittering.
Lead is shielded and as short as possible (60 inches).
Used special shielded case for the 9 pin connector too.

Where have you installed the interface:

- near the PC?
- near the Tx?
- just in the middle of the lead?

Which is the best place?

(Edited by RookieOne at 7:00 pm on Nov. 5, 2001)

[Stroker]

My interface is near the serial port. That works just fine. Its placed on the table beside my computer.
The powesupply in the computer may cause some noise.

[Mike Roberts]

Try a revision to the serial interface to move the switching point nearer the middle of the voltage swing. See my website for details under R/C projects>FMS>Circuit diagram
Mike
www.customelectronics.co.uk

[Vic Shaw]

To Mike Roberts
In your newer wiring diagram you have inserted a diode in the V+ line for power from PC to Tx.
Only for a custom made buddy box you say, because of the low voltage.
I have a Futaba 8UAPS with removable RF unit, if this unit is removed could it be run using the PC power supply through your revised interface.
I'm having trouble with the jitters and so making the newer interface.

[Mike Roberts]

Vic
The reason for the diode is that the RS232 Data lines go negative as well as positive. When they go negative (e.g. when FMS is not running) there is a possibility of damaging the transistors. The diode avoids this risk.

The special Micron dummy transmitter has a very low current consumption (1ma). This low power can just be taken from the RS232 data lines. Anything more would prevent the RS232 interface working. You can measure the current of your coder either directly or more accurately by measuring the voltage drop accross a 1k resistor in series with the supply.

Mike

[Vic Shaw]

Have made the interface with the newer values of components.
Channel 1 and 4 have improved but 2 and 3 seem worse for jitters.
Have not connected V+ yet waiting for your reply.
Vic

[Mike Roberts]

Vic
Something else you might try. Start with my circuit with the diode. First add a capacitor from the transistor end of the diode (+) to ground. This might give a more stable supply to the coder. (I wonder if some RS232 interfaces have lower power capability).
A second idea is to apply a separate 8 volt supply to the interface so it takes no power from the PC.
Let me know how this works.

[Mike Roberts]

Vic
I should have said - capacitor around 10uF should be about right.

[Vic Shaw]

Our E-mails are almost crossing.
Just tried a .022uF cap in position recommended (Because I had one).
Ch1 on my Skysport 4 TX became ch2 in calibration/mapping field on sim.
Ch2 became 4, Ch3 became 6 and Ch 4 became 8.
So that not the way to go.
Then tried a 1000pF (again because I had one) and that gave great improvement.
Steadier model in sim.
Have looked in my junk box and haven't got a 10uF cap.
Will get one, try it and report back.
Thanks for your help.

[Vic Shaw]

Tried turning off sound in FMS menu then rebotting machine that has made a difference.
Gliders are perfect, they don't have sound anyway.
Power models are much better but still twichy, maybe a little too twichy just like a real small model.
Will still try a 10uF cap from V+ to ground when I get one.