|
Home > Archive > Electrical Engineering > June 2006 > Mains interference: Digital TV problems
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Mains interference: Digital TV problems
|
|
| Dave Ryman 2006-06-10, 11:21 am |
| Hi,
New to this group.
We have the now common set-top box for digital terrestrial TV and
radio.
We get a lot of intereference: Picture break-up and loud clicks in
the sound. Sometimes making a programme unwatchable. We have managed to
reduce this: The aerial has been upgraded, and there's a filter on the
mains socket supplying the AV equipment.
However, we get serious problems when we have a fan running, the
washing machine is on, or we run the hot tap (It's one of those that
turns on a heater to heat the water as you use it).
I'm not sure whether the appliances concerned are generating radio
waves, or are spiking the mains supply.
From the research I've done, the next step to getting a clean signal
is upgrading the aerial cable, filtering the appliances in question, or
providing a dedicated (shielded) ring main for the AV equipment.
The first and last options are going to be a bit tricky, since the
aerial cable is routed in walls and all sorts, and I'm not convinced that
a decicated ring main will be cost-effective.
Is it possible to put chokes or filters of some sort on the
offending appliances? If so, advice on how to do it would be most welcome
(240v 50Hz supply).
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| Roy L. Fuchs 2006-06-10, 4:21 pm |
| On Sat, 10 Jun 2006 13:38:33 GMT, Dave Ryman
<dave_rymanNOSPAM@hotmail.com> Gave us:
>Hi,
> New to this group.
>
> We have the now common set-top box for digital terrestrial TV and
>radio.
>
> We get a lot of intereference: Picture break-up and loud clicks in
>the sound. Sometimes making a programme unwatchable. We have managed to
>reduce this: The aerial has been upgraded, and there's a filter on the
>mains socket supplying the AV equipment.
>
> However, we get serious problems when we have a fan running, the
>washing machine is on, or we run the hot tap (It's one of those that
>turns on a heater to heat the water as you use it).
>
> I'm not sure whether the appliances concerned are generating radio
>waves, or are spiking the mains supply.
>
> From the research I've done, the next step to getting a clean signal
>is upgrading the aerial cable, filtering the appliances in question, or
>providing a dedicated (shielded) ring main for the AV equipment.
>
> The first and last options are going to be a bit tricky, since the
>aerial cable is routed in walls and all sorts, and I'm not convinced that
>a decicated ring main will be cost-effective.
>
> Is it possible to put chokes or filters of some sort on the
>offending appliances? If so, advice on how to do it would be most welcome
>(240v 50Hz supply).
My "off air" HDTV reception is fine using a set top antenna. Try
one. You mat find that your in wall post wire installation of your
antenna feed has some "RF holes" in it. At least big enough to let
the noise from an AC appliance affect it.
Rabbit ears work great on my HDTV off air tuner.
| |
| Dave Ryman 2006-06-10, 5:21 pm |
| Roy L. Fuchs <roylfuchs@urfargingicehole.org> wrote in
news:ni4m82hhemp8l4l444kb654mjdvsb3hvjp@4ax.com:
> On Sat, 10 Jun 2006 13:38:33 GMT, Dave Ryman
> <dave_rymanNOSPAM@hotmail.com> Gave us:
>
>
> My "off air" HDTV reception is fine using a set top antenna. Try
> one. You mat find that your in wall post wire installation of your
> antenna feed has some "RF holes" in it. At least big enough to let
> the noise from an AC appliance affect it.
>
> Rabbit ears work great on my HDTV off air tuner.
>
I don't think we'd get enough gain from such an aerial - our geography is
such that an aerial on the ground floor probably won't be high enough.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| Roy L. Fuchs 2006-06-10, 8:21 pm |
| On Sat, 10 Jun 2006 19:29:02 GMT, Dave Ryman
<dave_rymanNOSPAM@hotmail.com> Gave us:
>Roy L. Fuchs <roylfuchs@urfargingicehole.org> wrote in
>news:ni4m82hhemp8l4l444kb654mjdvsb3hvjp@4ax.com:
>
>
>I don't think we'd get enough gain from such an aerial - our geography is
>such that an aerial on the ground floor probably won't be high enough.
OK, then is the coaxial cable run from you antenna to your tuner a
"home run"? That is a single cable run, with no breaks (unions) in
it. You have at least ONE as you are getting it from a wall outlet.
If you have enough slack in the wall, remove the wall plate, and bring
the cable out "flying lead" style.
Each break costs a couple db.
| |
| Dave Ryman 2006-06-10, 10:21 pm |
| Roy L. Fuchs <roylfuchs@urfargingicehole.org> wrote in
news:jbkm82prm6p84u6iqktrnuj8ce3ru2c9b9@4ax.com:
(snip)
>
> OK, then is the coaxial cable run from you antenna to your tuner a
> "home run"? That is a single cable run, with no breaks (unions) in
> it. You have at least ONE as you are getting it from a wall outlet.
> If you have enough slack in the wall, remove the wall plate, and bring
> the cable out "flying lead" style.
>
> Each break costs a couple db.
>
That's interesting - mind you, I always thought that standard coax
connectors were a bit hopeless. Any thoughts on BNC connectors?
Looks like I'm going to try replacing the whole length of cable - wish me
luck!
Thanks for all the advice.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| TimPerry 2006-06-10, 11:21 pm |
|
"Dave Ryman" <dave_rymanNOSPAM@hotmail.com> wrote in message
news:Xns97DF11BEC2C03daverymanhotmailcom@80.5.182.99...
> Roy L. Fuchs <roylfuchs@urfargingicehole.org> wrote in
> news:jbkm82prm6p84u6iqktrnuj8ce3ru2c9b9@4ax.com:
>
> (snip)
>
> That's interesting -
its also bullshit. conector loss might be about .1 dB.... granted having as
few splices as possible is a good thing.
mind you, I always thought that standard coax
> connectors were a bit hopeless. Any thoughts on BNC connectors?
>
> Looks like I'm going to try replacing the whole length of cable - wish me
> luck!
>
> Thanks for all the advice.
>
> --
> Regards,
> Dave
check the easy stuff first.... look for an open balun on the antenna.
| |
| Dave Ryman 2006-06-11, 12:21 am |
| "TimPerry" <timperry@noaspamadelphia.net> wrote in
news:mbCdnQUolIUf7xbZnZ2dnUVZ_sKdnZ2d@adelphia.com:
(snip)
>
> check the easy stuff first.... look for an open balun on the antenna.
>
>
>
Just looked "balun" up. Our aerial is a new one. Am I right in assuming
that, in our case, the balun is the small plastic box on it within which
the coax attaches? If so, ours is fine. If I'm talking B*llocks, then
someone please let me know!
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| TimPerry 2006-06-11, 1:21 am |
| Dave Ryman wrote:
> "TimPerry" <timperry@noaspamadelphia.net> wrote in
> news:mbCdnQUolIUf7xbZnZ2dnUVZ_sKdnZ2d@adelphia.com:
>
> (snip)
>
> Just looked "balun" up. Our aerial is a new one. Am I right in
> assuming that, in our case, the balun is the small plastic box on it
> within which the coax attaches? If so, ours is fine. If I'm talking
> B*llocks, then someone please let me know!
restating the problem:
" We have the now common set-top box for digital terrestrial TV and
radio.
We get a lot of intereference: Picture break-up and loud clicks in
the sound. Sometimes making a programme unwatchable. We have managed to
reduce this: The aerial has been upgraded, and there's a filter on the
mains socket supplying the AV equipment.
However, we get serious problems when we have a fan running, the
washing machine is on, or we run the hot tap (It's one of those that
turns on a heater to heat the water as you use it)."
to me this sounds like some kind of ground loop issue.
just try this: get 2 low cost baluns. they have an F connector at one end
and 2 spade terminals at the other. solder or bolt the spade lugs to the
matching lug on the other balun. this gives you a 75 ohm F connector at
each end. add a short male - male cable and insert at the receiver.
if this does not help then you can proceed to more drastic and expensive
measures.
| |
|
| Dave,
I assume you are in the UK.
It's possible that the signal strength of the digital signal is marginal -
which would cause the symptoms which you describe.
The digital transmitters run at much lower powers (~10kW ERP) as opposed to
the analogue transmitters (can be up to 1MW ERP) and although you may be
getting a satisfactory analogue signal the digital signal may well be near
to the receiver's threshold.
John
| |
| Roy L. Fuchs 2006-06-11, 10:21 am |
| On Sun, 11 Jun 2006 00:52:11 GMT, Dave Ryman
<dave_rymanNOSPAM@hotmail.com> Gave us:
>Roy L. Fuchs <roylfuchs@urfargingicehole.org> wrote in
>news:jbkm82prm6p84u6iqktrnuj8ce3ru2c9b9@4ax.com:
>
>(snip)
>
>That's interesting - mind you, I always thought that standard coax
>connectors were a bit hopeless. Any thoughts on BNC connectors?
>
>Looks like I'm going to try replacing the whole length of cable - wish me
>luck!
>
>Thanks for all the advice.
Standard F are fine done properly, and you will end up with more
breaks in the line with BNC as you'll have to convert back and forth
to hook up to all standard gear.
Yes, you must cut (prepare) all of your fittings carefully as well.
| |
| Roy L. Fuchs 2006-06-11, 10:21 am |
| On Sun, 11 Jun 2006 05:12:22 -0000, "John" <no@spam.tnx> Gave us:
>Dave,
>
>I assume you are in the UK.
>
>It's possible that the signal strength of the digital signal is marginal -
>which would cause the symptoms which you describe.
>
>The digital transmitters run at much lower powers (~10kW ERP) as opposed to
>the analogue transmitters (can be up to 1MW ERP) and although you may be
>getting a satisfactory analogue signal the digital signal may well be near
>to the receiver's threshold.
