It’s been a rather busy week, so I think it is understandable if the random post is a little more limited in scope than usual.
It seems that my content is being more and more widely discovered each day, but a side effect of it is that it seems to be more and more plagiarised each day. I just want to draw your attention to several things:
- It seems that there are some sites out there which are translating my content for a new audience without providing any attribution (and, at times, collecting revenue of their own for my content). While I’m not totally thrilled about this, I suppose that they (might) have put some effort into translating it and are opening the content to a new set of eyes, so it’s probably a net positive.
- Others have been stealing my images and using them in their advertisements or postings to prove a point without the context or a prominent linkback.
In both cases, I think it’s dangerous to just look at a few pictures, or a graph without also the context that I provide in the article. Some people think it’s too long, and they don’t bother reading (i.e. the TL;DR generation), but this means they can come to the completely wrong conclusion based on the data itself. Please, if you are sharing the content here, make sure you provide prominent link-backs to the content so that people can benefit from fully reading the article.
Furthermore, if you are translating the documents, please also make an effort to read the related articles, linked articles and updates. As this is a log that resembles an engineers’ logbook, I don’t go specifically looking for old articles which could be out of date in order to update it – I will often post a follow-up instead.
In some cases, this is evident from the statistics themselves – where people are being linked to an old article only despite there being a follow-up or update article linked at the top in large bold text. They are missing out on all the facts.
It is because of a lot of this, that moving forward, I would make a more concerted effort to watermark all images being posted to avoid their misuse.
Sydney Trains Meltdown
At around 5pm on Thursday 11th September, around the closing time for ElectroneX, an empty train derailment at Central station occurred with three wheels off the track. This caused rail chaos, with platform changes, but extensive delays across the Western and South lines as trains were banked up waiting to get around the loop. I managed to hop on a train to go out of the city, and for the most part it was a strange sight to be speeding along outbound with trains waiting at the minimum signal distances from each other in a line from Central radiating out to near Strathfield on both the inner west tracks and the western line tracks.
But the joy was shortlived, as my train passed Lidcombe, it stopped for a lengthy period of time as the trains ahead of it weren’t clearing signals fast enough due to lots of switching trains between tracks.
It seems that such meltdowns aren’t that common, but when they occur, widespread disruption happens.
Sydney Trains Station Upgrades
People who have been following along with my observations at Granville would have noted that the new terminus was almost ready for use as of last week. This week, on Wednesday 10th, it was already in service with cars in the new carpark and buses utilizing the new layup area. There were still staff in high-vis vests finalizing things.
Wednesday 10th September 2014
By Thursday, they were nowhere to be seen. We’re now officially in business!
Thursday 11th September 2014
If we compare this to what the terminus looked like before the upgrade, we can see the old dilapidated facilities, but also the decline of Granville as an important interconnection hub. It’s sad to think that they’ve not only lost many buses in the early 2000’s, but now they’ve also declined in train volume too.
A close look at Granville also sees the old fiberglass slat-based benches being replaced by new “personal-space-preserving/anti-hobo” metal benches from Street Furniture Australia. These ones are probably a better choice when it comes to long term maintenance, as the former fiberglass ones needed frequent repainting. Some might even have layers of red paint under the blue CityRail paint for when State Transit Authority was the guys in charge.
Another “improvement” at Granville seems to be the removal of an old concrete stanchion foundation on Platform 1.
The foundation has been filled in with bitumen tar, however, I do miss the old concrete foundation. It was a mound which many people waiting for the train used as an impromptu seat – myself included. Trains rarely stop at Platform 1 at Granville anymore, but when I did catch one earlier this week, I instinctively walked up and tried to sit on the mound, and promptly noticed its absence when I reached the end of the platform without finding it!
Work seems to be underway replacing tactile tiles across the network – this an example from Strathfield of what happens when the tactile tiles see too much weather and become brittle. It doesn’t help that people probably rub their shoes on it and snap them off …
In the meantime, it seems LED replacement of pole-mounted platform lights seem to follow a similar sort of methodology as the ones pictured earlier at Granville – it seems different stations are getting different solutions.
For example, these ones at Redfern seem to have their own photoelectric cells at the top to turn them off during day-time (but their thresholds don’t seem to be properly set).
A closer look of another one at Redfern that was off appears to show it uses a total of 36 high powered white LED packages.
