Signal Amplifiers (Amps, Preamps, Distro Amps) - See Chart in Post #1

[mlord]

Specs look (really) good, but one cannot believe them until proven in use. It is a distribution amp, though, not a mast-head pre-amp, so it won't do as well as a pre-amp given any normal amount of coax between antenna and set. Looks Great for indoor (attic) installs, though.

Send me one and I'll test/compare it here with the RC 0.4dB NF pre-amp I have.

(A pre-amp accepts power-over-coax; a distribution amp uses/requires a separately routed power connection).

[holl_ands]

Originally Posted by balm

STL:

Originally Posted by mclapp

Not including amp overload problems, a good low noise amp will almost always improve your reception regardless of the length of coax. The amps noise figure is usually a few db lower than most TV's or PC cards noise figure so you will gain that difference. I've seen a difference in using an amp on just a 20ft piece of coax to a common TV or digital converter box.


Originally Posted by STL

A pre-amp will not always be helpful in a setup since it introduces noise, can can be prone to overload (where strong signals cause intermodulation on adjacent frequencies), and can overload your tuner. In your specific case, the 6 dB of NF introduced will hinder reception if your setup would otherwise have less than 6 dB of loss. As I recall you don't have splitters and just have a straight coax from your antenna to tuner.

...Low-noise amplifiers introduce 2-3 dB of noise. So your amplifier isn't low noise.

Also, pre-amps do not increase reception but help overcome loss to splits and spans of coax. So if you don't use a splitter or more than 50 ft of RG6, I don't think a pre-amp will help.


Do these statements square-off...

Tuners have Noise Figures of about 5-10 dB, mostly due to inefficient voltage tuned varactor
filters on their input. This loss plus coax downlead & splitters are REDUCED by Preamp Gain.
Hence a Preamp with a 2-3 dB NF can ALWAYS help the overall System NF, thereby
improving the sensitivity for weak signals....if the intermods don't kill it.

HOWEVER, this only works when intermod products generated by Preamp are no
higher than the thermal noise floor. The maximum input signal levels I calculated in
PREAMP OVERLOAD spread sheet are the point at which the intermod noise begins to
equal the thermal noise. This also yields the maximum Spurious Free Dynamic Range
(SFDR), which perhaps best explains what you want to do. Be able to receive the
weakest signal levels without being killed by the strongest signal levels.
[Which explains why the high gain CM7777 is usually NOT suitable for urban areas.]

You can still operate a Preamp with input signal levels above calculated Max Sip.
But intermod noise levels increase 3 dB for every 1 dB above Max Sip, thereby
decreasing Dynamic Range from say 60-70 dB down to only 30-40 dB if operating
10 dB above Max Sip. [Don't forget Max Sip depends on number of strong signals.]

And yes, a Preamp can ALSO "overload" (actually "desensitize") your HDTV. So it is
important to also check to make sure the Sih dBm level is less than about -10 dBm
and perhaps lower than -15 dBm, since no one ever publishes tuner overload specs.

And if you use both a Preamp and a Distribution Amplifier, I can almost guarantee
you have desensitized your reception system unless you carefully calculated signal
levels and have the RIGHT amount of loss between Preamp and DistroAmp.
[But if you only want local stations, maybe you can live with reduced dynamic range.]

PS: I emailed Preamp Co's for NF and overload clarifications so I can update spread sheets.

- Tin Lee acknowledged some inconsistencies in their specs and will update website shortly.
Their NF specs come directly from transistor spec sheet vice actual Preamp test.
And yes, they use 46 dB Intermod (Cross-Mod) spec level, same as C-M and W-G.

- Kitz Technologies had nothing to add to their Max IP3 overload spec,
but did say "The noise figure is normally around .5 to .7 db.".

- Research Comms has not yet responded.

[balm]

holl_ands:

Thanks for that info. Now I will have to absorb, and do some more research to learn. Also I have to go look at the spreadsheets - but yes, I guess that's the ultimate goal, is to strengthen weaker signals, without letting the strong ones kill - the best of both worlds!