>
It all comes down to bit-error-rate with digital. Once it gets
above ten percent, anomalies start to appear. Picture gets artifacts,
sound drops out... entire image goes blank, no sound.
| |
| Dave Ryman 2006-06-11, 10:21 am |
| Roy L. Fuchs <roylfuchs@urfargingicehole.org> wrote in
news:cq3o82t8kv2515bt7d53q7in7iksmho07u@4ax.com:
> On Sun, 11 Jun 2006 00:52:11 GMT, Dave Ryman
> <dave_rymanNOSPAM@hotmail.com> Gave us:
>
>
> Standard F are fine done properly, and you will end up with more
> breaks in the line with BNC as you'll have to convert back and forth
> to hook up to all standard gear.
>
> Yes, you must cut (prepare) all of your fittings carefully as well.
>
By F-connector, I assume you mean the metal connectors that screw onto
the shielding and outer insulation of the coax, and have the wire sliding
into a centre pin. I have actually found one of these now that has a grub
screw to make contact with the main centre wire: The standard ones here
in the UK usually just depend on the wire making contact by sliding into
centre pin - very hit and miss. I trust the average water pipe connector
more - and that's saying something!
Thanks for that.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| Dave Ryman 2006-06-11, 10:21 am |
| "TimPerry" <timperry@noaspamadelphia.net> wrote in
news:9vedndrrue5uChbZnZ2dnUVZ_rednZ2d@adelphia.com:
> Dave Ryman wrote:
>
>
>
> restating the problem:
>
>
> " We have the now common set-top box for digital terrestrial TV and
> radio.
>
> We get a lot of intereference: Picture break-up and loud clicks in
> the sound. Sometimes making a programme unwatchable. We have managed
> to reduce this: The aerial has been upgraded, and there's a filter on
> the mains socket supplying the AV equipment.
>
> However, we get serious problems when we have a fan running, the
> washing machine is on, or we run the hot tap (It's one of those that
> turns on a heater to heat the water as you use it)."
>
> to me this sounds like some kind of ground loop issue.
>
> just try this: get 2 low cost baluns. they have an F connector at
> one end and 2 spade terminals at the other. solder or bolt the spade
> lugs to the matching lug on the other balun. this gives you a 75 ohm
> F connector at each end. add a short male - male cable and insert at
> the receiver.
>
> if this does not help then you can proceed to more drastic and
> expensive measures.
>
>
>
>
That sounds like it's worth a try. I've being researching baluns, and my
head now hurts.
Thanks.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| Dave Ryman 2006-06-11, 10:21 am |
| "John" <no@spam.tnx> wrote in
news:cLqdnQRV0_beOxbZnZ2dnUVZ8t2dnZ2d@pipex.net:
> Dave,
>
> I assume you are in the UK.
>
> It's possible that the signal strength of the digital signal is
> marginal - which would cause the symptoms which you describe.
>
> The digital transmitters run at much lower powers (~10kW ERP) as
> opposed to the analogue transmitters (can be up to 1MW ERP) and
> although you may be getting a satisfactory analogue signal the digital
> signal may well be near to the receiver's threshold.
>
> John
>
>
>
Yep, in the UK. I suspected the power used to transmit digital was poor,
but didn't realise it was just 10% of the analogue signal - no wonder
there's problems! And they want to turn the analogue off in a few years:
They must have shares in the companies that make aerials, coax and
connectors.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
|
|
"Dave Ryman" <dave_rymanNOSPAM@hotmail.com> wrote in message
news:Xns97DF9006DD04Bdaverymanhotmailcom@62.253.170.163...
> "John" <no@spam.tnx> wrote in
> news:cLqdnQRV0_beOxbZnZ2dnUVZ8t2dnZ2d@pipex.net:
>
>
> Yep, in the UK. I suspected the power used to transmit digital was poor,
> but didn't realise it was just 10% of the analogue signal - no wonder
> there's problems! And they want to turn the analogue off in a few years:
> They must have shares in the companies that make aerials, coax and
> connectors.
I live in a margin area and the digital signal does tend to drop in and out
exactly as you described in your OP. It's also weather dependent - the
hotter summer months are worse than the cooler winter/spring/autumn.
I would firstly check all your connectors and ensure the centre of the coax
is soldered to the centre pin of the plug (for our US readers we don't use F
type connectors for Terrestrial TV, they're only found on satellite
equipment).
Then if the connections are sound, try to get the aerial checked by someone
who is competent - look for someone who is a member of the CAI Confederation
of Aerial Installers.
Like you I'm also wondering about the analogue turn-off. Hopefully they may
be able to up the power of the digital transmitters once the analogue TXs
are turned off and the spectrum is freed up.
John
| |
| Dave Ryman 2006-06-11, 1:21 pm |
| "John" <no@spam.tnx> wrote in news:X42dnTtOVJhUghHZRVnyhA@pipex.net:
(snip)
>
> Like you I'm also wondering about the analogue turn-off. Hopefully
> they may be able to up the power of the digital transmitters once the
> analogue TXs are turned off and the spectrum is freed up.
>
>
> John
>
>
>
I wouldn't bank on it - the whole thing seems to have been done on the
cheap with no real thought for the viewers - logos on-screen which risk
burning-in on your TV screen, a useless program guide, and equally useless
Teletext.
Thank god for DVDs!
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| TimPerry 2006-06-11, 1:21 pm |
| Dave Ryman wrote:
> "TimPerry" <timperry@noaspamadelphia.net> wrote in
> news:9vedndrrue5uChbZnZ2dnUVZ_rednZ2d@adelphia.com:
>
>
> That sounds like it's worth a try. I've being researching baluns, and
> my head now hurts.
>
> Thanks.
i had assumed that you had a solid signal except for the interfearance
problem.
another possibility is to add a inline amplifier at the antenna. this boosts
the signal a bit. it is then attenuated by the length of coax. a inline
section is added at the receiver end to power the amplifier.
it looks something like this http://www.audio-discounters.com/bia-20.html
check to see if there are any splitters in line and remove if possable.
i understand you have to license TVs in the UK. would that apply to the
digital box too?
| |
| Adrian C 2006-06-11, 1:21 pm |
| John wrote:
> Then if the connections are sound, try to get the aerial checked by someone
> who is competent - look for someone who is a member of the CAI Confederation
> of Aerial Installers.
CAI membership doesn't count for much. Their members give *very*
variable service. Some of the best aerial installation companies out
there are not members so go by recommendation and read up on the
subject. If it costs more than £100 just to get a decent aerial signal
you may as well go FreeSat.
See uk.tech.digital-tv where this would be more on-topic.
--
Adrian C
| |
|
|
> CAI membership doesn't count for much. Their members give *very*
> variable service. Some of the best aerial installation companies out there
> are not members so go by recommendation and read up on the subject. If it
> costs more than £100 just to get a decent aerial signal you may as well go
> FreeSat.
I always considered CAI installers as being in the upper part of the quality
range - perhaps mistakenly.
For a marginal area Freesat is probably the way to go. I've also heard some
horror stories about Sky's installers but that's for another thread.
Good Luck
John
| |
| Roy L. Fuchs 2006-06-11, 5:21 pm |
| On Sun, 11 Jun 2006 13:07:28 GMT, Dave Ryman
<dave_rymanNOSPAM@hotmail.com> Gave us:
>Yep, in the UK. I suspected the power used to transmit digital was poor,
>but didn't realise it was just 10% of the analogue signal - no wonder
>there's problems!
Not really. A digital signal is easier to retain all of the signal
due to the fact that a lot of FEC gets sent with the signal.
> And they want to turn the analogue off in a few years:
>They must have shares in the companies that make aerials, coax and
>connectors.
I use a regular antenna. I only miss on one local station. I
actually used to get it from 20 miles farther away... over hills.
| |
|
|
"Dave Ryman" <dave_rymanNOSPAM@hotmail.com> wrote in message
news:Xns97DE9521329E0daverymanhotmailcom@80.5.182.99...
> Hi,
> New to this group.
>
> We have the now common set-top box for digital terrestrial TV and
> radio.
>
> We get a lot of intereference: Picture break-up and loud clicks in
> the sound.
That is an indication of a weak signal - I had all the symptoms, and the CC
fixed it using
a pre-amp.
(The digital device can normally fill in any small holes, if it has enough
to work with orginally. If it gets more spurious signal with the weak
desired signal -fan hash, e.g., , it will manifest itself more strongly)
Check the "db" of the incoming signal to make sure you have the required
signal strength.
Sometimes making a programme unwatchable. We have managed to
> reduce this: The aerial has been upgraded, and there's a filter on the
> mains socket supplying the AV equipment.
>
> However, we get serious problems when we have a fan running, the
> washing machine is on, or we run the hot tap (It's one of those that
> turns on a heater to heat the water as you use it).
>
> I'm not sure whether the appliances concerned are generating radio
> waves, or are spiking the mains supply.
>
> From the research I've done, the next step to getting a clean signal
> is upgrading the aerial cable, filtering the appliances in question, or
> providing a dedicated (shielded) ring main for the AV equipment.
>
> The first and last options are going to be a bit tricky, since the
> aerial cable is routed in walls and all sorts, and I'm not convinced that
> a decicated ring main will be cost-effective.
>
> Is it possible to put chokes or filters of some sort on the
> offending appliances? If so, advice on how to do it would be most welcome
> (240v 50Hz supply).
>
> --
> Regards,
> Dave
>
> dave_ryman@hotmailNOSPAM.com
> http://welcome.to/daves.website
> http://travel.to/formula.one
| |
|
| One other item - each splitter degrades the signal by 3 db minimum (cuts it
in half) just to split - so you get one-fourth of the input out of each leg
of a two-way splitter - so minimize the splitters
(and to minimize noise - if you connect to two input devices with input
coax -dvd, vcr, cable box, etc, - you need to break the metal shield on the
coax at the input to one of the devices or the resultant loop of coax
shield-chassis-coax shield/ chassis-coax shield will act like a loop antenna
for nearby interference)
"Dave Ryman" <dave_rymanNOSPAM@hotmail.com> wrote in message
news:Xns97DE9521329E0daverymanhotmailcom@80.5.182.99...
