Saturday, September 11, 2010

Truing a Puller

One of my workshop's operating principles is "you can never have too many pullers". So when I saw this 8" puller on special at Princess Auto I made sure I got myself one.

These inexpensive pullers from offshore do give you a lot of metal for your money, but they tend to come up a bit short on precision, and are likely to need a little attention before they'll work as well as they ought to.

The shortcoming with 3-jaw pullers that I keep seeing is in the uniformity of the lengths of the jaws. When the lengths are uneven, the tool sets up crookedly. The defect doesn't render the puller useless, but it does make it clumsy to set up (and pullers are clumsy enough things to set up when they're perfect) and it does impair the puller's effectiveness -- the puller's force is off at an angle to the subject axis; that's not helpful.

To check the jaw length evenness, dismount the jaws and rack them up on a close-fitting bolt or length of threaded rod. Any length unevenness will be obvious.



I think I'll have to ask you to trust me on this, but that near-side jaw at the lower right is shorter than the other two. The puller will set up crookedly. This puller is actually not too bad in this regard; I have a 6" puller that was much worse. (Why the factory cannot or will not amend its methodology to correct this is beyond me.)

You need to also examine the jaws at the other end, and assemble the jaws using the other hole and examine both ends. In this case, the other end was good enough not to need attention; there's just this one short jaw that wants grinding to even it out with the others.

Here's the rig I came up with for grinding puller jaws:



That angle grinder has an M8 threaded hole on either side for a handle, so it was a simple matter to make this prop arrangement that presents the wheel at an angle suitable for grinding the hook-faces of puller jaws. The next photo shows the process in action.

This is not precision machine shop work by any stretch of the imagination; it's rough, nasty, 'by guess and by gosh' work. You'll find it a bumpy ride as particles lodge is the wheel now and then and kick the work. Check your progress often, and don't expect a pretty outcome. The result pictured below is where I stopped and said "good enough".



As I said, it's not pretty. The sheer size of these jaws made this job more difficult than others I've done, but I'll have a better puller for having done this.

The final photograph is of the jaws of a 6" puller that had very uneven jaw lengths, much worse than this one's. On that puller, one jaw was much longer than the other two, to the point where I would have seriously weakened the other jaws had I ground them to align with the long jaw.

I got around that by fashioning a shim that I could crimp onto the face of the long jaw. It worked well. I pulled a brake drum with this unit the other week and the puller set up nice and straight. It certainly wouldn't have prior to the grinding and the addition of the shim.



# # #

FEEDBACK

# # #

Grinding Wheel Dresser

At the risk of appearing to be a shill for Lee Valley, I must give credit where it's due and report on their diamond grinding wheel dresser No. 70M5003.

I'm impressed. The thing works nicely.

The new wheel pictured was almost 0.020" out-of-round, so I thought it would make a good test subject. With very little work I got the wheel to within 0.005" of round. I dressed two other wheels while I was at it and both came clean and 'smooth' easily.

The tool works best when it's swept from side-to-side, using the full length of the dressing head. You might want to wear a shop apron while using it. By the time I was done with the out-of-round wheel, I needed a good vacuuming in front.

- - -

Addendum

This pretty much renders my clunky old star-wheel style dresser obsolete. I must find a good use to put it to.

# # #

FEEDBACK

# # #

Monday, September 6, 2010

A Micrometer Stand

When you acquire a micrometer and begin making use of it, you soon discover that miking tiny parts with it is a very clumsy bit of business. You truly need a third hand, and the stand pictured does the job.

Before I built this, I looked into getting a ready-made one. They can be had, but not nearby or inexpensively. It somewhat surprised me that such a useful accessory for a micrometer would be such an obscure item.

Anyway, since I much prefer working to shopping, I went on a scrounge around my workshop and came up with the makings of a stand in short order.

The base is a steel, 3 1/4" diameter pulley. The pulley is heavy enough that the stand isn't tippy. A 4" spring clamp serves nicely for holding the micrometer. Two 3/4" corner braces and a few fasteners and washers provide the hinged mounting arrangement that permits adjustment of the micrometer's viewing angle.

