Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: 1 ... 4 5 [6] 7 8 ... 12

Author Topic: SSFP Mk 3  (Read 70428 times)

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 31171
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: SSFP Mk 3
« Reply #75 on: September 26, 2014, 09:49:56 PM »

Ah, I see. Thank you.  :)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

les-70

  • Kitizen
  • ****
  • Posts: 1254
Re: SSFP Mk 3
« Reply #76 on: September 27, 2014, 06:57:51 PM »

  I said I would compare the Mark3 with the BT NTE-2005 ADSl v1.0 faceplate made by Pressac.  The NTE-2005 has from my looking a the circuit no "RF3" type component and connector A and B from the drop wire go straight to the adsl line. 

 Here are the Mark 3 stats.

Max:   Upstream rate = 23663 Kbps, Downstream rate = 77788 Kbps
Bearer:   0, Upstream rate = 20000 Kbps, Downstream rate = 72030 Kbps
      Down      Up
SNR (dB):    7.9       6.6
Attn(dB):    17.5       0.0
Pwr(dBm):    13.3       5.1

  Here are the NTE-2005 stats

Max:   Upstream rate = 23936 Kbps, Downstream rate = 79268 Kbps
Bearer:   0, Upstream rate = 20000 Kbps, Downstream rate = 72030 Kbps
      Down      Up
SNR (dB):    8.3       6.7
Attn(dB):    17.5       0.0
Pwr(dBm):    13.3       4.9

  With the absence of the "RF3" type component I was expecting a higher attainable and indeed it is 1.5 Mb/s higher  :). My previous impression was that the Mark 2 had a more aggressive "RF3" than the Mark 3 so I guess that a Mark 2 to NTE-2005 swap should show a slightly bigger improvement. I attach the respective snrm and bitloadings per tone.    After 45 mins the error rates (with the speed capped) look less  :)  but a reliable error rate test will take a few days to do properly.

 Very Tentative conclusion 

 Both the Mark 2 and Mark 3 may have common mode filtering that has a negative impact impact on some (possibly many) lines.  Almost all modems are pretty good at rejecting common mode and I have long wondered why BT thought they could do better.  The RF2/3 may have had a purpose when people could hear radio channels on their phone calls but the adsl/vdsl filter in a face plate should stop RF on the drop wires reaching the phone (I think!).



« Last Edit: September 27, 2014, 08:07:36 PM by les-70 »
Logged

NewtronStar

  • Kitizen
  • ****
  • Posts: 4844
Re: SSFP Mk 3
« Reply #77 on: September 27, 2014, 11:10:23 PM »

but a reliable error rate test will take a few days to do properly.

I would not expect the MK2 or MK3 to increase your broadband rate if your error rates are low, the only thing they can do is to "try" and lower thoses errors 24/7, I am going to tell you those devices are useless to combat against external RFI (radio frequency interference) , but they do seem to fair better with internal EMI (Electromagnetic interference).
Logged

JGO

  • Reg Member
  • ***
  • Posts: 729
Re: SSFP Mk 3
« Reply #78 on: September 28, 2014, 08:03:40 AM »

Very Tentative conclusion 

 Both the Mark 2 and Mark 3 may have common mode filtering that has a negative impact impact on some (possibly many) lines.  Almost all modems are pretty good at rejecting common mode and I have long wondered why BT thought they could do better.  The RF2/3 may have had a purpose when people could hear radio channels on their phone calls but the adsl/vdsl filter in a face plate should stop RF on the drop wires reaching the phone (I think!).

It needs figures from many caes !
 
 Yes, modems can reject common mode to some degree, but when that rejection isn't adequate for the situation,  a common mode rejector such as the RF3  improves it further, particularly when installed correctly.
The aim is improved S/(N+I) ratio, so improving performance may involve a bit of loss to give a nett improvement.  Anyone who is looking for a miraculous device with no snags will be disappointed ! 