> Hi,
> New to this group.
>
> We have the now common set-top box for digital terrestrial TV and
> radio.
>
> We get a lot of intereference: Picture break-up and loud clicks in
> the sound. Sometimes making a programme unwatchable. We have managed to
> reduce this: The aerial has been upgraded, and there's a filter on the
> mains socket supplying the AV equipment.
>
> However, we get serious problems when we have a fan running, the
> washing machine is on, or we run the hot tap (It's one of those that
> turns on a heater to heat the water as you use it).
>
> I'm not sure whether the appliances concerned are generating radio
> waves, or are spiking the mains supply.
>
> From the research I've done, the next step to getting a clean signal
> is upgrading the aerial cable, filtering the appliances in question, or
> providing a dedicated (shielded) ring main for the AV equipment.
>
> The first and last options are going to be a bit tricky, since the
> aerial cable is routed in walls and all sorts, and I'm not convinced that
> a decicated ring main will be cost-effective.
>
> Is it possible to put chokes or filters of some sort on the
> offending appliances? If so, advice on how to do it would be most welcome
> (240v 50Hz supply).
>
> --
> Regards,
> Dave
>
> dave_ryman@hotmailNOSPAM.com
> http://welcome.to/daves.website
> http://travel.to/formula.one
| |
| Floyd L. Davidson 2006-06-11, 7:21 pm |
| "hob" <dehoberg@comcast.net> wrote:
>One other item - each splitter degrades the signal by 3 db minimum (cuts it
That is correct. You should have stopped there!
>in half) just to split - so you get one-fourth of the input out of each leg
>of a two-way splitter - so minimize the splitters
If it were 1/4 to each leg, that would be a 6 dB drop, but it is
3 dB as you stated originally. That is half the power to each
leg. (In fact it is probably more like 3.5 dB, as there is
approximately 0.5 dB loss in the hybrid itself.)
> (and to minimize noise - if you connect to two input devices with input
>coax -dvd, vcr, cable box, etc, - you need to break the metal shield on the
>coax at the input to one of the devices or the resultant loop of coax
>shield-chassis-coax shield/ chassis-coax shield will act like a loop antenna
>for nearby interference)
That is not true. Do *not* break the outer shield on a coax
cable. If you do, then you *will* get an antenna instead of a
transmission line.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Roy L. Fuchs 2006-06-11, 7:21 pm |
| On Sun, 11 Jun 2006 13:22:48 -0800, floyd@apaflo.com (Floyd L.
Davidson) Gave us:
>"hob" <dehoberg@comcast.net> wrote:
>
>That is correct. You should have stopped there!
>
>
>If it were 1/4 to each leg, that would be a 6 dB drop, but it is
>3 dB as you stated originally. That is half the power to each
>leg. (In fact it is probably more like 3.5 dB, as there is
>approximately 0.5 dB loss in the hybrid itself.)
>
>
>That is not true. Do *not* break the outer shield on a coax
>cable. If you do, then you *will* get an antenna instead of a
>transmission line.
I think that he was thinking of an audio patch cord there.
| |
|
|
"Floyd L. Davidson" <floyd@apaflo.com> wrote in message
news:87pshfqtef.fld@apaflo.com...
> "hob" <dehoberg@comcast.net> wrote:
it[color=darkred]
>
> That is correct. You should have stopped there!
>
leg[color=darkred]
>
> If it were 1/4 to each leg, that would be a 6 dB drop, but it is
> 3 dB as you stated originally. That is half the power to each
> leg. (In fact it is probably more like 3.5 dB, as there is
> approximately 0.5 dB loss in the hybrid itself.)
No, I did mean 3 db for the fact that the device is in the line trying to
separate a signal, and
the original signal is split in half because it is a two-way splitter (or in
thirds because it is a three-way splitter)
thus giving 6 db between input and either output leg on a two way splitter.
Why?
First, if the splitter was only 3 db, as you say, it would be a perfect
device - each leg output being half the input. Sorry, 100% efficient
devices are not possible, especially in RF. And there would be no isolation
and no impedance matching.
Second, the equations predict a "cost" of 3 db to split a signal in a
perfect splitter - either from impedance mismatch externally to the fed
devices, or due to the parts needed to match impedance internally so as to
have matched external impedance.
Passive splitters are theoretically at best 50% efficient (check the
equations - we had to do them in lab) - that means half the power is lost in
the split, not half the power goes out each leg.
A detailed lay explanation can be found at
http://www.minicircuits.com/appnote/an10006.pdf
>
the[color=darkred]
antenna[color=darkred]
>
> That is not true. Do *not* break the outer shield on a coax
> cable. If you do, then you *will* get an antenna instead of a
> transmission line.
Having worked as the cable grounding specialist for the largest
ballistic missle warning system, with its multimegawatt transmitters and
"many db" amplifiers all around, I can assure you that ALL of our coax that
could create a loop due to shield path was made without the shield in place
on ONE end only, and that I have PERSONALLY removed coax with shields
connected at both ends that were put in by otherwise competent techs, coax
that was seriously degrading portions of that system, and replaced those
leads with coax with the shield grounded at ONE end in order to restore the
system.
I DID mean to make the poster's shields end up in a "spider" grounded
system, using the system input coax as the mother lead, and breaking the end
loop made by coax shields and chassis-power grounds at one point in the coax
shield.
OK, perhaps he has two-prong rather than three-prong power plugs, and he
has floating chassis grounds rather than fixed, and he could get by with
having a weak loop if his digital signal was strong enough for the
discriminator to weed out too-sharp rise-time noise and only pass pro forma
pulses - but it sure sounds like he does not have a strong enough
signal-noise ratio for his (crap?) discriminator, and he needs whatever he
can get.
fwiw...
>
> --
> Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
> Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Roy L. Fuchs 2006-06-11, 11:21 pm |
| On Sun, 11 Jun 2006 19:18:46 -0500, "hob" <dehoberg@comcast.net> Gave
us:
snippitude
> but it sure sounds like he does not have a strong enough
>signal-noise ratio for his (crap?) discriminator, and he needs whatever he
>can get.
>
So now you are saying that he needs a new/better tuner?
Bit-error-rate... that is the term you are looking for.
| |
| Floyd L. Davidson 2006-06-12, 12:21 am |
| "hob" <dehoberg@comcast.net> wrote:
>"Floyd L. Davidson" <floyd@apaflo.com> wrote:
>
>thus giving 6 db between input and either output leg on a two way splitter.
>
>Why?
>
>First, if the splitter was only 3 db, as you say, it would be a perfect
>device - each leg output being half the input. Sorry, 100% efficient
>devices are not possible, especially in RF. And there would be no isolation
>and no impedance matching.
>
>Second, the equations predict a "cost" of 3 db to split a signal in a
>perfect splitter - either from impedance mismatch externally to the fed
>devices, or due to the parts needed to match impedance internally so as to
>have matched external impedance.
> Passive splitters are theoretically at best 50% efficient (check the
>equations - we had to do them in lab) - that means half the power is lost in
>the split, not half the power goes out each leg.
>
>A detailed lay explanation can be found at
>
> http://www.minicircuits.com/appnote/an10006.pdf
Not too bad, but you failed to actually *read* it. It does
*not* say the trans-hybrid loss will be 3 dB. What it actually
says is that it will be "less than a 3dB loss". As I noted, it
is usually about 0.5 dB. That is a *long* ways from 3 dB.
Typical insertion loss for a hybrid is between 3.5 and 4.5 dB.
>
> Having worked as the cable grounding specialist for the largest
>ballistic missle warning system, with its multimegawatt transmitters and
>"many db" amplifiers all around, I can assure you that ALL of our coax that
>could create a loop due to shield path was made without the shield in place
>on ONE end only, and that I have PERSONALLY removed coax with shields
>connected at both ends that were put in by otherwise competent techs, coax
>that was seriously degrading portions of that system, and replaced those
>leads with coax with the shield grounded at ONE end in order to restore the
>system.
Me thinks you are blowing blue smoke. What you describe might
work for certain applications, but it *won't* work in the one
under discussion.
And I would be *very* surprised if it can be made to work with
"multimegawatt transmitters".
It appears that you don't understand what ground loops are or
how they work, not to mention having not a clue what a
transmission line does or does not do.
> I DID mean to make the poster's shields end up in a "spider" grounded
>system, using the system input coax as the mother lead, and breaking the end
>loop made by coax shields and chassis-power grounds at one point in the coax
>shield.
That is a really dumb idea. I realize that a lot of people use
that for audio, where the "transmission line" is very very very
short compared to the wavelength of the signal. It isn't the
best way to do it there either, but it actually can work. But
doing that at RF is absurd.
> OK, perhaps he has two-prong rather than three-prong power plugs, and he
>has floating chassis grounds rather than fixed, and he could get by with
>having a weak loop if his digital signal was strong enough for the
>discriminator to weed out too-sharp rise-time noise and only pass pro forma
>pulses - but it sure sounds like he does not have a strong enough
>signal-noise ratio for his (crap?) discriminator, and he needs whatever he
>can get.
Now I'm *positive* that you are blowing blue smoke. That's meaningless
verbiage.
>fwiw...
Nothing.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Dave Ryman 2006-06-12, 5:21 am |
| "TimPerry" <timperry@noaspamadelphia.net> wrote in
news:j8idnSZc88CjohHZnZ2dnUVZ_o2dnZ2d@adelphia.com:
> Dave Ryman wrote:
>
> i had assumed that you had a solid signal except for the interfearance
> problem.