The screw for the pivot-point is Loctited into a threaded hole in the lower corner brace, so the screw never loosens when the wing nut is loosened for adjustment. I should add a second screw or an interference-fit pin where the clamp is attached to its corner brace, so the clamp can't twist out of position sideways.

I won't hold my breath waiting to win any industrial design awards for it, but for the price it's been serving me well. (By the way, that oversize 1/4"-20 wing nut was once the cover fastener for a Ford Pinto's air cleaner. Think what you will of the Ford Pinto, they were good basic transportation, and they had fine air cleaner cover wing nuts.)

- - -
Thursday, September 9, 2010

Addendum

I finally got around to doing what I should have done in the first place, I added a roll pin to the clamp/corner-brace attachment point so the clamp can't get torqued out of position.

This is a handy technique for securing interfaces against rotation when you have very little room for an additional threaded fastener. Drill one small hole, install a roll pin, problem solved.

That roll pin is tiny. It's 1/16" diameter x 3/16" long, but it's all that's needed.

# # #

FEEDBACK

# # #

Sunday, September 5, 2010

A Watering Can Repair

The old garage-sale-find watering can pictured is a nice piece of gear. There's a brand name on it; "Garden Club". It must be pretty old because it was made in the U.S.A., and the U.S.A. doesn't do that sort of thing anymore.

The body of it is still in fine condition, but the nozzle is past it -- embrittled and cracked in many places. The nozzle in the foreground is a good salvaged one that I can adapt once I get the duct tape adhesive muck off the can's spout. (I'm not a big fan of duct tape. I've spent quite enough time cleaning up what it leaves behind when it's removed, and I don't recall ever having enjoyed it.)

The difficulty here is that the replacement nozzle came from a can that had a considerably larger diameter spout-end on it than this can has. I could just slather the parts with a silicone gasket maker and assemble them that way, but it would be a poor job. The key to using adhesives and sealants is to never ask too much of them.

What's needed is a bushing of sorts to take up the space between the two diameters. That will minimize the amount of adhesive required, and make for a snug, sturdy interface. Sometimes you get lucky with these situations, and it turns out that some piece of standard-dimension thin-walled tubing is close enough to what's needed. But here, I'm dealing with a couple of very odd diameters, so I'll have to fabricate what I need.

- - -

Here's what I came up with:

That band was cut from a scrap of 0.020" thick aluminum window casing material, and rolled around a 9/16" socket wrench to form it. The replacement nozzle can just be slipped on over the band with a bit of force applied, so I've succeeded in taking up the bulk of the gap with solid material. The 1/8" diameter holes perforating the band are key to getting the strongest possible outcome here. I'll be using epoxy as an adhesive/filler.

What the holes do is they effectively turn two interfaces into one. The epoxy filling the male thread on the spout, and the epoxy filling the female thread in the nozzle won't be entirely isolated from one another. When the epoxy has cured, at every hole there'll be a tough, contiguous span of epoxy through the hole. So, there won't be two epoxy-adhered interfaces; there will be one aluminum-reinforced epoxy-adhered interface -- big difference. In normal use, the repair should be near indestructible.

Here's a shot of the inside of the nozzle (the sprinkler head is a snap fit and can be pried off fairly easily). You can make out the edge of the filler band, and see that the epoxy has formed a full and reasonably uniform fillet all around. That nozzle's not going anywhere without the can.

I'll leave that to fully cure overnight, snap the sprinkler head back on and photograph the finished item.

- - -

And here's the finished watering can out with its pals:

A good outcome. You'd never know to look at it that the can and nozzle weren't made for one another. And the purchase price was all of twenty-five cents, no sales tax.

* * *

Nothing Lasts Forever -- WEDNESDAY, MAY 16, 2018

'Went to use the watering can, and discovered that the nozzle had acquired a nasty fracture.


'Don't know what happened there. It appears that either the nozzle shrank, or the epoxy adhesive/filler swelled up. In any event, what I have now is an insecure, leaky nozzle. I can fix that up so the watering can remains useable, but it won't be the most elegant repair I've ever done.

- - -

And here we are.


