Visited: ElectroneX Expo (5-6th September 2018)

Every two years, the ElectroneX Expo comes to Sydney. This expo is like a trade show for those involved in electronics design and assembly with a number of stalls from relevant vendors, free talks and is held in conjunction with conferences from the SMCBA (Surface Mount and Circuit Board Association). This time, it was held at Rosehill Gardens rather than at the Australian Technology Park. This was closer to home, but entry to the venue wasn’t as straightforward as we had hoped, with a few people coming by bus from Parramatta getting a little lost (and eventually, decided to follow me … an equally clueless bloke with a plan – get to the train station to cross the tracks).

The Expo

Rather surprisingly, this time, the show seemed more vibrant than in previous years. It seemed there was more show-floor area, more exhibitors and crucially, more people. Arriving at the registration hall, it was unusual to find myself queueing behind a number of people to get my badge. It was only a few minutes past opening!

Unlike previous years, I took a few more photos of the event, but not in a serious effort to cover the event. After all, it was a chance to see “old friends” in the industry who I might not know by name and new technology alike. I was there to help, learn and enjoy – and I think I was able to achieve that quite well.

In what follows, I’ll discuss a little bit about the companies I talked with and the things I learned about their operation. Note that links and discussion are provided as a courtesy to the reader and their mention does not represent my endorsement of their company or services. The post will only mention companies which I’ve conversed with and had some interest in – unfortunately, not everyone will be mentioned but the list of exhibitors and stuff on show is online.

One category of companies on the floor were contract manufacturing companies who were willing to take designs and bill-of-materials (BOMs) and source, fabricate, populate and test. Exhibiting this time were OnTrack who were based locally in Mascot but do their PCBs overseas, Leach (SZ) Co. Ltd. who are based in Shenzhen with a long track record of experience and GPC Electronics who manage the process with Asian manufacturers and have done work for some rather recognisable brands. Owing to the issues surrounding products and intellectual property, it seems that all companies are now willing to sign NDAs, however how “watertight” this may be remains to be seen.

This year, when it comes to electronics component suppliers, it seems a few big names are around again. Mouser were on show, with a small stand and a few representatives. My favourite element14 was back again this year, with a small stand as well and mostly fresh faces. I’ve had some concerns about their acquisition by Avnet, but it seems that after a period of uncertainty, the Australian arm is actually doing well. That’s great news for us “down under”. I was also told about a few new products “on the way” which should be interesting for those looking for good value, but I’ll keep my lips sealed so as not to spoil any potential surprise. ADM Instrument Engineering were back again this year, representing MeanWell power supplies as our local supplier. It’s good to be reminded that we do have “local” options for these things. One group I wasn’t aware of was Fairmont Marketing who are able to obtain components through multiple sources, claiming to get a “better price”. They were also here representing Mornsun power converters and ICs. Finally, for those who are interested in passives, magnetics and connectors – Würth Electronik are back again this year, reminding us all that we can order directly from them and save some money along the way. Finally, there was also Altronics, although I’ve not dealt with them much.

There were a number of specialist component manufacturers on show as well. As with previous years, it was nice to see both OKW Rolec and Hammond Manufacturing at the show again with a large sample of enclosures on show and a few free samples (if you’d like). The Hammond case I got last time became my Hearing Loop Receiver. As with previous years, it seems like the main showbag was sponsored by WAGO Technologies, a very reputable connector company. There was Mastercut Technologies, a local company who specialise in fabricating thin metal parts such as signs, solder mask stencils and shims. I also spoke to Reid Industrial Graphic Products who are a local manufacturer of a number of interesting things including flexible electronics substrates/sensors/input panels. When it comes to specialist PCBs, it was good to see Lintek (another local company) on show with their mesmerizingly-complex and fine designs. So many of these companies are local, it proved to be surprising to me.

When it came to embedded systems, it’s no big surprise to see Congatec on show – it seems they’re at almost every show I end up attending. As usual, the wall of Intel-based and i.MX based boards was on show. This year, I came across Glyn High-Tech Distribution (a company I wasn’t aware of) who seem to carry a lot of different embedded systems development boards and system-on-modules, including wireless solutions. I also stopped off at Outerspace Design, an industrial-design company that helps bring prototypes to market. They showed off a smart dog-treat dispenser which plays games with dogs … which seemed pretty cute, almost like a miniature cement mixer.

