First of all, congratulation, you just passed an
important milestone. You can use this experience to
improve your construction.
Transistor with the same part number could (and usually do)
have quite different parameters and especially gain.
That's why for some applications they have to be matched.
Transistor with poor gain can definitelly affect the
amplitude of oscillations and thus range. Gain
decreases with higher fequency in all transistors.
What makes one transistor "high frequency" is ability
to maintain reasonable gain levels as operating
frequency increases. Oscillator is simply amplifier
with positive loop. This means that part of the output
signal is returned to the input of the amplifier to
drive oscillations. If the gain is one, whole output
signal must be used just to keep the circuit oscillating.
As soon as you take part of that signal to do something
else (like drive antena) oscillations will dim to nirvana.
But if you have gain of say 4, only 25% of the output
will be needed for driving oscillator and ca 75% can be
sent to antena (well it's a bit more complicated but
I hope you get the idea).
Bottom line is the higher gain, the more of the signal
is left for use and less is wasted on maintaining oscillations.
Usually max frequency listed in the specs is the highest
frequency that transistor will be able to produce
oscillations - but that's about it. Do not expect
transistor rated at 100MHz to be great choice for
transmitter tuned in 98MHz. You need some safety margin
and for 88-108MHz you should not even consider
anything that cannot oscillate with at least 300MHz.
Except maybe as proof of concept (duh...).
To get better range you will have to experiment a bit.
Why not simply try different transistor? Even different
model. It is also not clear what setup is used in range
mesurement. If you are talking 30feet range through
28feet thick concrete wall filled with metal scraps from
WWII (like submarines, tanks and all other little chunks
of metal), that's not bad range at all...
In my experience it is possible to achive ca 1/4mile
or so in open area with NO antena on sender but with short
antena on receiver (ca 2feet). When sender is fitted
with piece of wire as antena, range is bigger off course.
Now don't get too excited because not every single sender
would achive same range. The 2N3904 is supposed to work at
100MHz. According to datasheet it should have gain of 300
at this frequency but remember that gain can be reduced
even for good part if you overheat it while soldering it to PCB.
You might consider real high frequency transistor for Q2
(like 2SCxxxx for Japaneese parts or BFxxx for European).
But enough of transistors, I noticed one more thing and that's
the 1k resistor in the emiter circuit of the Q2. In my
designs this was usually 220 Ohm or so and there was another
2.2k resistor from base of Q2 to gnd. Everything else is the
same. You might want to experiment a bit to get the optimum
setup for given transistor. Another option is to put
additional stage (transistor) and amplify RF signal a bit...