NSW 'B' class mogul

[Jim]

Looking great David and I must say the 'Acme' thread looks brilliant. Making the form tool and successfully cutting the thread is a real feather in your cap. Well done.

Jim.

[Roger]

Hi David,
It looks like you've nearly cracked it, this is a really good effort.

I'm inclined to think that most of the issues are related to how well you can grind the tool with appropriate clearances and dimensions. I don't think machining HSS will ever get a sharp enough edge, so you may as well grind everything. That's a real challenge, but I'd start by making a fixture that holds the tool firmly and allows you to set an angle. A temporary guide clamped to the grinding wheel rest could be set so you're always presenting the tool at a known angle. Trying to do this freehand is not going to give an accurate result in my opinion.

Making a tap ought to be pretty much the same process, albeit with more challenging materials.

[Lisa]

A grinding rest like this is on my 'stuff to make someday' list: www.steves-workshop.co.uk/tools/grindingrest/grindingrest.htm

Good job on that thread though! I'm not sure I'd have even tried, so you're doing better than me.

[David]

I'd like to make Harold Hall's grinding rest one day but I'd even more like to buy one so I don't have to. At the moment my grinder is sitting unbolted on a flimsy toolchest, I don't even have have a stand for it, never mind a proper grinding table set up. That one Lisa linked to looks a lot quicker to make than Hall's one though so would be a good start.

It occurred to me I could go with a single start LH thread and it would look the same except at the ends and just take longer to go from end to end. I might do one of them as a fallback and then spend time on the double start.

[mbrown]

After making a single start square thread for my reverser I was very pleased with myself - but, just to see if I could, I had a go at a two start thread and the difference between 14 turns and seven turns between forward gear and reverse made such a difference to the feel of the reverser that it felt well worth it.

My screw is 7/32" diameter. I made a silver steel tap and it cut a good thread in the brass nut, although it took a great deal of forward pressure to get it started and the tap broke at the neck behind the thread on the final quarter turn. The nut fits the screw very slightly better in one orientation rather than the other suggesting that, like you, I didn't get the two threads precisely identical.

But if I can do it, it can't involve much wizardry, and I think it's worth it.

Malcolm

[GWR 101]

David / Malcolm. Great work,I think this is part of what our hobby is about. I made a two start thread for my Juliet's regulator and also made a tap out of silver steel. Not as difficult as your efforts but works well and worth the effort. Regards Paul

[David]

I do want to make the 2 start. I'll have another go soon. The tool is certainly sub-standard due to just being milled, and I was cutting at 200rpm because it was a LH thread so no danger of hitting the chuck. Perhaps that wasn't helping, although I can't see why it would make a difference.

I'll dress my grinding wheel to a square corner and if I don't have to throw it away try to touch up the flanks of the tool by hand. The good thing is being milled it's easy to make another one.

7/32... 5.5mm is tiny! Do you recall how deep the thread is? A square thread is wonderfully simple - tool width is 0.5 pitch, depth of cut 0.5 pitch.

[Roger]

Just a few thoughts on machining a tool for this....

Personally I'd go for either Silver Steel or Gauge Plate and then harden and temper it. It's easier to cut and you're likely to get a much better finish.

I presume you're holding the tool vertically in the mill and are adding the side relief angles in the program? By those I mean the ones that taper top to bottom when you look at the front of it in use.

If you hold the stock at an angle, you can machine off the top which will become the front clearance. Obviously you need to make the cuts slightly deeper than the thread depth to compensate.

If the tool is prone to breaking, you can pre-drill a radius at the sharp corner or 3D machine a small radius, still using a milling cutter with a square edge.

When using the tool, looking from the top, I don't think you need any side relief at all, I never use it on parting tools and this is no different.

And finally, if you're going to all that trouble, you could 3D machine an Acme thread tooth form instead which would make for a stronger tool



Alternatively, you could make a completely different style of form tool, the sort we used to use on production CAM Auto machines in the 70's. These have the profile created on a disc, with a cutout to create the top flat face of the tool. There's one like it on Figure 4 on this page
It could be just a small diameter piece of Silver Steel, and again, you could make an Acme thread. The diameter needs to be small to get the required side relief which only comes from the radius of the tool below the cutting edge. The Silver Steel rod could be say 10mm long and held in a simple split holder with a hole drilled across it from side to side and the front machined off to let the tool stand out from the hole.

