This is yet another repair I had to make on the Linotype at the MacKenzie Printery and Newspaper Museum.

The keyboard assembly has a row of 91 short pushrods, known as the Lower Reeds, which push up when a key is pressed, ultimately releasing a matrix from the magazine above. The magazine has 90 channels but there is also a lower reed for releasing spacebands.

These reeds are supported and aligned by a double comb made of brass (part number H-2484). The lower part of the comb has closed slots for the reeds to pass through, but the upper part of the comb is just open teeth to space the reeds apart. These teeth are vulnerable to damage if the keyboard is swung out on its pivot (e.g. for maintenance) when some of the reeds are raised (e.g. due to keyboard jams), unless the operator takes precautions to prevent the raised reeds from striking the lower ends of adjacent upper reeds.

The comb is mounted on a steel support bar (H-2485) by several screws (H-207), and the entire assembly has part number H-2580.

The keyboard at the museum has had three of these comb teeth broken off; two had been repaired in the past, and now I have repaired the third one in the same manner.

View from the top of the comb showing the bent and broken teeth.

Another view of the damaged teeth

The comb is formed from sheet brass, and it would be pretty much impossible to weld/braze/solder on a replacement tooth. However, the repair used on the other two teeth is fairly easy to repeat. A 1/16″ hole is drilled in the steel support bar exactly where the comb tooth should be, and a short piece of 1/16″ brass rod (I used brazing rod) is inserted. In my case the rod was a good enough fit in the hole to stay put, but if necessary a locking compound (such as Loctite™ 638) could be used to retain the rod.

The hole was not as accurately positioned as I would have liked, especially where it broke through the rear of the steel bar, but the tolerances in this particular area of the Linotype could be described as “sloppy” so it was easy to bend the rod to line it up good enough. The two teeth from a previous repair, as well as a couple of intact teeth on the comb, were also gently bent back into alignment.

Reinstalling this comb was a bit of a juggling act because of the need to keep 91 lower reeds from going astray. I think it was easier to install this bar and comb, and remove the other bar (which has a vertical stop for the reed motion) instead. I could then drop all the lower reeds into the comb, using a steel ruler to prevent them from falling out or dropping too low (the reeds have a notch in their side which limits their vertical movement). I could then use the ruler to raise all the reeds a bit so the bar could be reinstalled. I suspect an alternative might be to trip all the keyboard cams so they would hold the reeds up in such a way that the bar could be reinstalled.

The knife block on the Linotype at the Mackenzie Printery Museum was pretty much impossible to adjust due to dried out lubricant.

This block controls the position of the right knife which trims the thickness of the cast slug to its finished size, and it should be set by the operator to match the size of the mould being used. If it is set too wide (too large a size) no trimming will occur and you will have an overly-thick slug with a rough surface that may not lock up well in the forme. If it is set too narrow it will be trying to trim off too much metal and the slug will likely jam when ejected from the mould.

The actual knife position for each size is determined by selecting a set screw (which in turn has its own set screw to prevent its going out of adjustment); the slide that holds the knife is pushed against the end of the selected set screw by spring tension.

Because it would be difficult to rotate the selector wheel with the slide bumping along from one adjustment screw to the next, the knife block also includes a rotary cam which pushes the slide keeping it just clear of the adjustment screws, at which point the selector wheel can be easily turned.

To do this the knife block knob can rotate a bit without the selector wheel, engaging this cam, then rotate with the selector wheel to select a new size, and finally rotate back without the selector wheel to disengage the cam leaving the slide pressing against the newly selected set screw.

Engaging the rotary cam closes up the knife a bit, and this would be impossible in the case of a jammed slug, so there is another control on the knife block that allows the selector wheel lock to be disengaged without engaging the cam, allowing the user (with some difficulty) to select a wider position to clear the jam.

All this leads to a rather complex-looking mechanism between the knob, selector wheel, and selector wheel locking pin.

On this particular caster, the knob was seized onto its shaft by lubricant that had dried and turned sticky, so it was impossible to release the selector wheel lock and choose a new size. I remove the knife block and, in my basement workshop, I disassembled, cleaned, oiled, and reassembled it, and it now works like a charm.

The knife block mostly in pieces. The orange and violet bits are shims, which I assume are colour-coded by thickness.

A couple of the parts were difficult to assemble and disassemble (even had the block not had gummy lubricant). In particular, there is a two-part guide that forms the right hand side of the ejection chute for the slug (the two light-coloured parts on the left in the above photo). The hockey-stick-shaped piece screws to the slide that carries the knife, then the rectangular plate (which floats a bit and has a spring behind it) mounts over it, covering its mounting screws. The problem is that the shoulder screws that hold the floating plate go in from the back and there is very little room to get a screwdriver in to remove or install them. It would be much easier if the floating plate had holes in it over the mounting screws for the fixed guide; you could them screw the two together before installing either onto the block.

One thing that might make this easier, which I did not try, would be to remove/install the floating plate with the selector wheel and the guide bracket (the curved piece lower left in the photo) not installed. This might just give a better alignment to use a screwdriver on the shoulder screws. The downside of this is that you then must spend more time handling the heavy knife block with this rather fragile piece of sheet metal sticking out.

The knife block does not appear to have any facility to oil this knob and its shaft without at least removing the knob and its linkage to the selector wheel. There are a spot or two where you can apply oil and hope it wicks into all the necessary places, but no proper oil holes.

I have a YouTube video. showing reassembly of the block (and implicitly its disassembly), how it works, and with a few tips on servicing it.

One of the tasks I had in getting the Linotype at the Mackenzie Printery and Newspaper Museum working was to repair one of the keyboard reeds which I had broken (I blame the service manual for failing to warn of the possibility of this happening).

The Linotype typically contains 90 of these push rods, which rise from the operator’s keyboard to the top of the machine where they operate an escapement mechanism to release a matrix which determines the shape of the letter. Each reed operates one channel in the matrix magazine to release a particular letter.

The keyboard itself also contains short lower reeds which raise the upper reeds to release a matrix. The keyboard is mounted on a vertical pivot so it can be swung out for servicing, but if (due to a keyboard jam) one or more of the lower reeds is stuck in the raised position when the keyboard is swung out, they will strike the lower ends of nearby upper reeds. This can cause either the upper or lower reeds to be bent or broken off, or can damage the comb that keeps the reeds lined up properly. This is exactly what happened to this caster.

When swinging out the keyboard it is crucial to check for stuck lower reeds and manually raise the upper reeds as needed to clear the lower reeds, but the manual contains no such warning.

The lower reeds are soft enough steel that they can generally be bent back to their proper shape, but the upper reeds are hardened steel and so will break off rather than bend back.

I repaired the reed by welding on a new bottom end, after cutting the broken end off clear of any guides it runs in. I’ve posted a YouTube video showing the intermediate steps (though none of the actual work).

It took me several tries to repair this: My first patch was cut from mild steel, and it was only when trimming the broken end that I determined the reeds are hardened steel and so using a mild steel patch would probably not last. My second (and subsequent) patches were cut from O1 tool steel which can be hardened, but it took me three tries shaping the joint and welding it before I had something that could be machined to the proper shape. The steel was partially hardened by the welding heat, and such a thin part is very difficult to anneal because the metal cools too quickly in open air. I did some of the shaping with a saw and files, but I also had to use some grinding tools in my Dremel to shape the hardened parts.

After adjusting the S-bend to the appropriate offset I hardened and tempered the part and reinstalled it in the Linotype, where it has been working fine ever since.

I actually did this repair in June 2022; I’m just slow posting about it.

The comb that guides the lower reeds was also damaged, and I have addressed that repair in a separate post.