>
> another possibility is to add a inline amplifier at the antenna. this
> boosts the signal a bit. it is then attenuated by the length of coax.
> a inline section is added at the receiver end to power the amplifier.
Got an in-line amplifer already, thanks.
>
> it looks something like this
> http://www.audio-discounters.com/bia-20.html
>
> check to see if there are any splitters in line and remove if
> possable.
Need an inline spliiter, I'm afraid - these digital boxes only give one
output, so you need more than one digibox if you want to record a program
whilst watching another.
>
> i understand you have to license TVs in the UK. would that apply to
> the digital box too?
>
>
>
Yep, any TV that's capable of receiving and displaying a TV picture.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
| Dave Ryman 2006-06-12, 5:21 am |
| "hob" <dehoberg@comcast.net> wrote in
news:wdSdncL477zIGRHZnZ2dneKdnZydnZ2d@comcast.com:
>
> "Dave Ryman" <dave_rymanNOSPAM@hotmail.com> wrote in message
> news:Xns97DE9521329E0daverymanhotmailcom@80.5.182.99...
>
> That is an indication of a weak signal - I had all the symptoms, and
> the CC fixed it using
> a pre-amp.
>
> (The digital device can normally fill in any small holes, if it has
> enough
> to work with orginally. If it gets more spurious signal with the weak
> desired signal -fan hash, e.g., , it will manifest itself more
> strongly)
>
> Check the "db" of the incoming signal to make sure you have the
> required signal strength.
Signal strength is usually ok.
(snip)
>
>
>
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
|
|
"Floyd L. Davidson" <floyd@apaflo.com> wrote in message
news:87bqszqfbf.fld@apaflo.com...
> "hob" <dehoberg@comcast.net> wrote:
splitter.[color=darkred]
isolation[color=darkred]
to[color=darkred]
in[color=darkred]
>
> Not too bad, but you failed to actually *read* it.
speaking of not reading it -
see page 1 - insertion loss table
see page 4, last paragraph - "A signal applied to port A will be routed to
port S, minus a 3dB loss in the internal resistor;"
(you will note why that resistor has to be in there on page 2, and the
effect of mismatch on page 5 - note the 6db loss INSIDE the splitter for
mismatch)
I knew the loss was there before I read it - and it is the same after I read
it.
It seems you failed to understand what you read: You lose the 3 db
internally to match impedance, or you lose 3 db or more in the impedance
mismatch. Either way, it costs you 3db loss to insert it into the line.
(Having two outputs side by side is called a parallel ciruit, and their
impedance is half of what was there before the splitter was inserted. And
maximum power transfer ONLY occurs at matched impedance.)
> It does
> *not* say the trans-hybrid loss will be 3 dB. What it actually
> says is that it will be "less than a 3dB loss".
That out-of-context quote of yours refers to the mismatch effect on
insertion loss - read what it said just before that sentence.
It is talking about the effect of shorting out one of the ports -
"THE POWER LOSS WITHIN THE POWER SPLITTER WOULD BE... -6DB of the orginal
signal power [power in- at the input port]"
And IF you short out port B output, in a real splitter, the LOSS at port
A output (not the signal) is less than 3 db because some of the signal power
that was to go to port B (that cannot leave because of the short) is
reflected into port A.
As I noted, it
> is usually about 0.5 dB. That is a *long* ways from 3 dB.
No, in a real passive filter, it is closer to 3.5 db. The internal loss is
well known and accepted.
>
> Typical insertion loss for a hybrid is between 3.5 and 4.5 dB.
>
that is along way from .5 db you had just claimed. Apparently you are
unfamiliar with the term "insertion loss"
that[color=darkred]
place[color=darkred]
coax[color=darkred]
the[color=darkred]
>
> Me thinks you are blowing blue smoke. What you describe might
> work for certain applications, but it *won't* work in the one
> under discussion.
Special RF in those RG-59/58/6 coax lines of the poster? His Rf doesn't
follow the laws of physics?
So why don't you enlighten us and tell us WHY shield breaks and "spiders"
won't isolate the signal from RFI/EMI? There are a couple of authors on rf
grounding systems I will pass it on to.
> And I would be *very* surprised if it can be made to work with
"multimegawatt transmitters".
perhaps that is why they paid me to do it as a civilian specialist, and not
you
( the actual size and db of the equipment was classified, fwiw )
>
> It appears that you don't understand what ground loops are or
> how they work, not to mention having not a clue what a
> transmission line does or does not do.
>
Well, I was the BMEWs site grounding engineering tech for two years and have
no need to blow smoke - and I have a good deal of experience working with L
band and X-band grounds.
- but since you apparently have no real world experience, try a book on
grounding and look up "spider" systems.
end[color=darkred]
coax[color=darkred]
>
> That is a really dumb idea. I realize that a lot of people use
> that for audio, where the "transmission line" is very very very
> short compared to the wavelength of the signal. It isn't the
> best way to do it there either, but it actually can work. But
> doing that at RF is absurd.
>
Well, apparently you have zip experience working with RFI/EMI grounding
systems or antennas, or you wouldn't say such stupid things.
To make it very simple for you hobbyists - Ever see a TV antenna
collector/reflector rod? How long is it and what is that signal's
wavelength? And the length of loop made by equipment and connectors? And
what is that signal's wavelength?
And who uses coax for audio or worries about transmission line length vs
wavelength in audio cables? Monster cable buyers? That statement is so far
out of reality it says you are clueless in this field
your 60 cycle hum is not in the same field as RFI
he[color=darkred]
forma[color=darkred]
he[color=darkred]
>
> Now I'm *positive* that you are blowing blue smoke. That's meaningless
> verbiage.
>
That statement clearly and unequivocably says that you have absolutely no
clue how grounding is done inside equipment and that you have no real
professional experience whatsoever.
>
> Nothing.
>
> --
> Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
> Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
|
|
"Roy L. Fuchs" <roylfuchs@urfargingicehole.org> wrote in message
news:nrgp82hjsjmb6e8t9hrcq972mq1cv1pa6b@4ax.com...
> On Sun, 11 Jun 2006 19:18:46 -0500, "hob" <dehoberg@comcast.net> Gave
> us:
>
> snippitude
>
he[color=darkred]
> So now you are saying that he needs a new/better tuner?
>
> Bit-error-rate... that is the term you are looking for.
Not in the computer sense per se, but perhaps in the digital signal to ratio
sense -
What I am saying is that since digital "amplifiers" do not amplify a
signal as in an analog fashion, passing all the signal, rather they recreate
the signal at a higher level, isolating at each stage, they are susceptible
to creating extra bad bits in noisy signals. That is, while an analog amp
will give you most of what comes in, a digital output is only reproduced
timewise from what comes in. ( Removes the need for good filters and such.)
A digital amplifier's weakness, however, is that it is looking for a bit
to reproduce - and it only "recognizes" signals having a certain
risetime -it discriminates against signals rising too fast for the amp ( it
can't "trigger") and it discriminates against signals that rise too slowly
(the "gate" set by the clock closes before the voltage level rises enough to
trigger an output).
If you look at a real digital signal, you will see a series of pulses and
noise (non-signal) - if the noise is well below the level of the pulses,
the pulses drive the amplifier and low noise has little effect - the bits
are pretty much repeated and thus cleanly reproduced at a higher level.
If the level of the pulses are weak and thus are in the noise, then there
are too many false output bits created for a coherent signal. It does not
discriminate properly between noise and signal.
( A "better" tuner would fix the problem only because the word better
implies one that does not have the problem -)
a "better at handling low signal to noise ratio" tuner might fix the
problem, but normally you get what they give you and expect you to get the
signal up to a certain strength and above a certain signal to noise ratio
(if that ability to discriminate in high noise is what is called
bit-error-rate in digital signals, then that is the term.)
fwiw
| |
| Roy L. Fuchs 2006-06-12, 7:21 am |
| On Mon, 12 Jun 2006 02:57:45 -0500, "hob" <dehoberg@comcast.net> Gave
us:
> And who uses coax for audio or worries about transmission line length vs
>wavelength in audio cables? Monster cable buyers? That statement is so far
>out of reality it says you are clueless in this field
Nearly ALL audio RCA type patch cables are a pair of coaxial runs.
I cannot think off any that are not. Can you? Give the model and
brand name here:
| |
| Roy L. Fuchs 2006-06-12, 7:21 am |
| On Mon, 12 Jun 2006 02:57:45 -0500, "hob" <dehoberg@comcast.net> Gave
us:
>
>your 60 cycle hum is not in the same field as RFI
No shit, Dip Tracy.
| |
| Roy L. Fuchs 2006-06-12, 7:21 am |
| On Mon, 12 Jun 2006 03:30:34 -0500, "hob" <dehoberg@comcast.net> Gave
us:
>
>"Roy L. Fuchs" <roylfuchs@urfargingicehole.org> wrote in message
>news:nrgp82hjsjmb6e8t9hrcq972mq1cv1pa6b@4ax.com...
>he
>
>Not in the computer sense per se, but perhaps in the digital signal to ratio
>sense -
Wrong. It is not a term that has ANYTHING to do with computers.
It is a COMMON term used in MPEG2/digital audio video transmission
jargon.
ANY compressed, digital, A/V signal that uses FEC for correction of
bad data uses THIS TERM to refer to signal quality.
>
> What I am saying is that since digital "amplifiers" do not amplify a
>signal as in an analog fashion, passing all the signal, rather they recreate
>the signal at a higher level, isolating at each stage, they are susceptible
>to creating extra bad bits in noisy signals. That is, while an analog amp
>will give you most of what comes in, a digital output is only reproduced
>timewise from what comes in. ( Removes the need for good filters and such.)
>
> A digital amplifier's weakness, however, is that it is looking for a bit
>to reproduce - and it only "recognizes" signals having a certain
>risetime -it discriminates against signals rising too fast for the amp ( it
>can't "trigger") and it discriminates against signals that rise too slowly
>(the "gate" set by the clock closes before the voltage level rises enough to
>trigger an output).
