How to repair noisy RF Transmitter?

[Richman87]

Hi,


I have a voice activated RF Transmitter, ~23MHz, powered from AC lines (220V 50Hz)[no ground wire].

It is always activated with noise sound in the Receiver,But it works normally while I'm touching the Ground track of PCB.

- after touching the PCB several times on several points, it appears to work normally (without hand touching it), but I'm afraid that it won't work for a long time..


If anybody has encountered this problem, please help me..

 

[enjunear]

Hi,


I have a voice activated RF Transmitter, ~23MHz, powered from AC lines (220V 50Hz)[no ground wire].

It is always activated with noise sound in the Receiver,But it works normally while I'm touching the Ground track of PCB.

- after touching the PCB several times on several points, it appears to work normally (without hand touching it), but I'm afraid that it won't work for a long time..


If anybody has encountered this problem, please help me..

It sounds like you are providing a path to ground (human body = water + carbon = decent conductor). Ground your PCB, or at least add a few kOhm resistor from your PCB ground to earth ground. Depending on where/how the noise is getting into the detector circuit, adding a cap to filter off the noise is possible, although requires a LOT more math and time.

 

[BradtheRad]

You may be sending a static charge into the circuitry, activating it.

And even if it's not static, your body can pick up ambient 50 cycle hum, and then you conduct it to your circuitry, activating it.

Evidently your noise detector is very sensitive.

Presence of static charge (and ambient 50 cycle hum) can be detected several feet away by a simple circuit based on a JFET.

 

[Richman87]

I will try to clarify the problem more.

first, I have imported dozens of this device 'SOME OF THEM' suffering from this problem, and the Far Eastern manufacturer doesn't respond to my question :-?

so it's not a convenient solution to add a wire for grounding.

second, the device shouldn't transmit voice otherwise it reaches a threshold adjustable by the user.
(so it's called "voice-activated")

- In the case of faulty device, the transmitter is always sending noise sound even if there is no voice, and when I'm touching ground track of the PCB the noise disappears and the device functions normally, after removing my hand the noise comes back.

* I had tried the following method for some faulty devices:
"touching the PCB several times on several points,then it appears to work normally (without hand touching it)".

it worked ,but I think it's not reliable method. :???:

 

[biff44]

try a new microphone. You may have a break in the ground shield wiring.

 

[ALERTLINKS]

In oscillator section, increase the inductance of the coil by turning ferrite slug towards inside half to one turn, until noise disappears.

 

[BradtheRad]

Thank you for clarifying.

A unit mistakenly thinks it is detecting audio, so it turns on.

However it works properly while you touch the wiring.

Sometimes the problem stays gone after you remove your hand.

Sometimes the problem returns.

This is identical to the effect of hum/static from your hand.

I have seen a sensitive device put out square waves when I simply touch the input lead.

When I take away my hand, the device remains in the state it was at during the exact millisecond I removed my finger. I have tried repeatedly touching and removing my finger. Sometimes the ouput is high, sometimes the output is low.

Usually the charge drains away gradually. After a few seconds the device returns to its original state (if it's a digital device). Or it returns to quiescent mode (if it's an analog amplifying device).

When you touch the ground wire of the pcb, you are stabilizing the operation.

Your transmitter is probably adjusted to be very sensitive to any sound. Normally this is preferable. However being sensitive, it's also easily fooled by any static or magnetic interference, coming from the mic power circuit, or the mic cable, etc.

I think your mic pre-amplifier needs to be adjusted so it is not so sensitive.

 

[chuckey]

I would find the "voice operated switch" and try to adjust it so it is not so sensitive. Or slug it so it is not so fast that a single pulse of noise operates it, say a .5 -1 second delay. I don't think touching the PCB is very scientific.
Frank

 

[jiripolivka]

Repairing a finished product is often difficult. As I can see the transmitter is quartz-crystal controlled, I would recommend to start from it. Disconnect the other stages, connect a spectrum analyzer (if you have it, or a receiver with a signal-power indication) to the oscillator-output coil (not tuned), and tune the oscillator for a clean output.
Then connect the second stage, and test again.
The modulator stage as you explained is a "comparator" or VOX type, activated by a voice input, may need to adjust a reduced sensitivity. Also, check RF blocking of its input lead.
In general, all RF devices should be well enclosed in a conductive enclosure, with a RF connector well connected to it, and all DC or low-frequency connections well blocked to ground. Otherwise, unblocked wires do carry some RF signal out, and can as well pick stray signals in.

 

[davenn]

Repairing a finished product is often difficult. As I can see the transmitter is quartz-crystal controlled, I would recommend to start from it.

We dont know if the OP's circuit is crystal controlled, he has not given any info on the cct in use. for all we know it could be synthesised.
You are referring to the cct ALERTLINKS posted, which may have nothing to do with the OP's situation

Its really time for the OP to tell ua a lot more about his cct in use so that we dont have to play the 20 questions guessing game....

Richman87 how about some photos and circuit diag of what you are working with please

Dave