24-08-2016, 03:42 PM
The RF input input to the average TV should not be less than 1mV (0dB) to avoid unnecessary noise or, with older sets, simply running out of gain. The maximum should be not greater than 3mV (+10dBmV) to avoid overloading problems. 2mV (+6dB) is the ideal target. Don't worry if you're not familiar with the use of decibels, those three figures are the only conversions you need to know, and then only from a comprehension point of view, apart from that it is a simple matter of addition and subtraction.
If you are using an Aurora, the specification shows the output as +76dBµV but it is much more convenient to work in dBmV. 0dBmV = +60dBµV so the Aurora's output is +16dBmV (6.3mV).
If you are feeding just one set, a 10dB attenuator (pad) will reduce the output to +6dBmV, which is bang on target.
If you want to feed more sets - up to eight, say - it is quite straightforward if you take note of a few simple facts.
The loss of a two way splitter should be assumed to be 4dB. For a 4-way splitter, this increases to 8dB and for an 8-way splitter it is 11dB. These can be taken as worst case figures for good quality splitters covering up to 860, 950 or even 1000MHz but beware that these are becoming increasingly difficult to source as lots of suppliers only sell splitters that go up to 2.4GHz these days and, as you don't get anything for nothing in this world, the penalty is higher losses, including across the bands we are interested in. If you have to use these, check the specifications carefully and adjust the figures I've given accordingly.
Anyway, back to the plot.
The Aurora's output can be split in many different ways to suit different circumstances.
If you only want to feed 2 sets, a 2-way splitter will still produce a signal level in excess of the target maximum of +10dBmV (16dBmV - 4dB = 12dBmV) so a 3dB attenuator (pad) would bring it down to an acceptable level. However, a 4-way splitter will do the job just as well and gives you extra outputs for future expansion!
8 outputs is the most that an Aurora will reasonably feed without an amplifier and can be achieved either with a single 8-way splitter or you might find it more convenient, either due to the physical layout of your display area or just the availability of suitable splitters, to use a 2-way splitter feeding two 4-way splitters.
I've avoided mentioning cable losses so far as I'm assuming a fairly compact layout but they can't be avoided completely.
However, the good news is that, if you use a high quality cable such as CT100 and its successors, losses are very reasonable - just over 1dB on Ch. B1 for 25m, which will be much more than I would guess will be used in the average installation! The loss figures for the same cable length increases to 2.5dB @ Ch. B13, 4dB @ Ch. E32 and 4.9dB @ Ch. E68. For a VHF only installation, these losses are unlikely to bother you but I've made available a customisable cable loss tool in Excel which can be found at:
https://dl.dropboxusercontent.com/u/2430...ulator.xls
or: http://tinyurl.com/cablecalc2
You can change the data in the green highlighted cells to customise it to suit your own installation.
If you are fortunate enough to have two Auroras, one for Band I and the other for Band III, you will need to combine them. This can be done with a 2-way combiner (a splitter wired backwards) but you will lose 4dB in the process. Finding a Band I/Band III diplex filter can be difficult now that Band I has been abandoned.
If you have added, or are thinking of adding UHF to your network, you will have a similar choice but you will probably be better of running the UHF distribution alongside the VHF. This has a particular advantage that there will be no danger of harmonics from the Aurora(s) getting into your UHF channel and you can probably simplify the network sloghtly as you will own have to pipe it to the dual standard sets in your collection, which will have two aerial sockets, anyway!
Alternatively, you might want to consider using a diplex filter. Vision produce a filter that was originally specified down to 40MHz but is now only guaranteed down to 88MHz although the company say it should still work down to 40MHz but they no longer check the alignment down that far. We were lucky that they offered to align one specially for the Dulwich museum.
Ensure your UHF modulator has an output equal or greater than the Aurora, then use the same planning rules.
You may be interested to see how a more complex system fits together: here is the Dulwich BVWS Museum installation:
https://www.dropbox.com/s/e6lc0cc7c5qx05...2.png?dl=0
or http://tinyurl.com/dulwichbvws
Although it uses only one amplifier, it is capable of considerable expansion if ever needed in the future. Note the 8dB pad on the output of the amplifier and the spare terminated feeds!
I was told to design a system that would last 25 years - not that I'll be around that long - I've done my best, though! The odd diplex filter for Band I/Band III, which cuts off part of Chanel B5 is actually a Cable TV reverse/forward path combiner that I scrounged from Virgin Media!
