Davis Typewriter Works: 2007

Davis Typewriter Works - 2007

May 4, 2007. The 2007 season gets underway with this machine; Harris Visible No. 4, serial number 107931. This machine is unusual for two reasons: First, it is clearly labeled as a "Harris Visible" in all the proper places but is also clearly labeled as having been made by Rex; second, it can type several fractions, which you don't usually see. We were thinking this would be a good warmup, since it was apparently mostly dirty but had just a few problems. For example, the shift keys were balky, some of the keys stuck, the ribbon spool tops were all bent up, so forth and so on. We also thought it might be nice to detail the disassembly, since this particular range of machine can be had by anyone who waits long enough to find one.

First, remove the drawband from the carriage by pulling the clip on the carriage straight out to the right. Either clip it onto something or put a pair of vise grips on it so it doesn't zing back into the machine.

Remove the side and rear panels, held on by tiny screws; replace the screws in the body. Also remove the keytops from the margin release key and tabulator key, and remove the front panel. The ribbon reverse lever is mounted to, and comes off with, this panel. Remove the rear panel; the mainspring winder key comes off with it.

This machine surprised us; it has insulating or sound deadening material inside the side and rear panels, attached as you see here. The Rex Visible No. 4 we did last year didn't have this at all; now, we notice that in pictures you can spot presence or absence of this material by observing the two screws in the side panels. We'll have to check into just when that stopped.

You're now ready to remove the carriage; remove the two screws holding the carriage rail down, and the carriage will lift off (move the paper holders forward when you do this.) Unlock and raise the carriage levers that hold in the platen, and the platen will lift right out. The side frames of the carriage are topped by these levers, and the little locking levers are at the rear.

You'll want the top deck off next. Here, in this shot, Dave is pointing to the ribbon spool shaft, and a small locking collar attached to it right underneath the top deck. This prevents pulling the ribbon spool shafts right out the top. You can loosen these and the driven gears and get the shafts out, or else do what we did this time and get the shafts and spools off with the top deck. You must disconnect the back-spacer linkage on the right side of the machine, and must undo two screws that hold down an assembly that guides the ribbon vibrator; remove that assembly. The top deck is attached by four screws, right in the top; after you've done all this the deck should lift straight off. The space-bar link will just flop out safely.

At right, Dave is checking everything out to make sure we didn't miss any connections between the top deck and the upper action assembly in the Harris.

Here's the state of the machine after about an hour of disassembly and initial cleaning. After this we took the paper table off; it's held on by two long screws whose heads are hard to get to under the bend of the paper table. Use a quarter inch socket and go for the nut on the bottom of each and you'll have a much easier time. Note, in this shot, the upper deck assembly in the top right; it's upside-down with the ribbon spool shafts still installed.

At right, I'm fiddling with the quick-disconnect type-bar links after having found a problem; on this machine, for some reason we can't yet find, the Z and W key levers seem locked together. You press Z and W goes down too. I pulled the links off and found that the type bars work fine; the bind is in the bottom end (lower action unit) and we can't figure it out yet. I also pulled off the two key lever springs for the appropriate levers to try to see inside the contained part of the lower action unit, and the Z key freed up some, but not a lot. It looks like this one might have to come ALL the way apart now, so we hit the parts we'll need to free up with WD-40 and will let that sit overnight. Perhaps tomorrow we'll get it apart or else the problem will solve itself.... but I bet on the former.

May 5, 2007. The WD-40 application worked fairly well; the key levers all loosened up pretty well overnight, except for Z and W which were not only still sticky but apparently operating together. You push W, and Z goes down with it. Very liberal WD-40 application loosened up Z, but W was still extremely sticky. Close examination of the machine's internals revealed that the "W" primary key lever has actually frozen to the shaft on which all the primary key levers are mounted, so that depression of the W key attempts to rotate the entire shaft (which is supposed to be locked, rotation-free in the lower action unit.) We scribed around the edge of the contact between the key lever and the shaft as best as possible with an X-Acto knife and applied more penetrating oil, and hope that fixes it.

Further observation of the sound-deadening material found inside the machine reveals that it appears to have a fibrous surface layer (perhaps thick paper) and a main layer, possibly leather but also possibly rubber. We're not sure yet.

While the key levers were soaking the first time today in fresh solvent, I looked at flattening out the ribbon spool tops. They were really bent up, well out of anything like a planar nature; I applied an 8-ounce hammer with no effect. I was considering a bigger hammer, and asked Dave if the vise I put in the garage was handy.. but he made another suggestion, and after some discussion and much laughter I agreed that perhaps the best tool for flattening the ribbon spool tops was the one you see at right. I placed them on the ground behind the right front tire and rolled backwards very slowly in neutral. They're now perfectly flat. We imagine this is a first in typewriter restoration!

