Sender FM-Teil regeln ?

[ropp]

Sorry for not using german; if I find a translater I'll post it again in german ; Answers welcome in german too; reading it is not a problem.
-----------------------------------------------------------------

Hi ,

I want to verify my TX frequency.
I have a HP spectrum analyser which I can use.
As the signal send by my radio is modulated, it both sends the Xtal base tone (in my case 35.030Mhz) and the down shifted frequency
(as our RC FM is basically FSK, frequency shift key).
My questions :

-how much does a Futaba radio shifts down (negative shift keying) ?
-how precise should my frequency be ? How many ppm may it be off ?
-Would it better to "kill" the LF-signal to get a proper measurement ? (disconnect from HF-part)
-I measure "true air"; is it better to directly connect the spectrumm analyser to the radio ? If so, how ?
-Is a spectrum analyser (calibrated HP8691A) good enough to get a correct freq. measurement ? Wouldn't a counter be better ? Is so , how to connect to the HF-part of the radio ?
-If one changes channel ; is it needed to have the HF part recalibrated ? Is yes, what is tuned again ? Only the 2 frequencies or also other things ? Is so, what ?
-Tuning the HF , Is this something I can do myself using the spectrum analyser ? Do I need other things to ? Which ? What other parameters need to be checked/set apart from the frequencies ?

All inputs welcome .
Please no reactions on the legal aspects; I'm only interrested in the technical matter.
For your info, I'm familiar with electronics (engineering degree) but have not much experience in RF.

thanks,
Rob.

 

[Uwe Gartmann]

Hi Rob,

If you are unsure with your TX, send it to a service stations for your used brand. Ok, i agree, some things can be handled itself, but hey, do you repair your car with broken engine by yourself ;-)))

What I mean, HF has some magic. Without knowledge on that parts, you have no chance to be sure to make the 'right' thing.

But to give you some answers:

-how much does a Futaba radio shifts down (negative shift keying) ?

Not more then 5 kHz? ;-) (10kHz Channel separation in Europe)

-how precise should my frequency be ? How many ppm may it be off ?

in the end, it depends on the used receiver. Channel separation is one of the criteria ...

-I measure "true air"; is it better to directly connect the spectrumm analyser to the radio ? If so, how ?

For TRUE results, avoid any other influence

-Is a spectrum analyser (calibrated HP8691A) good enough to get a correct freq. measurement ? Wouldn't a counter be better ? Is so , how to connect to the HF-part of the radio ?

It depends on the resolution/time. Frequency-counters normaly needs stable signals.

-If one changes channel ; is it needed to have the HF part recalibrated ? Is yes, what is tuned again ? Only the 2 frequencies or also other things ? Is so, what ?

Normally, the parts are calibrated at factory. In the field, no changes are neccessary

-Tuning the HF , Is this something I can do myself using the spectrum analyser ? Do I need other things to ? Which ? What other parameters need to be checked/set apart from the frequencies ?

RC equipment is quit stupid (thats why some planes fall done even everythings looks ok), so don't care too much. If you use standard mainstream equipment, no problem under normal circumstances should happen.

As a reminder: Do NOT adjust your transmitter trying having better results than before. If you are unsure about performance, send it to the service station.

Greets
Uwe

 

[ropp]

Uwe,

Thanks for you reply.
Off course you're right about sending the transmitter in for service; those guys know what they do, they do it every day ... .
But:
I've been experiencing radio problems for a while already and I can't seem to find out what it is that goes wrong. Every now and then , I get a "glitch" (some channel that deflects full and comes back to neutral) .
I have experimented with several receivers in several airplanes and gliders.
I have receivers from futaba (transmitter is FUT FC18), single conversion(1) and dual conversion(3) (no PCM) and also 2 Schulze alpha-8 . I have seen problems with them all. Also changing channel (from 63 to 74) didn't bring improvement.
The problems with all my experimenting is that the glitch only comes every few flights; not every flight ; sometimes not even during several weeks > this makes it extra hard to compare the different setups; it takes several flights on different days to draw conclusions.
Seen the fact that the problems arrises with different receivers , the problem is either the transmitter or the used channel (XTAL ? , external influences ?) .
(other pilots in our club don't have similar problems)

The transmitter has already been in for service but it came back with the message "everything OK, HF is fine as it is".