I wired the nozzle in place with 0.047" diameter (No. 18 U.S. Steel Wire Gauge) stainless steel wire, and slathered the fracture with hot-melt glue. It isn't pretty, but it will serve.

A downside to the repair is that the nozzle isn't readily removeable, should one want to be able to pour water directly from the can's spout. But then, that was the case with my first repair as well, so I really haven't lost anything here.

# # #

FEEDBACK

# # #

Wing Screws

Everything has its converse -- vice has virtue; joy has sorrow; hex sockets have hex keys and wing nuts have wing screws.

Wing screws are easily fabricated when the need arises. The only complication is that most screw heads in any given thread series are too large in diameter to fit directly between the wings of a wing nut. (The examples in the photo are 10-24.) An exception to that is hex socket head screws, but I don't care for the finished appearance (upper left); the hex socket head is too bulky to look right.

But a few minutes at the lathe with an ordinary round head screw gets you a reduced diameter screw head that's in good proportion to the wing nut (upper right).

Assemble the two pieces with red Loctite threadlocker or CA adhesive, and tighten very firmly.

- - -

Addendum -- THURSDAY, OCTOBER 10, 2013

If you need an M4 wing screw, but only have 8-32 wing nuts, just run an M4 tap through an 8-32 wing nut and you're away. M4 and 8-32 are very near one another in diameter and pitch. (M4 is slightly smaller in diameter than 8-32.) The M4 thread you get from tapping an 8-32 wing nut is entirely adequate for this application.

# # #

FEEDBACK

# # #

Sunday, August 22, 2010

Roll Pins

Roll pins (aka 'rolled spring pins' or 'slotted spring tension pins') are a very good thing to have on hand. They come in a wide range of sizes; the two pictured give an idea. The large one is 3/8" diameter x 2" long. The small one is 3/32" diameter x 5/8" long. (There are metric dimensioned ones too, of course.)

The two shown here are from a 120-piece assortment I got from Princess Auto. Should I ever need a size or a quantity that the assortment doesn't provide for, I'll have to go to an industrial supply outfit like Spaenaur and buy a package quantity (typically 100 pieces).

The pins are made from quite hard spring steel, rolled into shape as the name implies. The key feature of them is the dimensioning. A roll pin's diameter as you receive it is always slightly oversize of nominal; e.g. a 1/8" (0.125") diameter roll pin will have an actual diameter of 0.131" to 0.135". To install one, drill a hole the nominal size and hammer in the pin. (One or both ends is/are always chamfered for ease of starting.) As the pin is driven, it contracts to fit the hole and effectively gives an interference fit of considerable strength.

An installed pin can be punched out and reused, which means that salvage can be a source of them. (A word of caution about punching them out: be sure to use an exact-size punch, and be especially careful if you're starting on a slightly protruding end of a pin. An undersize punch that manages to get down into the hollow centre of the pin can make a nasty mess of things. There are special punches available that have a centring point on them. You'd have to go to Spaenaur or a similar place for those; they're not something you're likely to find at the Home Depot.)

Uses are endless -- pinning gears or virtually anything to shafts; reinforcing glued up cylindrical assemblies; as motion-limiting stops in adjustable assemblies; what-have-you. Uses have a way of suggesting themselves when you have the pins at hand.

Occasionally, the hollow centre can be an undesirable feature; e.g. on a fertilizer spreader axle where corrosive material could get inside. In such cases, the centre can be filled with a length of soft steel wire or rod and epoxy or CA adhesive.

In closing, I must leave you with a word of caution about Princess Auto's assortment; the smaller diameter pins are poorly formed and seriously oversize, and will not install correctly in nominal size holes. To use them successfully, you have to mike them and select a suitable number size drill. For example, the 3/32" (0.09375") pins are about 0.113" in diameter. That's almost 0.020" oversize, and that's a lot for an interference fit. A No. 38 drill (0.1015" diameter) worked ok.

MONDAY, AUGUST 23, 2010

Addendum

Here's some information you may find helpful for dealing with oversize fractional inch diameter roll pins. Following is a list of all the available nominal diameters with some added information. Each entry is in the form:

Nominal Fractional Diameter; Nominal Decimal Diameter; Maximum Allowed Decimal Diameter*; Difference (Difference Rounded Off to the Nearest Thou)

* i.e. the maximum diameter that a given pin should have from the factory. A pin whose diameter exceeds this figure will be troublesome to install in a nominal-size hole.

