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 ... 7 8 [9] 10 11 ... 44

Author Topic: FTTC - Distance from the Cabinet - Realistic speed estimate  (Read 253952 times)

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #120 on: August 19, 2011, 12:05:30 AM »

First catch the rat . . . :P

O.k. What I did was to ignore the fact that were are talking about VDSL2 frequencies and to plug a selection of line lengths into Kitz' Maximum Speed Calculator. Once I had the typical attenuation values for the two line lengths of interest at ADSL frequencies, I made a rather crude approximation that, over the same line lengths, the relationship of the attenuation values tends to be linear. Crude scribbled sums on my notepad resulted in the 15 to 18 dB range being ringed at the bottom.

I would be interested to know if Mr Pag has any real observed data that we could use but considering what he has recently written, I suspect that none is available. :(
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.

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #121 on: August 19, 2011, 12:37:25 AM »

Just for curiosity, I tried the Kitz ADSL calculator the other day, using my 30.2 dB attenuation.

The result was 2.2 km or 1.4 miles, almost 3 times my likely line length, but I had no idea what effect different frequencies/SNR etc. would have when considering VDSL2.

razpag's readings were of great interest as they gave me a slight indication of what I should be expecting, especially as most of the line was aluminium (which has been suggested for my line's probable age).

Ideally he'll do a 0.5 mile installation tomorrow & post the results here :)

On Monday, when my line is all fixed again, I'll post my results & brag about my super duper speeds. ;D

EDIT:

Some sort of a record including theoretical & actual results would be quite useful for cases like mine (& no doubt for others in the same position).

If anyone knows where there is such a record for FTTC (I can't find one anywhere), please let me know.

Alternatively, if anyone wishes to post their stats (not at all easy to obtain), I'll start one off.
I could do with a bit of help regarding theoretical values though as I haven't sussed out how to calculate them for FTTC.

Paul.
« Last Edit: August 19, 2011, 11:56:13 AM by Bald_Eagle1 »
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #122 on: August 19, 2011, 11:45:06 AM »

@ razpag,

I have just read the older thread regarding the Eclipse test. I'm glad you mentioned it.

Useful info to add to my checklist for Monday..................just in case.

Woosh test? Could this be relevant?

Paul.
« Last Edit: August 19, 2011, 11:47:37 AM by Bald_Eagle1 »
Logged

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #123 on: August 19, 2011, 12:40:19 PM »

@ razpag,

I have just read the older thread regarding the Eclipse test. I'm glad you mentioned it.

Useful info to add to my checklist for Monday..................just in case.

Woosh test? Could this be relevant?

Paul.

The engineer will NOT have WOOSH access. This is a facility we used to have (don't get me started) but now only certain elements of Openreach along with BT Wholesale, have this access. It's basically a tool that allows them to 'talk' to your Hub/Router to see current stats. But the best bit is, it also has the RRT which can view your circuit historically. If you recall, I mentioned somewehere about how the 'baseline' is logged on RRT which should show the attenuation when the broadband was first connected.

I honestly don't know if other ISP's can view WHOOSH, or have their own form of it ??? 
Logged

jeffbb

  • Kitizen
  • ****
  • Posts: 2329
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #124 on: August 19, 2011, 07:58:17 PM »

Hi
@BC No need to scribble I used the same method but just something near what you would expect say 10db then adjust until 0.8 distance was shown.result 11db. Saves the old grey matter  :)

more info  at http://www.internode.on.net/residential/adsl_broadband/easy_broadband/performance/including (fromKitz acknowledgement.)

It only goes up to 24000kbps , but you can see that even at 24000kbps the distance is much more critical ,anything above  900 Metres then there is a rapid loss of Synch rate . So it would seem reasonable to assume on a nominal 40000kbps there would be some losses in synch rate starting at about 500 to 600 metres Metres  to the cabinet (FTTC)
Regards Jeff
Logged
zen user

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #125 on: August 19, 2011, 09:26:09 PM »

Jeff,

Thank you for the link. Once I had removed the trailing "/including", I then saw the page with the graph. So using your means of estimation, what sort of DS attenuation do you think Baldy Mr Eagle should have with a 805 metre D-side pair?
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.

