Around two weeks ago, a friend of mine offered me some of their unwanted old technology. Among the pile on offer was a fax machine. My friend was surprised when I leapt at the chance to grab it – who would want a fax machine nowadays?
But I’ve always been a sucker for old-fashioned telephony and voice-band modems. They were the way we managed to turn a medium meant for voice communication into something that carried a more versatile stream of data, which in the case of fax, meant carrying a facsimile of a document to be reproduced on the other end. Now that the era of the landline is almost over, these technologies are relegated to obsolescence.
Because of this, I’ve been interested in preserving the experience in some way. I started up (and subsequently concluded) Project FAX in the hopes of collecting some faxes, but it seems that across its nearly two-year tenure, there were only a few successful transmissions. I also looked at the Bureau of Meteorology’s Weather by Fax before it shut down. But what I was missing was a fax machine, but that wasn’t always the case.
In fact, we did own a fax machine ourselves, a Sharp FO-365 direct thermal printing unit with automatic paper cutter. It was an expensive machine, but during its life at home, we probably only ever sent ten or so pages from it. It wasn’t until many years later after I had set-up Cheyenne Bitware and WinFax Pro for faxing from the PC using a fax modem that I cleared out our local Tandy of thermal paper rolls at 10c per 100m roll (on clearance) and used them in the unit as a makeshift low-resolution printer for my high-school notes. Later, I even used it to print 1.5m-long artistic banners as the continuous roll paper allowed the machine to just keep going.
In the end, the machine was junked not because it stopped working, but because the anti-curl system had attracted a nest of cockroaches because of its warmth. Besides that, I had run out of thermal paper and I wasn’t going to spend up to get more especially when it’s not very permanent compared to a plain-paper laser!
After losing that machine, I did salvage another machine from a neighbour when they threw it out on the street. Unfortunately, that Mitsubishi Electric unit was even older and just plain refused to work. This time, I hope I’m a little luckier.
The Machine
The unit is a Panasonic KX-F1010AL plain paper fax, missing its source paper tray support and output wire baskets. This style of machine is a common compact desktop model, the sort that you might see in a home-office or small business situation. The unit operates as a telephone, speakerphone and fax machine.
One thing that might seem to be missing is the document scanner inlet – that’s under a flap to protect the paper path from dust. A nice touch, but this unit can only handle up to 15 sheets on its auto-document feeder. At least it has one, so you could use it as a crude 100/200dpi black and white ADF scanner, albeit not a speedy one!
The front of the machine is where the paper is ejected – the lower section for printed material with the print face-up and the upper section for the scanned originals with the print face-down. Ordinarily a wire-basket attachment keeps the two separated.
The handset is a rather plain unit, with a speaker in the bottom section for speakerphone operation.
That side of the machine almost looks like an attachment to the machine, jutting out from the side and having the handset cord plugging into the base instead.
The rear of the machine has an IEC socket for power, a switch for tone/pulse mode and connections for the phone line and an extension. This extension is “switched”, thus has been blanked off, only to be used if you have downstream phones or answering machines and you want the machine to auto-detect faxes or seize the line to receive faxes thus preventing other devices from interfering with transmission or reception.
Above that is the plain paper input hopper, which is spring-loaded and has to be “opened” using the lever on the left for loading paper. Ordinarily a plastic tray support would be placed into the slots to hold the paper at the appropriate angle.
How does the printing work and what do you do in case of a paper jam? Well, there’s a release latch on the side of the machine which allows the whole top of the machine to lift up like a car bonnet. This is slightly different to the old Sharp FO-365 I had where the document scanner and thermal printing sections were in two different compartments.
Inside, we see the thermal transfer film ribbon cartridge. This is actually quite similar to direct thermal printing faxes, but instead of heating the specially coated thermal paper, plain-paper units use a plastic ribbon containing a waxy black substance to transfer it to the plain paper passing underneath. I didn’t realise it until now, having never used a plain-paper fax before, but this is very similar to how many battery-operated plastic labelmakers work as well! Of course, there were also units using inkjet and laser technologies.
Removing the cartridge shows the paper inlet path from the hopper, where it passes over a rubber roller which “sandwiches” the film ribbon and paper against the thermal printing head.
The cartridge consists of a frame and colour-coded gears. The film travels at the same rate as the paper, so a 100m roll such as this one would be good for approximately 336 pages, unless you use film saving modes which omit printing blank lines and essentially “compress” your printed output. The film is wrapped onto notched cardboard cores, so that the gears can engage the cores and pull the film through the unit. The film is intended to be used once, as any pigment transferred leaves a “blank spot” on the film. As a result, the film contains a negative impression of all your printed faxes – dispose with care!