1/16"; 0.0625"; 0.069"; 0.0065" (0.007")

5/64"; 0.078125"; 0.086"; 0.007875" (0.008")

3/32"; 0.09375; 0.103"; 0.00925" (0.009")

1/8"; 0.125"; 0.135"; 0.010" (0.010")

5/32"; 0.15625"; 0.167"; 0.01075" (0.011")

3/16"; 0.1875"; 0.199"; 0.0115" (0.12")

7/32"; 0.21875"; 0.232"; 0.01325" (0.013")

1/4"; 0.250"; 0.264"; 0.014" (0.014")

5/16"; 0.3125"; 0.330"; 0.0175" (0.018")

3/8"; 0.375"; 0.395"; 0.020"; (0.020")

1/2"; 0.500"; 0.524"; 0.024" (0.024")

- - -

SUNDAY, JULY 3, 2011

Addendum -- Application Examples

Here's an example of a pin anchoring a stud at the end of a rod.

And here's a roll pin serving as a valve handle.

Scroll down in either and you'll get to it.

# # #

FEEDBACK

# # #

Thursday, August 12, 2010

Somewhere, a Saucepan Needs a Handle

Right about here, actually.

The saucepan pictured is quite a nice piece of cookware, but its handle went south a long time ago.

The handle attachment bracket that's spot-welded to the side of the pan is still perfectly sound. The bracket contains a square M5 nut for attaching the handle with a machine screw.

I have an idea how to go about this that will make a nice illustration of how common garden hose fittings can be employed as structural mechanical parts. The outcome won't be the most aesthetically pleasing thing I've ever done, but it will serve the purpose and keep this lovely piece of stainless steel manufacture out of the landfill. Here's a photo of some of what I'll be working with:

The hose fitting cap will be drilled for an M5 screw, and securely attached to the pan's handle bracket.

I'll strip the ferrule and hose remnant off the male hose fitting, and wed it to the end of the hardwood broomstick piece. I'll have a sturdy replacement handle for the saucepan with a unique feature -- it can be easily detached to make washing the saucepan easier.

The first thing to do is drill that cap through the centre for the attachment screw. A 3/16" hole will give me a slight interference fit for a 5mm screw, which is exactly what I want to have here -- minimal screw clearance so the final assembly will be as snug as possible. That raises an interesting point about inch vs. metric in the workshop. A little digression is in order.

Inch versus Metric Measure

There's a fallacy contained in the above sub-title; the word 'versus'. The two systems are not mutually exclusive. In working with both of them, I've come to regard them as a complement to one another; in a sense, I tend to see the two as one big seamless system that I happily put to use however it suits me. The 3/16" hole for a 5mm screw just mentioned is a small example. One encounters many such mixed uses of inch and metric in manufactured goods. My old Raleigh Grand Prix bike and my ancient 8" Beaver table saw both exhibit mixes of inch and metric dimensions/items where it suited the designers' purposes.

I grew up with fractional inch measure, and it will always be the system that I'm naturally inclined to think in. But in the workshop, one is well advised to be fully conversant with and equipped for the use of both systems.

I say that quite calmly now in spite of the fact that when Ottawa began to seriously embark on converting Canada to metric decades ago, it got my back up something fierce. Ottawa's embrace of metric measure had nothing to do with technical progress. It was a transparent, ham-fisted exercise in social engineering, and an opportunity to jettison another piece of our 'colonial' past. If nothing else, imperial measure is a cultural artifact of great beauty and utility. Ottawa would have had us discard it as if it were a worthless piece of trash. Ottawa can be pig-ignorant and tree-stump-stupid at times.

There, I got that off my chest. Back to the saucepan. This is a workshop blog, after all, not the op-ed page of the Globe & Star-Post.

