Tech Tip of the Month
Yet another little task not often looked forward to with any relish is that of cutting the counterbalances into the crankshaft web (con-rod end profiling, 'shaft web cut-outs... is there any task we do look forward to?) I'll offer up my reason for this reluctance. Some parts take a lot of effort to produce, and effort equals time. As the process nears completion, the time investment is large and the opportunity to double the time taken by stuffing up the one in hand reaches a maximum--especially if some final operation requires a less than ideal set-up. Cutting the crank web cut-outs is one of those as it involves an interrupted cut coupled to a problem in workholding.
The purpose of cutting a relief in the crankshaft web is to achieve static balance. It is impossible to achieve perfect dynamic balance on single cylinder engines like ours. The best we can expect is a reasonable compromise by attempting to counter-balance one-half of the reciprocating mass of the engine, namely the piston, wrist-pin, and conrod little-end [LCM]. This is generally achieved by removing metal from the web so that the crankpin will sit horizontal when a weight equal to one-half of the reciprocating mass is acting through the axis of the crankpin. The photo here (from the 5cc Sparey Project shows a shaft set on knife-edges with a donut of the correct weight slipped over the crank-pin. Sadly, even with massive cut-outs, the shaft is not within coo-ee of being balanced. At this point, one can try adding lightness to the piston, but then you run the danger of having an easily distortable piston and subsequent loss of compression. I've had compression punch the crown out of one such light weight piston--a most unnerving occurence!
Obviously, the metal must be removed from the half of the crank web that carries the crank-pin. There are a number of ways of doing this; the picture here shows five of the most common approaches. I'll describe these before looking at workholding. The names used are in no way "official". I've made them up to be descriptive (I hope). Note there are other ways. These are merely the most common.
-
[A] Crescent Cut-outs
- A large diameter cutter is used to take a bight out of the web on either side of the crank-pin. The center of the arc is arranged to minimise the amount of metal near the pin, consistent with retaining some strength in the web. The lower extent will be on, or slightly below the diameter of the web. A variant on this scheme is to drill holes in the web either side of the pin. I've even seen old designs that advocate filling these holes with aluminium. The idea being to achieve mass reduction without volume reduction. Another variant is to drill the holes on the opposite side and plug them with lead, tungsten, depleted uranium, or some other easily obtainable heavy metal.
[B] Flat Segment Cut-offs
- This is similar to [A], but does not require a large diameter cutter and offers some other options for work holding. The key point on this approach, and that of [A], is that the cut-out can extend beyond the diameter line (shown just a trifle exaggerated in the 3D rendering) to improve balance. The reason is that up to a point, the amount of metal being removed by the cut on the "heavy" side is less than the amount being removed on the "light" side, so the ratio heavy to light continues to increase until that critical point.
[C] Obtuse Angle Cut-outs
- This is a variation on [B], that retains all of the mass on the "heavy" side of the web, so instead of an arc, we have an obtuse angle. The downside is machining is more complex and you need to be careful that the intersection is not so sharp that it induces cracking. The style was used on a lot of old sparkers, and some more recent designs as well. One extreme I saw on an old design reduced the obtuse angle to a right angle!
[D] Peripheral Crescent Cut-outs
- Yet another variation on [A], except the cut-out crescents are made in the edges of the web only. A thin ring is then generally shrunk over the web diameter to seal the edge. The objective is the have the effect of a full web on crankcase volume, while achieving some degree of balance. This approach has been used on commercial engines and high performance engines. It does require a thicker web than usual.
[E] Web Thinning
- A common "full-size" practice is to bolt (or otherwise secure) a weight onto the web opposite the crank-pin. Model designers frequently achieve the same effect by machining away the web concentric with the crank-pin. The result is more mass where we need it. Some designs supplement this with [A] or [B] type cut-outs. The down side is complication of the conrod profile to avoid having it clobbered by the counter-weight.
Now onto workholding for making those cut-outs. This is usually the last operation to be performed on the shaft, so the opportunity for an accident to ruin hours of work is at a maximum. Here's how Bert Streigler setup to make type [B] cut-outs. The shaft in question is a Brown shaft and the cut-out is per this pioneer engine design. Bert says:
-
Here is a way I came up with. I don't have much in the way of measuring equipment, so I often resort to graphic set-up methods. One pic shows the little drawing, but I have since reverted to not even using that and simply do a drawing, then measure what pin is needed to get what I want. The three parts used in the set up are shown, and consist of a drawing to determine the pin size, a soft aluminum shield with a rather large cut out to allow the clamping of the crankweb without denting the inevitable radius where the web joins the shaft, and the required spacer pin.
The second pic shows the set up. All I do is press down on the shaft to be sure it is in good contact with the top of the vise jaw, then roll it slightly until it is in good contact with the crankpin and then simply clamp down on it with the vise and machine. For the other side, just loosen the vise jaw slightly, roll the shaft the other way and clamp down again. This gives a perfectly symmetrical cut on both sides. All clamping is done on the web.
Ken Croft used a similar method for the crankshaft of the MS 1.24 (we don't always use the same method):
-
I don't know how you chaps do it, but when I have to cut slices off a crank disc for balancing, I so far have only a botched method for setting up. This has involved drawing the crank disc in TCad and printing it out. I then cut out the drawing, glue it to the disc and then set it up by eye under the mill. The results is a long way from perfect and I am never satisfied.
