splitter quality:

Well, I have my antenna and gear apart to add a rotor and I noticed that I needed to replace the splitter I had combining the 2 4221HD's on the mast. The old caulk I used to seal the coax going into a weather proof box split and the box filled with water. It still worked, but I didn't want to take a chance it would fail from corrosion/other in a few weeks. In seeing how varied people's experiences with other parts are, I decided to test a series of splitters I had here to see if there was much difference.

I had 4 splitters laying around, and used my Rogers/Scientific Atlanta 3200 cable box diagnostics screens to see which gave the best signal strength readings. The baseline was taken using a cable from the jack in the wall, through an inline F connector and another cable on to the STB. The result was -7dB QAM, and -8 dB on FDC. I then inserted the splitters for the F connector, and connected a second TV to the splitter tuned to a single channel and took my readings from there. The tests were using a series of about 15 channels from 2-200 on the STB. Channel 2 was the channel on the second TV pulling analogue cable signal. I'm sure the test could have covered more frequencies and been more scientific, but I think it does give a good idea of where the splitters rank, and if I missed something obvious to the experts, someone will surely show the error of my methods and I can try this again sometime. I also reversed the legs to see if there were differences. All were identical.

The first, a Monster Cable 5-1000mhz unit. I've had this on there for a while as I needed one quick a few years back when I got the second TV in my office and started splitting off the living room. A hefty unit, with as I recall a hefty price, it appears all that heft is for the casing and marketing and not the electronics in the design. It turned out to be the worst for loss I had on hand. Signal strength was reported at -12 dB on FDC and -11 dB on QAM across the channel range.

Next, a splitter I found in a bin at Active surplus in Toronto. They had a bin full of them and I grabbed a pair. The "Yesat" 5-2500Mhz all power pass splitter came in at -12dB on the FDC, and -10dB on QAM. Both units were identical.

Next, an Ideal "digital" splitter from Home Depot. Lowes also carries them, and I think Rona. It seemed nice to support a local employer like Ideal too, although they are still made in (feigned surprise) China. -12 dB FDC, and -10 dB QAM. My waterlogged one was of this variety.

Finally, came the one I figured would lose hands down. The above are all PCB designs, claim low loss, claim to be better than the cheapies, and this one is your dollar store special, with the small wires soldered around inside to make the split connections and glowing in all it's fake gold plated glory. The "Video CV-51303" from HRS-Global. Carried at "Everything for a Dollar" stores. It came in with a signal strength of -9 dB FDC, -9 dB QAM.

I tested another. Then picked up 2 more at the store this morning on my way to Tim Horton's, and they came in the same (one showed -10 FDC, but it turned out to be DOA as it was the one that would short out every time it was shaken). The verdict to me was clear...the cases may be cheap, the manufacturing methods suspect, but the result is better than the others for 1/5 the price. Even if you get a duff unit like the one above, you could buy a mitt full of them for the price of the cheapest of the others I tested, and away you go. I put on a little liquid electric tape to ensure the backer was sealed tight, a little magic seal on the connections after screwing on the new cables and in the new waterproof box it went with new silicone and a weep hole this time.

Results: HRS-Global CV-51303 from dollar store (far right) Tie: Ideal from Home Depot (far left) Yesat from Active Surplus (next from left) Monster Cable splitter (next from right)Again, not a lab quality test, but I hope it serves as a starting point for others to try their own tests.

http://img.photobucket.com/albums/v193/bgrieger/splitters1.jpg

When I add splitters between the antenna and my TVs, signal strength increases. Can splitters act as attenuators?

For my location, there is a transmitter 2.5 miles away, while others are 35 to 70 miles away.

It should also be noted that this affect only happened when I wasn't using a pre-amp; when I used a pre-amp, the use of splitters did not change signal strength +/-.

Can splitters act as attenuators?
Yes. They split the signal in half, so each side of the splitter gets half of the original signal strength. About a 3dB attenuation, plus the 0.5dB connection loss.

-ml

Twinz,

Thanks for this. I needed a splitter to make an antenna and the dude from The Source was trying to upsell me the 29$ HDTV splitter... I bought the cheap 6$ one that looks like the winner in your tests.

I feel better about my decision now.