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #126 on: August 20, 2011, 07:19:31 AM »

Addendum to my post further up this thread re: the attenaution from the PCP to the EU's. It was as said, 13.5dB on a 488mtr length of wire ......... and 95% of this was made up of Aliminium. The EU still received 39.9Meg on the DS.

 ;D

Just curious razpag, did you obtain the 488m length from the JDSU, or by some other means?

488m = 0.3 miles (approx)

0.5 miles = 805m (approx)

Roughly, what do you reckon the attenuation should be for 0.5 miles (805m)?

Paul.

Apologies, I missed this post for some reason ??

I acquired the total metreage, by adding together the different lengths of cable in the ground from PCP to DP. These lengths are shown on our 'Network Records' programme installed on our laptops. We then have a 'ruler' function that allows you to manually measure from point A to point B, in this case from the DP to the premises which is the length of the dropwire in effect. The figue I quoted will have a +/- factor, but it won't be that far out, believe me. ;)

Just one other point that I forgot to mention. The quoted attenuation figure on this job (13.5dB) taken at the EU's with my JDSU will be higher than the true figure reported once the EU's kit is plugged in. The reason being the resistances in the JDSU are higher than standard Hubs/Routers. So, the actual reading testing via WHOOSH may be approx 11dB. 
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #127 on: August 20, 2011, 07:48:34 AM »


I acquired the total metreage, by adding together the different lengths of cable in the ground from PCP to DP. These lengths are shown on our 'Network Records' programme installed on our laptops. We then have a 'ruler' function that allows you to manually measure from point A to point B, in this case from the DP to the premises which is the length of the dropwire in effect. The figue I quoted will have a +/- factor, but it won't be that far out, believe me. ;)


Cheers razpag,

Do I take it that all engineers carry laptops with access to the Network Records programme?

If so, Monday's engineer will be able to confirm once & for all, the actual line length & actual line route from the PCP to my home then?

This would then remove any doubt regarding "theoretical" speed estimates based on reported line length rather than actual line length etc.

Do you have any idea of the line attenuation thresholds where BT actually accept a "reportable fault"?
i.e. Assuming 800m line length, are my 30.2 dB DS & 52.9 dB US line attenuations "within specification" & therefore not classed as a reportable fault?

Also, I believe the JDSU is able to report line length (possibly calculated based upon actual line conditions)?

Did you notice that measurement from your recent install job, to compare against the Network records?

I'm sure you must have taken a peek at your own JDSU line stats at your home.
How do they actually compare against theoretical estimated speeds obtained from BT's online line checker, attenuation SNR etc?

Paul.
« Last Edit: August 20, 2011, 08:06:17 AM by Bald_Eagle1 »
Logged

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #128 on: August 20, 2011, 01:29:21 PM »

Not enough time to respond in full, as off out on t'lash with t'lads in 10 mins. Just waiting for my taxi to show up. If I forget to respond more comprehensively later, then yes, all engineers have Network Records on their laptops, (and this will give you a 99% accurate distance you are from the PCP). Whether they can use them or not, is a flip of a coin. It generally depends on the engineers skill-set as to the probability of using Network Records. I'm talking generally here, so any OR pedants please take it as so. ;D

Will try to go more in- depth tomorrow Paul, as no doubt I'll be talking scrotums later tonight.  ;) ;D

Logged

jeffbb

  • Kitizen
  • ****
  • Posts: 2329
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #129 on: August 20, 2011, 07:11:51 PM »

Hi
quote @BC:what sort of DS attenuation do you think Baldy Mr Eagle should have with a 805 metre D-side pair?
As per my previous post 11db ( RP's caculations and Kitz's calculations agree :))
The graph is used to show the effects of distance/attenuation on synch rate for a given connection  8000,24000 Kbps. The attenuation is not affected by the the max rate for the line .

