good radio, bad radio

Dunno, but I got a philips brand unit - FM with digital scanning. Background hiss is unbearably loud - louder than the program in many instances. Post purchase googling found many reviews stating the same.

SWMBO has a fairly old Roberts small portable (jogable?) which I've repaired several times - the soldering on the earphone socket fatigues. I've tried several others, but nothing comes close in terms of sensitivity and noise on FM.

I had a Philips jogable, the FM was crap but the long wave was better than any other I've tried. I gave it to a cricket fan so he could listen to the Test Match at work on BBC R4 198kHz.

But yes, you'd think they'd all be the same.

Cheers

--
Syd

--

Sort of along the same lines... Several of my friends have brought their new weather alert radios to me for setup and programming. they all seem to receive the NOAA station(s) in the area, but the AM and FM bands are practically useless. They can't recieve anything except the most powerful FM stations, and fewer of the AM stations. Granted, AM is on the wane in my area (the most powerful being about 5KW), but all my other radios can recieve them pretty well. FM reception on other radios is great. Has anyone else noticed lack of sensitivity on the broadcast bands on these weather alert radios? I bought one of them last year, na d returned it because of this problem. I bought another (different brand) a couple of months later... same problem. I returned it as well. Are there any weather radios worth having? I realize the main features of these radios are the weather alerts and NOAA broadcasts, but in an emergency, most of the information you need for survival will come from broadcast radio stations, not NOAA.

Dave M

No? Any RF product that is directly cloned from a reference design is almost guaranteed to have fit, compliance, performance, and cost problems. Very few reference designs are ready for production.

Your problem, sir?

Your unspecified model receivers may or may not have had an all digital single chip receiver, probably made by Silicon Labs. For example: When the chip has an A/D converter in the receiver front end, the RF amplifier needs to have an AGC (automagic gain control) in order to for the A/D to use all the available input resolution in both AM and FM. The result is usually a compromise RF amplifier, with some variations in sensitivity. These chips have only been around for about a year or two, but have resulted in a variety of fairly similar receiver features. Depending on implementation, I would expect some variations in performance.

If you are expecting audiophile quality from such a radio, you will be disappointed. Well, that's not quite true. The DSP IF (intermediate frequency) and demodulator sections work VERY well at getting rid of the adjacent channel garbage and some of the self-interference produced by HD Radio (IBOC) FM stations. If you have a strong local FM stations that is transmitting in HD, but has somewhat distorted audio, on a different receiver, try it with your armband receivers and see if it's any better. If it has a DSP IF and demodulator, it probably will sound better.

The marginal antenna usually found on such miniature devices also can be a factor in sensitivity. For FM, it's usually the ground lead of the earphone cord, which can be random lengths and therefore random impedances. Such armband receivers come with an ear buds, with a known length of cord. If the earphone is exchanged for a different model, the antenna matching, and therefore the field strength sensitivity will change. You might try different earphones, with different cable lengths, to see if it makes a difference. The ideal length cored would be 0.25 wavelengths at 100 MHz, which is 750 cm long, which is rather long for a armband radio. Details and design guide:

If you want to play with it on the bench, there are kits and development boards available for the Si4735 chip:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

If it has reset and scan buttons then it's almost certainly based on the TDA7088 chip, a successor of the famous TDA7000. That chip is over 20 years old but does the job nicely. Scanning is done in analog way by slowly charging a capacitor, sending that voltage to the varactor tuning then stopping when a station is found. IF is set at 70 KHz which is unusual but won't require ceramic filters or IF transformers and keeps cost low. A search for that chip on google images brings lots of schematics.

So I would assume that chip must have a rather good sample and hold circuit in it? I mean, what if you want to sit there for a while?

Years ago I made a frequency to analog converter that would react in change per Sin. I used bimos op-amps, 4066 as the switch, edge trigger transistor circuit to activate the analog switch for the sample and hold refresh. The lowest drift I could hold was ~2mv/s.

That circuit worked well for what it was designed for, and this was before the internist came to my place. All we had was college access main frame and dialup back then, 300 baud, 9600 on a good day :)

Jamie

Don't need S&H, PLL holds it on target. Standard fare for FM, plus the PLL demodulates. Make it so the PLL filter cap (integrator) has a slight pull-up bias, and the PLL only has to 'tug' on it (open collector style) to keep it steady. Disable the PLL and frequency drifts up; enable and it captures the next one. Finally, a comparator with hysteresis detects when it's at the top of the band, resetting it to the bottom (holding SEEK reduces it to a relaxation oscillator).

Tim

-- Deep Friar: a very philosophical monk. Website:

"Jamie" wrote >

Jeff Liebermann schrieb:

Hello,

a calculation error may exist, my HP 32S and mee found only 75 cm.

Bye

The better FM tuners use DSP for less adjacent-channel interference and for more noise suppression. But I wonder if an armband type radio would go to that expense.

Argh. The correct length for 1/4 wavelenth at 100 Mhz is 75 cm or 750 mm. Thanks.

I found a box of assorted ear buds, earphones, and cell phone "headsets". Measuring the cable lengths, I find most of them between

100 - 120 cm and another group between 45 - 60 cm. I then tested all of them on the FM radio in my Droid-X2 smartphone, which also uses the earphone ground as an antenna. There was some variation in signal strength, but nothing spectacular. The cords near 120 cm seemed to work the best, while the shorter 45 - 60 cm cords definately had problems. That Droid-X2 FM antenna was probably designed to use the Motorola EH15 headset, but I can't find one to measure the cord length.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

The Silicon Labs DSP IF and demodulator come with the all digital AM/FM receiver chips. For example, note the block diagram of what's inside the chip: Incidentally, the DSP is 24 bits and 48 KHz sampling.

Hacking the SI4734: Note the built in AM antenna tuner.

List of SW radios with the Si4734 DSP chips (Jan 2011):

Also, note that their Weather receiver chip includes DSP and a built in SAME decoder:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

I'm prototyping a new 80m foxhunt receiver using the Si4735. The manufacturer has provided (under NDA) some SSB software for the DSP, which allows off-tuning to produce a whistle... nice. It would be nicer if the software wasn't encrypted and the DSP was documented.

The QFN is a 3mmx3mm chip, quite small enough for an arm-band receiver.

Clifford Heath.