Designing a Rolling Mill

[Gundog48]

After playing around with the idea of getting a power hammer or press, I've decided that a far more sensible idea would be to make a rolling mill, as it would be quieter and smaller than the alternatives, and would actually do what I need it to better. I'll mainly be drawing out and tapering knives and swords which this should be most useful in doing.

I've only just started working out how I'll achieve this, so I drew my idea of the top roller today. Drawing something properly makes it easier to understand how you'll do something practically. I was having problems with many parts of the top roller design, but doing this has cleared much of it up. I'll be using half shaft for my rollers, which if I'm lucky, will be 4140 which would be ideal for rollers, if not, mild steel would do it just fine anyway! I'll be able to get access to a lathe after talking to my tech teacher at school, there's no way I'd be able to do this without. I'll be taking the half shaft down to 3/4" at the ends to form the shaft, and boring out a 3/4" hole in some 1.1/4" 660 bronze rod for the bearing. Doing this answered many problems, but threw up some more, so I'd like to ask these questions:

• The Shear Pin, I put the shear pin in to prevent damage to the rest of the drive train if it got stuck, any idea what materal or size it should be? It needs to take the strain of normal operation, but break if it jams
• The drive train- obviously it needs to be geared down, what is the... cheapest way of doing this? I was thinking of using a very small sprocket on the motor shaft and a much bigger one on the top roller (I'll need to do the RPM calculations) Would this principle work?
• Bearing- I have joined the bronze to an aluminum block with pins, is there another way of joining it? All I really have is a MIG welder. Do you think my current setup is up to it? And will the aluminum block be hard enough. Aluminum is more available to me and easier to work with.
• The Screw Jack- I need to do a lot of research on this, but quite frankly I have no idea how it works. I've never used one before, nor do I know how they work. Could someone explain the basics, or how I would operate it to perform a taper, for example?
• The Design- Please point out any other problems in my design, I'm a student that has never manufactured anything this heavy before, please point out any problems you see, it's probably riddled! I noticed my mistake with the bearing blocks, I'll be mounting them horizontally instead so they don't get in the way of the bottom roller.
  

As you can see, my drawing is awful, and my handwriting much worse than that! Hopefully you can get the idea of what I'm getting at.

[Wayne]

Some where on the web there are details from an Australian smith who has already designed a set of rollers, it may be worth checking them out

[Wayne]

This is the chap

Link removed at the request of that sites owner. He does not want linking to his site.

[Gundog48]

Thanks Wayne, I'll be getting a copy of that soon. No need to reinvent the wheel, although this section will likely change from the design based on what I have to work with!

[george m.]

Based on my experience of 45 years ago when working in the alloy bar mill at US Steel's South Works in Chicago the one thing that you absolutely want is an easily accessible kill switch. If something unplanned happens you want to be able to shut things down FAST. I suggest some sort of big red button or lever switch. If it were me I'd think long and hard about guards for the moving bits.

Safely,
George M.

[Robert Yates]

that is why many wheel rollers have foot switches so they can be controlled and stopped by moving ones foot off the foot bar and the machine stops immediately .

SAM

[BIGGUNDOCTOR]

Have you looked into just buying a used rolling mill? There is a ton of used equipment on the market. I averaged 10 cents on the dollar for my machine shop equipment when I had my machine and fab shop. Some stuff is going for scrap prices due to the low/no demand here in the states. I know one guy who hauled 800,000# of equipment to the scrap yard in the 90's due to no market.

[Daniel.85]

Have you looked into the mcdonald rolling mill plans? just google it.

[JNewman]

I am pretty sure the McDonald plans have the stock feed towards you. This is MUCH safer than rolls that pull away from you. I just read on here or the NWBA forum recently about someone having a set of tongs run through their rolls, if he had been holding a cold end of the stock directly that easily could have been his arm. If you you look at all the rolls on the Bulldog spade videos any where the smith is holding the stock it is fed towards him. This can be done with movable rolls like the McDonald rolls or D shaped rolls like the ones in the Bulldog videos.

About 20 years ago I worked on the repair of a printing press where a workers arm was pulled into the press. The rollers pulled the flesh off his arm right to and including the shoulder. This was working in a large shop with others around to hit the Estop and call 911. In a typical single person blacksmith shop an injury like that would likely be a fatality.

