F terminator caps reduce loss with oversized splitter? - Page 2 - AVS Forum
Forum Jump: 
 1Likes
 
Thread Tools
Old 06-26-2014, 02:09 PM
AVS Special Member
 
domino92024's Avatar
 
Join Date: Jan 2007
Posts: 3,472
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 41 Post(s)
Liked: 176
Quote:
Originally Posted by SAM64 View Post
It's also temperature related. but not a function of temperature, nor is it a function of frequency.
I'm guessing it's too complicated for you to grasp, so lets just say yes, you're right, you have a perfect understanding of RF circuits.
[Chuckle] I'll accept your sarcastic response. I think we have different definitions of "related" and "function of."


re·lat·ed

/rɪˈleɪtɪd/ Show Spelled [ri-ley-tid]
adjective 1. associated; connected.



What does "function of" mean?

Annotation category:
Chapter 3


The number of hours you spend toiling away in Butler library may be a function of the number of classes you're taking. It's also likely to be a function of whether you have a computer in your dorm room, whether you have a printer in your dorm room, and whether your roommates are loud and obnoxious and won't let you study.
domino92024 is offline  
Sponsored Links
Advertisement
 
Old 06-26-2014, 04:32 PM
AVS Special Member
 
egnlsn's Avatar
 
Join Date: May 2007
Location: Taylorsville, UT
Posts: 2,234
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
Does the insertion loss of passive devices increase as the frequency increases? Yes. Does anyone care why that's so? No. Most people just care and accept that that is the way it is. It doesn't even ruin their day. Is it because of the inefficiency of the device? Could be. Because of imperfect and/or inefficient materials, less than perfect design, workmanship, and/or whatever else, they lose more than the 3.01dB that a perfect 2-way splitter should lose. Who even cares why. Does it affect how an r.f. distribution system is designed and built? No. A1l most people care about is that the higher the frequency, the greater the loss.

Same thing with increase/decrease in attenuation due to changes in temperature. Sure, attenuation changes 0.11% per 1 degree change in temperature, but the vast majority of people don't even care why.

CIAO!

Ed N.

Last edited by egnlsn; 06-26-2014 at 04:45 PM.
egnlsn is offline  
Old 06-26-2014, 05:03 PM
 
SAM64's Avatar
 
Join Date: Apr 2011
Posts: 1,592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 74
Quote:
Originally Posted by domino92024 View Post

And yes, splitter loss is indeed higher at higher frequencies, believe it or not.
If you look at the graphs posted previously, attenuation is also higher at lower frequencies.
There is no predictable relationship between frequency and attenuation.
SAM64 is offline  
Old 06-27-2014, 08:21 AM
Senior Member
 
ADTech's Avatar
 
Join Date: Jun 2010
Location: St Louis
Posts: 464
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 36
Quote:
Originally Posted by SAM64 View Post
If you look at the graphs posted previously, attenuation is also higher at lower frequencies.
There is no predictable relationship between frequency and attenuation.
Look again at the graphs in post #26 and the final graph in post #30 . The others in post #30 are not properly terminated and the reflected signals from the open ports is causing an additive effect on the S21 trace.

For properly terminated splitters, attenuation does increase with frequency.

Tech support for Antennas Direct
ADTech is offline  
Old 06-27-2014, 08:24 AM
 
SAM64's Avatar
 
Join Date: Apr 2011
Posts: 1,592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 74
Quote:
Originally Posted by ADTech View Post
Look again at the graphs in post #26 and the final graph in post #30 . The others in post #30 are not properly terminated and the reflected signals from the open ports is causing an additive effect on the S21 trace.

For properly terminated splitters, attenuation does increase with frequency.

What is the mathematical 'function' that describes this?
SAM64 is offline  
Old 06-27-2014, 08:27 AM
Senior Member
 
ADTech's Avatar
 
Join Date: Jun 2010
Location: St Louis
Posts: 464
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 36
I just measure them and present the performance as measured.

I suppose that someone with the proper mathematical and RF theory could come up with the proper math if they wished. In any event, it would differ from one device manufacturer's implementation to the next.

Tech support for Antennas Direct
ADTech is offline  
Old 06-27-2014, 08:55 AM
AVS Special Member
 
AntAltMike's Avatar
 
Join Date: Nov 2001
Location: College Park, MD
Posts: 3,598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 29
Maybe ADTech can get his hands on some of those European 450 -1,750 MHz splitters that were commonplace a decade ago and scope out those. I was doing a large distribution system installation a few hundred miles from home and ran out of cable splitters, so I popped in half a dozen of those and had to return to the site to find and remove all of them a few days later. They decimated channels 4 and cable channel 17, which means they most likely developed steep notches at those frequencies that were harmonics of one another. That system didn't have a superband channel 31, but if it did, I bet there would have been a notch there, too.

Last edited by AntAltMike; 06-27-2014 at 10:04 AM.
AntAltMike is offline  
Old 06-27-2014, 09:14 AM
 
SAM64's Avatar
 
Join Date: Apr 2011
Posts: 1,592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 74
Quote:
Originally Posted by ADTech View Post
I just measure them and present the performance as measured.

