Here is a list of all the postings Iain Downs has made in our forums. Click on a thread name to jump to the thread.
|Thread: Steam Engine Number One|
So the latest item on the production line was the DValve.
This was a fairly straightforward piece of milling, marred in the first instance by my measure (wrong) once and cut... oh dear ... approach.
Second times the charm.
My original drawing had a slot and bar approach to connecting the eccentric rod, but I decided to change it for a gap with nuts and washers (which is what my main source the Stuart 5A has). It may be that the ridge is now not high enough, but the DValve is also easy enough to remake, so I'm not too worried.
Also attempted in this period was the valve chest. Sadly, this failed part way through due to the cutters pulling out of the collet and digging too deep, spoiling the part. It looks like the root cause it not enough drawbar tension, so I hope it will work better this next time!
|Thread: MIlling cutter pulling out of collet|
Thanks (as usual) for all the feedback.
I've checked the cutter I've been using. This shaft measures around 0.01 mm ( 2 tenths) undersized (9.99mm) though it's possible that the flutes are a tiny bit smaller - not easy to measure.
It is a tight sliding fit in the collet which measures (best I can with calipers) pretty spot on 10mm.
Good news I thought. I cleaned both with denatured alcohol and mounted them in the quill.
I've got a new torque wrench but had to make suitable socket to go over the 12mm drawbar. I would have happily bought one but couldn't find such a thing.
The socket was constructed from some mild steel bar drilled one end at 9.5mm and milled to 12mm the other and then squared up with 2mm carbide endmills. I had several approaches to this each of which lost me one end mill. In the end I was able to take 2mm deep cuts going round the square and it worked. For various reasons the socket is a bit of a mess, so no photos!
I applied what I believe to be around 28Nm of force and proceeded to mill the same piece.
10mm deep. I started at 1 mm depth of cut, but got brave and ended up with 2mm depth of cut. I cut 3 stripes at 2mm (6mm in total) conventional milling and then 3 more 2mm stripes climb milling.
As best I can tell there is was no movement between cuts.
I'm not sure if you can tell from the photo, but there is almost no ridge to the finger touch on the top. Scraping a finger nail there is the slightest hesitation. In short this is about as good as it gets.
So it would seem that my earlier error was simply not to tighten the drawbar enough.
Having said that 28 Nm felt like a LOT at the end of the torque wrench and it was VERY hard to undo with the 6 inch spanner I normally use.
Is my general solution just to get a longer spanner? I'm not particularly keen on messing with a torque wrench to set tools.
Michael: I've not actually measured the bore of the new collet, but 10mm end mills (not reground) and edge finders fit in nice and snugly and don't drop out (unlike in it's predecessor) so I'm content that this is reasonably sized.
Martin: There's plenty of thread on the drawbar. There's more thread on the drawbar than it sticks up by when a collet is inserted and the drawbar fully unscrewed.
I will, however double check that.
I've ordered a torque wrench so I can apply some science to this, but will need to make an adapter between a 3/8ths socket and the 12mm square spanner nut thing on the top of the drawbar.
Thanks for even more thoughts!
Jason - I'm not sure what you mean by 'out of thread'. The drawbar screws down about 40mm into the collet. Having said that some collets are still sticky at the end. I keep losening the setscrew in the spindle (which appears to be the issue - or more correctly R8 collets which have a keyway that's not to spec). Do you think this could be a contributor?
Thanks again. In fact the collect is relatively new bought to replace one which was a tad oversized.
Jason - I'd actually planned to do that (cut the boss on the lathe) originally, but then thought that I would trim off the bulk of the material to minimise the interrupted cut. Now it's more a question of working out the right technique as a matter of principle.
Sadly, I am now in the day job, so will try and do some more tests tonight. I believe I no longer have a torque wrench, so I may need to leave that aspect until Mr Bezos can deliver something.
It looks like I measured once and cut twice with the text of my original post.
The boss is 16 mm not 6mm and I was (as some of you have realised) suggesting use an ER32 collet chuck rather than the R8 collets I was using here.
I've had a similar problem with trying to tap with this machine is well = the shank of the tap doesn't seem to be gripped well by an R8 collet.
I will try ER32 once I've found some stock and taken it back to size!
I will double check the dimensions of the mill tomorrow.
This has happened a couple of times, but today ruined a part I was several hours into .
The part was a 75x52x22 (mm) mild steel block stood upright in the vice.
The task was to remove to a depth of 10mm except for a spigot which would ultimately become a 6mm dia boss.
