Precision diameters

[Howard LewisParticipant @howardlewis46836]

The clearances and dimensions being quoted here are those required for full scale diesel fuel injection equipment, (operating , currently 1500 bar and beyond, with Ra of 1 – 2 microinch CLA). To be measured accurately, and consistently, the workpiece and the measuring equipment need to be "soaked" in an environment where temperature and humidity are VERY closely controlled. This is usually the Calibration area within the already closely controlled atmosphere of the Standards Room.

From what I have been told, minute "cuts" with tipped tools are closer to burnishing than cutting. (Look at the tip under high magnification, the edge is relatively blunt)

At the speeds at which these engines are likely to run, leakage will be minimal. Leakage, for a given clearance, is a function of pressure and time. (The pressures will be there, not as high as with F.I.E, but with very little time, even on the two stroke cycle). On full size engines, the fit of the piston rings to the grooves has a great effect on blowby, as does the ring gap. The gap needs to be minimal, but without ring ends butting, to prevent ring breakage and bore damage. Below 3,000 rpm, about 1% of the theoretical 100% volumetric efficiency air intake of the engine, is about as good as you can get, for blowby.

The power decrease noted with lapped piston/bore combinations relative to machined combinations may well be down to increased friction, as the finer finish will result in greater contact areas. (The contact will be on the peaks of the finish, so machined parts will have fewer and narrower peaks, and fluid friction from lubricant will be decreased).

At least with two stroke and glowplug engines you don't have to worry too much about oil control!

Piston attitude will be affected by clearance, a large clearance short skirt piston, with large clearance, will rock more, and so be more prone ring and bore scuffing. Conversely, a close clearance can result in seizure of the top lands, or skirt, depending upon piston shape. (Which is why full scale pistons are oval with barrel shaped profiles, determined by measurement of temperatures at various points along and around the piston)

Sadly, in model sizes, such luxuries are unlikely to be available. So you do the best that you can, and work empirically.

Howard

[Neil LickfoldParticipant @neillickfold44316]

Andrew,

The alloy has been specially formulated for the model engines at a very great expense by an individual who has always been looking for the best. He has been kind enough to share this technology to others so that more can enjoy the benefits of this material. It does not fall into any of the commercial number series of alloys, and it is very expensive.

From looking at a Kallenberger internal wheel spindle, it is not as controlled as the main head stock spindle. There is a very small amount of movement detectable with the the fine dti, the head stock has none detectable. As far as I am aware, there are no high speed super precision bearings. To get then to run fast,they have to have a little more clearance, Most precision bearing set ups have some form of spring loading on them to allow for heat expansion etc. Often running them at the highest rpm range, the work piece does not come out as round as it would if it were run at a slower speed. On my Myford setup, I can not run it in the high speed range as it will create heat in the head stock tapered bearing.

Howard,

You are very correct and bring up a lot of very valid points. Diesel injector making was one of the most difficult to produce parts in a production enviroment.I used to work for a company that made injector pump parts and repaired them.

Most normal inserts that you buy are not ground to a sharp edge. There are ground sharp edged inserts for cutting Aluminium. Like wise for the PCD cutters, There are a series that are sharpened with a positive rake geometry that will take micro cuts when sharp. PCD needs to be EDM ground and some are a lot better at it than others .To control the piston, the bottom of the skirt is made bigger than the seal band diameter, as they are a tapered liner, the skirt diameter fits at about the diameter of the transfer ports, giving the piston very little room to rock around. Traditionally the pistons were more of a barreled shape,or were parallel to the pin and then slightly tapered to the skirt, although these did work, the newer design with a large skirt performs better. Another aspect is the pistons have a small relief after the seal band area, about .1 inches long, to force the piston to behave like a ring. This also improves over all efficiency. Model engines get away with stuff that full size just can not, but at the same time create their own unique sets of issues. An interesting experiment was to take the set up that worked in a model diesel engine and try to run it as a glow engine. It did not per well at all. It did run however, but was too tight and showed signs of over heating. But the glow engine did not make good enough compression to get the engine to run as a diesel on the fuel supplied. A special brew with more ether was needed to get it to start. it was interesting to do and see what was the results.

Like I said, my setup is not good enough to make the pistons for the modern racing diesels, certainly good enough for a saturday afternoon park flier.

[Anonymous]

Neil: Thanks for the information; it tells me all that I need to know.

Regards,

Andrew

[Neil LickfoldParticipant @neillickfold44316]

An update on piston trimming and taking very fine cuts. So, since I got this postitive geometry PCD insert, I can now a make and fit the pistons for the F2C Diesel race engines. This cutter allows me to trim the piston dry, get a really good finish, and allows for less than 1um cuts on a piston. Cutting dry is a big time saver as they need to be clean and dry to get an idea of the fit point in the liner. I also added a continuous drip oil feed to the saddle to keep the bed wet with oil all the time. Makes it very consistent on the saddle movement.

Neil

Saddle oiler

Neil

[john carruthersParticipant @johncarruthers46255]

Before testing optics I let the temperature of the work piece equalise for 30 minutes in a bucket of water that's been in the room for a day or so, might speed up testing for you?

[John HaineParticipant @johnhaine32865]

An interesting thread this. I've always been frustrated trying to turn to reasonable tolerances on diameter, especially since converting my Super 7 to CNC some years back. In the last couple of months I decided to do something about this. The typical problem I was getting was: chucking a piece of material; taking a light skim cut to the end to get a clean diameter to measure at some sensible value of X; measure the diameter and reset the tool coordinate to correspond; then use the Mach 3 wizard to cut to the desired diameter; then measure and find the diameter was out, sometimes 0.1mm to big (which is OK), but sometimes too small (not OK!).

In another thread I described a home switch to allow me to set the X axis to a consistent zero, and I have further modified this to reduce the chances of wear and tarnishing. And I also decided that the top slide had to go, as I could see significant deflection of the tool using a dial indicator on a test bar, just winding the feed handle. So I made a fixed and rigid toolpost to take my Dickson tool holder. Separately I've built a probing system for the lathe to sense when a tool touches a test bar which can be used to measure the tool offsets, which are then stored in Mach 3. It can also be used to zero the Z axis to the end of a bar in the chuck (previously I used the top slide to pick up the end of the bar by feel).

The best diameter reference I have is a Mitutoyo digital micrometer which consistently reads to 1 micron. I don't know how accurate it is not having any distance standards like gauge blocks.

Now I can switch on the lathe, home the X axis, select and mount a tool, chuck a bar, find its end, the fire off the wizard to turn to a required diameter.

So yesterday I went through this process to turn down 15 mm of a piece of steel bar from ~ 26 mm to 25 mm using a freshly ground tangential tool, in 3 cuts, the finishing DOC being 0.1 mm. Measured diameter after cutting was 24.996 mm. Next I turned a short spigot on the same bar, going down from ~25 to target 6 mm diameter for 7.5 mm length. Rough DOC was 0.5 mm DOC, finishing cut 0.1 mm. On previous experience I would have expected to be +/- 0.1 mm or more by the end. This time the final diameter was 6.007 mm.

On my lathe the cross slide screw has been replaced with a 5 mm pitch ball screw driven directly by a stepper. Micro-stepping is enabled and steps/mm is 1280, so assuming perfect micro-stepping the resolution is ~0.8 microns per step. Limited evidence, but it seems that it is possible to turn directly to accurate diameter without using the angled top-slide trick.

[Author]

Posts