The front section comprises the paper rollers responsible for pulling in the original document in front of the line-CCD sensor and illumination. One of the rollers is beginning to craze and disintegrate – not unsurprising given the age of the machine.
The top side houses a paper sensor for the auto-document-feeder and some springs to push it against the rollers, as well as the thermal printing head.
Fax Me!
To demonstrate the machine, I had to see whether I could get it to work. As it turns out, this machine wasn’t all that healthy – the LCD had completely failed, making it difficult to configure anything on it, and the RTC battery had failed resetting all of its settings. The speakerphone and onboard speaker had also failed, resulting in no audio feedback. Despite this, I still managed to get it to work and also managed to configure some settings via a remote-setting capability intended for dealers to troubleshoot issues with customer machines, detailed in the service manual. This can be accessed by making a phone call, pressing the menu button and having the far end dial in 9000*, followed by pairs of [option code]* and [value]*.
To make it “work” as you would on a regular old telephone line, I relied on my Linksys PAP2T in a back-to-back configuration, such that it was on line 101. The other end was connected to a USB fax modem on line 100, connected to a VMWare virtual machine running Windows 2000 Professional so I could use Symantec WinFax PRO. This solution allowed me to support the more complex modes this machine supports, including MMR ECM, although I did not have any support for superfine resolution. The machine is only capable of 9600bps transmission and reception.
So what does faxing using this machine look like? Watch the video!
The first half of the video shows receiving and sending a page with the modem sounds recorded via tapping the phone line and synchronising it to the video. One thing you might notice is that the transmission ends before all of the fax is printed – this is because the printing mechanism is a little slow, but is also the reason why most fax machines have internal memory to buffer against this. The negotiation time between pages provides the mechanism some time to catch up, however. The memory also allows a machine which has run out of paper or suffered a jam/misfeed to buffer up some of the transmission while the user corrects the error. When the fax transmits the page, most of the page is transmitted but then the fax halts for a bit – this is due to the error corrected mode (ECM) of operation, which checks the remote side for errors after a certain amount of data has been transferred before resuming.
The second half of the video shows the same, but listening to the machine’s operation. The machine is “rhythmic”, operating with a stepper motor. The sound depends on the content of the fax to some degree, but it is a very mechanical beast. One with an operating speaker would likely play the line audio through to Phase B, and beep to indicate completion or error.
Ordinary G3 faxes are generally black-and-white, usually in either standard (~100dpi) or fine (~200dpi) resolution. Some machines were capable of superfine (~400dpi) and half-tone modes for photographic imagery. Later Super-G3 faxes offered higher transmission speeds, higher resolutions and even colour. In this case, I decided to fax a full A4 page image of the Meganebashi Bridge from my recent trip to Nagasaki. This fax took 15 minutes to receive, with very fine halftone dithering by WinFax PRO producing a decent result given the constraint.
In fact, the first time I tried, the fax machine suffered a paper mis-feed, so it decided to receive into memory. I corrected the paper input, but by the time the machine decided to start printing, it was because it had overflowed its internal memory! So much for their “20-something” pages capacity – that’s based on a simple letter, not something this detailed.
The printed result from the fax is scaled down slightly, partly because of how such auto-document feeder scanners work. They obtain the length based on sensors that detect the edge of the paper by contact. If the rollers slip on the document slightly or are slightly out of shape, then the transmitted image can be longer than a regular sheet of A4 paper which might cause one page to be reproduced as two separate pages on the other end. Direct thermal printing faxes using roll paper don’t have an issue with arbitrary length pages, but plain paper units do. I took the printed output and faxed it back to the computer which resulted in some degradation of the image, but it’s still quite good. Note the addition of a header line, which is very commonly inserted by fax machines based on the programmed “logo” and “TSID” information.
Depending on your settings, you can select whether you want to print a transmission report after transmitting or receiving a fax. This provides a log of the fax status, but this machine provides only a very brief report – other machines can provide much more specific details such as resolution, speed, bit error rates, etc.
As faxes can be received without human intervention, around the clock, it was customary to prefix every transmission with a cover page that provided some information about the upcoming fax and its length. This was particularly useful in case the transmission was interrupted, so you could spot an incomplete document, or in the case where someone else dials in and busies-out the fax machine while your machine was waiting to retry, allowing the operator to “shuffle” the pages back into the right order. This particular cover page can be generated by the machine itself – quite a neat trick.
Using the machine definitely bought back memories of faxing … but so did checking the onboard help.
Onboard Help
Something which this unit had that my old Sharp FO-365 didn’t is the onboard help feature. Given I had no LCD on this machine, it was perhaps interesting but not all that useful.
