diy hammer seems weak

[JamesBBrauer]

 

I've been collecting parts since 2009, and finally got around to welding up a new air hammer.  The anvil is two 2"x3" solid rectangles welded in a T, with 6"x6"x1" plates holding hand ground tool steel dies.  The hammer portion weighs in at around 25 lbs.  I'm running a Mac 4-way valve with 24v dc pilot that I coded and wired to an Arduino microcontroller to get a variable stroke length with a knob, then a foot switch to start hammering.  The air cylinder is a 5" bore x 6" stroke run to my 5hp two-stage air compressor with some 1/2" hose and fittings straight off the tank.  I'm spec'ing all this out, cuz it doesn't work all that well, and I could use some advice from somebody who has built one that works.

 

I hook it all up, and it makes lots of noise, but doesn't do much damage to an orange heat 3/8" rod.  It sort of flattens it, but in one heat I can do more work with a hammer than this machine.  To troubleshoot, I want to try hand hammering on the anvil, but I suspect that isn't the problem.  I think I need a longer cylinder, or maybe more mass on the hammer head.  I've seen longer cylinders in most of the photos I used as a reference, but I don't get the difference a longer one would make.  It seems like anything over about a 2" stroke, and this thing is hitting as hard as it is going to. 

 

- James B

[MOblacksmith0530]

I have not "built" one myself but have extensive experience with air cylinders in machinery and I would say your cylinder is too big. It will take too much air/time to exhaust one end of the cylinder and fill the other end and it is just cycling without force. I have seen hammers with 1.5 inch cylinders with an 8 inch stroke that will really move the metal. I dont understand the "controller" reference you state so I don't know what effect it will have on your hammer. Air flow is king! you need to move volumes of air quickly to make a hammer work so think about that. The only other thing I would mention is to make sure you don't have a friction binding somewhere.

[macbruce]

A 200 lb chambersburg utility for instance needs 70-80 cubic ft. per minute at 90 psi and has a 5'' bore with a 16'' stroke as near as I can recall. A good 5hp compressor puts out 17.5 cfm. If you were to lengthen the stroke it would run even more sluggish. I reckon a 2''-2 1/2'' cyl with a 10''-12'' stroke is what you need.

[doc]

I agree with Macbruce and I also believe your correct in thinking you need more mass on your tup.

[JamesBBrauer]

Thanks for the input. 

 

The CFM requirements for that chambersburg is really good info.  If I reduce my cylinder volume by about 2/3s, with a 2" cyl, I'll get closer to the air:volume ratio of a store-bought hammer, and hopefully to be able to up the airflow and move faster.  I didn' t really know what that ratio was till reading this.  It seems like the first stroke hits harder, so the notion of cycling without doing much work makes good sense.  After I get a new cyl installed, I might look at how much room I have to add mass.  I got that 5" fatty for $10 bucks, but maybe it wasn't such a good deal for what I am trying to do.

 

And the digital controller is similar to a PLC.  It reads two potentiomenters, and uses the numbers control the on/off time for the pilot valve.  The one pot controls the stroke length, and the other pot is a fine adjustment to make the down stroke a couple hundred micro-seconds longer than the up cycle, so the shortened stroke keeps the hamer striking the anvil, not oscillitaing the hammer in the air at the top of the stroke.  It is switching a Mac 4-way valve on and off with a single signal.

 

- James B

[Rich Hale]

I have not built any hammer at all but will share thougths about a lot of them I hav eseen on here: Air and mechanical.. ALmos tall of them are nto buitl strong enough. My lg 50  weighs close to 1800 lbs.  Many of the hammer pics I have seen  on here in pics have steel sq tube frames.. Home built hammers that seem to work well have really thick solid base plates,  I have noted some have plates half inch or maybe a little thicker.

it would tme like half inch lines are wayyy to small for this. You did not say how long they are but that may also limit flow.  I have used a lot of hand air tools and if the hose is small or long it takes a lot away from the tool. the fitting sizes are a big part of that also...the air flow through all valving may need a close look.

[mike-hr]

You said the cyinder in use has a six inch stroke. It's hard to tell from your picture, but it looks like you have at least six inches of clear space, leading me to think it ain't hitting hard, because it's bottomed out before it hits the anvil...

[ptree]

I have not built an air hammer built have a 30+ year career with machines and lots of valve experience.

Those 1/2" hoses go thru a hose barb fitting? check the ID if is probably too small.

You don't mention the size of the MAC valve, but it too os probably too small for a 5" cylinder.