 
Logged

les-70

  • Kitizen
  • ****
  • Posts: 1254
Re: SSFP Mk 3
« Reply #79 on: September 28, 2014, 09:33:44 AM »


It needs figures from many caes !
 
 Yes, modems can reject common mode to some degree, but when that rejection isn't adequate for the situation,  a common mode rejector such as the RF3  improves it further, particularly when installed correctly.
The aim is improved S/(N+I) ratio, so improving performance may involve a bit of loss to give a nett improvement.  Anyone who is looking for a miraculous device with no snags will be disappointed ! 

  I did try to express the same cautions but thanks for repeating them as they are important. I fully agree on the many cases and I may change my tentative conclusions, for just my line, after a few days of testing.   I also would expect that on some lines an "RF3" type device may improve overall stability even if not the sync speed.

  My concern is that the general use of the Mark 2 and Mark 3 SSFP may be forcing everyone to have something that is detrimental to the performance of relatively good lines vdsl lines that use the D3 band.  Lines that don't use the D3 band would I suspect not notice much if any difference.

  After 15 hours the error rates witht the NTE-2005 are still lower than previously seen on my line with a 72Mb/s sync cap. 

« Last Edit: September 28, 2014, 09:36:14 AM by les-70 »
Logged

JGO

  • Reg Member
  • ***
  • Posts: 729
Re: SSFP Mk 3
« Reply #80 on: September 28, 2014, 11:35:36 AM »

If we had some figures on the RF3's  loss vs frequency that may answer your questions. Judging by photos there are several RF3 versions about too.

Remember xDSL uses an unlicensed radio receiver so there is no guarantee of zero interference and AIUI BT have no obligation to help.

BT's devices do help say 90% of cases of interference but compromise between performance and convenience so are not always optimum, (particularly when people install jammers  such as Plasma TVs', PLAs !! )  There is always a compromise in design; the RF3 could have a lower insertion loss, slightly benefiting maybe 1% of users,  at the expense of size and cost. 
Logged

boost

  • Reg Member
  • ***
  • Posts: 765
Re: SSFP Mk 3
« Reply #81 on: September 28, 2014, 11:46:46 AM »

Thanks for sharing this les.

Some of the QLNs I see (ADSL2+) are a right state and if common mode interference isn't to blame, I guess there's an even bigger x talk issue going on than we realise!
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 31171
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: SSFP Mk 3
« Reply #82 on: September 28, 2014, 12:56:41 PM »

To be utterly systematic about it, one would need to test:
  • an NTE-2005
  • an RF3
  • a Mk 1 SSFP
  • a Mk 2 SSFP
  • a Mk 3 SSFP
in all the possible permutations.  ;)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

les-70

  • Kitizen
  • ****
  • Posts: 1254
Re: SSFP Mk 3
« Reply #83 on: September 28, 2014, 04:11:46 PM »

  I have all of those. I purchased with NTE-2005 and Mark3.  A Mark 1 came with an ebay HG612 and the Mark 2 with the FTTC install.

  Both the NTE-2005 and a Mark 1 were compared by me previously.  I think that for my line they are both "perfect" as far as they go. :--  The drop wires A and B go straight to RJ11 with no filtering.  I could not see any difference the between going through them and straight to the master socket and I also could not detect the addition of the house wiring or a ring and telephone call when either of these are used.

  The Mark 2 and 3 both have the extra toroid component that acts as an RF3 but giving relative to the RF3 relatively weak impact in the vdsl frequency bands. I would say they are RF3's designed to be acceptable on vdsl.   Whilst they may be acceptable it looks like the Mark 2 component took about 2-2.5 Mb/s off my attainable (~78Mb/s) due to attenuation in the higher frequencies mainly the D3 band.  BT may have recognized this and the Mark 3 gives a smaller reduction in attainable ( 1.0-1.5 Mb/s) on my line.  (see previous post for details and post http://forum.kitz.co.uk/index.php?topic=13420.msg259441#msg259441  to see by comparision how the RF3 itself hits a good vdsl line quite hard.)