If you meant what sort of synch to expect at 800 metres then with a little bit of fiddling and analysing the graph and extrapulating for a 40000 Kbps ,that if the line was top quality might make 35000Kbps .But that is only a guesstimate, probably a little lower in real life say about 30000Kbps.
Regards Jeff
Logged
zen user

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #130 on: August 20, 2011, 09:39:52 PM »

Purr. (Translation: 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.

waltergmw

  • Kitizen
  • ****
  • Posts: 2776
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #131 on: August 21, 2011, 12:48:05 AM »

@ Jefbb,

To a first approximation, your speed estimation is roughly what Paul was achieving just after his FTTC service was installed.

Kind regards,
Walter
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #132 on: August 21, 2011, 08:25:16 AM »

@ razpag

Plusnet's notes on my ticket regarding Monday's visit include this comment:-
"PSTN engineer booked to check for HR fault on D-side which may be affecting broadband
speed."

Just to TRY to ensure that no stone is left unturned, from what I have read, this is my understanding of what the engineer should be checking:-

a) TDR test on the D-side from the original PCP

b) From the JDSU, FTTC Connection speed, Line Attenuation, SNR, line length from the PCP

c) TDR test from my master socket

d) From the JDSU, FTTC Connection speed, Line Attenuation, SNR, line length from my master socket on arrival

e) Line length & line route from Network Records, accessible via the engineer's laptop

f) Results of the Eclipse test

g) From the JDSU, FTTC Connection speed, Line Attenuation, SNR, line length from my master socket on departure

h) A request by the engineer to BT to reset my profile.

i) Has a "lift & shift" been carried out at the PCP

I will run BT Performance Tests & speedtest.net speed test in the morning, & assuming the engineer calls me to say he is on his way, immediately run the tests again (1 hour wait between BT Performance Tests).

Question 1) Have I missed anything, or incorrectly listed anything?

Question 2) Can the engineer manually "reset" anything at the DSLAM in the PCP?

Question 3) As a PSTN engineer has been booked, will he actually automatically run the FTTC tests, or just look for a "voice" fault?

Question 4) If, as previously, the engineer "politely" refuses to, or doesn't know how to carry out any broadband related tests and/or remedial work, what can I do about it?

Question 5) If all the relevant tests have been carried out & everything is "LTOK", what else, if anything, can be explored?

Question 6) Will the engineer have access, either on site or via phone contact with BT, to records of the line condition & FTTC speeds from the original installation date (24/06/2011)? If so, what is the name of they sytem in which these records are stored?
Plusnet have confirmed they do not have access to these details.

I may be worrying unnecessarily. However, as communications to the engineer at the sharp end don't appear to filter through in full, I am concerned that either the "wrong" type of engineer will be visiting again, or that his/her instructions for the visit are not as comprehensive as perhaps they could be.

To everyone:-

Thank you all so much for your help, comments, advice, time spent behind the scenes loking into this matter for me etc.
I only recently registered with Kitz & am amazed at the time & effort you put in for complete strangers who immediately bombard you with questions (& write long rampling posts).

I have learned so much in such a short space of time, & I hope that I may be able to pass some of this knowledge/understanding on to other unfortunate souls in the same boat as me.

Hopefully, one of my posts on Monday will be to confirm that a "fault" was located & repaired & that I have ended up with a stable line & regained my previously fantastic FTTC speeds.

Best regards to all,

Paul.
Logged

Bald_Eagle1

  • Helpful
  • Kitizen
  • *
  • Posts: 2721
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #133 on: August 21, 2011, 09:05:43 AM »


If you meant what sort of synch to expect at 800 metres then with a little bit of fiddling and analysing the graph and extrapulating for a 40000 Kbps ,that if the line was top quality might make 35000Kbps .But that is only a guesstimate, probably a little lower in real life say about 30000Kbps.
Regards Jeff


@jeffbb

I have downloaded the spreadsheet (version 1.4) that I believe the Kitz Calculator was based on. I have attached it for reference.

