FM
Transmitters FM exciters FM Transmitter FM Exciter Frequency
Modulation transistor Radio VHF UHF kit PLL diy Panaxis Ramsey
Broadcast RF Circuit electronic schema amateur ham
My View on FM Transmitter Kits I
have made
Here will go a very biased view of Transmitter kits and RF Amplifier kits. It has to be biased as it will be loaded with my personal opinions on the quality of the kits I have personally built --
Documentation
Quality of Components
Quality of design
Cost
Support from vendor
A bit about me :
I fell in love with transmitters when I was 8 years old. My Papa took me to San Francisco on vacation in 1969 and one of his friends there gave me his handheld CB radio to play around with. So there I was, with this CB in my hand, being dragged around while Mama and Papa rushed around from one shopping mall to another. I started contacting various people and explaining them my view of America from a child's perspective. I knew then that I just had to investigate this technology further.
So by now I have built about 11 transmitters but only two from a kit, maybe 20 Audio Amplifiers from Elector / Philips, about 25 more Audio Amps/preamps/Equalisers of my own design and 4 Loudspeaker cabinets. Besides a whole bunch of Fuzz, Phasers, Tremolos, Sustains for the Guitar. I also wrote a program in Visual Basic for simulating Loudspeaker cabinets.
In other words, I have about as much experience as the average RF experimenter. I do not own any special equipment besides a Digital Multimeter and a VSWR meter and the SONY IC90 PRO Digital PLL Scanner. And once in a while, I can visit the local University that is stacked with Oscilloscopes and Gurus who really know their stuff.
Hence I think my personal experience in building various kits would be very representative of the experiences of the vast majority of kit builders who have average technical expertise and average kit building experience, with the most basic of test equipment.
( I am not affiliated with any Kit Manufacturer. I am not American or British either !!! ;)
08th August '98 with further comments added on 21st
August '98
My review of the Broadcast Warehouse 1 Watt PLL no-tune Exciter
with LCD
The bottom line first :
Well-designed kit. State of the art technology. Extreme ease of
operation due to no-tune design and push button frequency
settings. Highly recommended.
Scott Masters, Chief Designer at Broadcast Warehouse recently submitted me for review one of the first of his new 1 Watt PLL Exciter with LCD kits. I understand that this kit has been redone and improved since its earlier limited release.
For ease in reference, let's call this kit the "bw0001", as that’s what is written on the PCB. ( 31 October 98 : Broadcast Warehouse renamed this kit to KITLCD1W)
Shipping was smooth sailing :
I found that the staff at Broadcast Warehouse were very
good at following shipping instructions to assist me to get the
kit through my country's tough Customs Department. They sent the
PCB and Documentation in one parcel and all the electronic
components in another parcel. Upon my request, the Customs
declaration said "For Evaluation purposes, Not for
resale" to make it easier to get through Customs.
Seeing that Low power transmitting is considered illegal in most countries, I was pleased by Broadcast Warehouse' flexible approach in shipping, which can make all the difference between a budding transmiteer ever receiving his kit or not.
Upon receipt of the kit, it was nice to see the high quality components supplied with the kit. The PCB is a double-sided board made of Black Oxide. It has plated through holes (PTH) which make for better soldering. It has the component placements screen printed on the top, making it easy to build the kit.
The main board is
only about 1.5 inch by 5.5 inch. The compact design looks neat
while the ample ground plane and screened oscillator coil still
avoid RF interactions between components. But beginners do need
to have a sharp eye and a steady soldering hand as the solder
pads are quite close to each other and small as well at places.
It can also be hard to extricate/re-insert a soldered part due to
the Plated Through Holes, so you need to place components right
the first time.
The four MPSH10 RF transistors and the final stage 2N4427 transistor supplied were all made by Motorola.
The LCD is a module made by Hitachi. It has the driver ICs already soldered in at the back of the LCD. The LCD Module attaches on to the main board using plastic standoffs.
The Capacitors supplied are of good quality but the markings are so small and faint that you will need a magnifying glass to correctly identify them.
Documentation consists only of a single A4 sheet, thankfully printed on both sides. It does not contain the circuit diagram :( That and the tiny circuit board with closely spaced solder pads could make repairs difficult. Luckily Broadcast Warehouse does mention at their site that "All products carry a 12 month warranty". Broadcast Warehouse confirmed that this warranty means that they will replace any component that goes bad during construction or next 12 months, given that there are no obvious signs of abuse (reverse polarity power supply, running without an antenna etc). If a kit does not work on building it, Broadcast Warehouse will repair it for the standard kit construction charges. It's only due to that strong safety net that I can suggest building this advanced kit even to a beginner.