Saving the best to last was the category of test equipment manufacturers and distributors, with a good representation across the board, another surprise for a “small” market such as ours. At Trio Test and Measurement, there were a number of different devices from a range of brands. There was a number of “value segment” Siglent equipment which didn’t look too bad at a glance, alongside some Keysight (formerly Agilent) equipment. Interesting how they’ve just plastered the brand over the Agilent logo. But that DC electronic load underneath looks familiar … I suspect the B&K Precision Model 8600 I have is actually a Maynuo rebadge.

Surprisingly, the Keysight representation doesn’t end there – Keysight had their own stand where they showed off a number of their new oscilloscopes including their new Streamline Series USB-connected devices. It seems maybe traditional vendors are at long-last, challenging the likes of Pico Technologies rather than leaving that market untapped. They were also demonstrating the FieldFox, which I’ve seen several years back. I just didn’t know back then that it was a cable-tester as well with VNA capabilities – not just a portable spectrum analyser. I had some time with the representative chasing several “narrowband” signals near a UMTS carrier in the ~850-900Mhz band, wondering what they were. I suspected they may be the result of unfiltered harmonics or intermodulation and the representative proffered the suggestion they could be wireless microphone signals. Reducing the RBW, we could see it was a little asymmetric but similar to NFM in width, but a decode produced just noise. Reflecting back on it, that may be expected if the microphone is a digital system – however, such microphones should operate in the 520-694Mhz region nowadays. Whatever it was would have very likely been quite out of band.

National Instruments had some representation, but I didn’t have the time to stop off and take a peek. I probably should have stopped, but I decided to make a bee-line to a stand that was showing GW Instek and Newton’s 4th equipment. GW Instek make a lot of power supplies, which are commonly deployed in teaching labs, so their brand is familiar. That being said, I didn’t know they also offered various digital storage oscilloscopes and mixed signal oscilloscopes as well. I didn’t have the chance to play much with them, but it seemed fairly similar to what Rigol might offer. At the low-end, it’s a bit of a dogfight, with other brands like Siglent, Hantek, Uni-Trend and others trying to break into the market – but this means that consumers are finally able to afford a decent oscilloscope at a good price.

The next stop was the Emona Instruments stand to check out the latest from Rigol Technologies (the almost undisputed “value” choice) and a VNA from Pico Technologies. It seems that since my review of the Rohde & Schwarz RTM3004, some people were wondering if Rigol actually had a comparable “value” offering. I managed to spot an MSO7054 on the show floor which has a lower bandwidth (500Mhz) but claims a higher 10GSa/s sample rate (interleaved, with only 2.5GSa/s in four channel mode and only 8-bits) and deeper memory at up to 500Mpts (with 100Mpts standard). It had some superficial similarities – 10.1” touch-screen operation (but only 1024×600 resolution) and support for USB mouse, with three ports (which is nice) instead of one as on the RTM3004 necessitating the use of a hub. On the downside, the button panel layout and labelling was a little ugly and misaligned in places. Not the prettiest of units in my opinion. Not having lived with such a unit, I have no idea whether it has any issues – but on paper, it isn’t quite as well spec’ed as the RTM3004 and the price didn’t seem too special either.

Vicom were back as well, representing the Tektronix and Keithley brands (as they’re basically one-and-the-same now). The Keithley half of the stand had a number of interesting products including the 2400 SourceMeter unit – a “four-quadrant” power supply that measures as well. They also had their DMM6500/DMM7510 (I can’t remember which) touch-screen high-accuracy DMM that could also do graphing. Of course, you bring out your most powerful stuff for a show.

On the Tektronix side, it was quite exciting too – the RSA306B was on show (one I’m familiar with, as I had tested the RSA306 in the past) alongside a TTR500 PC-connected VNA. The representative was having a little difficulty in setting up a nice split-screen with Smith-chart with VectorVu-PC, so I had five minutes to have a tinker. I’d have to agree, as a user that’s never used the software nor read the manual, that the menus are a bit confusing as a “nest of nested menus”, but I was able to get there in the end. It seems that Tektronix have no problems embracing PC-connected instruments, at least in the RF arena. But the big attraction was the 5-series MSO with its massive touch-capable screen and (at the show) 6-channel input. Not a cheap unit, for sure, but mighty impressive to see they’re not being left behind when it comes to the user interface design.