>
> If you look at a real digital signal, you will see a series of pulses and
>noise (non-signal) - if the noise is well below the level of the pulses,
>the pulses drive the amplifier and low noise has little effect - the bits
>are pretty much repeated and thus cleanly reproduced at a higher level.
> If the level of the pulses are weak and thus are in the noise, then there
>are too many false output bits created for a coherent signal. It does not
>discriminate properly between noise and signal.
>
> ( A "better" tuner would fix the problem only because the word better
>implies one that does not have the problem -)
> a "better at handling low signal to noise ratio" tuner might fix the
>problem, but normally you get what they give you and expect you to get the
>signal up to a certain strength and above a certain signal to noise ratio
>
>(if that ability to discriminate in high noise is what is called
>bit-error-rate in digital signals, then that is the term.)
You keep forgetting about the FEC. There is enough of it such that
even missing segments of the signal can be replaced. Up to generally
about ten percent.
| |
| Floyd L. Davidson 2006-06-12, 9:21 am |
| "hob" <dehoberg@comcast.net> wrote:
>"Floyd L. Davidson" <floyd@apaflo.com> wrote in message
>news:87bqszqfbf.fld@apaflo.com...
>
>speaking of not reading it -
Now that you've read it, I have to explain it to you???
>see page 1 - insertion loss table
Okay, you didn't read that one apparently:
insertion loss as follows:
Number of Theoretical
Output Ports Insertion Loss (dB)
2 3.0
....
Not 6 dB, just 3 dB.
But, lets stop the foolishness right here, and do a definitive
reality check. Here are a few URLs all of which lead to the
technical specifications for a 2-way splitter. Every single one
of them says the insertion loss is ~3.5 dB, not 6 dB.
Monster Cable TGHZ2RF GHertz 2-Way Splitter
dB Loss peer output is:
5MHz-1GHz: -3.4 dB
http://www.electronicexpress.com/product?prod_id=1396
Allen Tel Coaxial Splitter - 2 Way
3.5dB insertion loss
http://www.twacomm.com/catalog/dept_id_1078.htm
Channel Vision Hybrid Splitters
HS-2 2-way splitter/combiner
3.5 dB insertion loss
http://www.channelvision.com/index/49
Now, if you can find any 2-way splitters that are advertised as
having 6 dB insertion loss, I'd be happy to see them.
Here is a short and simple explanation:
INSERTION/THRU-LOSS
When a 2-way splitter is inserted in the line, the signal
level at the output ports will be reduced by approximately 3.5
dB (4.0 dB). This loss is referred to as thru-loss or
insertion loss. For calculation purposes it is best to utilize
4.0 dB as the actual loss. If a 4-way splitter had been used
than resulting loss would be 6.5 dB (7.0 dB).
www.blondertongue.com/media/pdfs/te.../DistDesign.pdf
Hence, we might disagree on exactly what mistake you made in reading
the material in the reference you presented (or I suppose you could
claim it is simply wrong), but one thing we don't need to argue
further about is just what the insertion loss of a 2-way splitter
is: it *is* 3.5 dB.
So lets see what you had to say, and figure out where the error is.
>see page 4, last paragraph - "A signal applied to port A will be routed to
>port S, minus a 3dB loss in the internal resistor;"
That is how it works as a *combiner*.
However, a combiner is nothing other than a splitter where each
of two ports (that are isolated from each other, hence no signal
goes from Port A above to Port B) split each signal between two
other ports (Port S and the resistor in this instance).
A signal applied to either Port A or Port B does exactly the
same thing, which is output at Port S with a 3 dB loss. The
other half of the power goes to the resistor (which is actually
nothing other than a dummy load placed on one of a splitter's
two output ports). Note that a signal input to Port A does
appear at Port B (or at least not to the degree that it actually
is isolated).
And of course that is identical to inputting a signal into Port
S, which is then split between Port A and Port B equally, with
*nothing* going to the resistor because that port is isolated by
more than 20 dB (assuming all ports are correctly terminated).
> (you will note why that resistor has to be in there on page 2, and the
>effect of mismatch on page 5 - note the 6db loss INSIDE the splitter for
>mismatch)
That of course is not correct. As that document clearly states,
there is a 3 dB loss, not 6. When a signal is input to Port S,
it is split half and half between Port A and Port B, and *no*
signal appears across the resistor, which is isolated from Port
S.
>I knew the loss was there before I read it - and it is the same after I read
>it.
If you would *read* what it says instead of looking for snippets
of text that you believe support what you think it should say,
you might do better an understanding how a hybrid works.
>It seems you failed to understand what you read: You lose the 3 db
>internally to match impedance, or you lose 3 db or more in the impedance
>mismatch. Either way, it costs you 3db loss to insert it into the line.
Nothing in that URL says that. And there are no 2-way splitters
being marketed that list in their technical specifications an
insertion loss of greater than 4 dB.
> (Having two outputs side by side is called a parallel ciruit, and their
>impedance is half of what was there before the splitter was inserted. And
>maximum power transfer ONLY occurs at matched impedance.)
What do you think happens if the load impedance on each of these
ports is doubled???? (Hint: NOTHING!)
>
>That out-of-context quote of yours refers to the mismatch effect on
>insertion loss - read what it said just before that sentence.
> It is talking about the effect of shorting out one of the ports -
>"THE POWER LOSS WITHIN THE POWER SPLITTER WOULD BE... -6DB of the orginal
>signal power [power in- at the input port]"
Go back an look at what it is saying.
> And IF you short out port B output, in a real splitter, the LOSS at port
>A output (not the signal) is less than 3 db because some of the signal power
>that was to go to port B (that cannot leave because of the short) is
>reflected into port A.
Given that we *know* the insertion loss is actually 3.5 dB, not
6 dB, I'll leave it to you to figure out the mistake you made in
analyzing that paragraph.
If you do accept that your analysis *can't* be correct, but
cannot figure out what it should be, I'll be happy to explain it
in different terms that the reference we are using. You cited
it, but I'd say that it is terribly confusing, and is not
helping you to understand this at all.
>
>No, in a real passive filter, it is closer to 3.5 db. The internal loss is
>well known and accepted.
It is 3.5 dB in virtually *every* commercially available splitter you
are going to find on the market!
I have no idea what you mean by "a real passive filter". Real
passive filters can have almost any insertion loss, depending on
how they are designed.
>
>that is along way from .5 db you had just claimed. Apparently you are
>unfamiliar with the term "insertion loss"
I said the 0.5 dB was hybrid loss. Along with the 3 dB loss
from the two way split, a typical 2-way splitter has an
insertion loss of 3.5 dB. It happens that some (not the kind
used for RF at TV frequencies) may have as much as 1 dB hybrid
loss, which would then make for an insertion loss of 4.5 dB.
>
>Special RF in those RG-59/58/6 coax lines of the poster? His Rf doesn't
>follow the laws of physics?
Because it is RF, what you describe *WILL* *NOT* *WORK*. Go
blow that smoke up someone else's rear. You don't know what you
are talking about with grounding and transmission lines any more
than you understand hybrid splitters.
> So why don't you enlighten us and tell us WHY shield breaks and "spiders"
>won't isolate the signal from RFI/EMI? There are a couple of authors on rf
>grounding systems I will pass it on to.
If you have a broken shield on a coaxial transmission line... you
don't have a transmission line! It's called an antenna.
If your purpose is *not* a transmission line, but instead is a
shielded cable (for example with DC leads inside the shield),
there is a huge difference. That difference might be expressed
as the simple fact that a transmission line keeps the RF inside,
while a shielded wire tries to keep it outside. That is way over
simplified, but...
>"multimegawatt transmitters".
>
>perhaps that is why they paid me to do it as a civilian specialist, and not
>you
Giggle snort!
>( the actual size and db of the equipment was classified, fwiw )
Choke laugh, cough!
>
>Well, I was the BMEWs site grounding engineering tech for two years and have
>no need to blow smoke - and I have a good deal of experience working with L
>band and X-band grounds.
In two years I'm sure you learned all about how to brew coffee.
I've seen people who had 10 times your experience who *still*
didn't understand it!
[snip]
>Well, apparently you have zip experience working with RFI/EMI grounding
>systems or antennas, or you wouldn't say such stupid things.
Please go to a nice library and find almost any decent book on
transmission line theory. Try learning how it works first, then
mouth off.
[snip]
> And who uses coax for audio or worries about transmission line length vs
>wavelength in audio cables? Monster cable buyers? That statement is so far
>out of reality it says you are clueless in this field
Think of the telecommunications industry...
>
>That statement clearly and unequivocably says that you have absolutely no
>clue how grounding is done inside equipment and that you have no real
>professional experience whatsoever.
Or perhaps that I actually do know about it, and can
uncategorically state that you are babbling nonsense.
Here's the best phrase of all:
"to weed out too-sharp rise-time noise and only pass pro
forma pulses"
Reminds me of a guy who picked up a box of round headed rivets
used to put together metal shelving to store electronics parts,
who assumed it was some sort of electronic gizmo, and said
"What's this?". To the absolute delight of all the techs
watching, it was a facilities mechanic who instantly piped up
with "Oh, those are bi-directional diodes!" The new supervisor
sat the box down, said "Oh...", and continued on his way.
Didn't they kick you upstairs into management too? To keep you
from touching anything critical?
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Floyd L. Davidson 2006-06-12, 10:21 am |
| "hob" <dehoberg@comcast.net> wrote:
>"Roy L. Fuchs" <roylfuchs@urfargingicehole.org> wrote in message
>news:nrgp82hjsjmb6e8t9hrcq972mq1cv1pa6b@4ax.com...
>he
>
>Not in the computer sense per se, but perhaps in the digital signal to ratio
>sense -
Eh? Are you going to run off at the mouth with *another* one????
