Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Frequency Drift

Status
Not open for further replies.

zachtheterrible

Active Member
I built a very small transmitter, a little bigger than a pink eraser, and it can transmit 1/4 a mile. I'm having some trouble with frequency drift though. What are some causes of frequency drift? My transmitter uses a parralel LC circuit btw. It's not me putting my hand near the coil or touching anything on the circuit. Could a bad capacitor in the LC circuit cause this?
 
That's the tough one....

We are talking about very simple devices where each stage does more than
one task making it difficult to setup so some luck is required as well.
I've built tons of FM transmiters and sometimes it's better to put the
"misbehaved" unit away and simply make new one. Usually they
work very decently or stable enough if you follow few simple rules.
One thing to remember is that stable oscilator needs good quality parts,
all connection should be as short as possible (in my case all components
are sitting tightly on PCB and the PCB is ca 25x15mm or 1"x5/8"),
do not expect oscillator to work too hard to get the bigger range.
The more you load it, the less stable it is. Add another stage to boost
signal if required. For example most od the simple FM transmitters
use small capacitor as feedback from emitter to collector and emitter
is connected to battery through resistor. Don't just lower the value of
that emitter resistor and don't increase battery voltage too much.
You might get biger range but terrible stability. Use transistor with
higher gain. The bigger the better... Just remember we are not talking
about gain at audio frequencies, we need transistor with good gain
for operation at ca. 100MHz. The bare minimum is to have transistor
that can oscillate at 300MHz. Make sure not to cook components.
Transistors are very sensitive to temperature and with so short
connections they will be stressed thermally. Even though they don't
nececarily burn and they still work, thermal shock will reduce gain to
some extent - and we don't want that to happen. So before soldering
try to make sure that wires are clean (scrape the surface with the knife
if you have to) so that soldering is as fast as possible.
Capacitor in the circuit should be ceramic and have black tip.
Coil should use a bit thicker copper wire, not that tiny 0.2mm diameter
enameled wire you took from wallwart (remember "skin effect"). Try to use
0.6-0.8mm diameter (or even 1mm).
Once you have whole thing tuned the way you like it, you might want
to seal it with wax or similar so the LC circuit is mechaically more stable
and less sensitive to ambient air changes (temperature, moisture).
Add voltage regulator for the oscillator, decoupling caps etc.
And don't use that mickey mouse FM reciever from dollar store for testing.
Maybe problem you experience is not transmitter related.
My priority was always stability and then then range so 9V transmitters
with two transistrs would only occasionally break 300-400m range
without antena on transmitter (and this is in open). On the other hand,
they work quite nicelly (installed in pastic box because of missing antena,
all transmitted energy comes from coil, very little drift with temperature
and proximity of the hand or body).
For example unit sitting on the desk will drift so little after placed
in shirt pocket that no adjustment on receiver is required.
My advice is to use best judgement and experiment, build many and
keep the good ones for yourself (unfortunatelly they don't always come
out the same even if you have same PCBs and same batch of components).

Wish you luck...
 
What are some causes of frequency drift?
Temperature variation is a common cause of frequency drift when not compensated for. And it's not limited to the external environment. Even at room temperature, your circuit will self-heat to some extent (more so in a transmitter than a receiver, because of power levels), junction temperatures will vary, and this may be contributing to your problem.

One thing to help diagnose temperature-related drift, note if it seems to be slow and constant in one direction after powering up the unit. What is the nature of the drift you're seeing?

edit: just saw your previous thread, where you didn't think operating temp was a problem. Have you tried aiming a hairdryer (poor man's heat gun) at the circuit to see if the drift increases? A can of compressed air (held upside down) and a heat gun are low-cost and invaluable tools for checking temperature related problems down to a component level.
 
That was REEEEALY helpful. Thanx so very much. Here are a couple of things that may be the problem. If you could analyze them, maybe you'll have an idea what's wrong. Thanx again

When I tested my oscillation circuit's transistor, my meter said hfe was about 20. I thought it was supposed to be about 100 or so. I'm using 2n3904 transistor also.

i've built this same transmitter before, and it worked fine. I was just building a new one w/ 1/4 watt resistors instead of 1/2. I made a change to it though. I took an IC socket, and cut it so that there is just two sockets. The socket is in parralel w/ the cap of the LC circuit, so basically, by plugging in different caps, I get a huge range of frequencies. I disconnected this part of the circuit to see if it was the problem, and it wasn't.

I also made a smaller inductor so that I could get frequencies above the 108 MHZ limit.

My indcutor has about a 1/2" bit of wire connecting it to another part of the circuit. Nigel Goodwin said that this is alright. Second opinion???

My cap in the LC is the tunable kind, it goes from 0-4 PF.

Maybe I did fry the transistor. I used a 30 watt soldering iron, and kept it on the leads of the transistor for more than 3 seconds @ a time.

I haven't tried doing any temperature tests, but I highly doubt that that is it cuz a previous transmitter that I built that was almost exactly the same didn't have any problems wutsoever. Also, what laroche said makes me think that it is not temperature because it will drift within a matter of a couple of seconds.
 
The tuning capacitor value (4pF max) is very low, the L/C ratio bad.
Try with 4...20pF capacitor and smaller inductivty.
 
I made my circuit so that I could put in different capacitors and add to the 0-4 pf. I have tried caps like 18, 20, 30, etc. and they made no difference.
 
I disconnected the trimmer cap AND the ic socket thing, then soldered a regular capacitor in place of these two. Same thing. I even tried different value. When I do get it working, it seems that the frequency drifts, and then it gets to the point where wherever I turn the dial, I cant hear anything.

How long do you have to have a 30 watt soldering iron on a lead of a transistor in a t0-92 case before it ruins it?

Also, could a defective capacitor that connects to the emitter and collector of a transistor be the problem?
 
I think that I may have found the problem w/ my transmitter. If there is a wire that is within a hair's breadth of touching an inductor, will it mess up the LC circuit's oscillations? Cuz my pwr supply line actually touches the inductor but it is copper inside of rubber, or plastic, or whatever coats the wire, so it is not actually touching. Is this the problem??? Does it sound reasonable???
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top