But of all the stands, the one I was most eager to see was the Rohde & Schwarz stand. They sure didn’t disappoint! There was a lot on show, including the RTM3004 oscilloscope (which I recently reviewed, and am still waiting for a few firmware fixes for) – I felt at home as soon as I saw it. But seeing as I never had a chance to see what their other units were like, it was actually nice to have some time to play with an RTO-series unit (a much higher-end, compliance-capable unit) and examine the menu layout and play with the history navigation. I’d have to say, navigating between history records with the dial was actually quite a pleasurable experience compared to pressing a touch-button on the screen. There was the “flagship” RTP unit, which boasts rather impressive specifications – I didn’t play much with it but I did notice that the exhaust air was quite a bit warm so the unit is probably working quite hard. At the lower-end, they had an RTE-series unit as well, but it wasn’t the star. Alongside these oscilloscopes, it was also impressive to see a microwave signal generator (I suspect it was the SMB100A) hooked up to an FSH portable spectrum analyzer showing a 10Ghz signal or so. I flexed the coax, the signal changed … I even got to tweak the modulation width setting on the generator. It’s nice to have a chance to go hands-on.

So I didn’t let the chance escape – I took my time to play with the FPC1500 1-port VNA – a unit offered for RoadTest that I missed out on. Its wide screen and performance was quite impressive, although the menu system being different to the other equipment did take a little getting used to at first. I also got to play with the buttons on the HMP4030 power supply and explore the menu. This was a unit I applied to RoadTest, but am awaiting the outcome. It was a nice outcome to be able to preview the unit in person – the unit on display still bears the Hameg brand, which R&S acquired. They also had a more advanced HMC-series supply on show with a colour LCD, which made the HMP4030 feel a little dated in comparison. Definitely have to thank the representative for being so accommodating of my inquisitive nature … I managed to crash the ZNC-series (?) VNA on show as well but apparently that’s a known issue that’s awaiting resolution. Sometimes I just feel that some companies might be a little lax on their QA procedures – after all, test equipment is one category of equipment that probably shouldn’t have bugs … but every vendor is guilty to some degree – I can definitely say the same of products I’ve tested from the “majors”.

Also of mention is Silicon Chip, the Australian electronics magazine that still survives through to today when its peers have long perished. I suppose if you enjoy electronics and you can afford it, it might be a good idea to take out a subscription. Another is the UNSW Sunswift car, which seems to be getting plenty of mileage at trade shows.

In the end, the haul was pretty substantial despite trying my best to resist taking things home.

What is missing from the shot is about 1.5kg of catalogues, magazines and promotional material.

Conclusion

ElectroneX for 2018 has come and gone – the next chance to see it in Sydney will be in 2020. That being said, this year’s show was unexpectedly good – in number of exhibitors, in the amount of things on show and in the number of people attending. It’s refreshing to see signs that the electronics market in Australia is still alive and great to see familiar faces even if I’m not on a first name basis with them.

Best of all, to everyone who I spoke with – it was a pleasure to chat with you all, thanks for your time! See you in another two years!

Posted in Electronics, Event | Tagged , , , , , | Leave a comment

Tech Flashback: Landline Sounds & Telephone Teardowns

I’ve always been interesting in telephones and networks, but I’ve never really thought of becoming a telephone collector. Rather unexpectedly, through a process of not-throwing-things-out (it’s not hoarding, I swear!), I now have a few phones but not much to do with most of them.

Then I realised the main reason why – I’ve not had access to a real landline for the better part of a decade. After ditching dial-up for broadband, we went straight to “naked” ADSL2+ which meant no line-rental cost, but also no actual landline. We were able to compensate for this with SIP VoIP through an ATA, but even that got very little usage. As a result, I realised there are many sounds and experiences we’ve since lost with the transition to more modern forms of communication.

