![[a real schematic]](/filipg/LINK/70mw.gif)
[GET]. If you don't have
graphics:
Latest version from ftp://netacsys.com/pub/web/mycal/fmfaq.html
This file is Copyright (C) 1993, 1994, 1995, 1996 by Mycal <mycal@netacsys.com>
I grant free copying rights via BBS's, USENET, Internet FTP sites. Distribution via printed media or CDROM requires the author's prior permission.
Once more If you have any info, I stress **ANY*** about this subject please drop me a byte or two.
Have fun, mycal@netacsys.com
The information contained in the file is in no way complete, nor do I take any responsibility for its accuracy. With that in mind, along with the above paragraph I must say:
THIS FILE FOR INFORMATIONAL PURPOSES ONLY
THIS FILE FOR INFORMATIONAL PURPOSES ONLY
Also I am working on a some Postscript files that will contain instructions on how to build some equipment that is to complex for ASCII art. You might want to check the FTPable archives every month or so.
I would like to say thanx to all the people that contributed to the information in this file. The list has grown quite long, and some of the contributors would like to remain anonymous. So for now I am going to forgo all the names, if this is not up to your liking please let me know.
People can FTP into dg-rtp.dg.com with user "anonymous" and password "<your e-mail address>". In the fm10 directory you'll currently find three sub-directories and a file:
The price for this packet is $10 in the US and $12 overseas.
Address to send the $$ to is:
Note: this packet is in a constant state of change, more info could be added at any time.
Other places to scan, sci.electronics and the amateur radio groups.
Free Radio Berkeley now has an archive on CRL. It can be accessed by FTP'ing to site ftp.crl.com and looking in the directory /usrs/ro/frbspd
To use it, send a mail messages with the word:
help
in the body of the message to:
This will give you all the information you need to use this service.
A NOTE on power meters. I recommeded the above power meters because they are inexpensive and most people that are interested in hacking a FM-10 have very limited funds. These are by no means accurate, but they will give you some idea what is going on. If you can spend the money you can get an accurate power meter that is designed for this band, but the cost is 10-15 times more.
Basically you want to create a non-inductive 50-ohm load. This can be done with regular carbon resistors, or by buying pre built Amateur or CB radio loads. For low power ( <.5 watt ) an ethernet terminator works well (check temp when using if it gets very hot lower input power, if it is still cool you may be able to go up to .75 watt)
Most CB loads use a 2-watt carbon 50-ohm resistor.
You can build your own, as wimpy or as studly as you want by running resistors in parallel to create 50-ohms. ie, 2 100 ohm 1/4 watt resistors will create a 1/2 watt 50 ohm load...
Do not use 50 ohm wire wound resistors, they are not 50 ohms at radio frequencies.
-----
---------\ Sample Dummy Load where -****- = 50 ohm
---|--****-- / mounted in UHF connector. carbon
--------- resistor
-----
--probe tip-----||-+---/\/\/\-------to center of 50 ohm coax.
.01uf | 4.7M +------to braid
___ |
\_/ |
--- diode |
| |
cliplead for gnd---+-------------+
Anyway, the diode should ideally be
a Schottky diode (low rf capacitance). Although a 1n914 will work.
To use, just hook up to your digital Meter, set on DC voltage.
You will get very close to RMS RF Voltage. (this probe was specified for a
10Meg Ohm meter).To calculate power into a KNOWN purely resistive load (a.k.a. a dummy) use:
This is also useful for checking inputs and outputs of low power RF units, since the inexpensive power meters don't seem to do real well below 1watt.
So I dusted off the old stacks of Radio-Electronics and found two articles that may be of intrest.
In the January 1993 issue they have a "build your own digital voice changer" using a simple Real time digital signal processor. I think that this design is very simalar to the voice changing telephones. It basically raises or lowers your voice pitch. A place called LNS Technologies @ 1-800 -886-7150 sells the kits for $59.
In there September 1992 issues they have a "build this dsp voice-effects board" using a little more complex, programmable, real time digital signal processor. The software they include contains a harmonizer, echo, reverb, and pitch. The kit is sold by American Disributors Inc for $105 @ 1-800-877-0510. You can also write your won software but the programmer is several hundred $$.
DC electronics has a Robot Voice Kit for $15. I don't know how well this works or what it sounds like, but it clames to be adjustable for many different effects.
If you want an updated,
[GET]. If you don't have
graphics:
+12v
|
/
\ R1 <strong>*</strong>220 ohms(1/2 watt)
/
\
|
=== <strong>**</strong> F2
|
R2 9k | C2
+-/\/\/\/---+-------||-----> output
| /
<strong>**</strong> F1 === C /
| | /
C1 | B |/ <----------MPS2222A (276-2009)
in --||--+---|\ E -or- 2N4401 (better)
^ | -> -+ -or- MPS918 (best)
| |
| GND
currently on board
* you can also use 2 440 ohm 1/4 watt resistors run in parallel** Optional but recommended, Ferrite beads, help prevent oscillation, and help the amplifier produce more gain. Slip over one end of the resistor.
