From : "Kit Guy" Ramsey FM25B Review by FM Guy: I just completed one of the latest FM25B kits and wanted to share my test results and comments. I have the following observations both good and bad: It's reasonably easy to assemble. I'm a skilled electronics tech and completed the main PC board in a bit over an hour. It takes another few minutes to put it all together in the case for about 1.5 hours total. Ramsey is really conservative in their time estimate of 4 hours for advanced kit builders. All parts were included along with two extra caps for European pre-emphasis. It worked first try. The instructions are OK. I have a few gripes. They don't put the part values on the PCB (there's not a lot of room), just the reference designators. That wouldn't be so bad but they also don't give you a parts list sorted by reference designator. So either way you approach the assembly, you have to either hunt for the reference designator on the PCB (i.e. R7) or you find R7 on the PCB and have to hunt it down in the parts list to find out what value it is. Either method adds a lot of assembly time as you're always searching for the parts. At least they include a schematic which is more than some other kit companies do. They have you install parts in a somewhat unusual order. Normally kits have you install the stuff lowest to the board first so you can flip the board over and use a piece of foam or similar to hold the parts flush to the board for soldering. Ramsey doesn't do this, in fact, they have you install some of the largest parts first which is a bit awkward but it still works out OK. The surface mount (SMT) parts are pre-installed which is a big help. They have you install the PIC microcontroller socket very early in the instructions and have you immediately insert the PIC into the socket. This is a bad idea. First of all, it could be more easily damaged by ESD (static) that way. Second, it's wise to assemble the kit, power it up, and verify the 5 volt power supply is working correctly, there are no shorts, etc., BEFORE putting the PIC in the socket. Otherwise, if you had a stray short, or say put the 5 volt regulator in backwards (or it was defective or the wrong part) you could easily apply ~12 volts to the PIC and destroy it. It's a programmed part only available from Ramsey so this would be especially frustrating. Once it's assembled, you have to adjust a coil for 9 volts DC at a test point with the frequency set to 107.9Mhz. A DMM is necessary. At the correct voltage, the lock LED was flashing randomly while the manual says it should be on solid. It turns out it sometimes likes to flash at ANY coil setting at the highest frequencies. Sometimes it settles down and will show a solid lock with no audio input (it's supposed to flicker as a modulation indicator), sometimes it won't. Small adjustments in the coil value don't seem to affect it much. The random flashing, as it turns out, doesn't seem to affect the performance and it stops at lower frequencies. The above adjustment also takes a while to settle down as the PLL does its thing and the manual doesn't mention this. If you quickly adjust it to 9 volts, and go back and check it 20 seconds later, it will be way off. As kits go, most folks should be able to put the FM25B together and have it work if they have basic soldering skills, can use a DMM, etc. If you're not a skilled kit builder, just take your time, follow the step-by-step instructions to the letter and double or triple check everything before you power it up. If you're worried about your soldering skills, don't have a DMM, etc, then buy the assembled version. Now, onto testing the FM25B... I checked the frequency on a decent counter. A 90.0 Mhz setting was displayed as 90.00034 Mhz which is plenty accurate. I then checked the output at 90.0 Mhz on a spectrum analyzer for spurious content. The 2nd harmonic (180 Mhz) was the worst at -48db. This is not as good as some of the more expensive transmitters but at this power level, it's not likely to cause much trouble either. The general spurious garbage was typically -60db or so down from the carrier which is decent enough. The RF output with 15 volts applied to the DC input jack, and the RF level turned all the way up, was 900 mV RMS at both the upper and lower frequency extremes. This is 18 mW into 50 ohms --a bit short of the 25 mW claimed by Ramsey but not likely to be noticed. The FM25B uses a "capacitive multiplier" to feed the output stage and it drops about 2.5 volts. So keep in mind 14.5 volts is required at the DC input to get 12 volts to the output stage. The entire FM25B draws about 100 mA so the unregulated 500 mA 12 volt adapter supplies about 15 volts. If you run it from a true 12 volts (say a battery or regulated supply) the output power drops to 16 mW. Below 10 or 11 volts it will lose lock at higher frequencies and power drops off rapidly. This is normal in this sort of design. It should still work quite well running from