We did a quick job today too. Here you see a TOWER CONSTELLATION, which is Sears model 871.1500 with serial number 6LTT109739. The machine had two type-bar links off, and the corresponding keytops on upside down - Dave noted that the operator probably tried to free the stuck-down key levers at one point and pulled the keytops off, and for some reason put them on upside down and said "the heck with this" as the type face on this machine is REALLY clean. I also noted that the ribbon selector lever has three dimples next to it in the metal surface, but only the center one (white) and the lower one (blue) have any color applied. Moreover, the lever doesn't move beside the upper dot. Inspection reveals a mechanical stop applied to the machine to block out the upper position; in reality, when compared with a true three-color machine, this machine actually types in what would be "red" all the time but of course takes only an all-black ribbon.

The type-bar links that were off were those for the two right-most type bars, which is fortunate. I've had to re-apply twenty or more in a single Smith-Corona machine like this one before. The important things to remember when working on a Smith-Corona with type-bar links off are: First, start from the center and work outwards when reconnecting. Second, use other type bars to gently jam the disconnected type-bar UP near the print point. Third, ensure you straighten the link before reapplying. Fourth, have a pair of really thin, long pliers and a small flat-tip screwdriver to steer links through the various mechanisms in the machine that they like to ricochet into/behind/under when they pop off. The links are really like a pair of tweezers on the type-bar end, spring-loaded to stay shut and with a pin in one side and a hole in the other; the driven part of the type-bar is just a hole. So, you have to slightly spread open the end of the link and pop it onto the type-bar. Also, remember that often you can more easily get the link up near the type-bar by working from underneath the machine; you can sometimes see more easily. Other than these things to remember, the type-bar links are pretty easy to put back on actually and if you remember these things you should have no problem doing this yourself. The machine had a few other sticky functions/operations but these freed up with simple repetitive operation and it's now ready for just a good exterior cleaning and it will be fully operational.

May 10, 2007. Yesterday, we liberally soaked the "W" primary key lever's mounting on the primary shaft with WD-40 again, and were dismayed to find it still stuck. We managed to get long pliers on either end of the shaft, and the W key still barely moved - although it was getting better. We put on more solvent for an hour and then used a screwdriver and hammer to try to force the lever to move sideways, along the shaft, just a tiny bit. Yes, we beat on it -- and it freed up! More solvent was applied; today, I applied machine oil to the juncture and worked the key until it was as smooth and rapid as the others. While that was going on, on both days, we really cleaned up the machine.

I cleaned up the machine itself, while Dave did his chrome polishing magic on the carriage. I also cleaned the platen as best as possible, but it's hard as rock. Having gotten a good deal of cleaning done, I decided that cleaning the upper surface of the top deck would be easier with it on the machine, so I reapplied it. You have to reconnect the back-spacer linkage (on the right side) and the connection for the space bar to trip the escapement (done from the rear) when you put the deck back on. We also reapplied the bracket that holds and guides the ribbon vibrator. For a couple days now, it will be a matter of lots more cleaning and polishing... but you can see the difference already!

May 11, 2007. The Works were rather occupied with non-typewriter projects today, but I decided to put the panels back on the Harris. As I was doing so, I noticed something surprising, which you see Dave pointing to in this shot. It's a boss, cast into the frame of the Harris -- and it's for the lever-type ribbon selector found on Rex Visible machines made 1918 or later! This we consider to be more proof that Rex continued making machines with the Harris name for Sears after the Rex takeover in 1916. We brought last year's Rex Visible down for comparison.

It is known that no Harris Visible No. 4 machine has any kind of ribbon selector, and that the few known Harris Visible No. 5 machines have the side-button type which became standard equipment on the Rex Visible No. 4 at the beginning of 1916. Later, about 1918 the Rex Visible No. 4 changes to a front-mounted lever type ribbon selector. On the right is last year's project machine, returned to the workshop and with the right side panel removed. Dave is pointing to the ribbon selector lever shaft, which actually moves the whole ribbon bail left and right. When the bail is moved to either color position, the link to the ribbon vibrator is moved into one of two eccentrics that are different and which determine vibrator lift; centered, or stencil position actually centers the link between the eccentrics. Here we see that a special mounting is applied inside to guide and support the selector lever shaft, and provide a set of three detents for its action.

Long ago, during the making of the Harris Visible article on my site, I noted that many Harris machines had a notch in the decorative bead on the front panels, exactly where a Rex would have a hole for a ribbon selector .. but wondered why that was. It now seems clear that this is because the panel could either be punched, for a Rex Visible No. 4 with the lever type ribbon selector, or else not punched for a Sears-distributed Harris Visible No. 4 which never has a ribbon selector of any kind. Both Harris Visible No. 4 machines I have here with me are in the 100,000 serial block, and both have the break in the bead. As we know from this machine, one of them is actually clearly labeled as having been made by Rex even though it's a Harris Visible No. 4. At left, Dave is holding the front panel from the Rex-built Harris Visible No. 4 in front of the Rex Visible No. 4 we did last year, showing clearly both the break in the bead and the corresponding ribbon selector on the Rex. We're now certain beyond doubt that the serial number breakdown given on my Harris Visible article's page 6 (new this week) is nearly correct. Obviously both were made roughly concurrently, even though this Rex has a serial in the 40,000 range and the Harris is in the 100,000 range.