I built this thing right on the underside of the FM-10 kit, C1 is the cap that currently goes to the RCA ant jack, the 9k and the 220 ohm resistor have to be bought, note that if you cannot find 220 ohms you can make one by using 2 440 ohm resistors in parallel, and that a 10k will work in place of the 9k but yields poorer performance (-5%).
The MPS2222A is from Radio Shack part number 276-2009, use this part! if you substitute it for a 2N2222A you will get only half the gain. Be very careful to get the leads in the correct orientation!
I have found that a 2N4401 can be used in place of the MPS2222A with a little better performance, about 5mw more. I think the 2N4401 can be found at Radio Shack too. The MPS918 can produce up to 150mw.
C2 is of the same value of C1, I took the one that goes to the on board antenna pad.
Important! the value for R1 that seems to be optimal is 220 ohms, but it is very close to the sat point, If the amp. seems noisy (interferes with the TV etc.) back this value off to 240 ohms. If you lower this value below 205 ohms the power meter may read higher power but this will not be true, the transistor will be spewing all kinds of junk and the power meter will mistake this for higher output (in reality the signal we want will drop considerably)
Well that's it, effective range with a good antenna should be a little over double.
Tuning the amp for proper operation on 3-meters can be tricky. I suggest the following setup:
FM xmitter -> SWR Meter -> PA-1 -> Power Meter -> 50 ohm dummy load
Starting with C1 and C3 about 1/2 turn from closed, tune C2 and then C4 for maximum power output. If the SWR is much over 2:1, you will need to adjust C1, C2, and C4 to reduce it to an acceptable level. Watch the output power while you do this. Sometimes, tuning the trimmers for minimum SWR will peak the power output, whereas other times the power output will drop markedly. Avoid trimming the capacitors that have a large negative effect on the power output. After several iterations of adjustment, you should have decent power gain and low input SWR.
Note: Watch for sudden jumps in power output that you can't linearly tune through. If you encounter this happening, chances are your SWR between the transmitter and amplifier will go way up, too. This is a sign that your amplifier is oscillating and you will need to tune it out of this region for proper operation.
use the formulas out of your FM-10 manual 234/freq=length of rod.
Also If you have an SWR meter you can cut the rods a little longer and start clipping the ends off a little until you get the best SWR reading.
|
| | | = brazing rod
| | |
| |
----- ----- ----- = UHF connector
| - | / - \ -
| | / \
| | / \
The final antenna should look like this:
| ^
| |
// \\ sky ground --horizon--
|
v
Try not to have anything metal near the radiator, this will effect the radiation pattern. The radiation pattern should look a lot like a doughnut surrounding the radiator, though deformed a bit.
I have been told that you can shorten the radiator and make the ground radials longer to lower your radiation angle, but I haven't tried this, nor do I know what this would do to the antenna impedance.
I've tried this antenna and it works great! It is better than my di-pole at home and you can drive to a high, optimal location for your broadcasts. Also with this setup you need very little coax cable. Line loss using RU-58u can be killer @ 100MHz.
You could also try a 5/8 wave length antenna, this would give you 2+db gain, or almost 2x power gain on transmit.
When you amplify a signal, you get unwanted byproducts these are called harmonics. The show up at multiples of your starting frequency. For example if you amplify a 50MHz signal you may get echo's on 100MHz, 150MHz 200MHz, 250MHz... If you interfear with your neighbors TV, the local fire department, or anyone else, you are just asking for trouble. If you are only on the FM Band, you will hardly be noticed.
Many FM transmitters are designed without adequate filtering on the output. Add to this the fact that most pirate radio operators lack spectrum analyzers with which to tune up their equipment, and it is easy to see that many stations are likely to be full of spurious outputs. You're a lot more likely to attract the attention of the FCC if you wind up interfering with all your neighbors television sets because of spurious emissions than you are if you emit a clean signal. While more elaborate 5- and 7-element filters offer greater attenuation of harmonics, I have found the following design to be simple and very useful. Furthermore, several of these filters may be chained together in series to produce even better filtering [**]
If you make L1 with the 18 gauge wire as indicated and use capacitors rated at 50 volts, the power handling capability of this filter will be around 20 watts. If you use 100 volt capacitors, it should be able to handle 50 watts or more. The insertion loss of the filter is generally less than 1dB if you stay below the F(0db) point.