 Please note that if your line is not using the higher frequency vdsl tones then I would expect that all the devices would test as near identical.  My line  had an atttainable of about 78Mb/s during my tests

I provisionally guess that my line does not need this extra common mode rejection and hence that I will be better off with without the toroid of the mark 2 and 3.   If my expectations work out I shall probably just do a little soldering and short out the toroid in the Mark 2.  I could equally use the NTE-2005 or Mark 1.   
« Last Edit: September 29, 2014, 02:41:27 PM by les-70 »
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 31171
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: SSFP Mk 3
« Reply #84 on: September 28, 2014, 05:24:46 PM »

Thank you for that explanation of your current observations and thoughts. In your situation I would be tempted just to use the Mk 1 SSFP, keeping the NTE-2005 in reserve and putting away the Mk 2 & Mk 3 in your grotto.

As I only have an ADSL2 service, the tones with frequencies much above 1 MHz are unusable. So your results are very valuable to know . . .  :)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

les-70

  • Kitizen
  • ****
  • Posts: 1254
Re: SSFP Mk 3
« Reply #85 on: September 28, 2014, 06:22:09 PM »

  Thanks for that b'cat.  Here is an aspect of the RF3's that I don't understand but does not seem good.  They seem to lift/create the QLN floor. I attach the Mark1 (now running) QLN which shows just a steady decline with increasing frequency.  Then the Mark 3 with its toroid RF3  and finally a real RF3 used between a Mark 2 and the modem.  i.e. a Mark 2 + RF3 although the RF3 will dominate.     They show that RF3 gives a raised floor at the higher frequencies, a weak effect with the Mark 3, but a strong impact with the RF3.

 Can anyone explain this? Whilst I have a very basic electronics understanding,  noise other than that of resistors is a bit beyond me and always looked a bit of a black art.

    Please ignore the variation in spikes between the plots.  The times on the plots are just the snapshot time and not the sync times to which the QLN refer.  Some look like evening but the RF3 is I am sure a daytime sync -it did not reduce my evening spikes.
« Last Edit: September 28, 2014, 06:25:45 PM by les-70 »
Logged

JGO

  • Reg Member
  • ***
  • Posts: 729
Re: SSFP Mk 3
« Reply #86 on: September 29, 2014, 10:26:34 AM »

There is an article on the web "The design of Ruthroff broadband voltage transformers - M Ehrenfried - G8JNJ"  which shows dips in the frequency response  due to differences in winding technique. (PDF doc 1.35Mb)    Something like this might explain the effects Les-70 is finding, since for the bandwidth ratio needed for VDSL, a Ruthroff technique seems essential. 

( for some reason the web address won't copy)
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 31171
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: SSFP Mk 3
« Reply #87 on: September 29, 2014, 04:35:48 PM »

Putting that literal string ("The design of Ruthroff broadband voltage transformers - M Ehrenfried - G8JNJ") into Google brings up a list, with a link to the downloadable PDF document at the top.

If I have interpreted Google's output correctly, the URL is http://g8jnj.webs.com/Balun%20construction.pdf (Perhaps it is the space character between the words "Balun" and "construction" that causes a problem.)

It can be found on G8JNJ's web-page, as the first entry following the Baluns, Tuners & Matching heading. ("Problems with broadband transformers using Iron Powder Cores - why Ferrite cores give better results")

It can also be found here.

  :) 
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

JGO

  • Reg Member
  • ***
  • Posts: 729
Re: SSFP Mk 3
« Reply #88 on: September 29, 2014, 04:59:05 PM »

Many thanks for the help B'cat.
 Yes that is the web page.  I just couldn't copy the address !!
Logged

burakkucat

  • Global Moderator
  • Senior Kitizen
  • *
  • Posts: 31171
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: SSFP Mk 3
« Reply #89 on: September 29, 2014, 05:15:48 PM »

Team-work prevails, once again!
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.
Pages: 1 ... 4 5 [6] 7 8 ... 12