Where to buy signal attenuators locally?

Radio Shack 6dB Attenuators
Old Part Numbers:
#15-1257 F jack to plug
#15-1258 F jack to jack

Replacing the RG-59 with RG-6 is your best shot.

If that doesn't fix it, some Radio Shacks may still have their Variable Attenuator in stock. While there, see if they still have the little FM Trap, as well. Try them both, individually.

I wouldn't put to much faith in changing from RG-59 to RG-6. Unless the cable you have is defective or some real poor quality stuff, you are going to only realize about .3 dB on channel 39 and even less on channel 7.

The problem is, RG-59 has terrible shielding. So, as the signal leisurely plods it's way thru the cable (at about 66% of the speed of light), the rest of the signal is blasting in at full speed, and they sum together at all sorts of wierd timings.

This blurs an analog signal, and it blurs the data of the OTA or Cable Digital signals, making them more likely to fail. Also, the poor shielding allows all sorts of household electrical noise to get in....my own STB radiates a good bit of noise, and the nearby satellite receiver spews digital garbage across the VHF high band.

Not all RG-59 has a "Nominal Velocity of Propagation" of 66%.
For some it's 78% (Belden # 9659, 8221, 9240)
But most any common co-ax has a "Nominal Velocity of Propagation" of 66% to 78%.

C'mon Ken, you know better than this.

You can good or bad quality RG-59 and RG-6.

The old, RG-59 commonly used twenty to forty years ago has copper braid shield and leaks like a sieve, causing the blurriness that Ken alluded to. If someone is in a region were an ignorant contractor or fledgling cable company had once made it their conductor of choice, then they may be working in environments that are not representative of the world in general. Whenever I run across copper braid shielded RG-59, it is a piece added onto the system by a do-it-yourselfer. I have no leakage problems with foil shielded RG-59.

As has been explained so many times, the signal meter numbers are not just unreliable measures of signal strength, they are so far removed from being indicative of signal strength that they are nearly useless in estimating it. They are the resultant of unique algorithms that reliably principally on the amount of error correction being performed. You can vary signal power a thousand fold without the receiver signal strength number changing.

If someone is 18 miles from full powered transmitters and is using any conventionally designed consumer grade antenna, they will not experience any signal level problems that can be remedied through amplification. When signal level numbers jump from 70 to 80% down to zero or thereabouts, it means that there is multipath interference that fluctuates around the threshold level of canceling the desired signal to the point of rendering it unusable. The remedy is an antenna remedy, either reaiming it within the attic, replacing it or putting it on the roof.

While the original poster now has a situation that satisfies him, he has no way of knowing how much of a performance margin he has, because digital signals, unlike analog signals, only display avalanche failure symptoms. If any channel craps out again, the only practical solution is to put the antenna on the roof, where the signal quality is invariably, significantly better.