You just cant teach an old dog new tricks

[ciladog]

For little over a year, Doc and I have been posting about a pneumatic hammer control that goes beyond what has been built in offering much more versatility and control. And yet with all the guys out there building hammers, it seems that none of them can think out of the box and come up with something that is an improvement over a simple linear arrangement of control valves. If you are capable of building a hammer in the first place, you are capable of building anything to control it. Imagination and ingenuity leads to evolution and it’s time to move on.

So I challenge you to think out of the box and come up with new and better controls and not just make your hammers go bam bam because it’s the easy thing to do.

A few months ago I posted a video on how the control on my hammer works. That video is linked here at the end of the post. I posted it to show an alternative to the usual linear controls of most of the hammers out there including the “commercially” made hammers. I’m sure the principal can be adapted to almost any hammer. But why stop there. Figure out something new and innovative.

I’m not taking credit for coming up with the initial design of the controls on my hammer, which goes to Thomas Troszak, maker of the Bull and the Phoenix. But I took it a few steps farther to make it more controllable and versatile. I don’t make hammers for sale, I don’t sell plans or anything else and I have no vested interest in anything you make but let’s get creative.

Now I know that what I have built can seem a bit confusing to some from emails I have received so I posted a video showing a short hammer demo. What’s happening inside the control box can be figured out from the second video.

Once the controls are set on my hammer for the type of forging I want to do, the stroke, force, and speed are controlled with the treadle just like the big self-contained hammers but even better. There are infinite possibilities that I haven’t even figured out yet.

Anyone with hammer experience that would like to come and try this hammer is welcome just PM me and we’ll set it up. Take note Sam. It’s located in northern NJ.

[doc]

Great demo of the versatility of your hammer controls. You've got many more choices than with a simple linear switched hammer.

[Ferguson]

I just looked at both videos right now. I had looked at the control video before.

I confess that I did not find your explanation very clear.

Let me see state what I think that I know, including some things about which I am in doubt.

The upper end of the control arm is coupled to the head position.

The lower end of the control arm is coupled to the treadle position.

You are using a two position, two pilot control valve, without a spring return.

When you push on the treadle you are both increasing the travel and changing the "center" position. (I will define the center position as the head position that is halfway between the two microswitch operation points). (By the way, I would see that as the primary advantage of the control arm system that you use).

The treadle is also coupled to an exhaust valve, as in the simple Kinyon control system.

As far as questions, I have a few.

I saw that you had some adjustment points to tune the hammer operation, but I was completely unclear on what those adjustment points affected, and how.

How did you adjust the hammer for "hard and slow"? I am asking both how you do it conceptually, as well as what adjustment causes that to happen. The concept of hard and slow seems like an oxymoron, but presumably I am missing something.

I noticed that Ron Kinyon's comments suggested that a two microswitch control system might be best suited to a hammer with a lot of head travel, more than 10 inches.

Richard

[ciladog]

Richard,

Your four assumptions are correct:
The upper end of the control arm is coupled to the head position.
The lower end of the control arm is coupled to the treadle position.
You are using a two position, two pilot control valve, without a spring return.
The treadle is also coupled to an exhaust valve.

There are adjustments inside the control box mounted on the moving plates that are spring loaded for the initial timing of the hammer. Let’s not talk about them right now.

There are three adjustment points that you see in the video. The first is where the control rod attaches to the ram. That changes the steepness of the rod. The second and third are at the top of the control box and they adjust the positions of the pilot valves relative to each other and to the control rod.

For light forging, I set the control rod at a less steep angle which makes the stroke shorter. As I depress the treadle, the stroke moves down the column. For heavy forging, I make the rod more steep which lengths the stroke.

There are a few things at play in the dwell of this hammer that are difficult to explain like the ‘slow and hard’ setting but you can see in the video that it is possible with a full open throttle. I found that there are some quirky properties of pilot and spool valves.

What happens if you have both pilot valves pressurized at the same time? Well, at first nothing happens. The valve that was pressurized first controls the spool. But as I depress the treadle, one pilot valve (the controlling valve) begins to transition from presser to exhaust and the spool moves to the center so as one side of the cylinder is exhausting and the other side is pressurizing very slowly (relatively speaking) until the ram has moved to the bottom of the stroke and then wham.

