In the days of analog TVs, it was said a powered antenna was garbage-in, garbage-out, and better spend time on mounting and finding that sweet spot.

Question:

With DTV, is it fair to say finding the sweet spot is not as critical as vs analog and that the digital circuitry is able to deal BETTER with the potential amplified garbage?

I guess what am asking is, do I have more leeways with DTV antennas as far as mounting location as long as I have a decent amp?

No.

And if you're not extremely far from the transmitters, an amp will often make reception worse, due to overloading and noise amplification. If this is the case, getting a good antenna matched for the channels used in your area and at your distance from the towers is more important. The sweet spot is still important.

An amp just amplifies the signal received by the antenna so you can send it over a longer cable run or feed more lines from a splitter. It does not improve the reception capability of the antenna.
With OTA digital it is important to know that multipath reception can cause problems so some highly directional antennas are now sold as "HD" antennas. It is also important to know that a digital tuner unlike an analog tuner will fail to receive a channel if the the signal is either too strong or too weak.

It does not improve the reception capability of the antenna

This is false. It improves the range (distance from the towers) or the ability to tune weak signals.

Its not a panacea, but it does have its usefulness. It cant create signal out of nothing, the elements are still of primary importance.

In fact almost all tuners have an amplification stage integrated within the tuner itself.

It cant create signal out of nothing, the elements are still of primary importance.


That is exactly what I thought I was saying.

A low noise amplifier can improve reception by the margin of difference between the LNA's NF (2 dB) and the receiver's NF (5-10 dB).

This can usually be an improvement of 3-8 dB.

In this limited circumstance, the LNA can effectively make the antenna "larger".

There are few things more confusing than talking about pre-amps and antennas. The concepts are simple enough once you grok the jist of the Friis equations, but stating things accurately and concisely in english - a major challenge.

Back to the OP - at the carrier level (radio frequency broadcasts), analog and digital are not that much different. Most, if not all, of the rules that applied for analog reception remain true for digital reception. An amp cannot make up for a bad antenna or bad placement, but once you've optimized reception with antenna alone, a good amp can improve things nicely.

The one factor that is improved over analog, at least with 5th and 6th generation ATSC tuners, is handling of (stable, non-changing) multi-path.

Bummer. U guys sound like u know what ur talking about, so I'll take your words for it.

I was hoping the super dupper digital processing is able to discriminate against noises/bad signals and compensate for less than perfect antenna placement, not at the carrier level but some higher, intelligent processing filter extract usable signal blah-blah.

I was hoping the super dupper digital processing is able to discriminate against noises/bad signals and compensate for less than perfect antenna placement, not at the carrier level but some higher, intelligent processing filter extract usable signal blah-blah.

The tuner is doing a lot of noise suppression and error correction in order to get such a superior picture over analog. The problem is it doesn't degrade gracefully once the signal quality falls below the design spec.

The tuner is doing a lot of noise suppression and error correction in order to get such a superior picture over analog. The problem is it doesn't degrade gracefully once the signal quality falls below the design spec.
Roger that. When ppl say digital is digital, when a digital cabling gets long enough to degrade the signal that the receiving when can't distinguish a 0 from 1, that's when it breaks down. Hopefully my less-than-perfect antenna placement won't be that out of wack.

If I was designing this thing, the transmitter should periodically send out a "calibration" signal. The Receiver knows what this signal "should" look like, then computes a compensation (differential signal?) out from actual signal received, then apply this compensation signal to subsequent data streams, kinda like that laser beam they shoot at the sky to compensate telescopes for atmospheric distortions, whatever they call that. I make it sound so simple of course.

8-VSB does that, sort of. From the "What Exactly is 8-VSB Anyway" page:

8-VSB employs a similar strategy of sync pulses and residual carriers that allows the receiver to "lock" onto the incoming signal and begin decoding, even in the presence of heavy ghosting and high noise levels.

The page goes on to explain how a tuner can use the info to find and compensate for ghosts (multi-path), but it's pretty heavy reading.

Unfortunately there is no way to compensate for simple thermal noise, and I don't think there ever will be (it's considered impossible, but I won't say never). For now, the only solution for thermal noise is to get enough signal, and the way to do that is to find a good antenna and a good placement.

Unfortunately there is no way to compensate for simple thermal noise

Supercooled preamps? :-)

Unfortunately the antenna is pointed at a hot horizon, so that's a major source of thermal noise. You'd have to supercool the earth.

I was hoping the super dupper digital processing intelligent processing filter extract usable signal blah-blah.

Adaptive equalizer

A preamp, separate or integrated with the antenna, will make up for splitter loss and cable loss. It will not make up for insufficient signal coming from the antenna itself, because it will add some noise of its own making the SNR (signal-to-noise ratio) worse. A minimum of 15 to 16 dB SNR is needed to maintain lock on a digital signal. Once you have met that minimum requirement, the tuner will maintain lock on the digital signal. Any further improvement in the SNR will not be obvious, but it will provide additional protection from signal deterioration.

The FEC (forward error correction) system in the tuner will correct errors in the digital stream caused by poor signal quality, but only up to a certain point. After that the signal will fall apart at the "digital cliff" giving tiles, freeze, and dropout.