Eventually I may have to post my TVFool reception results....

thanks again!

[uglydukwling]

First, some background. My original setup was a medium sized (9 foot boom) combination antenna and a Delhi 476 preamp about 27 feet above ground. I'm south of Chatham, Ontario on the Lake Erie shore on a heavily treed lot. I was getting all the Detroit ( 55 miles) and Cleveland (65 miles) channels. Reception ranged from near perfect (signal indicator in the green, almost no break-ups) on some channels, to bad (unwatchable, almost constant break-up) on others. I get reasonable reception on the Canadian channels (analog, of course)

I've tried several changes, none of which made a significant difference. By significant, I mean dramatic enough to be sure that it was real. I get quite a bit of random variation in signal, depending on time of day, weather and probably other factors I haven't identified.
I removed the preamp to see if it was working. Then I replaced the antenna with a CM4228hd.

I couldn't find any of the preferred CM or WG preamps in this area, but I happened to see a Philips 61112 in Forest City Surplus for $7.95. The specs aren't too impressive: gain 28 dB vhf, 23dB uhf; noise 4dB below 500MHz. But I thought it might be worth playing with for 8 bucks, even though I'll probably have to get a better one when I can find one. Does anyone have any experience with this pre-amp? It didn't seem to make much difference in my reception.
I didn't have enough room above the rotor to mount the preamp the suggested 3 feet below the antenna, so it's mounted behind the reflector. How important is the placement?

I expect to put up a tower in the spring to see over the trees. At that time I'll add a longer pipe above the rotor and also replace the existing coax with rg6. I'll also try a DBGH. I've got the parts cut and ready to assemble, but I've been waiting for my local plumbing supplier to get 1/2 inch pvc crosses.

[STL]

A practical application for tczernec's system loss calculator is to plot how much loss you are actually incuring with a pre-amped system as a function of cable, splitter, and tuner noise losses downstream of the pre-amp.

The following graph uses the CM 7777 UHF gain and noise specifications, assumes a 0 dB balun loss, and a 6 dB tuner loss (typical).



One can correct the above graph for balun loss by shifting the curve vertically upwards by an amount equal to that of the balun loss in dB. Any cable loss before the pre-amp can be dealt with in the same way.

One can correct the above graph for higher (lower) tuner noise by shifting the curve horizontally leftwards (rightwards) by an amount equal to that of the difference of the actual tuner noise and 6 dB, in dB.

What's interesting of this graph is that the first 13 dB of gross cable loss after the pre-amp will translate into an actual 0.5 dB loss when using a CM 7777 (assuming of course no pre-amp overload). A further 3 dB of gross line loss to 16 dB will cost another actual 0.5 dB of system loss. A further 2 dB of gross line loss to 18 dB will cost another actual 0.5 dB of system loss and you will now be adding actual system loss at a faster rate per additional gross line loss.

As the gross transmission loss extends past the pre-amp gain, the relationship between incremental gross line loss to incremental actual system loss becomes 1:1; thus making the pre-amp ineffective at reducing further system loss when the line loss exceeds the pre-amp gain. This makes sense.

But what may surprise some is that gross cable and splitter loss after the pre-amp does cost you some signal (even though you have a pre-amp!) and this signal cost will increase as gross cable loss approaches and extends past the pre-amp gain factor.

So for those spanning coax and splitters while remaining under the pre-amp gain in dB, you may be adding non-negligeable loss to your system.

STL

[roger1818]

STL. Very interesting! I hadn't thought of this before, but it makes total sense since you don't tend to see sharp transitions in real life. It would be interesting to see how different pre-amps compare (I assume the graph is predominantly dependent on the gain of the amp).

I would also be interested to see what effect adding a distribution amp to the system would make (assuming of course no distribution amp overload). This would obviously be dependent on where the distribution amp was put in the system, but it might (and I may be wrong here) be a better way to overcome Gross Transmission Loss in high loss systems. To be a valid comparison, the total gain (pre-amp + distribution amp) would need to be the same as the pre-amp only system. You might also want to assume the distribution amp is after exactly half of the Gross Transmission Loss. The other factor of course is two amps will likely be noisier than one.

