the Papertrail – Page 22

Monotype Computer Control: Diagnosing Air Flow Problems

Posted on 2017/06/04 by kpmartin Posted in Caster Computer Control, Kevin — 3 Comments ↓

One observation I have made while testing out my Monotype composition caster computer interface is that not all its ports deliver the same amount of air flow. On some, the air pin in the caster lifts and drops crisply in response to actuating the air valve, while others are sluggish and a few don’t move at all.

Some of these differences are caused by differences in the air lines and valving in the caster. Some lines are longer than others. Some may be partly clogged with thick oil. Some may have more leakage around the air pin. And some channels drive more than a single pin (for instance, N also has to drive the extended-matcase valve which redirects I and L to NI and NL).

However I also expected that some of the problems were in my computer-control valve module. There could be restrictions in the air passages, and the solenoid valves themselves (which are 3518-year-old surplus) may have problems too. In order to pinpoint the problems I set up an airflow gauge on the supply line to the valve module.

Good airflow gauges are expensive, but for about $7 on Amazon I found a flow gauge made for shielding gas for welding. These gauges rely on the flow of the gas to raise a small ball in a tapered column until the gap around the ball was sufficient to let the flow pass. They are calibrated for specific gases but that didn’t matter to me; all I was interested in was comparing the relative flow among the 31 ports.

The gauge had fittings appropriate for shielding gas: The inlet connector fits the outlet of the regulator, and the outlet has a barbed fitting to accept a flexible gas hose. I wanted to use compressed air fittings (specifically ARO-style quick disconnects). Neither of the fittings on the gauge was a familiar thread.

I disassembled the gauge and tapped the interior of the inlet and outlet to 5/16NC. This would not normally be a good choice for something that should be airtight, but given the constraints of how much metal was available and what matching threads I could cut this was the best I could come up with.

The body of the gauge with new threads inside the outlet (top) and inlet (left). The third leg (right) is for a flow-control valve.

I made a couple of adapters from 5/16NC exterior threads to 1/4MPT from brass rod, and fitted them to the gauge body using lots of thread sealant.

Two quick-and-dirty adapters from 5/16NC to 1/4NPT

After everything was assembled here’s what I had:

The assembled gauge, with quick disconnects installed.

Once this was done I could start measuring air flow in the valve module’s ports.

Tororo-aoi sprouts

Posted on 2017/06/02 by kpmartin Posted in Gardening, Home-grown supplies, Kevin — No Comments ↓

I previously mentioned purchasing some tororo-aoi seeds from two suppliers, and I planted five of each to see how well they sprout.

After a week in a covered tray on our enclosed porch, all five seeds from Richter’s have sprouted. There is no sign yet of the other seeds sprouting, though.

The young plants already have a taproot extending out the bottom of the peat pellet, so I should put them in the ground right away to let the root develop well.

This root is quite fragile and may be damaged when I transplant these. I may also seed some directly into the soil alongside these to see if there is any difference in growth.

Search for Tororo-aoi

Posted on 2017/05/26 by kpmartin Posted in For Sale, Gardening, Home-grown supplies, Kevin — 2 Comments ↓

This spring I did a small foray into AliExpress to see what interesting products I could find. I ended up ordering three products: kozo (Broussonetia Papyrifera) seeds, Abelmoschus Manihot gum powder, and tororo-aoi (Abelmoschus Manihot) seeds. The gum powder was disappointing (I hope to post on this later) and the seeds just arrived in the last few days. The tororo-aoi seeds arrived after the AliExpress 60-day warranty period so I am applying for my money back on those. Due to their late arrival, I also purchased some seeds for the same plant from Richter’s, a specialty plant supplier just outside the Toronto area. The tororo-aoi seeds from AliExpress and from Richter’s arrived the same day.

First, the seeds from Richter’s Herbs:

The common name they use on the packet is “Sunset Hibiscus”, and the latin name is Abelmoschus manihot. Their web page for this plant also identifies it as Huang shu kui or Yellow sichuan mallow. Other sources also identify this same plant as being tororo-aoi, the source of neri formation aid used for Japanese papermaking, and as aibika, sunset muskmallow, or hibiscus manihot. This packet contained about 28 seeds, and cost me $3.25 plus $6 shipping plus 13% sales tax, and arrived within a week of ordering.

I think the “sunset” in some of the common names refers to the yellow colour of the flowers, distinct from the usual pale blue, pink, or violet hues of most of the mallow family.

Next, the seeds purchased from Miss Flower’s Store through AliExpress:

The common name on the packet is “Golden Treemallow”, and the latin name is Lavetera arborea. The seller’s web page identifies them as “Abelmoschus Manihot Seeds Golden Treemallow Flower Seeds” and shows a description and photos that match Abelmoschus Manihot so I’m not entirely sure what seeds I actually have! I received 5 packets with nominally 15 seeds in each (I counted 20 in the packet I opened) for $2.56, shipping included, but because of the late arrival I’m asking for a refund (aren’t I cheap?).