Whereas these at Lidcombe seem more similar to the ones at Granville, and seem to contain more LED packages. This is probably because they have been installed based on the expected lighting requirements.
I don’t know whether it’s intentional but it seems that the LED lights at stations are running for most of the day as well. Maybe it’s for testing or burning in, but at times, the platform seems plenty bright but the lights are still on.
I had the opportunity to get a close-up look at the LED lights at Redfern and it seems they are Philips Master LEDtube Value 19W tubes. They claim to be visually indistinguishable from regular fluorescent, but hey, I could tell the difference the day it was put in! The colour rendition and the light spread of LED tubes tend to be slightly different, and some have hot-spots, but to Philips’ credit, their tubes are very nicely diffused.
A closer look around Redfern station had me seeing about two dead tubes, whereas at Granville, I identified two dead tubes as well (one flickering one went out entirely). It seems that the LED replacements might be suffering infant mortality due to a failure of components or constructions despite their long claimed lifetimes.
Because of scheduling of trains, this week, I had to take a train outwards to Parramatta and change for a train to return back to Granville. This gave me an opportunity to examine the luminaires at Parramatta, which seem to have been entirely replaced.
A new sort of luminaire with a clear glass front and, what appears to be, a waterproof design have been employed. The tubes inside are LED tubes, and of a different vendor than those seen to date.
I tried to take as best of a photo as I could, but I determined that these are eSAVE 22w units, with a manufacturer LED lifetime of 50,000 hours. I suppose this means eSAVE isn’t standing by this when it comes to the whole assembly, which could mean disappointment in the future if the electronics fail ahead of time. The LED chips are from Epistar, a Taiwanese LED chip manufacturer which I’ve seen others deal with, but aren’t the first name to pop in my head for quality.
I didn’t have much time at the station, but I didn’t see any dead tubes in my short walk, which was good.
There was also an OSRAM LED floodlight to illuminate the subway station escalator entrance. I think it’s only a matter of time before all the stations across the network go LED for their lighting, to improve reliability, reduce maintenance and power costs.
It seems like the white LED market is a big one, and it has reached maturity in terms of being economical for businesses. The number of these tubes that are being installed will only continue to rise and more companies will want in on the profits. I just hope that they do reach their reliability expectations, and don’t sour the early adopters, as well as have good electrical characteristics to avoid contributing to the problem of low power factor distorting loads.
Other Changes Across the Network
It seems they’re starting to embrace advertising as a source of revenue, as Central Station’s interchange area is now completely plastered with advertisements, even on the pillars. I’ve never seen it at this scale.
They did do something less complete for advertising the Opal card, but it seems now they’re just leveraging the normal advertising channels with billboard advertisements as well as “on train” advertising.
I’ve also noticed some changes to the carriage numbering of the former outer-suburban Tangara carriages (i.e. the ones with reversible seats, toilets and luggage racks). Formerly, these had the prefix O added to their carriage numbers to denote their outer-suburban status, and would take the form of OD68xx or ON59xx/ONL59xx. The O seems to have been dropped now, as I boarded one with the O scrubbed from outside and inside!
Further looks at Tangara carriages seem to show that they have had their lighting replaced with LEDs as well. The original fluorescent tubes always gave the trains a “warm” yellow light feel, but the one I boarded felt a little different.
It’s not visible in the photo, but a close up examination as in the photo below …
… shows a multi-faceted corn-cob style construction tube with LEDs arranged in a diamond pattern (i.e. :·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·:·).
I wonder who makes these LED tubes, and whether there had to be some special re-engineering, as I expect the ballasts in trains to be specialized for use with the 1500V DC electrification and different AC buses on board, so they might not provide the same output as the retrofit tubes expect (i.e. mains AC) but would be suitable for regular fluorescent tubes.
I mentioned in my last random post that signage has been changing to take the orange colouration – here are two examples:
Finally, they have also updated the signage on Opal buses with new labels containing a blue footer that informs you where to get your Opal card.
I have noticed that some Opal enabled buses have difficulty in enabling the readers at each stop. Maybe they’re having GPS difficulties, or they’re not using GPS at all. Maybe they’re using wheel measured distance? I saw some changing from Please Wait to Tap on/off here just before the stop, and then changing to Please Wait when the doors open. It seems that the driver’s console allows them to manually select the stop and that “enables” the readers.