You will notice references to both 4 & 8-way taps and splitters and wonder what the difference is. Nothing is the answer! In this case they are same physical units, it is just their function that is different and the tap symbols don't need all the connectional information that the splitters do. Beware, though: all splitters can be used as taps but the reverse is not always true!
If you are using an Aurora, the specification shows the output as +76dBµV but it is much more convenient to work in dBmV. 0dBmV = +60dBµV so the Aurora's output is +16dBmV (6.3mV).
If you are feeding just one set, a 10dB attenuator (pad) will reduce the output to +6dBmV, which is bang on target.
If you want to feed more sets - up to eight, say - it is quite straightforward if you take note of a few simple facts.
The loss of a two way splitter should be assumed to be 4dB. For a 4-way splitter, this increases to 8dB and for an 8-way splitter it is 11dB. These can be taken as worst case figures for good quality splitters covering up to 860, 950 or even 1000MHz but beware that these are becoming increasingly difficult to source as lots of suppliers only sell splitters that go up to 2.4GHz these days and, as you don't get anything for nothing in this world, the penalty is higher losses, including across the bands we are interested in. If you have to use these, check the specifications carefully and adjust the figures I've given accordingly.
Anyway, back to the plot.
The Aurora's output can be split in many different ways to suit different circumstances.
If you only want to feed 2 sets, a 2-way splitter will still produce a signal level in excess of the target maximum of +10dBmV (16dBmV - 4dB = 12dBmV) so a 3dB attenuator (pad) would bring it down to an acceptable level. However, a 4-way splitter will do the job just as well and gives you extra outputs for future expansion!
8 outputs is the most that an Aurora will reasonably feed without an amplifier and can be achieved either with a single 8-way splitter or you might find it more convenient, either due to the physical layout of your display area or just the availability of suitable splitters, to use a 2-way splitter feeding two 4-way splitters.
I've avoided mentioning cable losses so far as I'm assuming a fairly compact layout but they can't be avoided completely.
However, the good news is that, if you use a high quality cable such as CT100 and its successors, losses are very reasonable - just over 1dB on Ch. B1 for 25m, which will be much more than I would guess will be used in the average installation! The loss figures for the same cable length increases to 2.5dB @ Ch. B13, 4dB @ Ch. E32 and 4.9dB @ Ch. E68. For a VHF only installation, these losses are unlikely to bother you but I've made available a customisable cable loss tool in Excel which can be found at:
https://dl.dropboxusercontent.com/u/2430...ulator.xls
or: http://tinyurl.com/cablecalc2
You can change the data in the green highlighted cells to customise it to suit your own installation.
If you are fortunate enough to have two Auroras, one for Band I and the other for Band III, you will need to combine them. This can be done with a 2-way combiner (a splitter wired backwards) but you will lose 4dB in the process. Finding a Band I/Band III diplex filter can be difficult now that Band I has been abandoned.
If you have added, or are thinking of adding UHF to your network, you will have a similar choice but you will probably be better of running the UHF distribution alongside the VHF. This has a particular advantage that there will be no danger of harmonics from the Aurora(s) getting into your UHF channel and you can probably simplify the network sloghtly as you will own have to pipe it to the dual standard sets in your collection, which will have two aerial sockets, anyway!
Alternatively, you might want to consider using a diplex filter. Vision produce a filter that was originally specified down to 40MHz but is now only guaranteed down to 88MHz although the company say it should still work down to 40MHz but they no longer check the alignment down that far. We were lucky that they offered to align one specially for the Dulwich museum.
Ensure your UHF modulator has an output equal or greater than the Aurora, then use the same planning rules.
You may be interested to see how a more complex system fits together: here is the Dulwich BVWS Museum installation:
https://www.dropbox.com/s/e6lc0cc7c5qx05...2.png?dl=0
or http://tinyurl.com/dulwichbvws
Although it uses only one amplifier, it is capable of considerable expansion if ever needed in the future. Note the 8dB pad on the output of the amplifier and the spare terminated feeds!
I was told to design a system that would last 25 years - not that I'll be around that long - I've done my best, though! The odd diplex filter for Band I/Band III, which cuts off part of Chanel B5 is actually a Cable TV reverse/forward path combiner that I scrounged from Virgin Media!
You will notice references to both 4 & 8-way taps and splitters and wonder what the difference is. Nothing is the answer! In this case they are same physical units, it is just their function that is different and the tap symbols don't need all the connectional information that the splitters do. Beware, though: all splitters can be used as taps but the reverse is not always true!