A power hammer seems like a dangerous machine because of the violence of the blows but machines like powered presses and rolls can be equally or MORE dangerous.

[Gundog48]

This one is designed to feed it towards you, as much for practicality's sake as safety. I've had some feedback from others about this, who have pointed out many problems. It'll be a complete redesign I think. I won't be building this for at least a year, so I'm in no hurry- I'd rather get it right!

[Mainely,Bob]

As someone with extensive experience in manufacturing and heavy fabrication I would suggest you look very closely at what type of machine you would want to be involved in an accident with. We are talking about the difference between one that mashes fingers or one that eats arms and more. We are also talking about the difference between a machine that will bite and relase ( a press/power hammer) as opposed to one that continues to drag you in even after the kill switch is hit (think inertia of moving parts).
I have seen some really fine examples of home made/modified machines and have great respect for the folks who built them but even the most safety minded among us usually stop the process short of proper guarding and other safety concerns. Sooner or later it always comes back to bite us, literally. I have a big ugly scar across my right kneecap from a Skillsaw that had it`s guard pinned back by another co-worker who failed to remove the wood wedge or tell the rest of the crew about the modification. I was lucky I didn`t lose function in that knee.
If you`re going to make your own powered rolling mill please seriously consider designing it so if something goes wrong it`ll spit you out rather than drag you in.

[Clodbuster]

The safety points and simplicity of piggybacking off someone else's proven design have been made on this, so I'll stick to answering your design questions for if you choose to continue:

The purpose of the shear pin is to be the weakest link in the system. In industry you commonly see keys, roll pins and grade 2 bolts used for this purpose. Since your shear pin is located in an area of low rpm, you can ignore the inertia of the rotating mass and size it based on the applied torque at that location - that will be the motor's maximum rated torque multiplied by the ratio of gear reduction. However, the way you laid this out doesn't leave much room for a shear pin - there's only 1/4" to work with. This puts a shear bolt around #6-32 in order to leave half the shaft diameter and that's going to be weak, plus quality varies widely on small stuff. I would recommend revising the design to have a collar on the drive sprocket slide over the 3/4" shaft. The geometry for this is much more favorable and it will be easier to dissasemble if a pin breaks and smears. If you get motor peak torque and gear reduction values I can calculate a shear pin size.

You're on track for the drive train - use sprockets to gear down. This reduces the speed and ups the torque which are both what you want. Keep in mind that what you drew shows the roller drive sprocket sitting out quite a ways from the frame - the closer you can bring it in the better for bearings and shafts.

On the blocks - Aluminum is fine for compression, but I would maybe redesign how you attach them as the bolts that hold them to the frame are horizontal and side loaded in shear. In practice your fabrication tolerances could give you fits here - the holes will have to be oversized for bolt clearance through the frame and this could let the block slip, causing your rollers to be misalligned. Also, with aluminum it will be difficult to apply enough torque to the bolts to secure it without stripping. I would recommend welding in a thrust block (piece of angle, etc.) on the outboard edges of the roller blocks to absorb that force so the bolts don't have to. Or, buy some pillow block type roller bearing/block sets and mount them to your thrust blocks.

I'm not quite sure what you're trying to do with the screw jack, I'll let others with more roller mill experience cover that.

The design looks pretty good overall, but I have some recommendations on servicability: Mount the aluminum blocks on the outboard sides of the frame so you can remove them without dissasembling the whole schebang. Also, make the hole through the frame large enough to withdraw the rollers through it when a block is off. Think about servicing the wheel bearings on your car and how the shaft end and paraphenelia are sized. Make sure you have shielding for the chain and sprockets, shielding to keep hands clear of the rollers, and I would highly recommend a deadman's switch/ foot pedal to operate this thing.

Good luck and welcome to the world of machine design!

[Frosty]

I won't go over the safety talk already written but please take it seriously, like your life and limbs depend on it.

One thing about your proposed design I really don't like is the size of the roll, it's smaller than the pillow blocks. Oh, NO pillow blocks to carry the shaft? Why the heck not. they're FAR better (IMHO) far less friction, more durable and WAY less work than bushings. Maybe if you were talking babbit bushings, MAYBE.

