Needle Seats
I've been thinking about needle valves and spray bars as I seem to have had to make several over the past month for my own engines and some restorations for friends. When I'm doing a replacement for an old engine, I try to maintain the external appearence, but have now settled on an "internal" design and method of manufacture that gives good, consistant results. Before this, I had made some that worked, but were difficult to adjust with any predictable results--and some that caused the engine to stop if you so much as touched them, though the engine could be restarted at the same setting that stopped it!
For the needle itself, I use hard, so-called "music wire"--the sort sold by hobby stores. Real music wire comes in coils and would be just about useless for our purposes. The needle is formed off-hand, first against a grinding wheel, then a sanding disk (on the other end of my bench grinder), then on a Scotch-Brite (tm) buffing belt (same end of bench grinder). The needle looks good, but is not a precision taper; ie, the point is almost certainly not on the axis of the wire, and the taper sholder will not be diametral--rather some kind of crazy ellipse. Used with a straight-thru hole in the spray-bar, such a needle would give poor results as mentioned above, but used with a seat, it's just fine.
The TurboCAD 3D cut-aways below show the arrangement required. Let's say the wire is 1/16" OD. First I turn the spraybar to shape on the side the will be threaded, then form the thread with a die. Always do this before drilling for the wire as the drag of the die is considerable and could distort the brass if it were already cored. Next drill the brass spraybar with a number drill that is just less than 1/16" to a point just beyond where the jet hole will be. Follow up with a #60 drill in the same hole for 1/8" to 1/4" of additional depth. The #60 bit will be centered by the point left from the 1/16" bit. At this stage, the needle wire should not fit in the hole.
In all these drilling operations, "peck" out the hole. I have a sash brush in my left hand and advance the tailstock with my right. Drill into the work about 1/16", withdraw completely, brush off the chips, and peck in again. If the drill ever squeeks, the flutes are loading up with chips and the drill is about to break, if it hasn't already. The peck method does not take long and keeps the core central. The internal finish will be cleaned up at the next step.
Make a D-bit reamer from the same music wire stock as the needle by grinding away 1/2 the diameter for about 1/8". The end should not be square, but raked about 5 degrees so normal clockwise rotation has the "point" leading into the hole. Use this to ream the last couple of thousandths from the hole down to the seat area, pecking away as before. The wire should now be a close sliding fit with no air leaks.
With the spraybar still on the stock, transfer to the drill press and cross drill #60, one side only, to form the jet. The fuel tube side of the bar can now be formed and center drilled with a fine #1 center drill, followed by a 1/16" bit to meet up with the #60 axial hole so that the seat passage is maybe 1/16" long.
The idea is that all fuel metering occurs around the seat and the point. If the point is a bit eccentric, it does not matter. Down-stream, the needle sholder never comes near the spraybar jet, so an elliptical needle sholder does not matter either. The close, reamed fit of the needle prevents air being sucked in from the thimble side and you have a needle that responds well and consistantly. Of course, we are talking sport class engines only here. High performance stuff is another matter, but the general idea of metering at a sholder rather than the jet still applies.
Some good observations there and a curious bit of venacular usage. Today, "bug" has entered the language as a general term indicating a problem, thanks to the ubiqutious nature of computers in our society and the problems we encounter daily with unexpected behaviour in our software. This usage of the word is attributed to Grace Hooper (later Rear Admiral Hooper), who discovered a dead moth in the contacts of a relay in the US Navy's Mark II computer at Havard University. She entered the term bug (and the dead critter) in the log and the rest is history. This was 1946; Roy Clough's article was published in 1950. I have no idea if he worked in the field, or if the term came to him from elsewhere. That it passed the editor of MAN as a term that readers would understand says it must have been in common usage by then. Amazing.
