Press fit and heat treatment question

WAG response would be that you need a new part or you'd need to fit slightly oversize pins in the existing disk, although the hardness of the disk may have been compromised by the heat soak, and could be annealed now.

Not knowing what the metal alloy(s) are, you should probably get a new replacement part from the machine maker.

Another possibility is that the pins may have been shrunk-fit into the disk.. which has little meaning now that the holes are oversize.

Some plastics are abrasive, and HSS pins may exhibit considerable wear if used to grind some plastics.. also, plastic may be more likely to stick to HSS. Carbide pins would likely be brittle (depending upon how far they protrude from the disk) but more durable/less wear, and less likely to have the plastic stick to them.

When you say "mils" while talking about dimensions of parts, it's unclear what sizes you're trying to describe.

Mils are common to film thicknesses.. heavy duty trash bags are 3 mils, or coatings thicknesses/paint coatings etc.

Mils isn't a shortened term meant to represent mm or millimeters (like tranny for transmission or the other s/he kind). Whereas a tranny fitting would generally be referred to as an adapter (or possibly a dressmaker term).

Millimeters are millimeters or MM/mm, so 118mm is about 1-3/4".

Mils are thousandths-of-an-inch, but 0.118" isn't expressed as 118 mils.. and 0.118" would be smaller than 1/8" in diameter. Maybe you're referring to the diameter of the pins, but it's not really clear.. 3mm is 0.1181").

Finally, if you expect to have frequent questions or comments regarding metalworking, home shop practices or similar topics (all welcome here in the RCM usenet newsgroup), get a news reader program, or use your email program with a news subscription or news feed service, instead of posting thru the "diybanter" or other online/web reader.

Wild_Bill Inscribed thus:

Surely you mean 4-3/4"

mill ==> 0.001 inch

How about cleaning off the pins and holes them super glueing them back in

But of course 4-3/4", Marilyn

Thanks for the replies. I was referring to the pins being 0.118" in diameter. I guess the term mils (=1/1000th of an inch) is not used for larger dimensions.

I am getting a quote from the manufacturer for a replacement disk, but I am guessing it is going to be in the $1000 to $2000 range (german made machine). The vendor did mention that the pins are press fitted as they are meant to be replaceable as the pins wear over time

Since I have nothing to lose, I may try just some two part epoxy to glue the pins onto the disk from the back (the pins stick out the other end).

I will look into the newsreader suggestion

Andy

[ ... ]

[ ... ]

Why not? The spacing of leads on the old DIP chip packages were specified as 100 mils (0.100"). It all depends on the field. Some use mils just as he has apparently used them.

Enjoy, DoN.

Yep, it depends upon the particular field when mils are commonly used.. and I mentioned a couple of examples. I've never heard of feeler gage thicknesses expressed as mils, either.. and yet they are.

IC pin spacings have commonly been specified in decimal/inch dimensions as far back as I can remember using them, in the '70s.. not as a component or board designer, just as catalog descriptions and datasheet specs. But pin spacings in highly compact dimensions are leaning toward metric dimensions more recently.

The OP is often the only one who has seen a part, and lots of folks seem to leave out descriptive details, which then require several exchanges of questions to get an idea of what the original question was.

Just sayin'

Well, you may have at least a couple of options.. and one consideration would be how much damage will occur if some pins came out while running the machine.

If making a new disk is within your capabilities (you still haven't stated if the disk is small or large/how thick or how many pins there are or how they're arranged), shrink fitting is performed at safe temperatures which wouldn't affect the hardness of either the disk or pins. Future replacement of the pins may be more difficult if they're shrunk-fit. You haven't mentioned how hot the pins may get during operation, which could be very detrimental to an epoxy product. Loctite makes compounds that are used for semi-permanent assembly of parts.

As I mentioned in a reply to DoN, you're the only one who can see the part.. If the disk is thin, epoxy is likely to fail. If there is much heat generated during operation, your options are fewer. Points or other features on the pins such as different lengths may affect the assembly method options.

Using slightly larger diameter pins in the original plate/disk may work, although unpredictable, since the original characteristics of the plate are unknown. Fabricating a new plate/pin assembly may be cost effective, but not as durable as an original replacement part. New pins wearing down may not be as bad as having them come out and damage an extruder or other part of the machine.

The size of the holes in the old plate/disk won't be reduced by reheating.. it would take some kind of stamping or swaging process to displace the metal around the holes, but that would make the metal thinner around the pins.

Using epoxy will probably fail. Using Loctite made for cylindrical fits might work, it needs to be heated to 450 degrees F in order to make it release. But I wouldn't count on any type of glue. Instead I would buy a .1245 dia reamer to ream the holes larger and some 1/8 diameter dowell pins and press them in. Dowell pins are available in the hardened condition (most common), soft, and in stainless steel. They are ground accurately to size and they are cheap. If the plate is quite thin, say 1/16 inch, a .1240 dia reamer would be better. Even a

1/8 inch thick plate might be better reamed to .1240. But any thicker .1245 to .1247 should hold the pins in. Eric

I haven't followed this whole thread, so this might have been mentioned, but...