Thanks for this. I needed a splitter to make an antenna and the dude from The Source was trying to upsell me the 29$ HDTV splitter... I bought the cheap 6$ one that looks like the winner in your tests.

It is amazing how much up-selling goes on with things like cables and splitters. It isn't rocket science here. As mlord said, the signal is being split in half so you are guaranteed a minimum of 3dB of loss according to the first law of thermodynamics. In practice, most splitters provide about 3.5 dB of loss so there really is not much room for improvement.

Over the weekend I replaced the $-Store splitter I used to join the two GH antennas in the attic before feeding them into a Winegard 8700, with this splitter

2 Way Splitter 1GHz DC Power Passive Gold HDTV 5 - 1000 Mhz Satellite Antenna Video 75 Ohm Coax Cable Digital Signal Component Combiner for UHF / VHF Aerial, DC Power Passing, Studio Grade, Part # Woods 7040
[SP2G01]

I immediately saw 4 to 5 dB increase in the readout on my Hauppauge on channel 66, an increase that was confirmed by the fact that I had to put a 6dB attenuator before the Hauppauge cards input to get rid of the input saturation effects I noticed on some of the analog UHF channels after the change. (my CRT TV was happy though)

What I like about this splitter is that it comes with dust caps and looks very well build. The attenuation factor of 3.7dB is also printed on the outputs . I should have ordered more than one of these - at US$0.95 èach it is a good deal - pitty about the shipping cost though.

PS1 Although the readout of the Hauppauge is tagged as SNR I have found that a 3dB increase in signal strength results in the SNR value increasing by a value of about 2.

PS2 The idea behind the Wingeard 8700 is more of an distribution amplifier than an antenna amplifier.

Just thought I share with the fellow forumites what a $-store 3 way splitter looks like in the inside. I bought it for a balun experiment and took this picture before I took the (minimalist) guts out - I will post something in the balun thread about my experiment

http://img5.imageshack.us/img5/1584/p1040369b.jpg

Quality stuff! What's that 40 gauge wire? Could you not drastically improve the quality by soldering some better wire between the centre conductors?

Just thought I share with the fellow forumites what a $-store 3 way splitter looks like in the inside.

So are the right two outputs shorted together? Not good. Just goes to show you just don't know what you are getting from the dollar store.

So are the right two outputs shorted together?

Exactly!

I've read many posts here about attenuators, but I'm still not clear on when you would need to use them. Most of the posts I've read seem to indicate they are used in conjunction with preamplifiers/amplifiers. So, hypothetically, if you have a weak signal and you go and put on a preamp/amp, then if you're overloading you put in an attenuator? A 3 dB unit cuts your signal in half, a 6 dB cuts it down to 1/4. Isn't that kind of drastic? Would that put you back in the original situation with a weak signal?

Use an attenuator/notch filter if you are very close to a transmission tower and it is causing problems for other stations/channels. I am 3 miles from the CHCH tower- when I do a straight feed to my TV, certain distant channels are weak, when I used a splitter (acted like an attenuator), the distant channels became stronger.

Legacy2009, have you read Post #14 in the OTA FAQ?

If so, I'd like to improve it if it is not clarifying the attenuation and amplification concepts well enough. :)

Yes I did read it, although it was a couple of months ago when I first got into OTA, and I was in information "overload" (no pun intended).

What I still don't get is that if you attenuate after amplification, wouldn't that also attenuate the weaker signals? If you need to reduce the signal of a particular channel, then you'd need a specific filter. The way I see it, the attenuator is wideband. But the way goforit describes it, does it attenuate primarily only the strongest signals?

Edit: After re-reading goforit's post, I see now that he said attenuator/notch filter, so both of those are being used in that situation, not just the attenuator. I think this is one of those counterintuitive things, you put in an attenuator, and the signals are stronger:)

if you attenuate after amplification, wouldn't that also attenuate the weaker signals?Yes it would, but hopefully we had done our planning properly so that the amplification is higher than the combined line losses and attenuation amount.

For example, if we apply 10dB of low noise amplification and we have line losses of 4 dB, we now have a net boost of just 6dB, so if we were to insert 6dB of attenuation we'd lose the boost we'd worked for and also unfortunately would have added some noise. This would mean that the weaker stations would probably be even worse off than with no amplification at all.