The machine is a VMR32L, A beefy thing for a hobby machine with R8 and a 1500W motor. Rigid enough that i can seriously consider climb milling with it.
MIlling cutter is a 10mm carbide running at full pelt (2200 rpm) - the cutter is a re-ground 'pro' cutter - possibly a little undersize because of it.
So I set the depth to 10mm an started taking 1mm - 1.5mm cuts. within a few passes I noticed that it was digging deeper on each pass so I stopped, tightened up the drawbar pretty tight and tried again.
Much the same thing happened, though with less creep this time. I ended up about 1.5mm too deep, so I'm going to have to junk the part.
The quill DRO showed little movement (around 2 thou).
The mill, by the way is not struggling with the cut and will quite happily cut this and more when there is no material at the bottom.
Am I expecting too much from this configuration? Should I be tightening harder? Would using R8 Collets give a better grip? I don't think there was oil in the collet / mill assembly, but could that be a contributor?
Looking forward to your answers.
|Thread: What to do when you lose something|
Last night I was reading in the lounge on my tablet. Watched a bit of telly without getting up. Half an hour later I went to collect the tablet to got up to bed. Still can't find it!
A couple of weeks ago I put down 2 m4 taps on my mill bench (I'm sure). Something stole them. No idea what.
Worst of all, a couple of years ago I lost my keys - pretty sure it was in the shed. Searched. Even tidied. Emptied the bin twice. Had to buy TWO electronic car keys (which was several months shed budget). 3 weeks ago a put an old coat on by accident and wondered why the electronic car key didn't work. Because the battery was flat after 2 years.
Hmm, thinking about it, I'm not sure where those keys are.
But yes, my ratio of looking for the thing I just put down to productive work is much too high!
|Thread: What Did You Do Today 2021|
Thanks for this, Michael!
That's very interesting. I spent a bit of time thinking about how to apply this on the gauge. The most likely implementation would be like this clamp (referenced from the wikipedia article).
There are two challenges that I see (well three, but we'll leave my technical ability out of it). The first is the one mentioned in the article, which is that the torque required to shift the differential screw may be significant. I don't think that would be an issue with the gauge, but I would need to build it to find out.
The one which concerns me more is that the centre threaded rod must remain static with respect to the outside holder. If there is any movement (as the differential screw is turned), then that will overwhelm changes from the differential.
Before I saw this diagram I was lying awake trying to think of ways to hold the nut (in the other model) rotationally rigid and was somewhat daunted by the size of the apparatus that would be required and the level of precision engineering. I'd imagined the nut running in some splines on the outer body - or something.
So am M4 / M3 combination seems to come with quite a few design / construction challenges which I've not quite got resolved. The idea above seems easier in this context if it works in practice.
I may well have a go at this at some point, but it will have to find it's place in my queue of half started projects.
Many thanks again!
Not today and not even this month, but I finally got round to finishing my surface gauge made mainly with a mystery metal.
I still have a bit of tuning to do -the first photo shows the cap screws (in the second photo) replaced with knurled finger screws, but the dovetail clamp is still a little tight so I plan to enlarge the hole a little.
The indicator in the lower picture is a micron (well 2 micron) indicated for Bangood which cost me £18 and appears to measure microns. (and yes, to all the professionals out there, microns are considerably beyond my and my tools capacity, but it's nice to dream!).
However, the M5 screw is a little too course to be able to zero the indicator and I may think some fine adjustment nearer the indicator.
|Thread: Steam Engine Number One|
I'm not leaving the work reporting behind for a plea for help.
I've not yet worked out how to get steam into and out of the valve chest.
There's actually a fair amount of meat in the cylinder to the exhaust port and the valve chest wall so I could put in a threaded hole.
Alternatively I could make / buy a boss and bolt it to the wall. In so far as I have a design source for this, it's the stuart 5A (being around the same size). That has bosses with half inch (12mm) or 3/8 (8mm?) 26 tpi (1mm?) threaded insides.
Having recently discovered compressed air, I was thinking of 1/4 BSP which is roughly in the same area.
Then how do I connect steam (/air) to it.
I would generally prefer to work in metric but a set of 1/4 BSP tap / dies are attractive now I have air in the shed.
Final question (for now). Can I use PTFE tape as a seal or would it melt with live steam?
These last few days / weeks I've been working on connecting the eccentric straps to the stephenson's link.
Not too many pictures I'm afraid, but..