[mugbuilder]

When screwcutting a two start thread I try to use a pitch that uses 1 or 3 on the chasing dial to engage the cut. Then I switch to 2 or 4 for the second cut. The two threads are then symetrical and evenly spaced. i did this on my 'B' class. My lathe has an imperial 8 TPI lead screw.
Barry

[mugbuilder]

That thread would look OK if it was dressed up with a fine file. Cameras and enlargements like that are very unforgiving.
Barry

[suctionhose]

Feb 8, 2019 7:37:38 GMT mugbuilder said:

When screwcutting a two start thread I try to use a pitch that uses 1 or 3 on the chasing dial to engage the cut. Then I switch to 2 or 4 for the second cut. The two threads are then symetrical and evenly spaced. i did this on my 'B' class. My lathe has an imperial 8 TPI lead screw.
Barry

just use the same number on the chasing dial and advance the compound slide by half the pitch to cut the second lead...

[David]

It quickly became obvious I can't dress the wheel unless I can run the dresser across it in a straight line and with the grinder not bolted down and no fixture to guide the dresser it was a lost cause. I was just moving the rounded corners further in.

Despite this I tried making yesterday's tool either more parallel or with some relief back from the cutting edge but ruined it. I expected that to happen so wasn't too put out.

I made another one, milling it a differently hoping to get it more parallel.

The double start thread turned out exactly the same. I was careful to move the compound slide as close to 0.75mm as I could, and I cut at the lowest speed. The cuts were clean. So I don't think it is my technique on the lathe, I think it is the tool itself. I don't use the thread chasing dial, I just don't open the half-nuts and run in reverse the get back to the start of the cut. I allow plenty of room to take up the backlash in the leadscrew before the tool meets the work.

If I have a lead of 3mm, trying for a pitch of 1.5mm, and the tool width is more than 0.75mm, would that cause the 2nd start to thin out the crests of the first one? Or perhaps the tool isn't quite set to the necessary helix angle so is carving out more material than it should?

Roger, I'm making the tools on the manual mill! That's why I'm going for a square thread rather than trapezoidal. I'm just milling away at the 4mm blank from each side until I get the cutting part as close to 0.75mm as I can. The flex in the tool as it gets cut down is such that go to the numbers, and then keep going bit by bit until I feels it's as close as possible without overshooting. I did consider trying to cut it with the CNC machine but I doubt that would work any better given the flex involved. The link is interesting, I think I've seen cutters like that in relation to making gears.

I decided to cut my losses for now and try a single start one as the initial working part. If that looks ok I'll have a go at a tap but I need to think about that - it's not yet clear in my mind how to machine it so the nut threads crests are shorter and more narrow to give the working clearances.

I also have the problem that the real thread is 52mm long and my lathe cut about 0.2mm wider at the far end of that than at the chuck. That's okay for the outside dia as I could turn it down at the end. But it also means the thread root is 0.2mm more shallow at that end even with a lot of spring cuts and a center. So the nut threads will have to have minimal crests.

I'll show the various threads and the tool at the club tomorrow and get some advice.

[mbrown]

I ground the tool on the stem of a broken 3/16" Slocombe drill and held it in a holder so that I could adjust the angle to suit the thread pitch. The angles were finished by eyes first. The pitch was 1/8" so easy to measure.

I cut the thread as slowly as possible, and only a couple of thou per pass, with several passes at the final setting. I had no luck with the silver steel for the tap until I brought in the travelling steady - as it's fingers bridged the crests of the thread it didn't need adjusting once set.

In the articles on the Allchin, Bill Hughes suggests cutting the tap in mild steel and case hardening it - which might work for a one-off so long as you don't use Free Cutting steel. I used FC steel for the screw and it threaded beautifully.

Hope this helps

Malcolm

[Roger]

Feb 8, 2019 11:17:45 GMT David said:

It quickly became obvious I can't dress the wheel unless I can run the dresser across it in a straight line and with the grinder not bolted down and no fixture to guide the dresser it was a lost cause. I was just moving the rounded corners further in.

