Member postings for Andrew Johnston

Soluble and neat oils are different, and have different purposes. Soluble oils are intended to cool with a little lubrication, whereas neat oils are intended to lubricate with a little cooling.

With carbide tooling it's a moot point as to whether cutting fluid is needed. I don't use it for general turning and milling, except for parting off. I use soluble oil for drilling with HSS tools and on the horizontal mill with HSS cutters. I also use it on the CNC mill irrespective of tooling, mainly to wash away the swarf. Anything other than flood coolant is a waste of time, especially if the primary purpose is cooling.

Whether fluid is needed depends upon the chipping characteristics of the material. For long chipping materials, like steel, as the chip is sheared it is deflected (creating a pressure point) by the top of the tool before curling up and possibly breaking. The fluid helps to cool and lubricate the chip as it passes over the tool increasing the life of the tool. For short chipping materials, like brass and cast iron, the chips come off as individual particles so don't exert the same pressure across the top of the tool. In addition, for cast iron, swarf plus oil creates a sticky mess.

Andrew

Today I've been doing the initial machining on one of my traction engine cylinder blocks. I have three sides, the top and the width of the flange done:

All machining has been done on the horizontal mill with an 80mm diameter insert cutter. Here's a typical set up:

The biggest cut was 2mm deep, full width, 420rpm and 420mm/min feed, not even a hint that the mill noticed it. I love it! And just to stir up the old versus new debate the horizontal mill cost me £175.

The iron casting machined beautifully, but was all over the place in terms of dimensions and squareness. At least I've ended up with all the sides and top square to each other and to dimension (in my CAD model) apart from the two faces parallel to the bores which are 15 thou under. The width of the parallel faces on the unmachined casting varied from 5 thou over nominal to 150 thou over, so somewhat on the skew. It's annoying to be a bit under, but it's not a problem in practice. Although I might tweak some of the port dimensions on the liners at a later stage.

The angle plates in the picture were indicated to be parallel to the Y-axis before using them. Just as well I checked, as setting them perpendicular to the front of the table with a set square meant they were many thou out. Since my horizontal mill is a universal (the table swivels for making helical cuts) I also indicated the table parallel to the column in X at the start.

Tomorrow I plan to fly the glider, before the bad weather sets in for the end of the week and the weekend.

Andrew

Feedrate is the key to parting off. On the centre lathe, with an insert parting blade, I never use less than 4 thou per rev power feed, Parting off is now, almost, a routine operation.

On the repetition lathe, with hand feed and a hand ground HSS steel blade, I've measured the chip thickness as between 8 and 10 thou.

To summarise; don't pussyfoot about!

Andrew

Put it aside for another 35 years.

Andrew

The answer is:

The access hole is 1/4" diameter.

Andrew

I've just done this for a clevis on my regulator, bottom middle:

Sadly I don't have a picture of the slot machining as I didn't think it merited one. I cut the slot with the steel bar vertical in the machine vice. I used a 6mm carbide endmill. First cuts were through the middle of the slot going 2.5mm deep per pass. I then shaved one side full depth (20mm) to get the side to the correct thickness, and then the other side to get the slot the correct width. For a hand knurling holder it wouldn't matter if the slot wasn't exactly central.

I considered using a 1/4" S&F cutter, but it was too much of a faff to set it up. That would have been on a large horizontal mill. I wouldn't consider using the S&F cutter on my vertical (Bridgeport) mill; it simply isn't rigid enough.

Andrew

This afternoon I did the final fettling on my cylinder castings before starting rough machining to clean up the main faces and get them square. This morning I created a first pass design of the starting (singling) valve for my traction engines:

And a sectional view:

The design will need some fine tuning and I might simplify it based on pictures of full size engines. Either the engines have been modified over the years or the original builders ignored the drawings coming out of the design office. Nothing new there then.

Andrew

Wot? I use the same feed rates for sliding and surfacing cuts (changing the feed gearbox) and I don't get crappier finishes on facing. With the caveat that sometimes the finish deteriorates at the centre. But that's down to material and surface speed issues, not feedrate. It's rare that I need to face an item right to the centre. Normally there's a centre hole so I'm not worried about the last little bit.

Andrew

I think SoD is getting confused between surface speed and feed rate. When facing ideally one would increase spindle speed to keep surface speed the same as the tool approaches the centre. Indeed on some materials with insert tooling one can see the surface finish change (gets worse) as the tool moves to the centre. On CNC lathes it's easy to use constant surface speed (CSS) when facing, via the G96 code. But the feedrate in terms of distance per rev stays constant. To a large extent surface speed and feedrate requirements are independent of each other on a lathe. Not so on a mill - there it's chip load rather than feed that stays constant with spindle speed.

I'd agree with Clive that the slower facing feeds are down to the practicalities, and cost, of the mechanism within the apron rather than any cutting considerations. My lathe has selectable feeds from 1 thou per rev to 80 thou per rev, with facing feeds half those. Never had the nerve to use feedrates much more than 20 thou/rev. Even with a 3hp motor and a geared headstock the lathe is power limited on high feed rates with sensible depths of cut.