After setting up my home PBX, it would make sense to run a few demos just for the sake of preserving the sound and feel of a landline, seeing as these are going to become practically extinct with the full cutover to NBN happening within the next year or two.

The Landline Experience

In the not so distant past, a telephone call was the most convenient way to interactively get in touch with someone who may be a long way away. At one point in time, almost everybody had a landline – only the most privileged may have a mobile number, whereas the lesser-privileged may have a pager number which you can call to leave a number or a message. The cost of having this land-line was the cost of “line rental”, which at this time is hovering somewhere about AU$30 per month with all calls billed additional to this. Owing to the economics, there’s no reason to have a landline when mobile telephone service can be had “all calls and SMS inclusive” and with 1GiB of data for AU$10 per month. The introduction of the NBN also means that copper land-lines are being deactivated as well.

To use a landline telephone, it’s a case of:

  • lifting the handset
  • listening for the dial-tone – if it’s there, the line is working and ready to accept digits
  • dialing the number – on older rotary pulse-dial phones, this means putting your finger into the corresponding number hole, rotating it to the finger rest and letting the dial return to its rest position, while it “clicks” the number onto the line at a rate of 10 pulses per second. On newer touch-tone (DTMF) phones, this means pushing the buttons that correspond to the telephone number in sequence.
  • listening for the ring-back tone – which lets you know that the dialled number is now ringing. Alternatively, you could get a busy indication to let you know that the line dialled is engaged so you can try again later.
  • hearing the phone get answered – then you can have your chat.
  • hanging up the phone – perhaps you’ve finished your conversation or the other end has replaced their phone “on hook” and you’re just listening to the tone indicating this, so you hang up your phone, clearing the line for an incoming call.

To receive a call:

  • the telephone will ring – depending on the phone, you might have a “real” bell, or an electronic “warbler” or selectable tones. This alerts you to an incoming call.
  • after the first ring, you can receive caller ID on suitably equipped phones – this only works if you’ve paid for the caller ID service and you have a phone with a caller ID decoder inbuilt. This sometimes doesn’t work if there’s too much line noise.
  • you can choose to answer the call by lifting the handset – then you can have your chat.
  • or you can ignore the call, which will ring … ring … ring … ring … ring …
  • hang up the phone once you’re done – this way you can receive another call.

None of this is too radically different to what you might have on a modern mobile telephone, but there are a number of key differences:

  • You don’t have a dial-tone anymore. You don’t hear the “hang-up” tone either.
  • You aren’t restricted to an old fashioned bell or limited electronic ringer.
  • Caller ID is taken for granted and just works “immediately” rather than being a paid option and only being received after the first ring.
  • There’s no mystery with caller ID either – if you didn’t pay for the service or didn’t have a decoder, you would answer every call just in case!
  • Voice quality is better on modern devices thanks to HD Voice between supported networks. On a traditional telephone, analog crackles, hiss, inducted noise from AM broadcast stations, etc would be sometimes audible along with level differences on longer lines.
  • You don’t have the same feeling of picking up the phone, punching in the number (or dialling the number on a rotary dial), or slamming down the phone on someone you didn’t want to talk to.
  • Voicemail is free – with landline phones, if you wanted to catch “missed” calls, you’d need to invest in an answering machine. Or if not, you’d need to pay extra for a service which records the last-missed-call number and allows you to retrieve it. In Australia, I still remember the TV campaign for this service, *10#.
  • There’s no such thing as a text message! But on the upside, landlines carry faxes, TTY and modem signals relatively well.

For those who may not have placed an international call before, it’s also of note that each country often has their own variations on tones to indicate call progress. As a result, Australian landlines have a particular “sound”, inspired from the British network, that you don’t get by default from some “emulated” phone lines from ATAs. If you’re interested, the database of tones from the ITU can be downloaded here.

Even though I don’t have a landline, I’ve configured my ATA to emulate one to a high level of fidelity, so here’s a call recorded from my modified cordless phone to my answering machine that shows the Australian dial-tones and ring tones.