Fortunately their web site also decrypts the pictorial information on the back of the seed packet:

Sowing Temperature : 15-30°C
Growing Temperature : 20-36°C
Spacing: 20×20cm
Grow up / Flower days : 60days
Height when grow up : 30cm

The seeds from both suppliers look like mallow-family seeds but are substantially different in size:

The smaller seeds on the left are from AliExpress, the larger ones on the right from Richter’s

I have now planted five of each seed in peat pellets, and placed them on our porch where hopefully daylight will keep them in that 15-30°C range for germination. I may also seed some directly into my garden.

The Richter’s packet identifies the plant as perennial in zones 8-11; we’re in zone 5b so we grow it as an annual and I think we just about get 60 days in the 20-36°C range so we should get some flowers. In any case, although the flowers are quite pretty, they are not essential to producing neri in the roots. Getting a few mature pods for seeds for next year would be nice, though.

I also wonder if removing spent flowers to prevent formation of seed pods would result it larger roots. I don’t think I will have enough plants to tell with any statistical certainty.

This plant is also generally held to require good soil moisture, something that is not reliable here in the summer, so I may have to keep them watered.

I still have 4 unopened packets of the suspect seeds from AliExpress, so if you’re interested in trying them out yourself and you can figure out how to pay me $1 for the postage I can send you a packet. Please note I don’t want to try shipping seeds across the border so this offer is only available to Canadian addresses.

Lubricating Oil for Old Printing Equipment

Posted on 2017/03/02 by kpmartin Posted in Documentation, Useful gadgets and ideas — No Comments ↓

Many old pieces of printing and typecasting equipment recommend (amongst other products) Mobil Vactra for lubricating some of the sliding surfaces. Locations like journal bearings (where a shaft rotates in a round smooth hole) retain their oil by surface tension. The oil doesn’t drain from the gap between the moving parts because to do so would increase its surface area too much.

Many other surfaces, though, have intermittent contact because a part either slides or rolls along another part that does not have the same shape or size. The immediate contact between the parts will retain oil by surface tension, and as long as the machine is operating the oil will be spread along the entire mating surfaces. Once the machine is stopped, however, the remaining surfaces not in direct contact are open and any oil present will just run off, given enough time, especially on vertical surfaces.

To limit this effect, special oils exist which contain a tackifier, making the oil somewhat sticky so that it remains on surfaces much longer. Mobil Vactra was such an oil, and so was recommended for sliding and rolling surfaces on a lot of printing equipment. It was also used on machine tools such as lathes and mills used in manufacturing. Tack is a distinct property from viscosity, and high tack is not a substitute for heavier oil.

However, in the late 1980’s Mobil started to remove (or perhaps change) the tackifier in their Vactra oil, with the result that it no longer effectively lubricated such parts. The earliest reference I can find to this is a message from user ‘icehd81’ on the Practical Machinist forum from December 8th, 2005. This user worked for the press manufacturer Goss and so seems like a relatively reliable source. His claim was that the oil was reformulated due to EPA requirements on the machining industry; I have seen others claim the tackifier was incompatible with the coolant/lubricant used for the cutting tools in the machinery. I suspect that the tackifier contaminated the machining fluid (which is mostly water), and when the machining fluid was disposed of, this contamination caused environmental problems (so both stories are the same).

In essence Mobil reformulated Vactra to suit a large chunk of their market (machining) without checking if the change would be a problem for another of their big markets (printing). Rather than admitting any mistake, they introduced a new product, Vacuoline, for the printing industry, and kept the reformulated Vactra.

Animation showing low tack of regular oil

Regular oil (not Vactra #2, however). Note how, when I pull my finger and thumb apart the oil quickly breaks into two round drops which then spread on my skin.

Animation of Vacuoline 1409 showing tackiness

Same test with Vacuoline 1409. Note how the oil stretches to a fine thread rather than immediately breaking away.

The net result, though, is that old equipment for which Vactra was recommended should use a Vacuoline product instead. It seems that Vactra #2 (ISO 68 weight) was the most popular recommendation, and its replacement is Vacuoline 1409.

So if you have any equipment from before the 1980’s which recommends Vactra #2, and you need to buy more oil, you should look for Vacuoline 1409 instead. Some people have claimed that you can get the same effect by adding some STP Oil Treatment to the oil, but I haven’t tried this because I now have a lifetime supply (20L) of Vacuoline.

Should the keta stretch the su?

Posted on 2017/02/28 by kpmartin Posted in Kevin, Moulds and Deckles, Sheet forming — 2 Comments ↓

After some thought, I would say “probably not.”