The indicator of signal quality (which is just as important as signal strength with digital) is called BER (bit error rate/ratio). The factors that reduce signal quality and increase BER are:

1. Improper signal level: A weak signal will cause a poor signal-to-noise ratio; a signal that is too strong can overload a tuner or preamp. A nearby FM transmitter can also cause overload, which would require an FM trap.
2. Reflections from multipath problems, static or dynamic.
3. Impulse noise in the reception area.

Andy Lee has an excellent post on his Official TV Fool forum thread that has two diagrams of what happens to the signal on its way to the tuner. The second diagram shows how the preamp adds its noise while adding gain which makes the SNR worse. Notice also that the antenna adds its gain without harming the SNR:
http://www.avsforum.com/avs-vb/showpost.php?p=15700679&postcount=397

Roger that. When ppl say digital is digital, when a digital cabling gets long enough to degrade the signal that the receiving when can't distinguish a 0 from 1, that's when it breaks down. Hopefully my less-than-perfect antenna placement won't be that out of wack.

If I was designing this thing, the transmitter should periodically send out a "calibration" signal. The Receiver knows what this signal "should" look like, then computes a compensation (differential signal?) out from actual signal received, then apply this compensation signal to subsequent data streams, kinda like that laser beam they shoot at the sky to compensate telescopes for atmospheric distortions, whatever they call that. I make it sound so simple of course.

The broadcasted RF signal is not a simple binary on-and-off signal, but a signal that is modulated to use multiple levels in a continuous waveform such that much of the analog signal analysis concepts still apply to the transmitted signal.
http://en.wikipedia.org/wiki/8VSB

There is, in fact, a calibration signal sent in ATSC broadcasts that is used for adaptive equalization by modern tuners to compensate for multipath "echo" effects. Tuners vary in their ability to adapt quickly as the multipath channel changes over time. The rate at which the training signal is rebroadcast may also be a limiting factor.

The multipath effect is greatly reduced by using a more directional antenna. Think of this as using a more directional microphone to reduce echos in a big room and making it easier to equalize.

This seems to apply to NTSC tuners, but some of it is still applicable. Are the ATSC tuners of today less noisy?

TV tuner sensitivity

Early TV sets, circa 1940, had very poor RF sensitivity. Bob Cooper recently commented that a receiver of that era required around 1,000 microvolts to produce a grainy image on the small screen, and RCA was recommending 5,000 microvolts. A modern TV set with 50 microvolts will produce a far better image than the 1936 version with 1,000 microvolts.

In the 1960s, TV tuner noise figures started to gradually improve, but were still inferior to some of today's sets. Typical TV tuner noise figures were around 10-13 dB. For this reason, low noise preamplifiers were more essential for 1950's and 60's TV sets.

The typical TV tuner noise figure for modern sets is 6-7 dB at 45-220 MHz VHF, and 10 dB at 470-860 MHz UHF.

For TV DX, the author uses a HS Publications D100 varicap 40-230 MHz TV tuner/RF converter. The D100 uses a Toshiba EG522F Mosfet varicap TV tuner. In most cases, I have found that a 2dB Mosfet tunable 45-70 MHz pre-amplifier offers little or no improvement on weak video signals. Why? Because at 45-70 MHz the RF noise figure of the EG522F TV tuner is low enough; hence external noise becomes the limiting factor. This can be proved by the simple test of plugging an outdoor aerial into the TV tuner's input, while watching a blank TV channel (noise on the screen). If noise on the TV screen shows an obvious increase, the tuner noise level is low enough, and hence a preamplifier will produce little or no improvement on weak signals. If however no increase in noise is observed, a preamplifier will be essential.

At frequencies above 88 MHz, external man-made and atmospheric noise is lower. For this reason, low noise pre-amplifiers are generally more beneficial for band 3 (170-230 MHz), and especially UHF TV.

During the early 1980's, the German company Telefunken introduced the ET021 Mosfet varicap TV tuner. The ET021 featured Mosfets in the RF and mixer stages of the VHF section, for use in areas subject to adjacent channel selectivity problems. Typical figures quoted for an interfering signal two channels away from the desired signal and to give a 1% cross-modulation on the desired signal are 100 mV as compared to a conventional varicap bipolar tuner with 25 mV (band 3 TV channels). The ET021 was a revolution in terms of strong signal handling and low noise RF performance.

The Toshiba EG522F varicap VHF/UHF TV tuner was introduced around 1987, and is also excellent in terms of strong-signal handling and low noise performance. Other Mosfet and GaAsfet TV tuners are currently available, for example, around 1989, Toshiba introduced a 3SK97 GaAs-MES-FET TV tuner.

http://www.geocities.com/toddemslie/bf981_preamp.html

This seems to apply to NTSC tuners, but some of it is still applicable. Are the ATSC tuners of today less noisy?

Dan,

This document may provide answers... In depth testing of CECBs....

http://www.nabfastroad.org/NAB-STV%20Digital%20Converter%20Box%20Evaluation/Converter_Box_report.doc.pdf