Call to Answering Machine

Caller ID on Landlines

There are a number of ways to implement caller ID on landlines which include low-speed modem data or DTMF tones in-between rings. For compatibility reasons, the first ring is sent as standard, but in-between the first and second ring, the caller ID data is sent. Australia seems to be using a “Bellcore” FSK based data at 300bps. As I’ve never had the chance to examine it – I decided to hook up my ATA’s FXS port to my Rohde & Schwarz RTM3004 to see what it looks like.

The Australian ringing signal is 25Hz sinusoidal (as far as I know), with a cadence of 0.4s on, 0.2s off, 0.4s on, 2s off.

In the case of the ATA, the line idles at close to -48V (as expected), with the ringing amplitude of about 85V above and below ground. The caller ID data is in there … but you can’t see it because of the wide voltage swings involved.

There’s a good hint of it if we zoom in on the DC-coupled recording, but it’s not likely we’d see the detail so lets switch over to AC-coupling and try again – letting the regular ring voltage clip.

The caller ID data burst can be seen in the middle of the rings – sloping upward due to the AC-coupling frequency response.

The signal starts with a 01010101… preamble sequence, the FSK is clear.

… towards the middle, we can see real-data being passed. I didn’t bother decoding it, but it doesn’t have a particularly big amplitude – about 1.6v peak-to-peak. It’s definitely enough to satisfy my curiosity.

“Generic” Extension Cord and Bell

The thing that really got me excited about telephony again was rediscovering this old “generic” extension lead and bell. I remember my parents using this when I was young – it’s a very “analog” piece of technology.

 

It’s quite a simple device – just one on-off slide switch on the side that allows you to use the unit just as an extension with no bell feature. The plug has approval number #RA87/146.

There is a modular jack at the other end for plugging in the phone, with only two pins available suggesting that the unit only carries the tip-ring pair. The rear had space for a label which has long been lost, as well as a provision for a pair of wall-mounting screws.

As such connectors are getting rare, I could connect the unit backwards through the modular plug as long as I don’t touch the pins on the other side. Or I could use one of these very rare adapters (Approval Number RA85/102) which converts from the 600 series to modular plug.

The inside is rather uncomplicated, as expected. The main “star” of the show is a bell from Chang Cheng.

The underside shows a rather interesting logo and the fact it is Made in Taiwan (Republic of China).

The PCB has a lot of holes but only very few components – two diodes, a resistor, a switch and a capacitor.

The capacitor is rated 1uF with a 250V rating.

It seems like the PCB allows for a number of other configurations, possibly to meet the needs of other countries’ telephone networks.

Inside the bell is a solenoid coil with a pin – it vibrates back and forth, striking the bell at two points from what I can tell.

Behold, the lovely sound of the analog bell – the way a phone’s supposed to ring.

Telstra TouchFone TF200

The Telstra TouchFone series are somewhat iconic, as a “rental” telephone from Telecom Australia well throughout the 90s. You’d see these around everywhere and the ring is somewhat recognisable as well. The phone features some speed-dial memories, which apparently uses some current from the phone line leading to the phone occasionally (potentially) disrupting dial-up connections. This particular unit was “ex-business” and was found discarded in a still-working but dirty condition. It’s got such a severe sun-tan that it probably needs some retr0bright … not that I can be bothered to do it.

The phone doesn’t have many controls around the side – just a ring-tone volume control. There is a vent at the top, to allow the ringing noise to escape the casing.

This unit is branded Telecom Australia, manufactured by Exicom, not a name that I immediately recognise. One of the rubber feet has been lost. On the underside, there is a three-position switch for ringer pitch, but only two seem to work. There is also an impedance switch and a tone (100ms/600ms) and decadic (pulse) dialling switch.

The phone itself is pretty light and opening it up reveals an AWA 87603 and AWA 87830 marking on the plastic moulds. That’s quite an interesting find along with the two speakers – one for the ringer, the other for the keypad “pip” which acknowledges every button press.

To save money, it seems they’ve extended the membrane keypad to add an additional button for the hook switch. This relies on a pivoting piece of plastic with a levered rounded protrusion that presses against the flexible plastic contact.

From the only thing I can see with a date code, I suspect this unit is Week 25 of 1991, making it 27 years old. It still works, although slightly crackly. Surprisingly it seems that the hybrid is already a “silicon” type, not being implemented in a wound transformer as you might expect of older telephones.