 

A hint: look at the air cylinder you intend to use for any project, and see what is the industry standard port size for that bore. So... If the cylinder has 1/2" pipe ports, then you need for everything in the system to be be AT LEAST that big.

 

Want to see if the cylinder is choking for air or binding up by exhaust not getting out? Take the lines off and direct pipe to the down port and use a ball valve, air on it hits hard? got an answer.

 

Most smaller spool valves have Cv factors far smaller than one would expect. Get a catalog and study Cv and what it means and learn about "Full Port" fittings.

 

If exhaust is the only ting binding you up a pair of quick exhaust valces at the cylinder ports will make for scarey fast.

 

Last but not least you need to study "Response time" on that valve. Every valve has a time to shift. While these are electically operated by solenoid, the solenoid is simply a small valve that allows air to then flow to a piston that pushes the spool. Back in the '70s I measured response time on new pnuematic valves in an R&D lab, and the time to send 100psig thru 100' of hard nylon tube of the correct size for the valve was in the 250 to 350 millisecond range for the very fastest valves much longer for more normal.

[macbruce]

Thanks for the input. 

 

The CFM requirements for that chambersburg is really good info.  If I reduce my cylinder volume by about 2/3s, with a 2" cyl, I'll get closer to the air:volume ratio of a store-bought hammer, and hopefully to be able to up the airflow and move faster.

 

2/3???  You missed it by alot. A 2'' cylinder is 3.14 cubic inches or 16% and a 5'' is 19.62 ci or 84%, far closer to 1/5...............There's plenty of tried and true designs and folks ready and willing to help but you've gotten off to a bad start. I think what you've done can be salvaged but don't hurry up and do it three or four times, there's too much knowledge here for that....... :)

[Frosty]

Mac's already posted a correction regarding your piston area. Still, you don't need anything like that much sq/in to move a 25lb. hammer smartly, even 2" is a little overkill.

 

Looking at the pic I'd have to say the wimpyness stems from not having enough mass in your anvil. The underweight ratio is 10:1 making for a 250lb anvil. And that's not counting the base plate unless it's really rigid thick and really solidly welded. The more efficient (IMHO) ratio calls for 20:1 making for a 500lb anvil. Take a look at a 25lb. Little Giant, their anvils are in the 500lb. range.

 

Running your supply lines a LITTLE over size compared to the cylinder ports is a good thing but too over sized and you'll  lose energy to compressing an unneeded quantity of air in the lines. Pressure will drop as it passes through the control valves and recompress at the cylinder ports, a little is good as it acts as a reserve but too much just sucks up energy. A LITTLE is good, a LOT is not.

 

Frosty The Lucky.

[metalmangeler]

Frosty, I don't understand how having your line huge could do any harm. It may not help if you have a bottle neck somewhere, but it should help reduce friction if nothing else.

[JamesBBrauer]

Now Mac/Frosty, I don't mean to be critical of you efforts to help me, but isnt' the volume of a cylinder the area of the piston times the stroke length?  118 cubic inches for a 5" cyl with 6" stoke less 2/3 of that size is pretty darn close (40 )  to the 38 cubic inch volume of a 2" cyl with a 12" stroke. 

 

And I learned the hard way, the part about using oversized air lines.  I plumbed 3/4" copper down to my wood shop, and the long length of pipe acts like a small air tank:  for anything without a regulator on the end, the first air is really high pressure, then you have to wait for the compressor to fill the line back up to get a consistent output. 

 

I actually got a new 2" cyl with a 12" stroke today, so hopefully I can get it pounding harder.  After reading ptree's comments, I'm pretty sure there was too much back pressure on the exhaust coming through the valve.  I couldn't find the Cv for the valve, but I did see how it effects the reduced volume of air that can go past the spool as the pressure goes up.  And this cyl/valve combo was trying to move a large volume of air at 130 psi.  If the new cylinder is still choking with this same valve, I might look at leaving the down stroke port open, and using a spring to return, or try a fast exhaust port.  I never knew such a thing existed.

 

Everything in the system (ports, barbs, fittings) is at least 1/2", but I did have a pretty good 20ft run of 1/2" hose.  My new 2" cylinder has 1/4" ports, so hopefully air supply and exhaust won't be a limiting factor.         

 

And my anvil is about 200 lbs if you count the die and top plates.  More would be better, but I'm hoping that this will get me by.  

 

Anyway, thanks of all the good advice so far.  I'll post an update when I get the new cylinder put in.