Despite this I tried making yesterday's tool either more parallel or with some relief back from the cutting edge but ruined it. I expected that to happen so wasn't too put out.

I made another one, milling it a differently hoping to get it more parallel.

The double start thread turned out exactly the same. I was careful to move the compound slide as close to 0.75mm as I could, and I cut at the lowest speed. The cuts were clean. So I don't think it is my technique on the lathe, I think it is the tool itself. I don't use the thread chasing dial, I just don't open the half-nuts and run in reverse the get back to the start of the cut. I allow plenty of room to take up the backlash in the leadscrew before the tool meets the work.

If I have a lead of 3mm, trying for a pitch of 1.5mm, and the tool width is more than 0.75mm, would that cause the 2nd start to thin out the crests of the first one? Or perhaps the tool isn't quite set to the necessary helix angle so is carving out more material than it should?

Roger, I'm making the tools on the manual mill! That's why I'm going for a square thread rather than trapezoidal. I'm just milling away at the 4mm blank from each side until I get the cutting part as close to 0.75mm as I can. The flex in the tool as it gets cut down is such that go to the numbers, and then keep going bit by bit until I feels it's as close as possible without overshooting. I did consider trying to cut it with the CNC machine but I doubt that would work any better given the flex involved. The link is interesting, I think I've seen cutters like that in relation to making gears.

I decided to cut my losses for now and try a single start one as the initial working part. If that looks ok I'll have a go at a tap but I need to think about that - it's not yet clear in my mind how to machine it so the nut threads crests are shorter and more narrow to give the working clearances.

I also have the problem that the real thread is 52mm long and my lathe cut about 0.2mm wider at the far end of that than at the chuck. That's okay for the outside dia as I could turn it down at the end. But it also means the thread root is 0.2mm more shallow at that end even with a lot of spring cuts and a center. So the nut threads will have to have minimal crests.

I'll show the various threads and the tool at the club tomorrow and get some advice.

If you only have 3mm of the tool sticking out above the vice, you won't have any flex to speak of. That's about all you need to machine the end of the bar. I'd definitely use the CNC mill to create the clearances too. Once you have the model and CAM output, you can make as many as you like.

I find it's usually easier to put the effort in up front on the tooling to make the job dead easy when you finally get to do it rather than struggle.

[David]

There are a couple of things working against me here regarding making the tool. (a) I don't have a 4mm 5C collet to hold the HSS which would give excellent support. (b) The vice on my milling machine has a step in the top of the jaws so even though it has a vertical V groove there is no support for about 6mm above the top of the jaws where they grip the blank.

Now I have that 'tool holder' made out of hex I can use that to hold the blank for machining. That hadn't occurred to me until I was writing this. The tool might rotate though, it's only two grub screws holding it in place.

[Roger]

Feb 9, 2019 1:02:52 GMT David said:

There are a couple of things working against me here regarding making the tool. (a) I don't have a 4mm 5C collet to hold the HSS which would give excellent support. (b) The vice on my milling machine has a step in the top of the jaws so even though it has a vertical V groove there is no support for about 6mm above the top of the jaws where they grip the blank.

Now I have that 'tool holder' made out of hex I can use that to hold the blank for machining. That hadn't occurred to me until I was writing this. The tool might rotate though, it's only two grub screws holding it in place.

So why not use a piece of  rectangular section Gauge Plate, you don't have to use HSS? If you set the piece at 10 degrees from the vertical in the vise, you can machine two angled flats that give the side clearance and create the tool width. I think you're making life unnecessarily hard for yourself.

[terrier060]

Great work David - I made a 2-start for my Princess loco and it really pays when you are on an up-and-down track! Besides an acme thread is a challenge and looks good when its finished.

[David]

I don't have any gauge plate, and I'm using the round so I can change the helix angle on the tool. I have no doubt I am making life hard for myself but I'm working to the best ideas I have at the time.

A frustrating weekend:

Took the red loco to the running day yesterday to test the injectors again. It was a mild day so perfect for testing.