I have no idea whatsover whether VDSL2 & ADSL2+ both function the same way regarding the effect of attenuation.

Entering an attenuation of the 13.5 dB that razpag quoted from his recent FTTC installation, the resulting calculated line length is 978m.
This is almost exactly twice the 488m that was gleaned from BT's Network Records.

The suggested ADSL2+ Attainable Rate from the calculation is 22612 kb/s.

This equates to approximately 94% of the maximum 24000 kb/s for ADSL2+

Assuming twice the 800m that I THINK is my physical line length, an attenuation of 22 DB gives a line length of 1593m.
For ADSL2+, this suggests an attainable rate of 19535 kb/s, roughly equal to 81.4% of the 24000 kb/s.

Again on an assumption that ADSL2+ & VDSL2 function the same, 81.4% of the 40000 kb/s for FTTC would be 32560 kb/s.

This may be pure coincidence, but initially I was actually achieving download speeds in this region, occasionally over 33000 kb/s.

As download speeds are always some value less than sync speed, is this more or less how you derived your estimated theoretical sync speed of 35000 kb/s for my circumstances?

EDIT:

Just as an aside, my 30.2 dB attenuation as reported by the JDSU gives an approximate line length of 2200m & an ADSL2+ attainable speed of 15132 kb/s (63% of the 24000 kb/s).

For FTTC, 63% of the 40000 kb/s would be 25220 kb/s. My IP Profile has gone up & down quite a lot in the last few weeks, & therefore the sync speeds must have also fluctuated.

If any of the above assumptions do stack up, it would reinforce our theory that my Line Attenuation MUST have increased since the original installation (also reinforcing the possibly intermittent high resistance fault theory) from when I was achieving downloads of around 32000-33000 kb/s.

All the above calculations have assumed a cable loss of 13.81 dB/km.
Is this a reasonable average value for a mixture of copper & aluminium as used in the UK, or is it a typical value for good quality copper?

Getting Monday's PSTN engineer to deal with this accordingly may well prove to be a completeley different ball game though.
It would probably take all his allocated 2 hours just to explain the issues
:lol:

Paul.

[attachment deleted by admin]
« Last Edit: August 21, 2011, 09:50:21 AM by Bald_Eagle1 »
Logged

razpag

  • Reg Member
  • ***
  • Posts: 622
Re: FTTC - Distance from the Cabinet - Realistic speed estimate
« Reply #134 on: August 21, 2011, 02:01:54 PM »

Hi

Rather than take the questions one at a time, I'll attempt to explain the skill-sets of different engineers and encompass your queries within the post.

You say you are getting a 'PSTN Engineer' on Monday. That is what we refer to as a 'Network Engineering visit', as opposed to a 'Broadband Engineering visit'. As such, the ECBT (Estimated come back time) will probably be less than the 2hours allocated to Broadband jobs. However, these timelines are only a guide as to how long historically, and as an average, a 'Noisy LTOK' type fault will take. Anything over this ECBT and the engineer will then be 'inefficient'. TBH, the eventual 'Clear-code' the engineer uses when closing the job, is what our performance system (i-POP) actually looks at and generates the stats from. Slightly off at a tangent, I know, but time is precious on tasks and I just want to make you aware that he/she won't want to be subject to the whole sordid tale, just make them aware in a potted format of whats gone on and what advice you've been given from engineers. My major concern is that you have a 'Network Engineering visit', so unless he's 1)Broadband trained and 2)Willing ........ he won't entertain anything you have to say about speeds, attenuation etc etc. Not because he's being awkward, but because he's not there to work on the DSL part of the circuit, just the PSTN (Telephone-speech).

The skill sets then. The lowest single skill (IMO) is the Frames Engineer. He's the guy in the exchange running 'Jumper wires' connecting up telephony and DSL and putting them through to the Bar/Pair (Fuses) that feed the Cabinet that in turn feeds the EU's premises.     