Pressing the blue HELP button prints this general help listing which gives you basic instructions on how to perform basic operations. The information can be followed to print further help pages.
The above is the help for how to use the onboard telephone directory.
This is the “how to setup” which prints a list of the feature codes and current settings. This repeats a lot of the information available in the list printing features.
This is the page about programming phone numbers – reminds me of the good old days prior to predictive text or T9 when we had to tap-tap-tap our way to the letter we’re interested in.
Finally, there is this one about printing a list. I think this is rather nifty use of the onboard character generators – after all, they’d have to have it for the purposes of generating the “header” line on the fax, plus there is G3 compression/decompression capability, so they could probably afford to squeeze all of this into the firmware ROM of the machine.
Of course, given the leads on this “list” page, I couldn’t help myself but print those out too.
The setup list is very much the same as the “how to setup your unit” list earlier.
Journal provides a history record of the faxes sent and received by the machine.
The printer test prints a full black and 1:4 pattern to identify any malfunctioning thermal head elements. In this case, everything looks perfect. The cover page was shown earlier.
But wait, there’s more. If you use the remote service code, it’s possible to print out a basic feature list, a different sort of journal …
… as well as the service data list which contains the less-common features as well as a total pages count. Note that it had been reset due to the failed RTC battery, but I did manage to scan 200+ pages with the unit just fine …
Rebooting the machine also resulted in some interesting output in a different font – this is mainly reserved for film warnings, but it consumes a whole page and the associated film!
Conclusion
With the slow erosion of landlines, the era of faxes is almost over. While faxing using a fax modem and software running on a PC is easy, real faxes were usually sent by desktop machines like the one in this article. Simple to use, straightforward in operation, they allowed for rapid transfer of the contents of a document over a phone line with such ease and speed that they would find themselves a fixture in most businesses and some homes.
This unit proved to be mostly functioning, save for its speaker and LCD, allowing for a demonstration of transmission and reception. Having never used a plain paper unit, it was a surprise to see the thermal transfer ribbon technology which is similar to that used by handheld labelmakers, and the “scaling” of received faxes as a “hack” to combat the possibility of slightly “long” pages requiring the use of a full sheet paper and associated ribbon.
While fax is definitely well under threat, I didn’t realise until I was going to use the machine just how much paper is under threat too … the transition to paperless is still happening in some sense, but when I think of sending a document, I don’t think of a “hard copy” document anymore!
Don’t you wish sometimes you were born in the 60s or 70s? I get the general sense that you really would have lived the life in those years. Or maybe if you had been born in the 60s you’d only have been invested in old obsolete stuff back from the 40s and 50s. Great blog anyhow.
Haha … well, it was a simpler, perhaps more dangerous time. I’m not sure I’d trust myself around tube equipment … especially with much dodgier test equipment. But at least things were perhaps engineered with a bit more “margin”, so there’s a good shot of having something work, and how it worked was perhaps more obvious than today’s “black boxes”. In return, we do get a lot of convenience for next to nothing nowadays – things like practical Bluetooth Low Energy “single chip” radios with such low power consumption and decent range would have been completely unfathomable that many years back!
– Gough
Do you have an example of fax using the film saving mode – I’m curious as to how the compressed output looks minus the blank lines.
Sadly, no. The failed LCD on this unit means that I’m essentially flying blind and the film save option is something that cannot be configured using the remote programming capability of this machine (along with other features like date/time, sender name/TSID). I haven’t found an easy way to open the machine without potentially breaking it either … otherwise I would perhaps try to sniff the LCD bus to determine what’s being displayed or try wiring in another potentially compatible display.
– Gough
Actually, after some blind fumbling, I managed to turn it on and … drumroll … it doesn’t do what I thought it does. It doesn’t seem to change the formatting of the fax at all, contrary to what I had heard. Perhaps I got the whole Matsushita Whiteline Skip (which is a proprietary compression method) confused with the paper/film save feature.
From what I can tell, it keeps the last few transmissions in memory, so that if it receives the same fax page again, it doesn’t print it. I guess that saves film especially if you’re being continually “bombed” with the same junk fax or a retry sends a page that has already been successfully received. That’s after sending the machine several faxes over and over to try and work out what, if anything, it did as the printed output looks visually identical.
I guess I learned something today … but even then, I can’t seem to get it to work consistently. I guess the detection method is probably not robust – perhaps it’s not based on image data but merely the metadata (e.g. caller ID, number of pages actually sent).
– Gough
Thanks for looking into it – and I’m glad you got to learn something extra along the way.