What I've considered is an eccentric roll, having watched them in action I frankly LOVE eccentric rolls over press rolls. how an eccentric works is the top (or bottom) roll is mounted eccentricly(SP?) on the shaft so at one point in a turn it's AWAY from the bottom (anvil) roll and 180* later it's at the minimum thickness of the piece you are making. It turns feeding material towards you but very slowly, maybe say 60-100 rpm. and there's all kinds of room on either side of the roll so you pass the hot material beside the roll and feed it in sideways. The top roll closes on it and feeds it to you and releases it as the eccentricity takes the roll AWAY from the anvil roll. You can either hold the work in place for the next smoosh or move it sideways out of the rolls.

The rolls don't need to be flat all the way across either, radiusing the edges is a good idea or the corners will damage the work. You can put a taper on part of the roll to form a bevel like a blade or it can have a raised ridge to form a fuller groove. There is literally no limit to how you can shape your rolls if you follow the rules.

Use an auto centering foot control, believe me if something goes wrong you're going to retreat from the machine and being an auto centering switch it's a deadman switch and will shut it down immediately. If you were to use a two way foot switch it'd make it really easy to back a stuck piece out of the rolls but I'd really rather just lift the top roll instead of taking any chance of me or anyone being caught.

Okay, your shaft is WAY too light, the larger the dia. the more durable it'll be regardless of what you use, bushings or bearings, they'll last lots longer. Better still it won't deflect nearly as much and believe me a 3/4" shaft is going to deflect at effective working pressures.

The way I prefer to make this kind of roller is pretty simple, just drill a hole in the roll stock the same diameter or a couple thousandths SMALLER than the shaft. Sure you'll want to broach a keyway but that's no biggy, it's harder to cut a keyway in the shaft. Sweat fitting the roll to the shaft is easy if it's only a couple thousandths interference fit, heat the roll just a couple hundred degrees and the shaft will drop in. Yeah, boiling water hot on a decent diameter roll like say 6-8" dia.

A 1" roll is going to have such a sharp bite it'll really take power or lots and lots of thin passes. A large diameter roll will move metal in a smoother press and you'll get more movement per pass for the same HP input. Larger dia. will want more down pressure but that's no biggy, if you use a house screw jack or similar. From what I've read guys aren't having such good luck using hydraulic cylinders for pressure. I don't know why I LIKE hydraulics, maybe they're not using large enough cylinders but I do NOT know.

Well, that's probably more than enough rambling for now. Making press rolls is a very duable thing but you can find yourself over your head very quickly, just HOPE it won't be while you're running it.

Frosty The Lucky.

[HWooldridge]

Frosty, how much eccentric throw are you referring to - something like 1/32 or more?

[Frosty]

There are factors that determine how much eccentricity you need. If you're forging 1/32 thick stock that would be enough. As I recall the videos I saw some were forging stock better than an inch. The shovel making video started with a hot blank being round bar about an inch in dia. and in one heat the smith forged out the shovel blade and the long wedge for the handle, took maybe 45 seconds. Then in the same heat he put it in a press to shape the blade and roll the cone for the handle, a very little hand work and it went into a big oil vat. The top roll looked flat with maybe 1/2" or larger radii on the corners but it may have been crowned and I couldn't see it.

So, I'm thinking I'd build one with 8-10 in dia. top roll and 1-2" of eccentricity. Lots of gear reduction though, 1728rpm to around 75rpm, give or take. But that's me. Large shafts and a good 6" of open space on each side for easy clear material access. I'm also thinking it may need less power if both rolls are powered, syncronized(SP?) of course.

Get you thinking Howard? I know I'm thinking about one again.

Say, when you going to make it up this way again? Deb and I'd sure like to host you again.

Frosty The Lucky.

[HWooldridge]

Ahhh, I getcha - it acts like a big fuller and you get a mechanical squeeze with each bite. Based on your description, 3" rollers might be a good start. I wonder if two sets of rollers would be any better - first pair is set to bite and pull while second pair is set at finish thickness...I don't know much about rolling mills so that may not be a feasible design method.

[Daniel.85]

I have been thinking about building a rolling mill and was wondering if it would be possible to turn small groves on the rollers with a lathe for drawing out stock. For example say I have 1/2" round and want to get it down to about 3/8, I would feed it though the specific grooves on the rollers. I know there are other ways of doing this with swages and powerhammers and such but how well do you think it would work with the rolling mill? Not sure if its already been done but Ive not seen it done.