Big Brother is watching you :
After I spent half a day staring at the PCB to draw out
the circuit diagram, I realized that the bw0001 obviously comes
from the same designer as its older brother - the Broadcast
Warehouse PLL exciter without the LCD. The basic building blocks
are the same, but that’s where the similarity ends. The
bw0001 is quite a different beast as it had different design
goals.
Broadcast Warehouse' ultimate vision and design goal for the new no-tune Exciter is obvious : It is meant to drive a no-tune RF Amplifier and Antenna that is broadband across the FM band. That way small Broadcasters need not be bothered by all types of setup procedures, meters and technology, they just can get their job done in the easiest possible way. Setup at different location for different purposes can be as easy as pushing the frequency up/down buttons, with no major adverse effects of changes in modulation level or power output across the entire FM band.
Also, now Broadcast Warehouse do not need to stock different transmitters tuned to different frequencies. They can ship the same unit to any customer worldwide and the customer can set up to his desired frequency in seconds. That should reduce costs for everybody.
Broadcast Warehouse have certainly proven they have the necessary technology. They are already selling no-tune 20, 100 and even 300 Watt Transmitters using the same no-tune PLL Exciter stage. A no-tune Broadband RF Amplifer has not too many compromises over the tuned variety, so it's ease-of-use should make it more popular in the future.
Their site even offers broadband Jaybeam folded dipole antennae that need no tuning across the 88-108 MHz band. (Yes, I realise that life is a compromise, and you loose something to gain something) (in this case, its loose gain ? ;)
In case the VSWR is unacceptably high due to an Antenna mismatch at a particular frequency, Broadcast Warehouse' other higher powered Transmitters automatically fold back the supply, protecting the output stages.
Like the earlier kit, the PLL is based on the ubiquitous Motorola MC145170 PLL frequency synthesizer IC along with a Microchip Technology PIC16F84 8-Bit CMOS Flash/EEPROM Microcontroller. The new chip on the block is a National Semiconductors' LF351 high speed JFET input Op-amp IC used as an active loop filter. Supply voltage to these chips is regulated by a 78L05 Voltage Regulator chip.
These ICs are supplied in the same polythene bag as the rest of the components, without the benefit of an antistatic bag.
The PIC has an IC socket, presumably allowing you to easily change it in the case of Broadcast Warehouse releasing new firmware. In which case, it would be a nice touch if the firmware release number was written on the IC for easy identification. The other chips are supplied without IC sockets as that increases lead length and with it, the chance to pick up or radiate stray RF signals. (Hobbyists : Don't fry those chips !!)
Comparison with the earlier kit highlights the differences in technology :
#1. As the bw0001 was designed to be a "no-tune" exciter, it does not have the manually tunable oscillator with its variable inductor as the earlier kit. Everyone loved to hate the variable inductor as hobbyists have the habit of breaking the core. You will also notice from other reviews on the Net that it can be quite difficult to tune the older design.
#2. Being a no-tune broadband design, the bw0001 obviously does not need the power LED circuitry that helped tune the last stages of the earlier kit. Almost constant power output within a 10% range is claimed across the FM band. ( See Marconi's review which shows power versus frequency charts)
#3. As the frequency locked condition is shown on bw0001's LCD panel, it does not need the lock LED and associated circuitry as earlier. Also, frequency selection is now via up/down push switches on the bw0001, which is more convenient than accessing and setting the DIL switches in the earlier design. You can also easily put a new set of up/down switches out on the front panel, just by connecting them in parallel to the original PCB mounted switches.
Automatic no-lock power-down feature :
The bw001 has an interesting circuitry to power up the
third MPSH10 RF stage only when the frequency is locked in. This
is a simple yet very effective idea not seen on other kits. To
explain it very simply without letting the cat completely out of
the bag: it is the same idea as a Lock LED. The PIC chip has a
port that sinks a few mA that can be used to light up a LED when
locked. But if you used the same PIC chip port to sink the
emitter current of one of the RF stages, turning it on only when
the frequency is locked, you have yourself a rather useful
feature that no other kit manufacturer has incorporated so far.
This automatic no-lock power-down feature is even more important in a no-tune design. On conventional systems, to change frequency, you would power down yourself and change switch settings and retune coils and and and and. But on a no-tune system, you have the possibility to change frequency in little steps by just pressing the frequency up/down buttons. With an Automatic no-lock power-down circuit, in theory, the transmitter will not transmit a signal at all the intermediate frequency steps while you reach the new desired frequency of operation.
#4. The new bw0001 has only one BB909 Varicap diode used to receive the PLL feedback voltage as well as the Audio Modulation. The earlier kit has two Varicaps, which many gurus actually consider better.