Here are the bits
And them (temporarily) connected
I@ve found that the straps no longer run quite as freely in the eccentrics. Not sure why that is - possibly some grit or could be orientation. I will have to tidy that up.
Also in this series of workshop trials, I've made the nut that connects the upper eccentric rod to the valve gear.
This was quite a complicated bit of work and here's what I did.
Firstly square up a piece of bronze. The final item is about 12mm square with the spigot being 12mm long and 6mm thick, but faintly curved along a 120mm radius. The centre hole for the axle for the upward part of the system is 4mm.
So I have a piece 12 x 20 x 20 (or so). My intent was to use a superglue chuck on a fixture on the rotary table. I figured that the longer the piece the better the adhesion. The stock was 2mm to high at the top and 2mm too deep at the bottom to provide a raft for holding and some excess on the top to clean off.
So with my squared stock I drilled and reamed a 4mm hole in the centre and did the same on my fixture plate (at 120mm out from the axis. A 4mm pin and a try square oriented the piece whilst the superglue set. Next day a light tap drove the pin out.
Cutting the curves (and try fitting as the width approach 6mm worked well. Next I took the length of the spigot to 12mm. This also worked well.
The I started to take the based down to 12mm and took too aggressive a cut and, 'pop' off came the piece from the fixture!
And it was going so well!
I'd originally intended to cut the round edges on the rotary table, but I gave up on that idea and, having trimmed the remaining top and bottom and sides in the mill, I (rather gingerly) rounded the ends with a small file. It's not something clickspring would be proud of, but it's good enough for me!
What I found interesting, was that the nut no longer fit the slot in the reversing link. I think that this is because I squeezed the base to hard in the vice in and it mushroomed. However a few strokes of the file in the slot and it runs nicely now.
Next step is the valve chest and D valve. Once that;s done I can check that the bits fit together and I'm nearly there!
|Thread: Boring aluminium - What am I getting wrong|
Another thing which caught me out, is that these brazed carbide tools seem to come unsharpened.
I had issues with boring with a similar head until I read this in the forum previously. Sharpening them on a green wheel and then touching up with a diamond hone from time to time has made all the difference.
I also found that if you don't tighten the grub screws up well, the tool tends to rotate, so even if you have it aligned correctly (as described above) after a couple of passes it's twisted out of alignment and rubs rather than cuts.
|Thread: Steam Engine Number One|
Will our intrepid novice manage to cock up his reversing link for the third time? Will any more carbide end mills suffer?
Read on to find out!
So I had a good deal of advice on this. Firstly I made the meat of the link a bit thicker. Then I tried to approach the outside and the slot a bit more gradually. Finally, I ran my carbide mill at top speed!
I took a different and more daring approach to this build. I decided that, rather than bond the fixture and piece and drill my marks through, I would drill and ream them separately and bolt together - no superglue chuck for this boy.
I started by marking them out on a surface table - not that I intended to use this to drill - I have a DRO and I'm not afraid to use it!
Here I am having reamed out the salient items on the piece
I was delighted (not to say a little surprised) that all my pins could push through both items at once!
Here we are (again) aligning the item up on the rotary table at zero degrees.
for the outline I used an 8mm carbide end mill running at 2200 rpm (thanks for the chance to try again Brett of ARC!). The first cut was 2mm outside the piece and 2mm deep, the next cut the same at a bit over 4mm to cut through
I then took 2 passes each 1mm closer to the finish dimensions. There was a bit of rattle at the outside arc which is quite a bit outside the support of the rotary table, but the finish was still decent.
Having done this, I think cut the slot. first cut in 1mm increments down the centre with a 4mm hss end mill. Then two passes of 0.25mm to the finish dimension. According to my calipers the slot is 6.03mm wide so I made it 1 though oversized. Amazing!
Here we have the raw piece, the fixture and the drawing and, finally, this is what it looks like after some tidying up on the belt sander and a bit of time with some 320 grit .
The astute reader will notice that the slot is a little higher than the end holes and that the eccentric rod holes aren't central to the meat of the legs.
Despite that I'm very happy with this piece and AT LAST, I can move on to another part.
|Thread: What Did You Do Today 2021|
Today I managed to puzzle my way through the worst Chinglish I've ever seen to find out how to set the top pressure in my compressor. Previously, it would run hit the end stop and keep running as the overpressure valve cut in. Take of a nasty plastic cover and turn the nut counter clockwise till it turns off a bit below 8 bar. None of those words being in the instruction.
Also grew my compressor tools skills slightly, but finally ordering a coiled extension with the right (Euro) connectors on. I now have an air gun by the mill and by the lathe.