[Ferguson]

Ok, so my assumptions were correct. I just noticed that the control arm is actually a telescoping arm.

Setting aside the question of the mechanism that you use to get "slow and hard", what is happening there? Is the head going down rapidly but then up slowly? The control valve is not a proportional device, it is a two-position device, so hard to see how that could be done, without introducing check valves and metering valves. I understand that the control valve position only changes if the two sides of the pilot have different pressures. Similarly, the microswitches are also two position devices, not proportional controls.

Putting on my engineering hat, any system using a two position control valve inevitably has a "bang-bang" control system, with some kind of hysteresis. In this context, hysteresis means that operation point in one direction is not the same as the operation point in the other direction. The furnace thermostat in your house has hysteresis, it might come on at 65 degrees, but not turn off until 67 degrees, a two degree hysteresis. Even with a single microswitch, the microswitch has internal hysteresis. Hysteresis is not necessarily a bad thing. Usually, some hysteresis is desired to reduce instability. As far as I can see, a system with two microswitches would allow a larger band of Hysteresis, and would allow adjustment of the hysteresis. If the two switches do not operate at the same time, that means that the head is driven in one direction longer before the valve reverses position. In fact, with a two pilot control valve, it does not matter if both sides are zero pressure or both sides are the same positive pressure, the result is the same, the valve stays where it was. If you had a simple control system, with two fixed microswitches, it would not matter which one was above the other; if the two microswitches do not operate at the same time, that increases the hysteresis, and the head travel.

There are a few things at play in the dwell of this hammer that are difficult to explain like the ‘slow and hard’ setting but you can see in the video that it is possible with a full open throttle. I found that there are some quirky properties of pilot and spool valves.

The mechanism that you have designed is obviously functional, but is somewhat complex. My concern would be that the more adjustments that something has, the harder it is to understand and to adjust, and the more likely that the user could mess it up by setting one control to the wrong position, and then adjusting another control to compensate, etc., ending up very far from the optimum solution. This is often a problem with commercial machines, where at least you know that the machine can be adjusted to operate correctly. This could be a bigger problem with home made machines, where the design and the adjustments might be such that it would be impossible to adjust it to operate correctly.

Richard

[Spears]

Hello Ciladog,

I like your inspiration for something new and innovative. I also like your desire for people to get involved and use their thinkers. However, you may be stuck in a world like I am. "If I want something done, it won't get done unless I do it myself". I hope some people with pneumatics knowledge jump in and help you out. Good luck. Spears.

[ciladog]

Richard,

By golly I think you are beginning to understand this. Your response and explanation are a breath of fresh air.

Hysteresis, yes. “If the two switches do not operate at the same time that means that the head is driven in one direction longer before the valve reverses position.”

Actually Richard the mechanism is not complex. It is very simple but how it functions can be complex to explain for some of the settings. There are many things to take into account if you want to explain how all the variable dwell settings are accomplished like inertia and pneumatic pressure rebound when a valve suddenly closes and a 120 lb ram changes direction in an instant. But the basic function is quite simple.

Those spring loaded adjustments in the control box that I mentioned in the previous post prevent screwing up the hammer if you adjust something wrong and the air cushion in the cylinder also are a sort of fail safe. The bottom line is the hammer works and the possibilities are endless although not necessary once the hammer does what you want it to do. None of that is possible with a linear valve system. Without some connection between the treadle and the control valves all you have is Fred Flintstone’s son (Bam Bam). ;)

[Ferguson]

I can see the advantage of linking the position of the microswitches to the foot pedal. Is it the Bull controls that use one microswitch, whose position is linked to the foot pedal? In some cases, that would allow a "clamp" function, if you hold the pedal down to the bottom. Not sure how useful the clamp position might be, seems like you could lose all the heat in the work while clamped. Perhaps more importantly, the hammer would tend to have a wider opening, especially at partial throttle, which could be very useful. I would expect that the head would go almost all the way up if you took your foot off the pedal.

I can also see the advantage of some kind of adjustment to the microswitch position, maybe move the switch up when using tooling, for example. Or maybe just to allow some tuning.