[STL]

This is getting VERY interesting.

I do not claim to be an expert here as I am just using tczernec's spreadsheet on the Friis formula and plugging numbers into it.

To answer your questions, Roger1818, the left-side slope is proportional to the ratio of amplifier gain to noise factor. So the slope is not a function of amplifier gain alone. Another property of the curve is the presence of an inflection point near the Gross Transmission Loss value equal to the amplifier gain in dB. The gain on the CM 7777 is 3 dB higher than on the RC 9254 [for UHF] and this explains why the CM's loss curve isn't as steep as the RC's curve on the right-side.

This brings to say that if you are using less than 17 dB's worth of cable and split loss, the RC amplifier will yield lower system noise. But if you are using more than 17 dB's worth of cable and split loss, the CM amplifier will yield lower system noise. This is very surprising, but is grounded in the Friis formula. So in other words, a higher noise factor amplifier, if it has higher gain, can lead to lower overall system noise if the gross transmission losses are significant enough. (Left side of my brain disagrees with this statement, but right side of brain doesn't!)

Here is the loss curve for the RC 9254.



The Friis formula is modular as you can cascade more loss or gain terms. So you could make such a curve for a distribution amplifier. Just remember to keep track of losses in the system before the distribution amp.

STL

[tczernec]

Very nice! Glad to see the calculator's coming in handy. It would also be interesting to see the two amps (or more!) on the same plot to see under which conditions they outperform each other on the system level.

Tom

[STL]

The actual losses cross over at 17 dB of gross transmission loss.

I'll try to add more pre-amps on this graph soon. At least these two seem to be generating a lot of interest from DHC members.

Definitely a useful spreadsheet!

STL

[STL]

I had to go all the way and add more common preamps! I hope the colours show neatly.

Choosing the best preamp for you is more complicated than picking the one with the lowest system loss given your gross line loss. You need to make sure that the input signal level is below the overload limit! For that, I refer you to holl_ands overload calculator spreadsheet.

STL

[tczernec]

I like it The AP-4800 is sounding more and more like what I'll be using.

Hey - what happened to the preamp chart? The post says it's been moved but when I click on the link it takes me to this thread - anyone else having that problem?

[stampeder]

Sorry, the preamp chart link is in Post #1

[mlord]

Wow. That chart, despite the difference in axis scales (vertical scale is exagerated), clearly shows the RC pre-amps superiority. Not that one needs a chart like that to appreciate a 0.4dB NF. But still!

Of course, throw in 100 feet of RG6, 2dB of added connector losses, and an 8-way splitter (TV in every room?), and.. well, the RC pre-amp still wins on that chart.

Kind of as expected, really.

Cheers

[balm]

Does anyone know if I would benefit from any pre-amp set-up based on the following TV fool (post June 2009) - using my DYi 4 bay bow-tie at 25-30 ft:


http://s638.photobucket.com/albums/u...adar-All21.png

Problem channels (also beware the report has a few channel # errors):

1) I am very close to locking on digital CBOT (analog #4, digital #25) from Ottawa, ON;

2) I can sometimes receive local - CBMT (analog #6, digital #20, eventually digital #21) ,CBFT, CFJP (analog #35, digital #42) - all digitals drop out with pixellisation, the digitals are heavily blocked by Mt-Royal;

3) Would like to get CJOH analog better (analog #8, eventually digital #45) from Cornwall, On;

4) Sometimes I get digital WNPI (analog #18, digital #23) but needs lots of height, and intermittent from Norwood, NY.

Bullet #1) would be nice to get. I live in the west island Montreal.

Thanks

[Kro]

Does anyone know if I would benefit from any pre-amp set-up based

Maybe Yes but, Montroyal station are very strong for you, depending on how good your attenna is at rejecting signal from the side.

If you have a long cable and or splitter, you might want to look at AP-2870 or HDP 269.