So, if enough gain and not too much noise have been added then applying the attenuator will knock down all the channels by only our planned increment. Using the example above, if we raise the low noise amplification to 20dB and then we subtract the line loss of 4dB, then insert a 6dB attenuator, we now would still enjoy a net boost of 10dB, meaning that the weaker stations as well as the stronger ones have been boosted by that same amount. If one or more of those strong stations are too strong we can take steps to reduce just those channels by using tools like notch filters.

You are correct that common, cheap barrel-type attenuators like those from The Source are wideband, affecting all the channels likewise. As CamDAB has reminded me by PM, there are variable-frequency attenuators (filters) out there, but I have to remind folks that such gear is not common and also not cheap. These days those are considered pro gear as used in MATV/CATV. :)

As far as combining amplification and attenuation being counter-intuitive, if you figure out your total signal budget and do the math there is a logical, recognizable pattern. Also see tczernec's Loss Calculator Spreadsheet in this post: http://www.digitalhome.ca/forum/showpost.php?p=877838&postcount=604. To use it, make sure to download it rather than using it online. ;)

Further to stampeder's post, broadband attenuators allow you to adjust the gain of the amplifier. In a perfect world you would be able to buy amplifiers with 1dB increments so that you could select exactly the right one for your needs. However this is not feasible (they don't even do this with professional broadcasting equipment) so the next best thing is to pick an amplifier with a larger gain than necessary and attenuate the signal as needed. Even amplifiers with a variable gain, use a built-in variable attenuator to reduce the amplification.

In the case where you have very strong local signals, notch filters are the optimal solution, but as stampeder said, they can be expensive so it is better to try broadband attenuators first.

Ok, it's making sense now... I've got a KitzTech pre-amp, now I see how the variable gain works (it's actually fixed gain with built-in variable attenuation)

And yes, all those custom notch filters, etc from places like Tin Lee are $$$$:eek:. I did get quotes from them for a few things a little while ago, but I decided to pass on it. I can see how it would be useful in some situations to get channel specific equipment, but then if a channel changes... that gear becomes useless. I guess it's all part of the fun of OTA.

Yes I did read it, although it was a couple of months ago when I first got into OTA, and I was in information "overload" (no pun intended).

What I still don't get is that if you attenuate after amplification, wouldn't that also attenuate the weaker signals? If you need to reduce the signal of a particular channel, then you'd need a specific filter. The way I see it, the attenuator is wideband. But the way goforit describes it, does it attenuate primarily only the strongest signals?

Edit: After re-reading goforit's post, I see now that he said attenuator/notch filter, so both of those are being used in that situation, not just the attenuator. I think this is one of those counterintuitive things, you put in an attenuator, and the signals are stronger:)
Yes, an attenuator reduces the signal level on ALL frequencies.

When in a high signal level situation (esp a few miles from a nearby transmitter), a SMALL
amount of attenuation can significantly reduce intermodulation distortion (IMD) noise.
For every 1 dB of attenuation of the desired signal, there is a 2 dB attenuation of the
second order IMD noise and 3 dB attenuation of the 3rd order IMD noise.

Hence inserting 4 dB of loss reduces FM second harmonic noise (in Hi-VHF Band) by
8 dB and 3rd order IMD (in any Band) by 12 dB. The goal is to optimize the overall
Signal-to-Noise Ratio by providing the best SFDR (Spurious Free Dynamic Range):
http://www.arl.army.mil/arlreports/2007/ARL-TR-4235.pdf
Strength of the strongest signals are reduced (along the linear Input/Output line)
until reaching the point of maximum SFDR, e.g. 3rd order IMD levels are equal
to the thermal noise level (see charts in the above document). Unfortunately,
most of us don't have a sensitive enough spectrum analyzer, so we gotta try a
few values of attenuation to find an optimum value.

When using a Preamp, esp. when also using a Distribution Amplifier, it is possible
to also overload the Tuner's input. Overload calculation spread sheets and links
to other information re SFDR are found here:
http://imageevent.com/holl_ands/files

Has anyone used this device?

Any opinions?

I want to join two antenna's... will this device work well...

plabonte, I got several hits in this thread using the Search This Thread tool to the upper right beside Thread Tools with "CC-7870" as the search term. ;)