[macbruce]

Ok, 3.14 square inches x 12= 37.68 cubic inches......19.62 sq inches x 6 = 117.72 CI or 37.68/117.72......Rounded off to say 118 divided by 20 you get almost 1/6.......The 2'' cyl with a 12'' stroke is a way closer choice but with a light 25 lb ram it's going to be very snappy, the opposit of what you had but the smallish ports may mitigate that a little and you can turn down the air pressure. Better more snappy than too sluggish.

Another thing that concerns me is you're ram guides don't look robust/heavy enough to me........Everybody is entitled to a good education, wether it be the school of hard knocks or otherwise, I should know better than most. Good luck with your project........ :)

 

Note: On my previous post I should have said square inches instead of cubic inches.

[JamesBBrauer]

Mac, that is a horrible photo so it is hard to tell, but the ram guide is pretty stout, and mostly 1/4".  But I'm going to rebuild it when I cut the cylinder off of there.  It is stout, but has too much slop for my liking, which I can adjust out, but then it binds.  I found some linear bearings in a box this week so I might try to rebuild it with those, or I have a couple scraps of that slippy white UHMW plastic I might use.  But that leads me to another question, since I'm attaching a new ram:

 

What is the best way to attach the hammer to the piston rod on the cylinder?  My hammer shaft is solid square stock, and the first try at this I welded a carefully aligned nut, the same thread as the rod, to the end of the hammer shaft, then locked them together.  Would I be better off with something like a clevis, and let the ram guide keep things aligned?  Or I was thinking of making two small square plates with a welded nut:  one that screws on the piston rod, and one welded onto the hammer shaft.  This way I bolt the corners of the plate together, and shim between the plates if the alignment is off.

[FieryFurnace]

I run a DIY air hammer, that I bought off of someone else. It's a 60 pound Kinyon style and is similar to yours. Mine runs off of a 2-inch by 13-inch cylinder at an average of 120 PSI. I am running a 7.5 HP, single phase compressor, with an 80 gallon tank. It pumps around 19 CFM. Downsizing you cylinder will help.

 

I would also recommend increasing the weight of your anvil post by quite a bit. Go for about 5-or 6-inches solid. You can sandwhich material onto what you already have in position. Increased mass equals more resistance to the hammer blows.

 

Making sure your hammer is mounted solidly is important too. If you have concrete, bolt that sucker down really good.

[macbruce]

If you use a nut welded on the ram make sure to use a jam nut above the threads on the rod and TIGHTEN that puppy.......I wouldn't use a clevis since yours is an in line set up....If you do, DON'T just slide the pin through and put a cotter pin in it, the slop will wallow the hole(s) over time.....Use a bolt and clamp the clevis to the lug.

[Francis Trez Cole]

use the bigger cylinder to compress air for the smaller cylinder you just might have a self contained air hammer

[JamesBBrauer]

use the bigger cylinder to compress air for the smaller cylinder you just might have a self contained air hammer

 

When I was about 9, I figured out that if you spun the little hobby motor backwards, it would light a little bulb.  Nobody could explain to my satisfaction why, if you attached two motor shafts and wired them together, they wouldn't perpetually spin on their own.  I tried and tried to attach the shafts with scotch tape, but never got it working.  Experiences like this are why I volunteer to judge science fairs now.

[Francis Trez Cole]

James I knew I read about it before

 

http://www.iforgeiron.com/topic/29133-a-shop-made-self-contained/?hl=%2Bself+%2Bcontained+%2Bhammer

[JamesBBrauer]

This thing is hammering metal to my satisfaction now.  Based on Mac's advice, I swapped out the existing cylinder with a smaller cylinder with a longer stroke.  Then I tightened up the bottom die, have the hammer sliding on a linear rails and bearings, and re-wrote my firmware.  I was able to pound out a foot for a traditional woodworking holdfast in one heat on 3/4 mild steel, and two heats with stainless. 

 

I'm not positive, but I swapped the supply and exhaust on my valve and it seemed to hit harder.  I wonder if those spools flow better in one direction than the other.  And I think I am still somewhat limited by the exhaust strokes.  When I dial a full stroke, it doesn't hit as hard as a 1/2 stroke.  I think the exhaust is building up in the hose between the cyl and valve, and  anything longer than 1/2 stroke has to to start pushing through the spool.  Floyd recommended a quick exhaust valve (thanks), and I may install one.  But for now, I need to spend some time hammering metal and grinding on my dies.

 

Thanks for all the advice.  I'll try to post a detailed write-up with clear photos of the build, and explain how I built the electronic controller for either dialing in reciprocating stroke lenght, or setting the foot switch as a single trip.