Here is a ![[schematic]](/filipg/LINK/lowpass1.gif)
[GET]
for a 3-element lowpass filter:
input o---+---uuu---+---o output
| L1 |
C1- -C2
- -
| |
gnd o---+---------+---o gnd
The following table summarizes 5 different variations of the 3-element
lowpass filter. I have constructed and tested some of these myself and
find that they work as advertised. The attenuation figures shown below
were produced by spice.
3-element Lowpass with ~50 ohms input/output impedance
======================================================
C1 C2 L1 F(0db) F(3db) F(6db) F(10db) F(20db) --------------------------------------------------------------- 33pF 33pF Note1 113MHZ 138MHZ 163MHZ 200MHZ 320MHZ 39pF 39pF Note2 98MHZ 119MHZ 140MHZ 173MHZ 293MHZ 47pF 47pF Note2 95MHZ 112MHZ 130MHZ 160MHZ 270MHZ 52pF 52pF Note2 92MHZ 107MHZ 125MHZ 152MHZ 255MHZ 56pF 56pF Note2 90MHZ 105MHZ 121MHZ 147MHZ 246MHZ 62pF 62pF Note2 87MHZ 101MHZ 116MHZ 140MHZ 235MHZNotes:
[**] You can make a 5 element filter using the above parameters by using the following configuration:
input o---+---uuu---+---uuu---+---o output
| L1 | L2 |
C1- -C2 - C3
- - -
| | |
gnd o---+---------+---------+---o gnd
Were C1=C3, C2 = (C1 * 2), L1 = L2, and the C1, L1 values are taken from
the above chart.[Also of interest is that the FM-10 puts out about 8-9mw and the 2nd harmonic is -25db off the fundamental (frequency we are broadcasting on). The FM-4 Kit by Ramsey puts out 130mw and the 2nd harmonic is only -12db off the fundamental, which means the 2nd harmonic of the FM-4 is about as powerful as the FM-10. db is log10, ie 3db is 2 times 6db is 4 times...]
On a 2nd note: I replaced c7 with a 68pF cap and found it much easyer to tune a rock solid 19KHz at the test point.
old set up new setup
c8 c1 xtl where c1=10pF and xtl=38KHz
|-||-| |-||-|\|-|
| c7 | | | v8=var cap
|-||-| | | c7=cap
| | | |
Remove C7 and C8, replace with 38KHz crystal and 10pF cap. Note that
the 10pF cap and the crystal are running series and the old cap setup
is running in parallel.Note: If you have an FM-10 you can call Ramsey and get this mod for free!
A much better pre-emphasis/input circuit is shown in the July 1992 issue of "Radio Electronics". Not only do they use 75K ohm resisters in their pre-emphasis, but they filter stray RF signals by inserting a .001 cap between pin 1 (of the BA1404) and ground, and pin 18 and ground.
It has been noted that the above mod may actually cause distortion on cheaper stereo receivers, since they were mass produced for the world market, they were designed for the European audio standard, which Japan and other Asian nations use too. Try it out, let me know what works for you.
Note: the FM-10a kit comes with both sets of parts so you can select European or American audio standard.
The FM-10 was designed to be inexpensive and cost-saving measures with components are inevitable. Disc ceramic capacitors are less expensive than silver-mica caps, and also much less stable. Simply replace c16 with a silver-mica, tantalum or negative temperature compensated disc cap of the same value. Even better is to cut the leads of the capacitor as short as possible and mount it directly on the bottom of the circut board.
Mouser Electronics Mouser Electronics Mouser Electronics
11433 Woodside Ave. 2401 Highway 287 North 12 Emery Ave.
Santee CA 92071 Mansfield TX 76063 Randolf NJ 07869
Mouser Electronics
370 Tomkins Court
Gilroy CA 95020
Catalog Subscriptions: (800) 992-9943 (Continental US only)
Sales & Service: (800-34-MOUSER
(800-346-6873) (US, Puerto R., Canada)
(817-483-9384) Fax
Giant Catalog! 239pages of parts!
Just about everything.
No min order for north america.
$100 min for overseas.
Panaxis Productions makes some very high quality FM transmitters. The last word in Transmitting, tons of kits.
(Right next to my old place of study Chico State!)
A little taste of there catalog:
A company called Progressive Concepts sells plans for a 88MHz to 108MHz amp. The power curves show that 12mw in will yield 2.5 watts, but can be driven harder for up to 12 watts. (I have not seen these plans)
Plans only in U.S., $16 (a bit spendy, ouch!)
The MRF239 can be used as direct replacement for the Ramsey 2 meter PA-1 kit. Cost is around $14 bucks.
Newark also has the 38KHz crystals for $2.90 ( I don't know Newark's address, this was sent to me in the mail, will try to find it though.)