I suppose that someone with the proper mathematical and RF theory could come up with the proper math if they wished. In any event, it would differ from one device manufacturer's implementation to the next.

So, as mentioned, there's no predictable relationship between frequency and attenuation.
SAM64 is offline  
Old 06-27-2014, 09:15 AM
AVS Special Member
 
AntAltMike's Avatar
 
Join Date: Nov 2001
Location: College Park, MD
Posts: 3,598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 29
Quote:
Originally Posted by egnlsn View Post
Does the insertion loss of passive devices increase as the frequency increases? Yes. Does anyone care why that's so? No. Most people just care and accept that that is the way it is. It doesn't even ruin their day.
It's ruined my day. I service a 120 room hotel that was built and wired back when it only had to carry the four local TV stations, channels 4, 5, 7 and 9, but now it instead distributes channel channels 14-53 (120-402 MHz). The signal at the beginning of each trunkline is flat at about 42dBmV, but by the time it travels through, in one instance, probably 80 feet of RG-11 sized cable splits in half and goes another 60 feet through RG59, the signal has developed a tilt of over 20dB.

I'm guessing that there are some 50 year old, VHF-rated splitters in the ceiling, but management won't left me find them and remove them, which would entail cutting several openings in a plaster ceiling. I have an "underground wiring finder" that would make it easy to locate where to cut, because each of the five trunklines simply splits in two, but even though they have had four managers and three building engineers since I have been their service person, none are willing to authorize that expense, so they instead pay me a few thousand dollars a year "robbing Peter to pay Paul" each time one guest complains that his TV picture is grainy, forcing me to swap out his correct value tap fur a stronger one and thereby further weakening the pass-through signal to the guest rooms below.

Last edited by AntAltMike; 06-27-2014 at 01:07 PM.
AntAltMike is offline  
Old 06-27-2014, 11:19 AM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,629
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 312
To the OP...
Whatever you decide, cap off unused ports.
Ratman is online now  
Old 06-27-2014, 12:20 PM
Senior Member
 
ADTech's Avatar
 
Join Date: Jun 2010
Location: St Louis
Posts: 464
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 36
Quote:
Originally Posted by SAM64 View Post
So, as mentioned, there's no predictable relationship between frequency and attenuation.
I suspect that if someone were suitably motivated, they could enter all the parts into a simulator like OrCAD/pSPICE and run a simulation of the circuit, they could predict its performance, with a reasonable degree of accuracy.

The basic building block of these splitters is a ferrite core transformer which will definitely have a relationship between frequency and its efficiency.


Quote:
Originally Posted by :
Maybe ADTech can get his hands on some of those European 450 -1,750 MHz splitters that were commonplace a decade ago and scope out those.
If you can get one to me, I'll characterize it. I like this new FieldFox....


Quote:
Originally Posted by :
To the OP...
Whatever you decide, cap off unused ports.
That's the "safest" route to take, following "best practices".

Tech support for Antennas Direct
ADTech is offline  
Old 06-27-2014, 12:32 PM
 
SAM64's Avatar
 
Join Date: Apr 2011
Posts: 1,592
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 19 Post(s)
Liked: 74
Quote:
Originally Posted by ADTech View Post
I suspect that if someone were suitably motivated, they could enter all the parts into a simulator like OrCAD/pSPICE and run a simulation of the circuit, they could predict its performance, with a reasonable degree of accuracy.

The basic building block of these splitters is a ferrite core transformer which will definitely have a relationship between frequency and its efficiency.
There'd be no need to run a simulation if attenuation was a function of frequency, as has been claimed.
I'm not disputing the fact that attenuation will be different at different frequencies in inexpensive, mass produced power dividers. I'm disputing the claim that attenuation is an inherent function of frequency in a power divider.
It is also possible to build a ferrite core transformer that is relatively flat across the desired passband, in which case, attenuation would change very little vs. frequency.

Attenuation is a function of length in a coax cable. Although attenuation will change vs. frequency, it is not a function of frequency.
I think it's the meaning of 'mathematical function' that is confusing some.
SAM64 is offline  
Old 06-27-2014, 01:14 PM
AVS Special Member
 
AntAltMike's Avatar
 
Join Date: Nov 2001
Location: College Park, MD
Posts: 3,598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 29
Well, attenuation tends to increase with frequency when the signal is subjected to inductance, but it also tends to decrease in response to an increase in capacitance.

Have you ever noticed that in relatively cheap, older signal processing devices (cough-Pico-cough), the air core inductors look nice and pretty but in the more expensive ones, they look shabby? I've been told by an old timer that in the expensive ones, they engineered the coil's induction value to be just a hair greater than the calculated, optimal value, and then "touch it up" by poking it with a non-inductive tool to reduce its actual inductance, and then they squirt it with whatever to hold it in place. I think you'd be hard pressed to develop a model that would accurately reflect the exact inductive reactance that takes place in a lot of passive RF components.