I have a su (bamboo screen for Japanese-style paper making) for which I made a keta (combined support frame and deckle) many years ago.

The su, or bamboo screen, used to make washi (Japanese-style paper)

The keta, or deckle frame, shown here open, which supports the su and contains the liquid pulp during sheet formation

The keta closed with the su in place. The combination is called a sugeta (a combination of su and keta with the unvoiced k replaced with a voiced g)

The keta closed with the su in place. The combination is called a sugeta (a combination of su and keta with the unvoiced k replaced with a voiced hard g)

I’ve taken the occasional attempt at making paper using this, with very marginal success. One problem I have is getting the sheet to couch off cleanly.

Even if there is full contact (i.e. no bubbles) between the post and the new sheet, the su seems to pick and tear the new sheet when lifted off. When I was trying to make paper when I was in PEI, I added a layer of no-see-um netting over the su, and only then managed to get sheets to couch off cleanly.

One thought occurred to me recently: My keta is a tiny bit too large, and the su has to stretch a bit for the keta to close. When dry, the su is a bit slack on the keta, but once wet, the bamboo splints swell and the silk threads shrink making the su a bit smaller than the keta. The edge sticks of the su are held by the closed keta, and the su must be stretched about 5mm for the keta to close.

At the time a tiny bit of stretch seemed like a good thing, to keep the su flat. But now I’m wondering if this might relate to my couching problems. The su is about 350mm wide, so 5mm represents about a 1.4% stretch. However, all the stretching occurs in the gaps between the splints, which represent about 20-30% of the total width. This means that the gaps expand by 4.5-7% when the su is placed in the keta.

When I form the sheet, at least some fibre goes into the gaps between the splints. When I remove the su from the keta and try couching it, the gaps close again by this same 4.5-7%, gripping at least some of the fibres so they cannot release from the su properly.

I have no good way of testing this theory other than to make a new keta. In the process I would also make it a bit narrower to better cover the sides of the su as another couching problem I have is fibre snagging on the ends of the splints. Finally, I would camber the top frame a tiny bit so it presses firmly on the edges of the su to produce a tight fibre-proof seal.

39th Grimsby Wayzgoose, April 29th 2017

Posted on 2017/02/19 by kpmartin Posted in Grimsby Wayzgoose, Past Events, Us — No Comments ↓

On April 29^th, 2017, the Grimsby Public Art Gallery in Grimsby, Ontario will host the 39^th annual Wayzgoose.

We’ve managed to snare a spot to sell our wares at this show. We will be selling handmade paper, marbled paper, a selection of bookbinding and marbling supplies, and new and used books on various book arts topics.

The show is open from 9am to 5pm, and everyone is welcome. Free admission.

Hollander Beater and Large Press (Still) Available in Toronto

Posted on 2017/02/11 by kpmartin Posted in Other people, Sold — No Comments ↓

Please see our updated older posting concerning a 2lb David Reina beater and 40-ton papermaking press available in Toronto.

Update: These items have been sold.

New Product Photos: Vats and Drying System

Posted on 2017/01/05 by kpmartin Posted in Things we sell — No Comments ↓

We’ve added a couple more pages describing more of our products:

The can also be found through the Products item in the main menu on each page.

Monotype Computer Control: A Long Overdue Update

Posted on 2016/12/30 by kpmartin Posted in American Typecasting Fellowship, Caster Computer Control, Kevin, Other people — 2 Comments ↓

My last post about this project was over a year ago, wherein I mentioned that I had purchased the aluminum stock to make the main parts of the valve unit. A lot has happened since then, to the point where I brought it along with me to the 2016 ATF conference to see how it fitted on other casters.

Here is how the valve body looked when I first tried it on my caster (July 2016):

The valve body installed on my caster as proof of concept.

Briefly, the compressed air comes in through the manifold and is distributed to the four pneumatic valves. The manifold also connects all the dump ports where the air is released when a valve is closed. Each valve controls eight air channels, which emerge on the far face of the valve arranged equally around a circle. The two labyrinth plates and the gasket between them carry the air from these four circular port arrangements to a single row of 31 ports on ⅛″ centers (one valve port is blind), to match the layout on the caster’s paper tower. An aluminum mouthpiece angles the ports by 45° so they meet up with the ports on the paper tower cross girt.

This is similar to the unit made by Krzysztof Słychań, but it uses machined channels rather than hoses and fittings to connect the valves to the caster ports.

The valve body has arms which hook onto one of the crossbars of the paper tower, and the weight of the valve unit provides enough pressure to seal a gasket between its mouthpiece and the paper tower cross girt.

Eventually all the electronics will be on a circuit board around the manifold, and an arm will link to the connecting hook to detect the part of the caster cycle where the air valves should be open. The unit will be fully self-contained, with just a power cord, compressed air inlet, and USB port. As with Bill Welliver’s interface, it will still use a computer to run it.