Here's the pan with its new handle receiver:

The hose fitting cap had a yellow zinc dichromate plating on it that was a bit corroded in places, so I brushed it with the wire wheel machine and that took the plating off entirely, leaving me with a naked steel part. I spray-painted it with grey primer and some automotive silver paint. What the heat resistance of ordinary spray paints is I have no idea, and I can't find a mention of it on any manufacturer's website. I guess it's a secret. Time and use will tell if the cap's paint job will prove durable.

The original M5 screw was longer than needed, and it had a combination slot/Phillips recess that I didn't care for on a screw that will need to be tightened very firmly. I replaced it with a 12mm long hex socket head stainless steel screw, and added a split lockwasher and flat washer, both stainless steel. (And there's another inch/metric mixture; No. 10 inch washers fit M5 screws perfectly.)

Stainless steel fasteners are outrageously costly, but in places like this where they're truly called for, hang the cost. I happened to have what I needed on hand, but would have gladly parted with a dollar or two to get stainless steel parts if I hadn't. That points up the value of salvage to a workshop. It's a good idea to cultivate an eye for good items that can be found in discarded machinery and the like. You can build quite a valuable stock of useful fasteners and material that way.

I used red Loctite thread locker on the screw, and really tightened it firmly. Handles that loosen are a nuisance I want no part of.

The Handle-End Fitting

This is the part that's to be wedded to the end of the wooden handle so the handle can be screwed into the fitting attached to the pan. I was working here with a die-cast male hose fitting, which complicated things a bit. Were I doing this over, I'd use a fully machined fitting.

The problem with a die-cast part like this is that in its intended application, the precision of the bore is of no importance -- it just has to let water flow freely. Consequently, the manufacturer takes no pains to ensure that the bore has a uniform, concentric diameter.

I'll try to summarize the construction of this part without glazing your eyes over.

The hose fitting's bore was something of a taper. I chucked the fitting in the lathe and bored it through 11mm diameter to get a more-or-less straight, consistent bore. The threaded rod is a piece of undersize 3/8"-16 rod with a 9mm diameter. (Not all threaded rod is created equal. This piece had quite flattened thread crests which made it undersize, and just what I needed for this.) I needed a bushing to take up the 2mm diameter difference, so I turned one from a length of thick-walled mild steel tubing that had a 9mm bore. With those three items done, I had the makings of what you see in the photograph. The missing piece of the puzzle was a suitable adhesive/filler to bond them all together with.

And that brings us to an epoxy product that's new to me; something known as 'J-B Weld'. The manufacturer claims that it will withstand 500 degrees Fahrenheit, so that should make it suitable for a saucepan handle application. I mixed up a batch of the stuff, slathered all the mating surfaces of the fitting components with it and clamped it up in the lathe so it would all be held straightly and concentrically while the J-B Weld hardened. Here's how the setup looked:

J-B Weld is a slow-hardening epoxy (fifteen hours for a full cure). I left it like that overnight, and the next day the outcome looked good. Just for good measure, I pinned the assembly through in two places with 3/32" roll pins, and now I've got my handle-end fitting as you see it in the preceding photograph.

Next, I have to fabricate the hardwood handle. That's going to be a bit of a challenge.

Boring and Turning Wood Parts

Wood turning can produce some remarkable items, but it's a process that needs to be approached with a great deal of forethought. Every wood turning project has its correct sequence of operations, and that sequence needs to be thought out before beginning. An oversight in planning a job can result in a situation where a necessary step gets rendered difficult or impossible to do because a preceding step was taken without regard for its effect.

The wood handle-blank needs to be bored to accept the mechanical interface part that I've just made. The difficulty for me here is that my metal lathe is too small to take the wood handle-blank in its hollow spindle. If I could set up the handle-blank that way, the boring job would be a breeze. But I have to come up with another way to do it.

Another complication is that I decided I'd best add a brass ferrule to the handle for strength of the final assembly. A brass coupling nut for a faucet supply tube fitting will serve for a ferrule. What I have to work out is a sequence to do it all in, keeping in mind how each step will affect the steps that must follow.

Here's a point-by-point list of how it was done:

1) Prepare the handle blank for turning between centres -- spur centre at the headstock; live centre with a point only at the tailstock, not a point/cup style of centre. (The books will tell you not to do that; that the tailstock end must be supported by both the point and the cup of a woodturning centre. I'd go along with that if I were turning a porch column here, but what I'm setting up for will be a very light bit of work, entirely safely supportable by a point centre only.)