Bit-error-rate is a term used for digital communications systems.
It actually has nothing to do with computers. I have no idea what
you mean by "digital signal to ratio sense" either.
"Bit error rate" is the standard measure of quality of service
for a digital data circuit.
> What I am saying is that since digital "amplifiers" do not amplify a
>signal as in an analog fashion, passing all the signal, rather they recreate
>the signal at a higher level, isolating at each stage, they are susceptible
>to creating extra bad bits in noisy signals.
Wow, you are not particularly articulate, but that is almost correct.
It won't add "extra bad bits", but it will generate errored bits.
>That is, while an analog amp
>will give you most of what comes in,
Oh, if only that were true... An analog amplifier can and
*will* add signals to the output which are not present at the
input.
The difference between analog and digital is that analog _noise_
is cumulative, while digital _errors_ are cumulative. Any
noise, no matter what the Signal-to-Noise Ratio, at the input is
guaranteed going to be included in the output of an analog
amplifier. A digital amplifier will not have any noise in the
output if the SNR is greater than a critical value. It happens
that the critical value for a digital amplifier is one at which
an analog amplifier is almost useless. Of course if the SNR is
low enough that errors do occur, they are cumulative (which is
where error correction comes into use).
Here is a chart, showing BER (Bit Error Rate) and thresholds for
SNR required to achieve them, for a binary transmission system:
Error Rate S/N dB Error Rate S/N dB
10e-2 13.5 10e-7 20.3
10e-3 16.0 10e-8 21.0
10e-4 17.5 10e-9 21.6
10e-5 18.7 10e-10 22.0
10e-6 19.6 1o3-11 22.2
>a digital output is only reproduced
>timewise from what comes in. ( Removes the need for good filters and such.)
I'm not sure what you mean by "reproduced timewise", but trust that
"good filters and such" *are* essential (to keeping the SNR below the
threshold referenced above).
> A digital amplifier's weakness, however, is that it is looking for a bit
>to reproduce - and it only "recognizes" signals having a certain
>risetime -it discriminates against signals rising too fast for the amp ( it
>can't "trigger") and it discriminates against signals that rise too slowly
>(the "gate" set by the clock closes before the voltage level rises enough to
>trigger an output).
Eh? Risetime may or may not have anything to do with how a
digital signal is decoded. It depends on just what parameter is
modulated. For example, typical v.90 modems are in fact
sensitive to phase hits, on the other hand a typically ISDN or
T1 is not.
> If you look at a real digital signal, you will see a series of pulses and
>noise (non-signal) - if the noise is well below the level of the pulses,
That does not describe "a real digital signal". You are
describing a voltage level encoded digital signal. T1 and ISDN
are two examples of types of digital tranmission systems than
have pulses as you describe (while a v.90 modem does not).
>the pulses drive the amplifier and low noise has little effect - the bits
>are pretty much repeated and thus cleanly reproduced at a higher level.
> If the level of the pulses are weak and thus are in the noise, then there
>are too many false output bits created for a coherent signal. It does not
>discriminate properly between noise and signal.
But lets do note that at the SNR where that signal will become
100% error free, an analog system would be absolutely useless
due to excessive noise! Typically, for example, a digital
transmission system can function with an SNR of 12 dB and
provide a channel equal to an analog system requiring a 60 dB
SNR.
> ( A "better" tuner would fix the problem only because the word better
>implies one that does not have the problem -)
The tuner of course is *not* looking at "the level of pulses" in
any way shape or form. It is receiveing a modulated RF signal
and *generating* pulses!
> a "better at handling low signal to noise ratio" tuner might fix the
>problem, but normally you get what they give you and expect you to get the
>signal up to a certain strength and above a certain signal to noise ratio
See above, relative to a 12 dB SNR being sufficient for a digital
system to provide the same quality as an analog transmission system
that has a 60 dB SNR.
>(if that ability to discriminate in high noise is what is called
>bit-error-rate in digital signals, then that is the term.)
>
>fwiw
Worthless. You should *ask* what this is about, not try to
fabricate what you think it could be.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
|
| you really don't know what an insertion loss is, do you?
You apparently think its a measure of what you get out of either port
relative to the input port - (try the price in signal for inserting that
splitter in the line.)
or anything about power transfer and impedance matching?
you think you can insert a parallel line in series into a line, create an
impedance mismatch, and still get 100% efficient transfer. You really need
to take a basic course in electronics.
<snip>
> Now, if you can find any 2-way splitters that are advertised as
> having 6 dB insertion loss, I'd be happy to see them.
they don't have 6db insertion loss - they have 3 db insertion loss - and the
halving of the signal in a two-way splitter makes for the 6 db drop in
strength at either output port from the input strength
nuff said - if you can't get the basics of electronics and don't even know
what the terms mean, you sure won't get anything technical - especially RF
grounding
( reread the article giving the basics of simple splitters with the proper
definition of insertion loss in mind)
"Floyd L. Davidson" <floyd@apaflo.com> wrote in message
news:87ejxuppym.fld@apaflo.com...
> "hob" <dehoberg@comcast.net> wrote:
>
> Now that you've read it, I have to explain it to you???
>
>
> Okay, you didn't read that one apparently:
>
> insertion loss as follows:
> Number of Theoretical
> Output Ports Insertion Loss (dB)
> 2 3.0
>
> ....
>
>
> Not 6 dB, just 3 dB.
>
> But, lets stop the foolishness right here, and do a definitive
> reality check. Here are a few URLs all of which lead to the
> technical specifications for a 2-way splitter. Every single one
> of them says the insertion loss is ~3.5 dB, not 6 dB.
>
> Monster Cable TGHZ2RF GHertz 2-Way Splitter
> dB Loss peer output is:
> 5MHz-1GHz: -3.4 dB
>
> http://www.electronicexpress.com/product?prod_id=1396
>
>
> Allen Tel Coaxial Splitter - 2 Way
> 3.5dB insertion loss
>
> http://www.twacomm.com/catalog/dept_id_1078.htm
>
> Channel Vision Hybrid Splitters
> HS-2 2-way splitter/combiner
> 3.5 dB insertion loss
>
> http://www.channelvision.com/index/49
>
> Now, if you can find any 2-way splitters that are advertised as
> having 6 dB insertion loss, I'd be happy to see them.
they don't have 6db insertion loss - they have 3 db insertion loss - and the
halving of the signal in a two-way splitter makes for the 6 db drop in
strength at either output port from the input strength
>
> Here is a short and simple explanation:
>
> INSERTION/THRU-LOSS
>
> When a 2-way splitter is inserted in the line, the signal
> level at the output ports will be reduced by approximately 3.5
> dB (4.0 dB). This loss is referred to as thru-loss or
> insertion loss. For calculation purposes it is best to utilize
> 4.0 dB as the actual loss. If a 4-way splitter had been used
> than resulting loss would be 6.5 dB (7.0 dB).
>
> www.blondertongue.com/media/pdfs/te.../DistDesign.pdf
>
> Hence, we might disagree on exactly what mistake you made in reading
> the material in the reference you presented (or I suppose you could
> claim it is simply wrong), but one thing we don't need to argue
> further about is just what the insertion loss of a 2-way splitter
> is: it *is* 3.5 dB.
>
> So lets see what you had to say, and figure out where the error is.
>
to[color=darkred]
>
> That is how it works as a *combiner*.
>
> However, a combiner is nothing other than a splitter where each
> of two ports (that are isolated from each other, hence no signal
> goes from Port A above to Port B) split each signal between two
> other ports (Port S and the resistor in this instance).
>
> A signal applied to either Port A or Port B does exactly the
> same thing, which is output at Port S with a 3 dB loss. The
> other half of the power goes to the resistor (which is actually
> nothing other than a dummy load placed on one of a splitter's
> two output ports). Note that a signal input to Port A does
> appear at Port B (or at least not to the degree that it actually
> is isolated).
>
> And of course that is identical to inputting a signal into Port
> S, which is then split between Port A and Port B equally, with
> *nothing* going to the resistor because that port is isolated by
> more than 20 dB (assuming all ports are correctly terminated).
>
>
> That of course is not correct. As that document clearly states,
> there is a 3 dB loss, not 6. When a signal is input to Port S,
> it is split half and half between Port A and Port B, and *no*
> signal appears across the resistor, which is isolated from Port
> S.
>
read[color=darkred]
>
> If you would *read* what it says instead of looking for snippets
> of text that you believe support what you think it should say,
> you might do better an understanding how a hybrid works.
>
>
> Nothing in that URL says that. And there are no 2-way splitters
> being marketed that list in their technical specifications an
> insertion loss of greater than 4 dB.
>
>
> What do you think happens if the load impedance on each of these
> ports is doubled???? (Hint: NOTHING!)
>
>
> Go back an look at what it is saying.
>
port[color=darkred]
power[color=darkred]
>
> Given that we *know* the insertion loss is actually 3.5 dB, not
> 6 dB, I'll leave it to you to figure out the mistake you made in
> analyzing that paragraph.
>
> If you do accept that your analysis *can't* be correct, but
> cannot figure out what it should be, I'll be happy to explain it
> in different terms that the reference we are using. You cited
> it, but I'd say that it is terribly confusing, and is not
> helping you to understand this at all.
>
is[color=darkred]
>
> It is 3.5 dB in virtually *every* commercially available splitter you
> are going to find on the market!
>
> I have no idea what you mean by "a real passive filter". Real
> passive filters can have almost any insertion loss, depending on
> how they are designed.
>
>
> I said the 0.5 dB was hybrid loss. Along with the 3 dB loss
> from the two way split, a typical 2-way splitter has an
> insertion loss of 3.5 dB. It happens that some (not the kind
> used for RF at TV frequencies) may have as much as 1 dB hybrid
> loss, which would then make for an insertion loss of 4.5 dB.