[Steve Sells]

Daniel see the McDonald Rolling mill its all in there, He only charges about $25 for the plans. These have worked great for about 40 years so most of us dont have to bother trying to reinvent the wheel. I love mine

[ThomasPowers]

As I recall the smaller in diameter the rolls are the more HP you need. I once saw a book on rolling mill design that had the angle of reduction/HP calculations in it.

I strongly suggest you work from a proven plan rather than try to throw one together and maybe have to totally rebuild it several times to get it to work right.

[HWooldridge]

I watched the McDonald videos on Youtube and they answered a lot of questions. I also believe going off an established set of plans is the wise approach.

[dkunkler]

I agree with Steve on the Hugh McDonald plans, it's a proven design. I've built one and like it. The plans have several pages of text on the hows and whys of the development and design. Even if you want to experiment with design and do your own thing, build the McDonald mill and learn how hot steel reacts in rollers, then go from there. "Stand on the shoulders of giants"

As I recall the smaller in diameter the rolls are the more HP you need. I once saw a book on rolling mill design that had the angle of reduction/HP calculations in it.

I agree Thomas, the smaller the rolls the steeper the "climb". Larger rollers reduce this but put more metal in contact with the rolls,increasing the force on the roll bearings and more contact absorbs more heat from the metal.

[Wes C.]

See my post in the smallsword thread.
Pictures of my rolling mill and a knife made with it.

[jason0012]

Why did McDonald not want a link to his website?

 

They did not link to McD they linked to "another smithiing site" that has informed us we MUST delete any links or referance to them. Ask Jock why, his orders to delete them.

[StahlUndEisen]

Guys,

 

In my day job I'm a metallurgical engineer at a hot strip mill that rolls everything from steel to cobalt-based super alloys, so I've been banging my head against rolling problems for years.  I'd like to say something about roll size.  Both your rolling force and torque are proportional to the contact area between the roll and the metal.  So, the larger the roll, the more torque and force are required.  You wouldn't believe how fast power requirements scale up with roll size!  For instance,  a flat rolling mill with 36" diameter rolls may need 15,000 HP at 60RPM. (And still stall.)  Now, one other thing.  If you are using a McDonald-type roll gap adjustment with the foot pedal, that's all the protection your mill needs.  If you have any other type of roll gap adjustment, I'd definitely design a shear pin or something like that into the mill drive.  It's kind of tough to imagine a cobble could get real bad on a McDonald mill, but then again, it's better not to take chances because cobbles can break things that you would never think could be broken.  Also, larger rolls equal larger heat sinks, so your work piece will lose heat that much faster, and making small billets, that can mean extra heats real quick.  Now, there's also another factor you might want to include in your equation.  That's roll bite angle.  For the most part, it's hard to feed steel into a mill if the roll bite angle gets bigger than 15 degrees.  Advantage:  big rolls means it's easier to take a bigger bite.  However, you've got two ways around this problem.  One is you could knurl your work rolls, or welding beads all the way around the roll would accomplish the same thing.  That extra texture on the roll face will draw the work piece in even if the usual roll bite angle limit is exceeded.  The other way is just to taper the end of your work piece.  Once the tapered end feeds into the mill, the friction will pull the thicker part in.  In the case of a McDonald-type mill, if you're trying to take a draught that's too large for your roll size (not likely), you could slightly fuller the part of the billet right by the handle, and bite down on that part.  I think the McDonald mill is a great practical design for a mill that one man can operate efficiently. 

[Steve Sells]

some nice things about the McDonald mill that solved some of the issues you posted.

 

1)   It does not operate the same way the large mills roll steel. There is no feeding in the traditional sense.   Just open the rollers, insert the hot steel, close and pull.  Traction traction is no issue,  Also  the texture would ruin pattern welded billets, but may create other patterns. 

2) There is no need for sheer pins.  The belt drive in the first reduction works as a clutch, it will slip before great jam and break the machine.

3)  No need to taper work.  This mill is not working the same as the large units which is a big reason it works so well for us.

4)  There is no need to make any changes in the McDponald Mill.  It works just fine as it is in the plans.

Most failures are from people making changes before they use one.  He mentions that in the plans also, along with a few examples. The only thing I wish I did differently is have 2 rounding tools cut into the roller off to one side. Not to interfere with normal rolling operations. One for 1/4 inch ribs and the other for 1/2 inch rounds. And I may add that some day.