The unit rings like a “warble” with both low setting and high setting.

“Generic” Cheap Handset

A very late model, just about the cheapest and most basic a telephone could get. The line cord reaches all the way into the handset which contains all the circuitry necessary to operate the telephone.

The keypad nestles inside the arched area of the handset, with the hook switch being a protrusion near the microphone which is held “closed” by the weight of the phone on a desk or in a wall cradle that has been long lost. Because this unit is “all in one”, I suppose this is a unit that you would especially avoid during a thunderstorm to prevent any chance of being “shocked” through the phone.

Because of its screw-less plastic construction, I didn’t open it up as I would have likely caused damage to it. However, I did make a recording of its weak, shrill-sounding tone.

Leader 724 PABX Desk Phone

This particular phone is intended for PABX use and was abandoned for disposal at a technology clean-up. Severely yellowed, it still works, although like the TF200, it seems to be a little crackly and susceptible to RF interference from Wi-Fi access points and DECT cordless telephones alike. The unit features speed-dial memories, a neon ringing indicator and mute capabilities.

The ringer can be set between off, low and high positions, with a “data port” on the side allowing for connection of dial-up modems to the phone without unplugging the phone.

The main line input is on the top side, with the handset cord coming out of the other.

A look at the rear shows a date code – probably Week 3 of 2004. The unit has a tone/pulse and line length selection on the rear with a ringer pitch potentiometer that felt quite stiff. There is a battery compartment, but a label over the top claimed that no battery is necessary – indeed there’s no contacts behind the cover. Despite saying it is for PABX, it operates just fine off the FXS port of the ATA – so it’s possible that it’s not impedance optimised for regular lines.

The construction inside seems to suggests that the design is rather adaptable, modular and somewhat kludged to meet whatever requirement is necessary. The PCB seems to accommodate a 9V battery due to the cut-out even though it’s not used. Other wires are “tacked” on for additional features, such as the neon indicator.

A cable tie also seems to secure a line filter for the input, but it’s not enough to stop interference from Wi-Fi access points.

The unit uses all-through-hole components on a paper type PCB, again, based on a silicon hybrid. This one has a Holtek HT9320H dialler chip. Surprisingly, there is a soldered-down Renata lithium coin cell (CR2430) that’s still reading about 3V even today. This is almost certainly used to hold the speed-dial memory. The hook switch seems to be a repurposed “power” type switch with the click-retention mechanism removed.

The front keypad is secured with so many screws, I couldn’t be bothered to disassemble it, so I’ve left it as it is. I don’t really want to break anything …

In case you’re wondering what’s inside the handset – not much. A speaker, an electret microphone and a bit of metal for weight and balance.

This is how the Leader 724 sounds like when it’s ringing.

Uniden XSA660 Cordless Phone

I tore down the XSA660 in an earlier post, but seeing as I mentioned radio scanning in a recent post, I thought I’d make some comment about early cordless phones and provide some sample recordings from an SDR.

Older cordless phones such as this operated in analog NFM in the 30-40Mhz band. While it was illegal to listen into telecommunications, I’d guess that a lot of scanner owners probably did listen in as it was an easy target for gossip. The joke was that your scanner neighbour would gift you a cordless phone as a way to learn all about you and perhaps even steal your identity. Such basic phones featured no protection – not even voice scrambling – against such evavesdropping which can happen even accidentally due to channel reuse. There was about 10 channels for use – the above shows me hopping between channels by hitting the scan button.

The only thing digital about these phones was the line control feature, including some sort of special rolling coding, so you couldn’t pick up your neighbour’s line and make calls on their line.

When listening on the base channel, it would sound like this – including the data-train paging the phone during ringing and the audio from the handset coming back out from the base for side-tine. Listening to the higher handset-to-base channel would sound like this, featuring only the handset audio and control signals. Likewise, this is what hitting the channel scan button during a call would sound like.

As a basic phone, you only had the choice of two ring tones – this one or this other one.