#5. The bw0001 has an IC used as an active loop filter while the earlier design used a passive loop filter. This is used to convert the pulses from the Phase/frequency detect pin from the MC145170 Synth chip into a DC Voltage. This DC error voltage controls the capacitance of the Varicap diode, thereby correcting the output frequency of the Exciter.
There are always small leakages of the loop control frequency into this DC error voltage. These signals start modulating the FM carrier, leading to spurious spikes around the carrier frequency. An active loop filter does a better job of converting the PD pin pulses into pure DC, leading to less spurii.
Also, as the Active Loop Filter can have gain, its output voltage can swing very close to the supply power supply, even though its input signal would be, at the maximum , the 5 Volt pulse signal it gets from the Synth.
On the other hand, a passive loop filter can only output, at the maximum, the same 5V level feeding the Synth chip.
If you design a VCO that has to tune from 88-108 MHz using the passive loop filter's maximum voltage swing from only 1-5 volts, it has to be more sensitive and gets harder to tune + shows more effects of spurii. The Active loop's voltage swing of 1-15 Volts makes the VCO less sensitive to spurii in the DC error voltage, leading to a better final RF signal.
#6. The bw0001 has an option for onboard Pre-emphasis to either European or USA standards. Built-in pre-emphasis can be a useful feature for beginners who directly feed the bw0001 with Mono signals from their CD players. If you are more professional than that, you would have options for pre-emphasis earlier in the audio chain, maybe in the Limiter or Stereo Coder. It always makes sense to pre-emphasise as early in the chain as possible as that would increase the signal to noise ratio at the high end of the Audio spectrum. In any case, you cannot pre-emphasise a stereo composite signal, you need to pre-emphasise before audio signals enter the encoding section.
If you do have other options to pre-emphasize or want to broadcast in stereo, you need to disable the pre-emphasis on the bw001. The current documentation is a bit fuzzy on how to achieve this. You need to omit/remove Capacitor C19 to disable pre-emphasis, not C11 as also mentioned in the current documentation.
#7. The bw0001 is harder to solder than the earlier kit, but how hard can soldering get ? On the other hand, the ease of setup to guaranteed performance levels without tuning is a great plus point for most Pirates as they do not own a Spectral Analyser.
To increase the thrill level in making the kit, one has also to wind a small toroid shaped ferrite core with 4 turns of wire supplied with the kit. This "4 to 1 transformer" is used as an impedance match between stage 3 driven by the last MPSH10 and Stage 4 based on the 2N4427. The core is held on to the board only by the wires coming of it. You have to be a bit careful that you do not bend the core after it is soldered on, as that would stress the wires holding it down.
The kit took me about three and a half hours to assemble, but I am a slow worker. It worked the first time around.
Setting the operating frequency
Upon startup, the unit displays Broadcast Warehouse's
web address for 7 seconds. It then displays the last set
frequency and says "Please wait" while the PLL tries to
lock into the displayed frequency. The unit takes about 20-25
seconds to lock the first time, which is a fair enough
compromise. When the PLL loop is locked, you guessed it, it says
"< LOCKED >".
If you want to change the frequency of operation, you have to press the up/down PCB push switches. Each press changes the frequency in steps of 0.1 MHz ie 100 KHz. SW1 which is physically higher up REDUCES the frequency while SW2 which is vertically below INCREASES the frequency, in slight contradiction to intuition. This should be changed to work as one would expect. (Scott has already incorporated my request in version 2 of the Firmware dated 17th August 1998.)
You can also change the requested frequency immediately after the unit stops displaying the web address, before the PLL locks the first time.
Like with most other electronics, keeping the push buttons pressed makes the unit go into fast forward or rapid rewind mode. This mode is so fast that I kept on over-shooting the frequency that I wanted to set. It can span the whole FM band in 3 seconds flat. Maybe this fast scanning speed should be reduced so that it takes 10 seconds or so to go across the band. (Scott has already incorporated my request in version 2 of the Firmware dated 17th August 1998.)
After initial lock, if you increase or decrease the frequency by 3-4 steps, loop lock is a achieved almost instantaneously, even for intermediate steps on the way. It would have been better to have the RF power forcibily kept down for a minimum of 5 seconds after release of any push switch. This would make sure that the unit only puts out full RF power when the user has well and trully made up his mind and is confident that he has set the right frequency. Actually, RF Power down during the entire tuning process is a requirement of the European Telecommunication Standards for FM Broadcasts. (According to Scott, this ergonomic point should be fixed soon.)
Where Less is Better:
The bw0001 has impressive claimed specifications for
Harmonic emission, at -60dB. This is mainly due to the 6
component LC circuit that forms the impedance match/filter at the
output of the last stage. The two inductors are pre-tuned and
need no tweaking.