That and a little bit of tidying up left me with a feeling of progress. I also started on the 3rd version of my reversing link and have yet to mess it up. Plenty of time yet.
|Thread: Steam Engine Number One|
'For the outside radius I would use the full height of the cutter and take say 0.5mm passes and if that is OK go up to 1mm working in to the finished radius.'
Do you mean to cut 4mm deep (depth of the metal) but start 1mm outside the finished edge of the piece, or do you mean to cut 0.5mm - 1mm deep on the finished edge - hence 4 - 8 passes?
I think one of the problems was that when I saw that the carbide cutter was cutting nice, I cut to full depth (4mm + 1mm inside the aluminium fixture) which I think dragged the cutter into the work.
This post will not cover me in glory, but the truth must out!
The task was to create the reversing gear (drawing a few posts back) which I was intending to do by mounting (superglue with tape method) my raw material on a long fixture, attach this to my rotary table and mill the outside through cartesian co-ordinates.
The idea is to drill / ream / mill the 6 and 8 mm locating / mounting / starting holes in X/Y co-ordinates and then switch to cartesian for the cutting.
My first attempt at this failed, mainly due to the superglue / tape chuck loosing grip (the tape side) and the piece slipping to the detriment of all.
I discarded this and tried again. This time I was going to bolt through the holes for the eccentric rods to keep it in place.
Here we have the aluminium fixture plate pinned to the centre of the rotary table and the first locating hole used to align the X Axis at 0 degrees.
Now, I bought the steel for the link from eBay advertised as mild steel, but that may have understated it's toughness. two 8mm HSS end mills pretty much refused to cut it and both had to be discarded. Also from eBay or equally suspect sources.
So I bit the bullet and ordered some quality mills from ARC. The 3 flute carbide cutter cut well (given that the first pass of the top had already been done by Mr. Dodgy mill). However, part way down the second leg the part started to vibrate something rotten and inspection revealed that the end mill had sheared.
IN a call with Brett at ARC he suggested that I was running this too slow and it was clearly user error, but despite this he is kindly sending me a replacement.
I carried on with a 2 flute HSS (again from ARC) which finished off nicely.
You can see the roughness from the original end mills along the top.
Next step was to mill the slot. In this case, I chose to follow some recommendations from Joe Pie and pre-drill (in my case with a slot dril) before a finishing cut.
This wasn't terribly successful, there was a lot of vibration (even with the table locked) and I had to finish off the slot by widening a bit.
Given all that had gone on I was reasonably pleased. I'd abandoned the idea of cleaning the ends with the rotary table and filed instead.
And then I took another look at the piece - spot the error?
Indeed! the slot goes right to the end and there is no hole for the control handle!
At that point I stopped for the night and a nice cup of tea.
Now I have a question. Ignoring various machining errors, it seems to me that there is not enough meat around the slot and possibly around the eccentric rod holes.
The design has 3.5mm above below the 6mm slot. That should be 13mm, but in fact it is 11.9. The slot is actually about 6.3mm and the lower meat is only a bit over 2mm.
The design calls for the meat round the eccentric rod holes to be 2mm.
Firstly, I think I need to pay more attention to how I mill this out. I think I should probably rough mill 0.5mm outside where I want to finish and then take a finishing cut.
Secondly, I think that the meat needs to be bigger. I think 4mm round the slot and 3 round the holes would be better.
Finally, of course, I should avoid merging the slot with the control handle hole!
I think I might have a go at this with just some aluminium plate to get my measurements and skills up to scratch and then try it again.
I'll also have a couple of goes at different ways of doing the slot.
Any feedback on sizes (or technique) would be much appreciated!
|Thread: Bandsaw overhang|
I have a little block just the height of the underside of the vice, which I use to support the part being sawn, just past the saw blade. Generally works OK, but I like the other ideas!
|Thread: Compressed Air|
My wobbler (1cc capacity? Maybe less?) worked rather nicely with a tyre inflator that came with a car. Would not do so well on a quarter size traction engine.
|Thread: Carbide Tooling|
Jason: Do you buy the cheap ARC tooling or the sumitomo?
old mart: Most of my problems came due to my 10mm collet being a bit oversized so anything would just fall out. I've now replaced the collet, which now holds a 10mm end mill OK when relaxed. But a mat is a very good idea.
What I'm trying to get my head round is how a part can continuously bash into a cutter at 20mph (interrupted cut) and that is good, but if you bump it at a snails pace it breaks. How can this be?
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.