2) Mount the work and square off the tailstock end with a parting tool as far as the centre will allow. Turn a 3/4" diameter land to accept the ferrule.

3) Dismount the blank and set it up for boring on the metal lathe. Here's how the setup looked:

With a somewhat unorthodox use of the steady-rest, I managed to get around the lathe's small spindle hole diameter. This boring arrangement worked well. The steady-rest is supporting a 3/4" bore ball bearing, which in turn is supporting the work.

4) Bore 9/16" diameter sufficient depth for the hose fitting's barbed nipple.

5) Bore 9mm diameter sufficient depth for the threaded rod.

6) Dismantle the boring setup. Do a trial fit of everything. Glue it all together with J-B Weld and mount it in the wood lathe to clamp it up like so:

This worked out nicely in that the cup of my wood lathe's live tail centre just fits the inside diameter of the hose fitting. I have a perfectly good setup here for turning the handle.

7) Turn the handle blank to a uniform cylinder to eliminate any out-of-roundness.

8) Determine the length dimension, and imagine/sketch the profile to be ended up with.

9) Turn and sand. (N.B. This is a nerve-racking place to be at. By this point, you have a lot of work invested. One blunder with a chisel can ruin the whole thing. Be sober as a judge and proceed with caution.)

10) Part off and dismount the finished handle.

And here's the handle out on the job. A good example of how humble, dirt-cheap bits of hardware and material can be put to very good use, but by no means a claim that it's 'easy'. A lot of thought went into this, and a lot of resources were brought to bear on it. Beware of publications that make repair/fabrication work appear to be a breeze. It's not.

TUESDAY, OCTOBER 5, 2010

Addendum

I appear to have the makings of a system here. A good thing, too, because our household just lost a frying pan handle to time and metal fatigue. Pictured below is the wreckage. Fortunately, no one was injured in the incident. A frying pan handle breaking while one is pouring hot bacon grease from the pan could be a frightful bit of business.

- - -

Here I have the opposite case to the saucepan handle -- the handle still exists, but the attachment point at the pan is a ruin. If I can salvage enough of what was a threaded boss, I can fit a hose fitting cap to the pan and attach the saucepan's handle to it.

- - -

Here's what I did to 'renew' the handle attachment boss, and prepare it to receive a new handle:

1) Saw off the boss remnant at an appropriate angle. (On an object like this one, you just have to eyeball and guesstimate angles as best you can.)

2) File the sawn-off boss smooth.

3) Drill through the pan's wall with a tap-size drill for a 1/4"-20 thread. (No. 7 drill -- 0.201" dia.)

4) Thread the hole 1/4"-20.

5) Bore through the centre of a machined, solid-brass hose fitting cap with a No. 7 drill and thread it 1/4"-20.

6) Round up a stainless steel screw and flat washer. Cut the screw to a length that's just a little longer than needed. (To cut a stainless steel screw, use a 1 1/4" diameter reinforced cut-off wheel in a Dremel hand grinder or equivalent. Stainless steel is hard stuff.) Do a trial fit of everything.

In the following photo you can see the prepared boss and hose fitting cap.

Next, I installed the screw w/flat washer in the cap very firmly with red Loctite applied, then I installed the cap/screw assembly in the pan's boss very, very firmly with red Loctite applied.

At the inner wall of the frying pan, I had a slightly protruding screw-end that needed to be ground flush with the pan's wall. I didn't want to do the grinding until the Loctite had cured, but I didn't want to wait for that overnight. I stuck the pan in the oven at 170 degrees Fahrenheit for about twenty minutes, then switched off the oven and let it cool on its own. (Heat is quite an effective accelerator for Loctite curing.)

I used the Dremel cut-off wheel to grind the protruding screw-end flush, swiped the handle from the saucepan and installed it, and here we are; the frying pan has a handle again.

- - -

Now I have to make another handle. I hope this is the last of broken pan handles around here. Mass-production is not an interest of mine.

# # #

FEEDBACK

# # #