So as it is a job I now have to do on the MS 1.2 copy that I am making, I thought about it and realised just how simple the job really is to do perfectly.
I drew the crank disc in TCad, then got TCad to take some measurements. I simply set the vertical pin to disc distance under the mill, using a flat ended probe and the downfeed register [I have a cheapo DRO on the mill].
I then end milled the required amount from the disc. I then rotated the disc and again set the pin to disc dimension the same as before, and removed an identical amount from the disc. The result is a perfectly symmetrical disc to exactly the required dimensions. Why did I not do it this way already?
Now I need to work out a simple method for scalloped cut-outs. That will not be so easy.
Bert and Ken have used two different measurement techniques to assure symetry, but their basic approach is the same. We all agree that gripping the web in a mill vise is not without risk, especially if the crank is small. What goes unsaid between experienced model engineers is that the cutter should be sharp, the speed high, and the cut made so the teeth are meeting the work offset to one side to produce "conventional" milling (not "climb" milling). Even so accidents can happen to the best of us.
Clamping on the shaft affords the opportunity for a more secure set-up, but requires care not to injure what is probably a finely finished surface. Since cutter and shaft axes will be in-line using this setup, the cutter will be attempting to turn the shaft, opening us to just as much risk as before, unless a positive anti-rotation aide can be employed. Here's how David Owen made the segment cut-offs for the run of GB5's he and Gordon Burford produced.
-
I generally draw one-offs up on the crankdisc itself as you have done. Then I hold them in a 5c collet in the machine vyce. Level them up agianst the line and register the pin height above the table with my vernier calipers. The actual dimension is meaningless. I trim down to the line as required and set the vertical stop. Then I turn the crank around until the pin is at the same setting on the other side, and cut down to the stop. I have used Bert's holding method but dislike it for the same reasons as your self.
On most cranks where the cut is a straight line diverging from the centre, removing metal below the centreline will result in greater loss above the centre, which is what we want. I've got the maths somewhere for this, showing the limits.
On cranks which have a scallop, or parallel cuts off the flank, the diameter is the limit. On this sort of crank, I mount in the 5c collet again, but with the shaft vertical.
On production shafts I have always drilled an 1/8" hole behind the crankpin as a register. You will see the three 1/8" dowel holes in the attached 5c image, which shows the GB 5cc shaft. The centre hole locates the shaft for roughing out the crankpin from the original full depth disc.
So I end up with a square crankpin about 1mm larger than the finished diameter and with about .25mm to come off the crankdisc face. All shafts are completed to this stage. Then the dowel is moved to a side hole and and one flank machined away from each shaft.
The dowel is shifted to the other pin and the second flank is cut. The scruffy looking piece of shim keeps most of the swarf out of the vertical collet. The shafts are then chucked in the large 5c offset fixture in the Hardinge and the crankpin is turned to plus 0.2mm and the face cleaned up at zero. Then the shafts are hardened and finally ground.
The 3C collets used the way David has, are the bees knees for this job. The collet is keyed into the closer, so it can't rotate. Then the shaft is keyed to the collet with a 1/8" pin in shear (maybe hardened), so the shaft is not going to rotate, hence no teeth-clenching required. For one-off shafts, drilling a collet for the pin is probably out, but a correct sized collet exerts tremendous clamping force on the work, so this is still the best approach. By the way, look at the pin in the center of the offset crank turning fixture shown in the second photo. This will also prevent unwanted rotation of the shaft during crank-pin turning. Well worthwhile, especially if you're making 150 of them by hand!
I generally make an offset fixture to hold the shaft for crank-pin turning. This fixture can double-up to hold the shaft while machining away the web (although 3C collets in a substantial holder would be far nicer, if I had any). The photo here shows the fixture made for the AHC Diesel that was re-used for the Sparey 5cc diesel. In this case, the shaft is secured by tightening up the prop nut, but a pin like David used with this 3C fixture would be good insurance. The holding fixture is gripped in the 3 jaw chuck, screwed to the rotary table mounted horizontally under the mill. It could also be held in the mill vise using a V-block.
The crankdisk is marked for the cut-out using a template like Bert's. In the case of the AHC and Sparey, a 3/4" end-mill was the largest available, so that sets the crescent radius. The cutter is positioned to touch as much of the marked line as possible while stationary and the mill axes zeroed. I then "nibble" away the web from the outside in using the down-feed, moving one mill axis between cuts, and advancing no more than 0.030" at at time, with tightly clenched teeth. When the "zero" point is reached, the quill is locked down, then the sides of the cut are cleaned up by winding along one axis from zero, then back to zero and winding the other handle. A sharp cutter will do this smoothly. A dull one will shudder and bump, and try to snatch the work. Even a sharp cutter will raise a light burr at the edge of the cut. This is dressed off with a small gringing stone in the Dremel hand-tool--taking great case not to cause a run-away that hits the crank-pin (more teeth clenching required).
Examination of a *lot* of engines and drawings in the "sport" class shows that below 2cc (0.15 cuin), most experienced designers do not bother to cut away the web. My own experience supports this. The amount of vibration experienced by small engines is not great, and theory suggests that the lower crankcase volume resulting from a full web will make the engine a better fuel pump--though given the ability of small engines to flood up, this is not often a major concern.
References:
Ageless Engines Has Moved
Editorial
)