Uniden WDECT3115 Cordless Phone

The next cordless phone I owned was one I purchased from a liquidator at a low price. It’s a pretty sophisticated unit with speakerphone ability and a large selection of ring tones. Unfortunately, it’s a WDECT unit, meaning it operates in 2.4Ghz and interferes with Wi-Fi. If it were not for this fact, I’d be more fond of it. It seemed to be a fairly popular model with a space-age design.

The handset is much more compact than the analog type, with a much shorter antenna due to the higher operating frequency. The frequency hopping and digital nature also makes eavesdropping extremely improbable. This particular unit has smaller buttons, a nice LCD screen for Caller ID and provision for a belt clip attachment and external earpiece.

The base has a very curvy shape, along with wall-mounting options. The underside accepts the power and telephone cables, but with a way to route the cables securely and neatly. It’s quite a good design, with charging indicator and find-button on the side.

The handset uses a three-AAA Ni-MH rechargeable cell for power. The base has a REN of 0.1, which is not unexpected due to the electronic nature of the telephone.

To my surprise, the unit is based on a DSP Group chipset (DUL36107UAB11AQC) – the same company making chipsets for modern VoIP ATAs.  The PCB is dated Week 52 of 2007, whereas the chipset is dated Week 48 of 2007.

The keypad is surprisingly well-done – having a separate plastic sheet with carbon contacts which interface with the silicone pad – this way, liquid ingress is less likely and replacement of the carbon contacts is a possibility.

Looking into the base, the unit has two antennas in the curved section which provides for some reception diversity, improving coverage. There is a corresponding DSP Group chipset.

The base actually has two boards – the rear board is the line interface, including some protection against surges and a tubular “reed” relay for switching the line.

There are a few test points on the back – I suspect CID+/- may be test points for caller ID signal.

The rear of the main board shows that there is an RF canned module with castellations, soldered to the mainboard. This suggests that the RF module may be pre-approved or shared between other units.

Compared to all other phones, this one is relatively advanced, featuring quite a few melodies – twenty to be exact:

While it doesn’t ring like a proper phone, it does ring in its own “retro” way, kind of like old polyphonic mobile phones used to.

Uniden DECT1015 Cordless Phone

I’ve also torn down the Uniden DECT1015 in the past, which is an even more modern but basic cordless phone. Unfortunately, there’s only two ring tones – Flicker and Clatter which sound somewhat crappy out of the beeper.

Conclusion

I never intended to collect telephones, but faced with a pile of phone gear that hadn’t seen much action, I felt a little nostalgic about the sounds of the landline network and telephones in general. Now I’ve got some decent recordings of the “real” ring of a telephone bell, to the electronic rings from the later units for a rather “retro” feel.

I guess you could load it onto a smartphone, but you still won’t have the same satisfaction of punching in a number or slamming down the phone …

Posted in Audio, Tech Flashback, Telecommunications | Tagged , , , , , , , | 4 Comments

Note: Installing Firefox on a Fresh Windows Install without using IE

It’s no big secret that Microsoft’s Internet Explorer and it’s closely related cousin, Edge, are not the world’s favourite browsers. They’re not even my favourite and for a number of good reasons.

In the battle for market share, Microsoft has pushed Edge upon the users of Windows 10 by giving it default placement, by captioning it as the “recommended” candidate, by resetting defaults occasionally after updates to favour it and by putting pop-ups that remind users that performance and battery life is “better” under Edge. Unfortunately, these reasons are not enough … and their validity is a little questionable.

But for better or worse, we will fire up either IE or Edge just once, to download our favourite browser. Kind of like what this popular comic by “Merryweathery” shows.

But to do that will lead to a barrage of advertising for the browser. I’d rather avoid confronting IE or Edge altogether.

Putting IE out of business

The problem is that it’s a bit of a chicken and egg issue. How does one get a new browser without using the only supplied browser? Unfortunately, there is always a dependency somewhere – you need at least some prior knowledge to make this happen.

In the old days, you could just fire up a command prompt terminal, open up FTP (which comes included in Windows), connect anonymously to a download server and initiate a transfer of an installer that way. Unfortunately, it seems like FTP mirrors are a bit of a dying breed as FTP is not a particularly secure protocol.  Even Mozilla’s ftp.mozilla.org no longer works, which is a shame, as that would have been an “easy” and familiar way to avoid IE-chan all together.