The rear labyrinth plate. Some passages penetrate to this layer directly from the valve ports and follow the channels to the outlet ports along the center line. The remaining passages follow channels in the front labyrinth plate directly to the output positions. This was the first plate I made, and you can see some misalignment between the holes and channels due to a loose clamp bolt on the mill.

Drilling the air passages in the mouthpiece. This was awkward to do because of the 45o bevel involved.

Drilling the air passages in the mouthpiece. This this involved some tricky work holding because of the 45° bevel involved.

The valve body is a close fit on the paper tower. I don’t want it too sloppy because this might allow too much misalignment between the output ports and the ports on the paper tower. On the other hand, too close a fit makes it too tight to put in. The only dimension I had to work from was the width between the side covers of the paper tower on my caster, and I have no idea what manufacturing tolerances are involved. To see how well it fit on other casters, I brought it along with me to the ATF conference in upstate New York

My Show and Tell goodies for the conference

My Show and Tell goodies for the ATF conference

The electronics for the controller included a power supply, a microcontroller in the form of a Maker conference badge, and the drivers for the valves wired up on a breadboard. Everything was mounted on an ugly piece of plywood. The bin also contained an unusual Monotype mould for making cored type (I think it was 24-point), and a few laminated caster adjustment cheat sheets to pass around.

The microcontroller was not yet ready to do actual casting work, but I had it programmed to run the air channels sequentially, holding each one for about a second before switching to the next.

I got to try the valve body for fit on three of Mike Bixler’s comp casters, but unfortunately it would only fit on one, and even there is was a tight fit. This was during the show-and-tell segment of the conference, and there were a few interested people having a look and asking questions. Unfortunately I seemed to be spending most of my time explaining that the software was incomplete and only running a test cycle, as that was ultimately the answer to several of the questions.

I also hooked up compressed air to the valve to see how well the caster reacted. Some of the air pins snapped up smartly when activated, but others were sluggish or just didn’t move at all. I had seen similar results on my own caster, but I didn’t have a record of which ports worked well and which did not.

So my take-home from this was that:

There is some interest in this from ATF members
I need to make the body a tad narrower to ensure it fits all casters
I need to look into poor air flow problems

Tagged with: Bill Welliver, Krzysztof Słychań

The 2016 ATF Conference, Day 4

Posted on 2016/12/30 by kpmartin Posted in American Typecasting Fellowship, Conferences and Meet-ups, Kevin, Past Events — No Comments ↓

Actually, there was no official Day 4 for the conference, but the Bixler foundry in Skaneateles and Virgin Wood Type in Rochester both had open houses again for those would couldn’t make the other times. Rochester was on my way home so I stopped off at Virgin to see their facilities.

They are located in a residential cul-de-sac, backing onto some freeway or other, with the dusty mechanized cutting taking place in the garage, and everything else in the house. The faces are cut using a router pantograph with a fixed spindle and moving work table. Several pieces of type are cut at once and later cut apart during finishing.

The pantograph: the spindle is just to the right of the lamp reflector. The tracer (in Jim Grieshaber’s hand) moves the work table through the linkage arms. A foot pedal raises and lowers the spindle.

The patterns for cutting are typically made by laminating cutout shapes (usually of thin plywood) onto a base. The cutouts are sometimes made by hand, but can also now be cut using a laser cutter. They have to be thick enough to resist wear from repeated passes of the tracer, and also to prevent the tracer from jumping up over the edge.

Although the wood used to make the type is end-grain, there is still a bit of wood fibre attached to the edges of the letters, so after separating the letters, the next step is to clean up the fuzzy edges left by the router bit. This is done by hand using a small fine flat file for the straight edges. The trick is to remove the fuzz without rounding over the edge.

Another finishing step is to clean up the inside corners which are too sharp an angle for the router bit to reach. For obtuse angles this is done with a sharp knife, but for acute angles a special punch is used. In either case the trimming is done almost by feel off the adjacent straight cuts so the trimmed edge aligns with the rest of the router cut.

When the type is finished, proofs are taken using carbon paper (so as not to ink the new type) on a flatbed press and the proofs are inspected prior to approving the type for shipment.

I’ve posted some YouTube videos of the process, showing the milling, edge cleanup, corner trimming, and proofing steps.

One product they have are chromatic faces, where there are two pieces of type for each letter, to be printed using transparent inks of two colours. The printing areas of the type overlap, so you get not only the two original ink colours, but also areas which show the combination of both colours. Although this could be extended to more than two colours, they’re currently only producing a two-colour version. There is an illustrated example at Virgin’s web site. These are made from a single multi-layer pattern to provide consistent register between the colours.