Last edited by AntAltMike; 06-27-2014 at 03:03 PM.
AntAltMike is offline  
Old 06-27-2014, 01:32 PM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,629
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 224 Post(s)
Liked: 312
This should be yet another ... "debate". I prefer term used in a Clint Eastwood movie.

"Best Practice" is the key term.
Don't over engineer. If you need 4 drops, use a 4-way splitter.
If you have to use a larger splitter, terminate unused ports.
RG6 is quite adequate.
Ratman is online now  
Old 06-27-2014, 11:15 PM
AVS Special Member
 
dr1394's Avatar
 
Join Date: May 2002
Location: Mizar 5
Posts: 3,171
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 21
Quote:
Originally Posted by SAM64 View Post
Attenuation is a function of length in a coax cable. Although attenuation will change vs. frequency, it is not a function of frequency.
Coax attenuation is a function of frequency. It can be simplified to:

a = K1 * sqrt(F) + K2 * F (dB/100 feet)

K1 is the resistive loss constant
K2 is the dielectric loss constant
F frequency in MHz

For example, Times Microwave LMR-400

(0.122290) * sqrt(F) + (0.000260) * F

http://www.timesmicrowave.com/cms/products/cables/lmr/

Ron

HD MPEG-2 Test Patterns http://www.w6rz.net
dr1394 is online now  
Old 07-02-2014, 07:45 PM
Advanced Member
 
nyctveng's Avatar
 
Join Date: Jun 2009
Location:
Posts: 839
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 63 Post(s)
Liked: 45
Quote:
Originally Posted by AntAltMike View Post
It's ruined my day. I service a 120 room hotel that was built and wired back when it only had to carry the four local TV stations, channels 4, 5, 7 and 9, but now it instead distributes channel channels 14-53 (120-402 MHz). The signal at the beginning of each trunkline is flat at about 42dBmV, but by the time it travels through, in one instance, probably 80 feet of RG-11 sized cable splits in half and goes another 60 feet through RG59, the signal has developed a tilt of over 20dB.

I'm guessing that there are some 50 year old, VHF-rated splitters in the ceiling, but management won't left me find them and remove them, which would entail cutting several openings in a plaster ceiling. I have an "underground wiring finder" that would make it easy to locate where to cut, because each of the five trunklines simply splits in two, but even though they have had four managers and three building engineers since I have been their service person, none are willing to authorize that expense, so they instead pay me a few thousand dollars a year "robbing Peter to pay Paul" each time one guest complains that his TV picture is grainy, forcing me to swap out his correct value tap fur a stronger one and thereby further weakening the pass-through signal to the guest rooms below.
As a former cable line tech that has seen the crap you describe, here are my recommendations: run the amp at an excessive tilt (highest channel used 20db above lowest channel used). As a comparison in the cable world typical tilts for 550mhz is 7, 750mhz is 10, 860mhz is 12. But since you have an extreme situation, the 20db tilt may help. Replace any passive equipment (barrels, splitters and fittings) you have access to. Anything rated to 1000MHz is fine. Good luck!
nyctveng is offline  
Old 07-03-2014, 09:47 AM
AVS Special Member
 
AntAltMike's Avatar
 
Join Date: Nov 2001
Location: College Park, MD
Posts: 3,598
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 31 Post(s)
Liked: 29
Quote:
Originally Posted by nyctveng View Post
As a former cable line tech that has seen the crap you describe, here are my recommendations: run the amp at an excessive tilt (highest channel used 20db above lowest channel used). As a comparison in the cable world typical tilts for 550mhz is 7, 750mhz is 10, 860mhz is 12. But since you have an extreme situation, the 20db tilt may help. Replace any passive equipment (barrels, splitters and fittings) you have access to. Anything rated to 1000MHz is fine. Good luck!
Not only did I do that, I put a Blonder Tongue BIDA distribution amplifier on the one of the five trunklines that incurs the most loss, while evenly splitting the output of the other BIDA.

I think a lot of people would be surprised how much, or should I say, how little an increase in signal power at the high end of the spectrum can be developed by increased tilting. Way back in the old days, we used to commonly "block tilt" our off-air VHF signals. In a typical market that had two VHF high and two VHF lowband channels, we could drive the same amplifier 2dB higher on highband if we simultaneously lowered the lowband by 4dB. But the range of that practice can't be extended, If we had eliminated the two VHF lowband channels completely, that would only enable us to increase the two VHF channels by 3dB.
AntAltMike is offline  
Old 07-03-2014, 07:04 PM
AVS Special Member
 
egnlsn's Avatar
 
Join Date: May 2007
Location: Taylorsville, UT
Posts: 2,234
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 31
Quote:
Originally Posted by AntAltMike View Post
It's ruined my day.
Mike, considering some of the headaches you keep finding yourself in, I think your day is ruined each morning when you first walk into the office...

CIAO!

Ed N.
egnlsn is offline  
 
Thread Tools


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off