>
>
> Because it is RF, what you describe *WILL* *NOT* *WORK*. Go
> blow that smoke up someone else's rear. You don't know what you
> are talking about with grounding and transmission lines any more
> than you understand hybrid splitters.
>
"spiders"[color=darkred]
rf[color=darkred]
>
> If you have a broken shield on a coaxial transmission line... you
> don't have a transmission line! It's called an antenna.
>
> If your purpose is *not* a transmission line, but instead is a
> shielded cable (for example with DC leads inside the shield),
> there is a huge difference. That difference might be expressed
> as the simple fact that a transmission line keeps the RF inside,
> while a shielded wire tries to keep it outside. That is way over
> simplified, but...
>
not[color=darkred]
>
> Giggle snort!
>
>
> Choke laugh, cough!
>
have[color=darkred]
L[color=darkred]
>
> In two years I'm sure you learned all about how to brew coffee.
> I've seen people who had 10 times your experience who *still*
> didn't understand it!
>
> [snip]
>
> Please go to a nice library and find almost any decent book on
> transmission line theory. Try learning how it works first, then
> mouth off.
>
> [snip]
vs[color=darkred]
far[color=darkred]
>
> Think of the telecommunications industry...
>
and he[color=darkred]
with[color=darkred]
forma[color=darkred]
whatever he[color=darkred]
>
> Or perhaps that I actually do know about it, and can
> uncategorically state that you are babbling nonsense.
>
> Here's the best phrase of all:
>
> "to weed out too-sharp rise-time noise and only pass pro
> forma pulses"
>
> Reminds me of a guy who picked up a box of round headed rivets
> used to put together metal shelving to store electronics parts,
> who assumed it was some sort of electronic gizmo, and said
> "What's this?". To the absolute delight of all the techs
> watching, it was a facilities mechanic who instantly piped up
> with "Oh, those are bi-directional diodes!" The new supervisor
> sat the box down, said "Oh...", and continued on his way.
>
> Didn't they kick you upstairs into management too? To keep you
> from touching anything critical?
>
> --
> Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
> Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
|
|
"Floyd L. Davidson" <floyd@apaflo.com> wrote in message
news:87ac8ipn7c.fld@apaflo.com...
> "hob" <dehoberg@comcast.net> wrote:
whatever[color=darkred]
ratio[color=darkred]
>
> Eh? Are you going to run off at the mouth with *another* one????
>
> Bit-error-rate is a term used for digital communications systems.
> It actually has nothing to do with computers. I have no idea what
> you mean by "digital signal to ratio sense" either.
>
> "Bit error rate" is the standard measure of quality of service
> for a digital data circuit.
>
recreate[color=darkred]
susceptible[color=darkred]
>
> Wow, you are not particularly articulate, but that is almost correct.
> It won't add "extra bad bits", but it will generate errored bits.
Utter parsing bullshit - "it won't add ...bits, it will generate...bits"
and the difference between additional unwanted bits and "errored bits" out
of a
functioning digital ampilifier is what, pray tell?
Apparently you are saying that given sufficient noise, the
sensing-repeating part of the amplifier won't add "extra bad bits", that is,
it won't add pulses to the output stream that are not in the input?
Talk about grasping.
>
>
> Oh, if only that were true... An analog amplifier can and
> *will* add signals to the output which are not present at the
> input.
Read - oh, I forgot, you havbe a problem reading... I said it gives you back
most og what comes in, NOT that is did not add signals (just as digital
does).
>
> The difference between analog and digital is that analog _noise_
> is cumulative, while digital _errors_ are cumulative. Any
> noise, no matter what the Signal-to-Noise Ratio, at the input is
> guaranteed going to be included in the output of an analog
> amplifier. A digital amplifier will not have any noise in the
> output if the SNR is greater than a critical value. It happens
> that the critical value for a digital amplifier is one at which
> an analog amplifier is almost useless. Of course if the SNR is
> low enough that errors do occur, they are cumulative (which is
> where error correction comes into use).
>
I have no probelm with that because that is what I said (re:timewise only)
> Here is a chart, showing BER (Bit Error Rate) and thresholds for
> SNR required to achieve them, for a binary transmission system:
>
> Error Rate S/N dB Error Rate S/N dB
>
> 10e-2 13.5 10e-7 20.3
> 10e-3 16.0 10e-8 21.0
> 10e-4 17.5 10e-9 21.6
> 10e-5 18.7 10e-10 22.0
> 10e-6 19.6 1o3-11 22.2
>
such.)[color=darkred]
>
> I'm not sure what you mean by "reproduced timewise", but trust that
> "good filters and such" *are* essential (to keeping the SNR below the
> threshold referenced above).
>
bit[color=darkred]
( it[color=darkred]
slowly[color=darkred]
to[color=darkred]
>
> Eh? Risetime may or may not have anything to do with how a
> digital signal is decoded.
Amplifiers discriminate based on the risetime of the digital pulse.
It depends on just what parameter is
> modulated. For example, typical v.90 modems are in fact
> sensitive to phase hits, on the other hand a typically ISDN or
> T1 is not.
>
Was not talking about phase here,. just rise time discrimination
and[color=darkred]
>
> That does not describe "a real digital signal". You are
> describing a voltage level encoded digital signal. T1 and ISDN
> are two examples of types of digital tranmission systems than
> have pulses as you describe (while a v.90 modem does not).
>
"noise" can be voltage or phase or whatever masks the intended signal
there[color=darkred]
not[color=darkred]
>
> But lets do note that at the SNR where that signal will become
> 100% error free, an analog system would be absolutely useless
> due to excessive noise! Typically, for example, a digital
> transmission system can function with an SNR of 12 dB and
> provide a channel equal to an analog system requiring a 60 dB
> SNR.
>
>
> The tuner of course is *not* looking at "the level of pulses" in
> any way shape or form. It is receiveing a modulated RF signal
> and *generating* pulses!
>
Apparently you have never designed or worked on digital amplifiers -all
amplifers see the input, and one common way to get a clean digital signal is
to "clip off the tops" of the pulses making up the bits above the noise and
passing them on (think a diode biased to stop everything near the noise and
pass everything above the noise - designers have been using that method for
nearly 40 years)
the[color=darkred]
>
> See above, relative to a 12 dB SNR being sufficient for a digital
> system to provide the same quality as an analog transmission system
> that has a 60 dB SNR.
>
don't care about an analog system here - it's digital.
>
> Worthless. You should *ask* what this is about, not try to
> fabricate what you think it could be.
What you hobbyists call things is not necessarily what engineers call things
or what they really are -
(And you who think I should ask about your narrow parsed definitions for
your convenience apparently think RCA audio cables are coaxial cables, and
you think insertion loss is total loss.)
>
> --
> Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
> Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
|
|
"Roy L. Fuchs" <roylfuchs@urfargingicehole.org> wrote in message
news:pjeq82dmgs1t8adlnsbmjvb6404uhd1l9b@4ax.com...
> On Mon, 12 Jun 2006 02:57:45 -0500, "hob" <dehoberg@comcast.net> Gave
> us:
>
vs[color=darkred]
far[color=darkred]
>
> Nearly ALL audio RCA type patch cables are a pair of coaxial runs.
>
> I cannot think off any that are not. Can you? Give the model and
> brand name here:
I can think of no RCA patch cables that are coaxial conductors.
coaxial cables are concentric conductors (co-axial conductors , hence the
name) that use fields to transfer energy near the conductors. Both the outer
and inner axial metal parts of a coaxial cable are ("same-direction")
conductors.
RCA audio cables are shielded conductors that use the center conductor to
carry energy. The outer metal part is not intended to carry the signal, just
to shield the conductor and sometimes provide a ground path.
( Some people undoubtedly sell coax for audio hookup and some fools buy it,
but some people also buy "super-pure" copper to carry a signal to a speaker
that has more inherent distortion than a thousand miles of romex.)
Yes, coax will conduct audio signal in its center conductor if you want to
waste your money, but an RCA audio cable is not coax and doesn't work like
it.
| |
| Roy L. Fuchs 2006-06-13, 1:21 am |
| On Mon, 12 Jun 2006 17:21:05 -0500, "hob" <dehoberg@comcast.net> Gave
us:
> you really don't know...
I know what a top posting Usenet retard is. That would be YOU,
dipshit.
It would seem that there are several aspects of reality that you need
to learn a BIT more about.
YOUR BIT error rate is too fucking high.
You are the weakest dink.... goodbye!
| |
| Roy L. Fuchs 2006-06-13, 1:21 am |
| On Mon, 12 Jun 2006 17:55:09 -0500, "hob" <dehoberg@comcast.net> Gave
us:
>Utter parsing bullshit - "it won't add ...bits, it will generate...bits"
Tell us... oh DIPSHITTO...
What is QAM 256?
| |
| Roy L. Fuchs 2006-06-13, 1:21 am |
| On Mon, 12 Jun 2006 17:55:09 -0500, "hob" <dehoberg@comcast.net> Gave
us:
>Amplifiers discriminate based on the risetime of the digital pulse.
Tell us, oh guru of digital transmission... what is QAM?
| |
| Roy L. Fuchs 2006-06-13, 1:21 am |
| On Mon, 12 Jun 2006 17:55:09 -0500, "hob" <dehoberg@comcast.net> Gave
us:
>Apparently you have never designed or worked on digital amplifiers -all
>amplifers see the input, and one common way to get a clean digital signal is
>to "clip off the tops" of the pulses making up the bits above the noise and
>passing them on (think a diode biased to stop everything near the noise and
>pass everything above the noise - designers have been using that method for
>nearly 40 years)
Straight modulated SCPC, perhaps... But there are NO digital
transmissions that are straight modulated anymore. (likely for nearly
40 years now).
Ever heard of PSK? How about QAM?