If you’re using Opal on the bus, you definitely have to see the driver if the readers aren’t operating when the doors open, otherwise you’re going to be charged default fare which is the maximum for that mode of travel and it won’t count towards the daily travel cap! Hopefully these things will get ironed out over time.
Also, in the randomest of random, I thought I’d throw the above image in. This is where two electrical segments meet on the overhead running wire/catenary. In order to isolate two 1500V DC circuits, they use this thing that sort of looks like a hook on the running contact wire. In the direction of travel, from left to right, the pantograph runs along the wire until it hits the hook. The hook then pushes the pantograph down and transfers it to the next running wire.
The catenaries above are terminated into an insulator joiner, whereas the running wire is threaded through the hook like arrangement. The body of the hook is conductive and connected to the right circuit, with the wire joining the insulating body of the hook.
You might be thinking that you’re shorting out two power supplies momentarily when the pantograph runs and touches the hook as well as the left running wire, and it seems you’re right to some extent. This is the main place where you will see bright blue flashes from the tops of trains.
The effect is somewhat current-limited by the fact that the current has to travel through the pantograph, and the contact material is graphite which can be made to conduct better in one direction than another. This same phenomenon is used in Variac brushes to limit the problem with shorting adjacent coils together. At least, that’s what I could deduce from what I know and seeing one in person.
Good Work E*Trade
This week, I got an e-mail from E*Trade about keeping your account secure.
While I’m pretty sure it’s a legitimate e-mail, if you can’t proofread your own e-mails properly, you are bound to confuse your own customers. Having such typos erodes customer confidence in the e-mail as they are one of the common traits of phishing e-mails!
So yeah. <sarcasm>Good job.</sarcasm>
Instead of writing long segments, I thought I’d just commentate on some of the recent news and discoveries in short form.
Most of the rumours came true, in which case, it’s rather uninspiring. New larger phones (who couldn’t see that coming), NFC payments, new watch. The phones are very pricey, but exclusivity is always the Apple drawcard. I’m not going to get involved with comparisons, as its really pointless now when we’re reaching the stagnation point with phones.
I suppose this is a big win for NFC especially because of the close working with banks, but also the fact that the majority of platforms now support it! This means that the institutions won’t be as hesitant as before in leveraging this technology.
I wonder whether the apps will be redeveloped, yet again, for a different aspect ratio screen for their iPhone 6+. At least it seems 16:9 aspect ratio is available, finally.
As usual, Apple is being hush-hush on the watch and battery life, most expect that it will need to be charged nightly. I don’t think it looks as good as I would have expected.
If anything, Steve Jobs is probably rotating in his grave much like a rotisserie chicken – many of this goes counter to his visionary path, and is a reflection of Tim Cook’s real influence showing through. Will people buy it? I don’t doubt that many will. But I won’t.
I think the more important milestone is the retirement of the iPod Classic, which signals an end to the invention that did the majority of the work in getting Apple noticed and reviving the company as a whole. Single function audio players have been almost entirely cannibalized by multi-functional smartphones, it would seem.
It’s been interesting to see the continued one-upmanship in the hard disk storage arena. A short time ago, Seagate announced 8Tb drives with 10Tb to be coming in around a year and it seems that Western Digital were just hanging back waiting to smash their competition with 8Tb and 10Tb helium models using Shingled Magnetic Recording.
It’s not an apples to apples comparison, of course, and the use of helium, more platters and a new recording technology is sort of cheating, but it does sound the death knell on plain Perpendicular Recording Technology. We’ve been through this several times – PRML+GMR based longitudinal recording, RLL, MFM for example. But it might mean some changes in how those drives are used due to Shingled Magnetic Recording peculiarities.