So what to do ? Buy a new one without knowing what is wrong ? Thats not me. I want to know what is wrong and I'm just looking for all possible info to help me with that . And if it takes to dive into the "mysteries of HF" , I'll do that.
The only thing I really want to avoid is to start experimenting with my HF and causing other pilots to get problems caused by my transmitter. So, don't worry , I will not start tweeking any pot or filter on the HF without knowing what I'm doing. I'm in the stage of verifying, with all possible means, if my transmitter really is OK as was stated by the service guys. And I guess that means checking the frequencies in the first place and see if the FSK shift is correct (enough but not too much).

btw : on RC-universe, I've found that Futaba shifts from F+1.5k to F-1.5k .

On a FC16, I see those two freq. bumps on the spectrum analyser, but I don't see them with my FC18 . If I zoom in (and minimise bandwith; minimum is 1kHz), I get a rather strange image . So, I'll measure again , disconnecting the LF signal , an forcing the 2 FSK frequencies seperatly to get stable readings.

note : all range checks have been OK; results above average in our club.

regards,
Rob.

 

[midgetmustang]

Hello Rob,

3Khz frequency shift is a recommended value. To see stable measurements, you have to disconnect the modulation voltage. Do see the correct transmitter frequency, first find the correct place where to disconnect the signal. Use a oszilloscope to find this. But, keep in mind, you are looking to see a problem, which occurs rarly, if at all, is a problem in your transmitter. Most likly, you will find, everything is ok. You said, or I understood, it is only one Servo-Channel which seems to show the Problem. As other Pilots don"t see the problem, I would expect the reason for the glitch inside your airplane. You said, you have seen the problem also with a Schulze receiver. Don"t they have this LED errorcounter? Check it after a glitch. Or check it after evey flight. Most common, problems are the transmision itself, not the transmitter.
We tried to narrow down a similar Problem from a friend, with rarely, but ugly glitches, and after some months of research, we didn"t find anything. Result: he changed the Transmitter and the problems where gone

cheers
mm

 

[midgetmustang]

Hello Rob,

it may be very hard to find the rootcause for this.

cheers

mm

 

[Uwe Gartmann]

Hello Rob, mm

Yes, try another TX if possible. Check the antenna wiring inside your TX too.
Sometimes cables / Stick-Potentiometers are defective or unvisible corrosion can happen.

cheers
Uwe

 

[radioglitch]

Rob,
in case both TX are using PPM and the same number of channels / same model type you should have the same / very similar readings on the analyzer. Try to check if changing the X-tal does make a difference . . . .

In case your spectrum analyzer does have a trigger input you can try to feed the ppm signal into. By inverting the the trigger you can select low/high frequency.

In case only some servo channels having trouble I would check the stick poti, cables and connectors. Try to check the TX PCB as well, all at different conditions (tilting the TX, cold, hot etc.)

Most common error for strange RX behavior in my opinion:
- empty / faulty / to small RX battery

 

[ropp]

I managed to verify the 2 frequencies by turning the modulation off .
Turning it off was actually rather easy : I just removed the ROM of the processor (is on a socket in the FC18). This gives the F+1.5k condition. For the negative shift, the LF signal (yellow wire) is to pulled to ground (black wire) ; this gives the F-1.5k condition .
I measured 35.031240 and 35.028255 , so 3k span but 250Hz out of center. It was easily corrected with the 2 pots in the HF part. One sets the span, the other shifts it (small problem : the one that sets the position , also slightly impacts the span).

Will this solve my problem ? I doubt it , but now at least I'm sure that the settings are correct.

One interesting thing : I also measured with another Xtal 35.140.
Before the correction it was 1kHz off ! After the correction still 750Hz.
Both Xtal are original Futaba.
This doesn't say much about the precision of the different Xtals , but it shows that a transmitter can only be set at maximum correctness for one single Xtal. When you swith channnel , the HF-tuning is gone... .
But then again : if this will have impact in practise is doubtfull.

btw: batteries : I also experimented using different RX-batteries > no impact ; all my batteries are fine .

greetings,
Rob.

 

[ropp]

For those interested , I also had a threath going on RC-universe :

http://www.rcuniverse.com/forum/m_4900819/tm.htm