There are two reasons those pins might have loosened in the first place, and one of them may still be a problem. First, it's possible, if the plate constrained the growth of the holes upon heating, that the pins simply compressed against the holes and displaced some metal. When it all cooled, the pins could be loose.

The other thing may sound strange, but, if I followed the situation accurately, is the more likely reason, IMO. Martensite -- the hardened state of the steel -- is less dense than the softened state (say, ferrite, to keep it simple) Pins in their hardened state are slightly larger in diameter than they are when they're annealed. The amount is very slight, but in something like this, they could be noticeably looser when the metal is soft. Again, it depends on how much the plate was hardened in the first place and how much its shape constrains the enlargement or shrinkage of the holes.

So you have to be careful when measuring and fitting the pins to do so after you've hardened them to their final state. If you fit them soft, and then harden them, you have a problem.

This is an old issue that used to come up in the making of drill jigs. In fact, it's the reason that Dick Moore invented the jig grinder, after having had success with the jig borer. You could get everything right with the borer, harden the tool, and then discover that the drill bushings wouldn't fit in their holes, and that the holes weren't in exactly the right dimensional relationship anymore.

The solution was to finish-grind the tool after hardening. It's pretty much the same situation -- again, if I read the early posts correctly.

Thanks for all the suggestions. I wanted to let you know where I am

Firstly: Here is picture of the machine almost identical to the one I have;

you look through the pictures, you can see that there are two plates with pins - one fixed and one rotating. It was the pins on the rotating plate that I ruined by heating to 600C. The pins go through the rotating plate

Well, I initially tried Loctite in the gaps and epoxy on the back side

- but the loctite did not cure and I did not have all the pins at exactly the same height, so it was scraping on the fixed plate and on the back side when rotating (the gaps are just a few mm's on both sides of the plate). So I burnt everything off in the oven - again and used just 2 part epoxy very liberally on the back side - all over the pins and plate. The back side does not see the chips that are grinding

So far after about 10-20 runs at~5000rpm, it seems to be holding - all the pins are still tight. SInce the gaps are so tight, there is no danger of the pins falling out, so I thought I can take the chance with this approach for now. I am not really a machinist to be able to rework the part and I do not know enough metallurgy to know what kind of heat treatment would work so this seemed to be the only approach for me

Thanks again

Andy

If you look through the pictures, you can see that there are two

Greetings Andy, I'm surprised that the epoxy is holding so well. And that the Loctite didn't cure. Loctite is an anaerobic curing chemical compound so maybe there was too much exposure to air for it to cure. I would still try to press 1/8 diameter pins into the plate after reaming out to .124 if the epoxy does fail. Apparently though the epoxy has a high enough compression strength and sticks well enough to take the strain. Good for you for trying your solution and reporting your success here. Nothing beats the empirical testing of your ideas, especially advice from folks like me who aren't holding your part in their hands. Cheers, Eric

jsw

"Jim Wilkins" fired this volley in news:jimcs3$gev$1 @dont-email.me:

I disagree with that. I think it depends upon the particular version you're using.

I use 747 "removable", and it cures on both aluminum AND stainless (and in combination).

It even cures in contact with plastics and aluminum, although it takes a few hours.

Lloyd

If you look through the pictures, you can see that there are two

Greetings again Andy, After thinking about Loctite I remembered that whenever I used it with stainless steel assembly I always used a primer because otherwise the Loctite would cure very slowly or not at all. Maybe that's why it didn't work for you. Eric

Loctite uses several different adhesives for their various products, but the original thread-locking and related products are anaerobic-cure acrylic adhesive. It does its intended job very well, but for the record, its shear strength and compression strength are a fraction of the strength of good epoxy.

However, there's a lot more to it than the bulk strength and ideal bonding strength of the adhesive itself. Epoxy can be tricky in terms of the actual adhesive bond, and the stronger it is, the more vulnerable it is to peel and cleavage failure.

Acrylic works better on surfaces that are less than perfectly clean. Epoxy needs near-perfect cleanliness unless it's formulated to work on less-clean surfaces.

One of the most interesting examples of the latter is the bead-type thread-locking epoxies used in industrical production. The resin and the hardener are each encapsulated in tiny, sealed beads. You mix the two types of beads together, as a coarse, dry powder, before use and then they're applied to the threads you want to lock, using a temporary adhesive (sometimes just starch and water) to hold the beads onto the threads. When the threads are engaged, the beads break and the epoxy begins to cure. It's an epoxy formulation that seems to work well even on oily surfaces. But they're not trying for really strong bonds, just the kind you get with acrylic Loctite thread-locker.

I think the reason the loctite did not cure is because I did not have an air tight seal between the pins and plate. I'm keeping my fingers crossed on the epoxy (3M DP-100 two part expoy is what I had in hand and used)

Well, good luck. If I remember the description of your job, it only requires shear strength. DP100 should be Ok for that. It's made for use on a wide variety of materials so it's probably somewhat flexible

-- generally a good thing. Like most fast-cure epoxies, it's not particularly strong or water-resistant. But it's probably all you need.

'Hope it holds.

BTW... I have NO idea where that number came from...(duh!)

It's #242 removable!

LLoyd