If one would combine the adjustable microswitch positon with the linking of the microswitch to the foot pedal, I would think that this would provide quite a bit of performance and flexibility. For light taps you only push the pedal down until the head just touches the work. The head travel would increase the further that you push the pedal down. Properly adjusted, the head upper position might stay roughly constant, while the head lower position would be proportional to pedal position. Without tying the microswitches to the foot pedal, the upper position of the head would be near the center if the pedal was only pushed down partway. The upper position of the head would only approach the upper limit when the pedal was more or less wide open, since the microswitch position would need to be near the center of the travel to avoid destroying the hammer. For heavier hitting you push further, and the head would be moving fast when it hits the work, especially since the throttle valve would be open further.

Adding the two microswitches could increase the speed of the head when it impacts the work, because that would increase the head travel, giving the head more time to accelerate before the valve reverses. This would mean that the head would be powered almost all the way down, instead of reversing the control valve earlier, which would decelerate the head before the hit.

For the absolute maximum hit, one would want the lower microswitch to reverse the control valve right before the head contacts the work. The upper microswitch would be set primarily to protect the hammer head from traveling too high, potentially damaging the hammer.

But there is something missing with this analysis. This relates to the spring effect of the air in the cylinder. Note that some people have mentioned the advantage of having a check valve between the air supply and the control valve. (I think that is where they put the check valve). That check valve converts the air in the cylinder to a spring, providing additional force to reverse the direction of the head. Think about it this way, assuming a nominal air pressure of 100 psi. If there is no check valve, the maximum force to reverse the head direction is 100 psi times the cylinder area. (disregarding any pressure drop through the control valve). Air could potentially get pumped back from the cylinder into the supply line, a waste of energy. If you add a check valve, the pressure in the cylinder would probably go over 100 psi when the head direction reverses, providing additional force to accelerate the head, at least as long as the pressure remains over 100 psi. One possible concern would be the pressure rating of the control valve; I am not sure if the commonly used control valves are rated to operate above 100 psi. Not sure how high the pressure could go; One might need to figure out a way to measure the peak pressure. The air spring effect might also tend to act as a cushion to protect the cylinder and the hammer from high forces if the head hits a mechanical stop in the up direction.

The spring effect would also increase the number of hits per minute, by speeding up the reversal of the head direction. After all, the work that the hammer can do is not just a function of the strength of the hit, but also a function of the number of hits per minute. If you had really strong hits, but only 60 hits per minute, that might not really be the best approach. You might be better off with slightly softer hits, but 120 hits per minute. A higher hit rate, but lower hit strength, might provide more control, even providing a planishing effect. To achieve a higher hit rate, one would presumably want to reduce the maximum travel by setting the two microswitches to operate at the same time.

Note that the new Kinyon design adds a leaf spring to the system, which would add an additional spring effect to the system.

Many helve hammers depend greatly on the spring coupling the eccentric (driven by the motor) to the arm. Many mechanical blacksmith hammers have a spring as well. (The little Giant hammers come to mind). I wrote an article about tuning helve hammers, pointing out that the resonant frequency of the head/spring system needed to match the maximum rpm of the eccentric, to achieve maximum hit strength.

Tuning helve hammers

This discussion thread is useful insofar as it forces both of us to think through how the pneumatics really work, and how they should work.

Richard

[ciladog]

This hammer is amazing. INCREDIBLE. I got the chance to run it and I have never seen a hammer with more control, and more options to adjust the type of blow it delivers. I have run Say Maks, Anyangs, bulls, Ironkiss and LG's and not seen it's equal in the ability to pound the snot out of steel or kiss it lightly, or for the control over the stroke. it looks complicated but watch the videos and think about it, it's the best limit switch hammer configuration I have ever seen, dare I say, unlimited?

Thanks Sam, it was a pleasure having you at the shop and you come back anytime.

[Steve Shimanek]

I wish i was closer to NJ so I could see this in person. My mentor has a Bull that is trying to destroy itself when the head goes up; I would love to modify his hammer with your adjustments. Excellent work! Any chance you will make fabrication instructions available? Aloha from Kauai, Steve

[ciladog]

I wish i was closer to NJ so I could see this in person. My mentor has a Bull that is trying to destroy itself when the head goes up; I would love to modify his hammer with your adjustments. Excellent work! Any chance you will make fabrication instructions available? Aloha from Kauai, Steve

The fabrications instructions are free on this website and you can PM me with questions. The reason the hammer slames up is that the throttle valve is not adjusted correctly.