Instead, we must do our downloading through HTTPS. Under Linux, there are always tools available – wget comes as standard. Under Windows, the tools are lesser known and sometimes need to be slightly abused to meet our goals.

You also need some prior knowledge – i.e. where the installer file is located (i.e. its URI). Unfortunately, determining this is not as simple as it could otherwise be, due to the script-heavy nature of the websites and tracking involved nowadays. The URI can also change when versions change, so it’s best to find an official URI that is a “metalink” to the latest version of a given browser.

My favourite browser is Firefox. After some work, I determined the best URI to use is:

https://download.mozilla.org/?product=firefox-stub

This is a metalink that will download the latest stub – this is an installer that then goes to download the newest version and finally install the browser. Unfortunately, trying to determine the same address for Google Chrome led me to understand that they seem to be generating “tracked” unique links with customised binaries to track installs – so no such luck there.

This isn’t particularly easy to remember, so there might be a temptation to shorten it. Unfortunately, the risk of doing that is to introduce a security issue where you might be misled to download another installer and get some malware instead. Best to stick with the proper URI – perhaps print it out, write it down on the installer CD, save it into a text file or download one of the scripts linked later onto your installation medium.

Method 1: Windows PowerShell

For later versions of Windows, PowerShell may be installed and available for use. This seems to provide the neatest way of performing the download, through the Invoke-WebRequest call:

Invoke-WebRequest -OutFile "Firefox Installer.exe" -Uri https://download.mozilla.org/?product=firefox-stub

Running this call will result in a pale-blue PowerShell window appearing for a short while as the file is downloaded, before exiting.

Method 2: Abusing certutil

If you don’t have access to PowerShell, the next best solution seems to be to abuse certutil to act as a downloader. The upside to this is that you don’t need administrative privileges to run – it can be run in a standard command prompt.

certutil -urlcache -split -f "https://download.mozilla.org/?product=firefox-stub" "Firefox Installer.exe"

The downside seems that it doesn’t print much progress status before exiting, but it does the job just fine.

Method 3: Using BITSADMIN

The third method, in case you can’t seem to use certutil, is to use BITSADMIN. This is apparently a depreciated method, but it uses the Background Intelligent Transfer Service (BITS) as used by Windows Update and other services to download the file. BITSADMIN appears to require administrative privileges to operate, otherwise you will get an error 0x80070005.

If you run the command as administrator, it will download the file into C:\ (as it requires a full path).

bitsadmin /transfer GetFirefox /download /priority normal https://download.mozilla.org/?product=firefox-stub "C:\Firefox Installer.exe"

It does show some status during downloading, but it also seems to be slower as it tries not to monopolise bandwidth.

After using one of the above three methods, we are rewarded with our precious stub installer file, which we can then execute and fully install the browser.

If you want to just download the scripts as a ZIP file, you can get them here.

Method 4: Get a Linux Live CD

I’m actually a strong advocate of having a Linux Live CD or USB stick around, as Linux is very useful and can “save your bacon” in many cases. This is equivalent to having a full operating system you can boot as an alternative to Windows. You can use whatever tools and browsers are in your Live CD and use that to download the browser of choice to a storage medium to install under Windows.

It’s like getting a jackhammer to crack a peanut … but it works … and I suspect you can get Chrome this way as well. But seeing as you’ve come so far, maybe it’s time to ditch Windows altogether or at least dual-boot?

Conclusion

As it turns out, through Invoke-WebRequest, abusing certutil or using BITSADMIN, it’s possible to grab an alternative browser without having to fire up IE at all. But this is predicated on a few things – that you have the knowledge of the URI of the installer package and the syntax which is required to make use of the tools.

I’ve written a few simple scripts to do it, which can be saved to your installation media as a convenience – but you could just as easily print out the one-liners for reference. Of course, there are no guarantees that the URI won’t change – if it does, then the scripts will likely stop working … and either you use another computer/OS to find out the new URI or you parse HTML in Notepad with your eyes to try and determine the new URI.

Or … give in and fire up IE once more …

Posted in Computing | Tagged , , , | 9 Comments