Andrew

It does on my lathe - Harrison M300. The leadscrew doesn't even turn with feeds selected. The leadscrew only runs when thread pitches are selected. But of course my lathe is industrial with a separate feed shaft running parallel to the leadscrew.

Andrew

On my lathe (M300) the cross feed is half the selected sliding feed. I suspect it's to do with the space available for convenient size gears in the apron.

Andrew

That's the official Burrell name, but I prefer to call it a singling valve.

On a compound the engine will not start without assistance when the high pressure piston is on dead centre. The solution is a starting valve which admits full steam pressure to the low pressure valve chest, bypassing the high pressure cylinder. This allows a double crank compound to start as the high and low pressure pistons are not on dead centre at the same time. But on a single crank compound (as I'm building) this doesn't work as both pistons are on dead centre together. The valve still admits full steam pressure to the low pressure valve chest. So the engine can run as a single using the low pressure cylinder - hence my name. Since the low pressure cylinder is larger than the high pressure cylinder more torque, if not power, should be available. The operating button is push to open the valve and is spring loaded closed. I've been told (on the TT forum) that this is how the SCC engines were operated. A momentary push of the button would give a short burst of more oooomph as needed.

None of the above is to be confused with a simpling valve. This is a valve that resides in the cylinder and diverts the high pressure cylinder exhaust to the chimney. Used in conjunction with a starting valve it converts a compound to two singles running in parallel.

Andrew

Thanks one and all for the advice. It's good to know I can use LG2. As said some bits will be made from brass, but the flange that bolts onto the cylinder will be PB102 as I've already got a piece of 5mm sheet. The body will be LG2 as I've got a suitable length in stock. It's a bit oversize, but I might as well use it rather than buy new stock. Money is tight at the moment; not sure I'm ever going to work again due to the current shambles. The valve itself will be made from stainless steel.

It's helpful to know that Easy-Flo is good, as that's what I'm using. I bought a number of rods some while back. Due to previous disasters I've been avoiding silver soldering where possible. But I'm on a roll now! I take note of the bigger clearances and will contact CuP in due course. Even if the solder doesn't flow as well a small fillet wouldn't be a problem as the original part was a casting.

Andrew

I've got as far as I can with the piston and valve rod glands for my traction engines:

The piston rod glands (top) are finished. But the valve rod glands (bottom) are still rough on the inner faces. I haven't machined the cylinder yet so I don't know exactly what diameter the mating recesses will be. So finishing will have to wait a while. Here are the piston rod glands in situ, with a small clearance on the slidebars as per full size and exactly as the CAD model shows:

Andrew

Sure will, and beautifully formed threads too.

Andrew

At least I'm not losing my marbles, ground or otherwise.

Andrew

Well that's odd; before I started breakfast I'm sure there was a post in this thread by JasonB explaining to Barrie and I that we knoweth not what we talk about with regard to SoD and grinders. But it now seems to have disappeared? May be SoD has more influence than we realise, and we should be careful in case the boys call round for a 'chat'.

Andrew

I'm in the process of designing the starting valve for my traction engines, based on fullsize. For those parts not in contact with steam I plan to use brass. But the body, which will be in contact with full pressure (wet) steam ideally needs to be bronze. The body will need to be silver soldered together from several smaller parts.

I've recently silver soldered some PB102 bronze with some success, but it's a right pain to machine. I've also machined some LG2 gunmetal with excellent results. The starting valve body is going to need some accurate and burr free internal machining, so I'd prefer to use LG2.

But LG2 contains lead, so the question is how does it silver solder? I've found a thread on this forum, from a while back, where it was said to silver solder well. But the internet seems to be confused on the subject. Some commercial organisations says it solders well, but is less good for brazing. I take soldering to mean relatively low temperature with tin/lead solder, whereas brazing includes silver soldering.

What does the collective think?

Andrew

Yes it would, if the two planes on which the V-blocks sit are parallel. One issue is that the movement of the end of the rod compared to the movement of the V-block is reduced by roughly the ratio of the length of the V-block compared to the rod. Which is rather unhelpful.

If the two planes on which the V-blocks sit are not parallel (quite likely if the plate isn't flat) then all bets are off, as the movement of one end of the rod will be influenced not only by the vertical movement of the V-block at the other end but also by it's tilt. Without other measurements it would be difficult to determine which effect was the cause of any movement of the rod end.

No idea where SoD got the statistic that grinding moves more metal in industry than anything else. Most of the machine shops I've used professionally over the years didn't have any grinding machines. Grinding is good for precision and high quality surface finishes, but it's a relatively slow process. As such it's to be avoided unless absolutely necessary. One of the factors driving the development of inserts that will turn or mill hardened steel was to avoid having to use grinders for finishing operations.

Andrew

Unless I've misunderstood what was measured I don't see that it tells you anything about how flat the plate is? If you put the rod and V-blocks on a large sphere and move it around the ends of the rod will stay at the same height relative to the surface.

Andrew