a 12 volt battery. Unlike some of the more expensive transmitters, this one doesn't shut off the RF output while it's seeking lock. So every time you power it up, or change the frequency while it's on, it sweeps across the FM band for a few seconds until it finds lock. You can hear birdies, blips, etc. on other FM stations as it does this. Not a big deal at this power level, but something to be aware of. The good news is it typically only takes a few seconds to lock upon power up. Ok, so enough technical stuff, how does it sound? Well... I guess that depends on how fussy you are. As others have reported, there's some buzz and hum that's audible during quieter passages and between songs. It seems to be due to RF leakage back into the circuitry. Moving the FM25B around, moving the audio input cable, grounding the transmitter, etc. change the level of buzz or even almost eliminate it. If you put it in a shielded case, or use a remote matched antenna (and remove the whip), I suspect it will be much less of a problem. I may try spraying the inside of the case with conductive paint for RF shielding. Others have put theirs in a metal case. The FM25B sounds reasonably good but far from great. It won't put your favorite FM station to shame as Ramsey claims unless your favorite station is a hundred miles away. The distortion steadily rises with increasing deviation (audio level) and at low audio levels the signal gets buried in assorted forms of noise. So you're between a rock and hard place when setting levels. Without using a compressor/limiter, it's hard to get anywhere close to the same overall volume as you get from commercial FM stations unless you crank up the audio level to where there's lots of distortion on the louder stuff. A compressor would help quite a bit here at the expense of compressing your audio just like commercial stations do and, depending on the compressor/limiter, some audible compression artifacts. Ramsey sells the STC1 compressor but I haven't tried it. Overall, it sounds like a somewhat weak FM station even when the receiver is right next to the FM25B. You get some "fuzz" and odd stereo artifacts at times. If you overdrive the audio input, it causes most receivers to drop in and out of stereo in step with the music (it's also well into audible distortion while it's doing this). It has a jumper to force it into mono. This requires power cycling the transmitter. In mono, as expected, it sounds better but still not great. If you're going to use a mono receiver, or are having signal quality problems, this is a useful option. The frequency response seems quite good. I haven't measured it yet (I might later, along with actual THD and S/N ratios). Likewise, the stereo separation is also OK. For an all-in-one FM chip, the BH1415F used in the FM25B does a decent job. Finally, some of you are probably wondering about the range. Using the built-in extendable whip antenna at the maximum RF power setting (18 mW) it manages to cover a good sized house and the yard around it with decent signal strength. It depends on what stations are nearby. If you live where the FM dial is crowded with stations your range is going to be less--especially with receivers that have a hard time ignoring stronger stations and receiving a weak one cleanly. If you have a whole 1 mhz area that's just dead hiss on your FM dial, say nothing from 90.0 to 91.0, and you set the FM25B to the middle (i.e. 90.5 Mhz) you should get pretty good results. As Ramsey suggests, the best way to increase the range is to use a better antenna. A simple tuned homemade dipole should do wonders for the range. The one thing to keep in mind, however, is you can't use an SWR meter with only 18 mW of power (most meters require at least 1 watt or so). So you have to use the formulas and hope your antenna is tuned, or perhaps borrow another transmitter or an amplifier to tune the antenna. Some of the above is being relatively critical. For say background music or use in a car, I think some people would be fairly happy with the FM25B. It's miles ahead of the little cheap MP3 FM adapters I've heard, non-PLL designs or even the old FM25. If you're expecting CD quality audio to pop out of your FM receiver on a high quality audio system, however, you're going to be disappointed. FM is far from CD quality even with a high-end transmitter and the FM25B is far from being a high-end transmitter. Overall, the FM25B is probably a decent value (the kit is currently on special for $120) if you're not too fussy. I'm not aware of anything that's significantly better that's not also significantly more expensive if you include everything you get with the FM25B. Most of the better transmitters are only mono exciters and require a stereo encoder. They also usually don't some with a case, power supply and antenna. The Broadcast Warehouse stereo 1 watt transmitter/encoder/limiter is a better all-in-one solution, but it's $500 without a case or power supply.