Get a clue, "top clipper boy".
| |
| Roy L. Fuchs 2006-06-13, 1:21 am |
| On Mon, 12 Jun 2006 18:37:57 -0500, "hob" <dehoberg@comcast.net> Gave
us:
>
>Yes, coax will conduct audio signal in its center conductor if you want to
>waste your money, but an RCA audio cable is not coax and doesn't work like
>it.
Even the tiny feeds up to my headphones are coax.
Co... axial. That's all that is required. It is a physical
structure, and YES, the audio uses the shield of the coax as a return.
It is typically referred to as a completed circuit.
| |
| Floyd L. Davidson 2006-06-13, 10:21 am |
| "hob" <dehoberg@comcast.net> wrote:
>"Floyd L. Davidson" <floyd@apaflo.com> wrote:
>
>Utter parsing bullshit - "it won't add ...bits, it will generate...bits"
It does not add bits. But it will cause errors in the bits that
exist.
>and the difference between additional unwanted bits and "errored bits" out of a
>functioning digital ampilifier is what, pray tell?
In most systems the data is framed, and adding even a single bit
would result in a loss of framing. At that point all data is
lost until framing is re-acquired. Hence, rather than a single
bit error, there would be a huge loss of all data.
> Apparently you are saying that given sufficient noise, the
>sensing-repeating part of the amplifier won't add "extra bad bits", that is,
>it won't add pulses to the output stream that are not in the input?
>
> Talk about grasping.
Gasping would be correct. I am utterly aghast at your responses.
>
>Amplifiers discriminate based on the risetime of the digital pulse.
Perhaps there are systems where that is true, but that is *not*
a necessary part of a digital system. For example, in a T1
Carrier system, where is *anything* based on rise time?
Even for a typical RS-232 port on a PC, rise time has nothing
to do with signal detection.
>
>Was not talking about phase here,. just rise time discrimination
If your system is sensitive to rise time, it *is* going to be
sensitive to phase hits.
You are stating that a "real digital signal" has pulses. That
is not necessarily true.
[color=darkred]
>
>"noise" can be voltage or phase or whatever masks the intended signal
You are the one who said "well below the level of the pulses".
That would, for example, exclude phase noise. Which is why I
pointed out that your definition of a "real digital signal" is
invalid.
>
>Apparently you have never designed or worked on digital amplifiers -all
>amplifers see the input, and one common way to get a clean digital signal is
>to "clip off the tops" of the pulses making up the bits above the noise and
>passing them on (think a diode biased to stop everything near the noise and
>pass everything above the noise - designers have been using that method for
>nearly 40 years)
Well, which do you want to talk about, a tuner or a digital amplifier?
The input stage in a tuner is *not* a digital amplifier.
>
>don't care about an analog system here - it's digital.
The input amplifier of the tuner is analog. The transmission
medium is analog. The modulation scheme is digital, not the input
signal.
> (And you who think I should ask about your narrow parsed definitions for
>your convenience apparently think RCA audio cables are coaxial cables, and
>you think insertion loss is total loss.)
I haven't said a thing about RCA audio cables, but certainly
most cables sold with RCS plugs on both ends are in fact coaxial
cables, simply because RCA connectors were specifically designed
for coaxial cable.
And in fact insertion loss of a device *is* total loss through
the device.
Insertion loss:
1. The loss resulting from the insertion of a device in a
transmission line, expressed as the reciprocal of the
ratio of the signal power delivered to that part of the
line following the device to the signal power delivered
to that same part before insertion. Note: Insertion
loss is usually expressed in dB.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Floyd L. Davidson 2006-06-13, 10:21 am |
| "hob" <dehoberg@comcast.net> wrote:
> you really don't know what an insertion loss is, do you?
Insertion Loss:
1. The loss resulting from the insertion of a device in a
transmission line, expressed as the reciprocal of the
ratio of the signal power delivered to that part of
the line following the device to the signal power
delivered to that same part before insertion. Note:
Insertion loss is usually expressed in dB.
> You apparently think its a measure of what you get out of either port
>relative to the input port - (try the price in signal for inserting that
>splitter in the line.)
In fact, that is true. See the definition of Insertion Loss.
>or anything about power transfer and impedance matching?
>
>you think you can insert a parallel line in series into a line, create an
A "parallel line in series into a line" is something that I don't understand.
Can you explain what that is supposed to be?
>impedance mismatch, and still get 100% efficient transfer. You really need
>to take a basic course in electronics.
You'll note that I've stated there is a hybrid loss in these splitters,
usually 0.5 dB, but perhaps as high as 1 dB. That is the lack of "100%
efficient transfer". The other 3 dB in the insertion loss is the fact
that the power is split two ways, each will see a 3 dB reduction in
power.
><snip>
>
>
>they don't have 6db insertion loss - they have 3 db insertion loss - and the
>halving of the signal in a two-way splitter makes for the 6 db drop in
>strength at either output port from the input strength
Wrong. If there was a 6 dB drop in strength at a port, that
would be a 6 dB insertion loss. In fact, as noted, the
insertion loss is 3.5 dB, and in fact if you do have correctly
match impedances, the output at each of the two ports will be
3.5 dB less than the input signal on a two-way splitter.
>nuff said - if you can't get the basics of electronics and don't even know
>what the terms mean, you sure won't get anything technical - especially RF
>grounding
You have demonstrated that to be the case.
>
>they don't have 6db insertion loss - they have 3 db insertion loss - and the
>halving of the signal in a two-way splitter makes for the 6 db drop in
>strength at either output port from the input strength
Giggle snort laugh cough. Who is it that doesn't understand
what?
[color=darkred]
See what it says there? Each port will have a signal that is
"approximately 3.5 dB (4.0 dB)" lower than the input signal.
*That* is insertion loss.
[color=darkred]
And the signal at the output ports is 3.5 dB less than the
signal at the input port. *That* is insertion loss.
[color=darkred]
I see that this went totally over your head, and you had no
response. The point, since you may have missed it, is that
you totally misread the material you cited.
A 2-way splitter is merely a 4 port hybrid with one port
terminated. The terminated port is isolated from the input
port, there is *no* 3 dB loss in the input signal associated
with that resistor.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Floyd L. Davidson 2006-06-13, 10:21 am |
| "hob" <dehoberg@comcast.net> wrote:
>"Roy L. Fuchs" <roylfuchs@urfargingicehole.org> wrote in message
>news:pjeq82dmgs1t8adlnsbmjvb6404uhd1l9b@4ax.com...
>vs
>far
>
>I can think of no RCA patch cables that are coaxial conductors.
You mean, I assume, patch cords using "RCA Plugs"? They are virtually
*all* made up with coaxial cables. That is because the plug is designed
specifically for coaxial cable. (It was originally called a "Phone Plug",
because RCA used it for the coaxial connection to a phonograph arm.)
>coaxial cables are concentric conductors (co-axial conductors , hence the
Okay. That is a *physical* description.
>name) that use fields to transfer energy near the conductors. Both the outer
>and inner axial metal parts of a coaxial cable are ("same-direction")
>conductors.
That is an electrical description of a transmission line. It happens
to be the way that transmission works when the physical device is
a "coaxial cable".
>RCA audio cables are shielded conductors that use the center conductor to
>carry energy.
Hence, they fit even *your* description of a coaxial cable.
>The outer metal part is not intended to carry the signal, just
>to shield the conductor and sometimes provide a ground path.
So? It is still physically a coaxial cable.
>( Some people undoubtedly sell coax for audio hookup and some fools buy it,
>but some people also buy "super-pure" copper to carry a signal to a speaker
>that has more inherent distortion than a thousand miles of romex.)
Coax is very commonly used for audio; primarily because it is a
very effective transmission line. In addition to fools buying,
smart folks do to! (RCA, for example, used coax for audio and
tha is how we happen to have a coaxial connector called a "RCA
Plug"!)
>Yes, coax will conduct audio signal in its center conductor if you want to
>waste your money, but an RCA audio cable is not coax and doesn't work like
>it.
Sure.
--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd@apaflo.com
| |
| Dave Ryman 2006-06-13, 2:21 pm |
| "TimPerry" <timperry@noaspamadelphia.net> wrote in
news:9vedndrrue5uChbZnZ2dnUVZ_rednZ2d@adelphia.com:
(snip)
>
> to me this sounds like some kind of ground loop issue.
>
> just try this: get 2 low cost baluns. they have an F connector at
> one end and 2 spade terminals at the other. solder or bolt the spade
> lugs to the matching lug on the other balun. this gives you a 75 ohm
> F connector at each end. add a short male - male cable and insert at
> the receiver.
>
> if this does not help then you can proceed to more drastic and
> expensive measures.
>
I've just tried this, and it works an absolute treat!! The Lion's share of
the interference has gone. Such a simple thing - I'm not even going to try
to pretend to understand the physics, I'll just apply Arthur C. Clarke's
definition of "Magic"!
Thanks very very much.
--
Regards,
Dave
dave_ryman@hotmailNOSPAM.com
http://welcome.to/daves.website
http://travel.to/formula.one
| |
|
|
"Roy L. Fuchs" <roylfuchs@urfargingicehole.org> wrote in message
news:vqds82h19p0u7d8u84bf6i6lqbl5o8vvop@4ax.com...
> On Mon, 12 Jun 2006 17:55:09 -0500, "hob" <dehoberg@comcast.net> Gave
> us:
>
is[color=darkred]
and[color=darkred]
and[color=darkred]
for[color=darkred]
>
> Straight modulated SCPC, perhaps... But there are NO digital
> transmissions that are straight modulated anymore. (likely for nearly
> 40 years now).
>
> Ever heard of PSK? How about QAM?
Quadruple Amplitude Modulation still has to pick signal out of the
background noise in order to use it.
You could call it RFU and still have to pull signal out of the background.
>
> Get a clue, "top clipper boy".
Get a clue, "no amp needed in my world boy"
| |
|
| |