It also seems that they are also game to sell drives at odd capacities which seems to be a great thing. Predominantly targeted at offline storage, it’s true that all drives are manufacturer tested and each head/platter combination has subtly different recording capacities due to track width, head response and defect differences. It’s no good for RAID of course, but as a plain drive, it’s not a bad idea. If only they could allow us to overclock our own drives by adjusting how much area is reserved for spare sectors, and tinkering with the encoding ECC protection level …
But that’s not all. Sandisk decided to launch a 512Gb Extreme Pro UHS-3 SDXC card. It’s big, fast and pricey! But at least it’s fast, since most of the time the larger cards tend to be the slower ones. It might mean that a 768Gb TLC SDXC card is achievable already …
It also seems to be a big trend for SSD manufacturers to ship performance enhancing software such as Samsung’s RAPID mode, and Plextor’s Plextool. Looking at this KitGuru review of the M6 Pro, a basic consumer might be mistaken that the software is magic. It’s definitely not. It’s only returning high numbers since the benchmark is taking in place in RAM. It’s as if you have a large ramdrive and you’re testing that. The writes aren’t committed to the drive, otherwise they will be subject to the SATA link bottleneck and overhead.
Likewise, while there’s a claimed write-coalescence benefit, a less widely publicised fact is the danger of these acceleration softwares. The issue of a RAM cache for writes was recognized early on when disk caching software was invented, and servers have insisted on having battery backup on RAM caches on RAID cards, or disabling caching on writes (write-immediate).
When there is an incentive to churn out fast numbers for benchmarks which occupy about 1Gb, the RAM cache is now much bigger than before which makes the potential for data loss much bigger than before. Some of them claim to have methods to reduce data loss, but it’s pretty clear that performance above the SATA III link rate and no data loss are mutually exclusive goals! It’s not possible to accept writes to the drives faster than the drive can write it and not store it elsewhere, and if that elsewhere is RAM, it will vanish if the power is lost!
It also means that if you have RAM issues, it’s pretty likely you will corrupt the drive or your writes as well! I make it a point to avoid this sort of software, it’s barely going to make a difference in real life especially if you don’t have loads of “vacant” RAM to expend.
Lithium Ion Dangers
I’m not sure as to whether the editors interpreted this story correctly, but it seems a drone operator has had a few batteries catch fire in his case after they were loaded into cargo on the plane. From the article, it seems that the plane had not taken off yet when it was caught, but then it goes on to mention pressure and temperature changes.
It is possible that many new drone operators aren’t as aware of the dangers of lithium-ion cells. The drone environment can expose cells to significant heating, heavy current draws and even physical impact shocks. These can all provoke borderline or damaged cells to let go and catch fire. More experienced RC helicopter flyers are advocating fire-proof areas or enclosures to charge cells and to monitor cells in case of unusual behaviour.
I think it’s definitely good advice to board with all the Li-Ion cells with you in carry on luggage. The lower mechanical shock and active human monitoring will avoid catastrophes like an uncontained cargo fire.
Screenshots of Microsoft Windows 9 Threshold Leaks
It seems that the tech world thrives on leaks, but the cynnic in me seems to believe the leaks to be an almost intentional release of information to tease the public and test the waters before actually releasing something. After all, they can just claim that the rumours and leaks were false … and get away with it!
Unfortunately, this isn’t really confidence inspiring when it comes to consumers and creates confusion. The same thing happens when new models are announced but they’re not actually available on the market yet. In some senses, that’s anti-competitive too.
Anyway, back on topic – alleged screenshots of the new Windows 9 leak, and it looks like an indecisive meld of Windows 7 and 8.1 as expected. The new start menu seems to be needlessly changed in order to allow the tiles to show, while also cramping the functional parts of the menu. Eugh. It also doesn’t look aesthetically pleasing.
The whole metro-apps running in windows also runs counter to their initial idea to generate a store of “tablet” style apps, but it recognizes the importance of efficient multitasking. Unfortunately, that’s just going to create UI uncertainties with different sized viewports, and I don’t see their app store as being a valuable source of applications for the desktop at this time.
It also apparently removes the charms bar, which is a feature I don’t mind because it means quick access to certain settings and shut down. Can’t they ever get anything right? As it turns out, I’m not your average consumer, so they probably don’t care what I think. But regardless, I think it’s a necessary upgrade, forced down our throats because of DirectX12 and USB 3.0 UASP support amongst other changes.
Stray J’s in E-mails
I’ve seen stray J’s at the end of the line in some e-mails and I never bothered looking it up until now. Turns out there’s quite a backstory, which might interest some readers.
As usual, my random posts are pretty random and don’t suit all, but it helps me clear my mind for more important things. This upcoming week, I have many things to prepare and get ready, and unfortunately that means less time for blogging. If it’s a bit quiet for a while, now you know why!