As the unit is a broadband unit, the RF stages are not tuned to pass only a desired frequency. They can easily amplify all the harmonics generated. So for a broadband unit, a final multi-stage low pass filter at the output is mandatory.
Having an integral RF filter onboard is great for those who will connect the kit directly to an antenna. Indeed with a claimed output power of greater than 800mW across the band, it would be the ideal kit for many small applications. You need nothing else to broadcast except a sound source, an Antenna and good humour.
But for those of us requiring more RF power, we would connect to an RF Amplifier that in all probability is running in tuned Class C mode. This kind of RF Amp almost totally rejects the harmonics at the input, but generates its own set of harmonics at the output. Which means that in case you run an Exciter into a Class C RF amplifier, the harmonic content of the Exciter hardly makes a difference to the final output. You would still need another multi-element filter at the end of the final stage anyway, only this time the filter has to be sturdier to be able to handle the extra power being handled.
The one main feature of an Exciter that does show through the other end of a Class C RF amplifier is the side bands and spurs on either side of the fundamental carrier. These are so close to the fundamental that they are close to impossible to filter out. The bw0001 shines out in this area as well, claiming spurious emissions of only -75dBc, making it one of the best speced/best featured Exciter kits available on the market today. According to the European Telecommunications Standards, all FM transmitters with an output of 10W to 1KW have to have spurious emissions of less than -75dBc, so this Exciter kit is more than adequate for driving professional Amplifiers upto 1KW of RF Power.
Mystery Component's role not crystal clear :
There is a mysterious component whose role is not
described in the documentation - a variable capacitor VC1. This
can be used by professionals for fine tuning the
crystal's frequency. Without fine tuning, the final RF
output can possibly be +/- 2.5 KHz from the displayed frequency
on the LCD. This is a very small error that needs to be
considered only by serious professionals. If you do not have a
frequency counter, the best bet is to leave the variable
capacitor at the half way position.
These are a few of my favourite mods….
A) There are three unmarked holes on the board quite close to the power supply entry point. Even though the documentation does not mention it, these holes can very conveniently be used to plug in an onboard 7812 or 7815 Voltage regulator. This will regulate supply to the whole board except the last RF power stage. It is recommended that a stabilised 12-15V 500mA source feed the Exciter, so in case your power supply is not regulated, you can add that 7812/7815 on to the board. Remember to cut the copper tracks at the point marked by a dash, else you will be bypassing the regulator.
This mod is also useful if you want to replace the final 2N4427 with a higher power, higher voltage device while yet feeding the rest of the board with 12-15V.
B) The bw0001's documentation issues a serious warning about reverse polarity power supply damaging the unit. I assume that an onboard 78XX Regulator will also protect the circuit it supplies against reverse voltage. But if you do not plug in the 78XX, the same holes can seat a normally reverse biased diode. In case of reverse polarity supply being connected to the unit, this diode will start conducting and will blow the fuse, unless it decides to blow first to protect the fuse.
C) You might also like to increase the value/voltage rating of the ripple smoothening capacitor C20 to the biggest value you can fit on the board. At only 33uF, it pre-supposes a good regulated supply. With 16V ratings, it is too close to the 15V supply for my liking.
Wish list :
A) Circuit Diagram to be included with the kit, basic block dia published on the Net.
(Scott has said on the Net that he will fax a hand drawn circuit diagram to anybody who asks)B) More documentation : Web link reference list for further info on the components, Some details of broadband antenna suitable for use with the kit. Test points and Voltage levels.
C) While setting frequency, RF power should come up only after 5 seconds of release of push button.
D) Availability of a 5-20 W broadband no-tune RF Amplifier kit with VSWR foldback circuitry for use with this Exciter, as an upgrade path.
E) Availability of the 20W no-tune PLL LCD Exciter in kit form for those directly requiring higher power
F) Broadcast Warehouse could offer a professional chassis to house the unit, with predrilled holes for the LCD and freq up/down buttons. It should be big enough to house the Exciter and a small RF Amp along with the necessary power supply.
G) Cost-effective way to read out the modulation level.
H) Any way to monitor VSWR of final stage of external RF Amp and use existing RF Power down mechanism to cut RF Power when the VSWR at final stage is bad ? Gurus out there, get thinking !!!
Wanna have your own page here, write
about your own experiences, own review, refute existing review,
add comments to existing review, publish your circuits ?
Contact
me
My review of the WaveMach
FMS2 Stereo PLL Exciter 20 Dec 1998
My review of the Mighty MAX-1 FM
Exciter 02 May1999
Page started on 18th May 1998. We have had page
loads since counter reset on 15 May 2000.