Now then, with a 'Noisy LTOK' type fault, there is an extremely high chance of just getting a CAL/OMI engineer. Customer Apparatus and Line/ One Man Installer. This is really a very basic skill which sees them only working from the top of the telegraph pole into the EU's premises. So if there's a fault on the underground cables, they will have to either 'further' the job to an engineer with the right skills, or try a 'D-side pair change' from PCP to the Telegraph Pole.

If your premises are fed via an underground feed, rather than overhead, there's a chance you may get a CSE Engineer (Customer Services Engineer). This animal is trained to work on all parts of the plant and as such sticks with the job unless he too has to 'further' the task for a reason, such as safety, no line plant, traffic lights etc etc.

FJ. This is a Faultsman Jointer and basically he works on only the underground plant.

The cream of the crop are CSE with Broadband/Fibre skilled engineers (well, I would say that  :blush:), who, even though they are attending on a 'Network' visit, will sometimes and with time permitting, carry out DSL checks as well, in order to alleviate the EU's worries that the line may still be faulty. It really is down to who the engineer is and the relevant skill-set he has.

I do know that Broadband/FTTC repair is tagged to the CAL/OMI queues in 'Works Manager'. This is the machine that automatically 'Pins' the work out to engineers. Myself and others have argued that only CSE's should be upskilled to BB/FTTC, as we do get underground faults from time to time. As time goes by though, more and more CSE's are getting this training and IMO, giving a better service to ISP's and their EU's. Just to add my own personal bitter point, the 'cream of the crop' engineers do all the jobs above, including the Frames, and get paid exactly the same as the Frames Engineer. Ridiculous state of affairs that even Frames Engineers agree with, >:(

As we know, you're getting a 'Network' visit, so a lot of your Broadband questions won't be getting answered I'm afraid. What I would personally do though if it was me attending on this 'Network' visit (so no Broadband faulting techniques), is to go straight to site. Strip the socket down so I'm effectively connecting straight onto the pair of wires coming from outside. Then my first job would be to perform a PQT (Pair Quality Test) which is a high-end test that can unfortunately still not see a 'HR'. Also bear in mind not all engineers can do a PQT. I would then spend 10 mins listening on 17070 (Quiet Line Test). If noise is heard, or not, I would then proceed to attach a TDR meter and try to find a 'HR'. If one doesn't present itself, I would then ring the PSTN number with the TDR meter still attatched and continue looking for HR's. The higher AC voltages that ringing generates, sometimes shows the 'HR' when it's in its infancy so to speak. For info ... TDR is always best when testing back towards a voltage, rather than on a 'dead' pair.
If that still doesn't show anything, I would then head to the Cabinet and disconnect the D-side and perform a 'Leg Balance' test. This does as it says on the tin. It measures each leg seperately and will always give an accurate reading no matter how small the 'HR'. Along with this, I would perform other tests such as looking for 'Battery' or 'Earth' contact faults, that a low-end test like 'Eclipse', 'RAT' and 'TAMS' can't see. There could for instance be 3/4/5 V of 'Battery Contact' on your line and 'Eclipse' will still give a 'LTOK' result. If everything passes and the line is quiet at the Cabinet, there's not much else I could actually do.
If it was a Broadband Engineering Visit, I would do all the above plus run a DSL test and look for errors etc etc.

Regarding 'Network Records' that you ask about. All engineers have them on their Laptops, but as I've said earlier it rather depends on the engineer as to whether he can use them or not. In my experience, CAL/OMI engineers don't know how to use them but there are exceptions to the rule. FJ's and CSE's will definitely know how to use them.

I hope this answers some of your questions, and apologies if I've missed something. Your queries about DSLAM's, Lift & Shift etc ,are fruitless however because of the type of task you say has been built against your premises. Me persoanlly, I would have rather you had a 'Broadband Engineering Visit' from a CSE. That way everything is encompassed within the skill-set. Lets not panic just yet though, as you may get a great engineer who does indeed find a fault of some kind and then it's happy days again. Best of luck Paul. ;) ;D
Logged
Pages: 1 ... 7 8 [9] 10 11 ... 44