1. When I connected the blower to the air line the plastic air hose shot off the fitting because it had perished and split. I wasn't strong enough to push the fitting into the hose where I cut it clean. Luckily there were lots of other ones around to borrow.

2. The container of lighting up wood had fallen over in the plastic bin of 'lighting up/blowing down' tools and all the kerosene had leaked out making a mess of everything and leaving the wood dry.

3. Steam came up quickly but as it did I noticed it blowing out of the gauge glass. The glass was cracked for the first time since I've owned it. It was fine last month so I don't know when it happened.

So that was not much of test. At least I got number of good laps on my father-in-law's loco.

Today I was trying the double start threads again. We spoke about various things that could be wrong yesterday and compound slide backlash seemed like a possibility and the advice was take it up in the correct direction and use a DTI to ensure I move it the right amount.

1. Find the DTI is gummed up so took it apart and cleaned it. That was time not expected to be required.

2. Did a RH 2S with a normal threading tool and it look okay. This was to check the basic DTI setup etc. Looked fine, if a bit weak because I was using a cheap no-name bendy DTI stand that you screw up to 'lock' it in position.

3. Did a LH 2S with the square tool. I tried to ensure I moved the compound slide the right way etc and had no backlash. Moved exactly 0.75mm according to the DTI, which agreed very closely with the graduated dial. The thread looked just like the others with 1 thin crest.

At this point I was pretty cranky so went to unload the red loco from the car as a distraction. No matter where I put the loco stand or the car, I could not get the 'bridge' of the stand at a vertical point where I could get the loco up onto it - my driveway has a pretty decent slope to it an the stand was made to suit a different car. This is a constant nuisance but I've managed it every other time for about 3 years! I finally lost my temper, smashed the loco stand into the ground a few times and in the only stoke of luck I had this weekend it twisted the angle where the castors are on the bridge side of the stand enough to bring it down to somewhere useful, while still barely allowing the castor wheels to rotate. Although they won't be swiveling anymore so it will be quite difficult to swing around.

4. Decided I'd better test things out a bit re backlash because maybe I still didn't have it right. Mucked about with the compound slide moving it both ways and seeing where the backlash was. It seems for a LH thread I should only wind the handle clockwise, moving the tool towards the chuck. If I do this and then try to push the tool back towards the chuck, simulating the cutting force, the DTI doesn't move.

5. Started another thread, got muddled up when moving for the 2nd start.

6. Decided to get methodical. Moved the compound slide towards the chuck, set zero on DTI, cut first start. Moved compound slide towards the chuck 0.75mm cut second start. 0.005" per cut to 0.030" depth on both starts. Same result as usual.

7. As for step 6 for the first start. Moved the compound slide 0.85mm towards the chuck before the second start. This created a thin crest on the trailing edge of the tool. That is what I would expect if things were going as they should.

8. Ground a little more relief from under the cutting edge in case that was the problem. This was probably a mistake, because...

9. This time I was going to move the compound slide 0.65mm for the second start. But the tool broke on the 2nd cut of the first start. I really wanted to see the result of this test!

10. Gave up in disgust and mowed the lawn. I expect to hear a crash from the workshop any minute now telling me the loco stand has collapsed. I've wasted about 1.5m of 12L14 hex so far.

I'm getting really quick at cutting these short test threads but I'm probably wearing my leadscrew out over that 30-40mm.

Obv I have to make another tool. I'll use the toolholder itself to support the tool and see if the two grub screws are enough to stop it twisting. If that works I might then be able to figure a way to do it in the CNC machine and cut the correct helix angle in from the start so I can stop guessing at it.

But not today - the workshop is locked, I have to cook dinner, and then I am going to watch other people making things.

Looking forward to going back to work tomorrow!

Sorry for the rant but it's been a bugger of a weekend.

[Jim]

We all have these days David. In my case I simply can't rest until I've sorted the problem, it drives me nuts and has at odd times found me out in the workshop in the dead of night having another go at the problem.

Barry will tell us 'It's just a hobby.' and it is...sometimes the you hit a day like the one you've had.

Jim

[simplyloco]

Have you considered taking up something more predictable and less hazardous? Base jumping for example?😎
John