Harrison L5 Metalworking Lathe (1950) Teardown & Overhaul

[Sideways]

Saddle and slides 3

With the cross slide mods complete, I made a new sheet steel insert to cover the front of the screw, thinning a section immediately above the gear on the left as clearance is very thin. This plate is removable so that I can slide it forward to access the oiling hole for the screw's plain bearing, but keep that covered otherwise.

The bed was cleaned and the rack removed for cleaning then replaced

Then refit the slide, bolt the apron back on, thread the two shafts through the apron and reseat them in the gearbox.

Finally bolting the right hand bearing block in place.

 

[Sideways]

Compound slide

Nothing complicated here. The compound slide screw and nut were in good condition. A good clean and stripping back to my chosen bare metal look for the parts that are going to be scratched by turnings. I can't remember now whether I cleaned, or replaced the thrust bearing on the compound slide ! Senior moment ...

Note that the L5 with just 9" swing is pretty simple up top to keep the slides as low profile as possible. The 11" capacity L5A and other models that followed it had a thicker cross slide with T slots that allow the compound slide to be completely removed very easily.

Here it is fitted and with an oversize quick change toolpost sitting on top.

The compound has two castings, lower with a round boss to anchor it to the cross slide and let it rotate

Upper has a fixed screw that the chosen toolholder drops onto and is then bolted down.

The slides between the two are dovetail again but the gib is plain not tapered with set screws to adjust the play

The feed screw and split nut are similar to the one on the cross slide, but simpler and no problems were found this time

There are two thrust bearings this time and bigger. The photo above shows the parts lined up in sequence for reassembly
Here is the front end of the screw. index plate bolted in place, sleeve pinned and ready for the dial to be slipped on and the handwheel fixed with a taper pin.

 

[Fergie 307]

Excellent insights Fergie. I didn't know those details.
My L5 doesn't have the mesh filter you mention or marks from having had one, but there are now a dozen flat neodymium magnets around the inside so I'm onboard with that method.
My big motor that I removed wasn't an original. It was 1980's, pre owned but not vintage
Your L5a looks great. Very clean, better gearbox and I like the splashback.
I'm registered for the groups.io Harrison mailing list which is mildly helpful but very limited because it doesn't intend to be a forum and has v limited space for image storage.
If there's another forum I'm unaware of it. I don't do facebook in any shape or form so if there is something there, I'll have to do without.
Please continue to advise and point out my mistakes as I post the rest of the cleanup !

it is the io group I was referring to, as you say not a forum and with limited space. Lots of knowledge amongst the members of all Harrison machines so if you run into a problem then there will usually be someone who knows the answer.

 

[Fergie 307]

Oh and the metric dials and screws are most likely from a 5a. These were made as a sort of hybrid with imperial leadscrew and gearbox but metric top and cross slides, Mine is configured in that way. Many parts are interchangeable between the L series machines. Some very according to size so a n L6 has a wider bed and and saddle and the screws are longer. Having said that the 5a screws ought to be a straight swap, s I doesn't explain yours being off. Like your solution though, ingenious.

 

[Sideways]

Tailstock

The L5 tailstock is MT2 taper, graduated, and inludes a feedscrew and pin. This means that it doesn't rotate as you wind it out with the handwheel.

There is nothing complex about the construction. Just a need to pay attention to the locking lever. This is cam action and needs to be adjusted to have it lock with the lever in the right position.

Note that alignment of the tailstock is done with a combination of adjusting screws and shims. I haven't done a full alignment on mine as I don't have all the kit I need but the process includes turning a test bar between centres and then clocking the cuts to look for taper which would indicate misalignment.

The head of the lathe is bolted down to the bed and doesn't include any special provision for aligning it as on more modern lathes. To test the alignment of head and tailstock bores I believe that you need ground test bars in MT2 for the tail and MT3 for the head.

Here are the teardown pictures of the tailstock

Lower casting and parts

underside

With the cam lock inserted

Lower portion that slides on the bed - upper side here that mates with the body of the tailstock (above)

And with the shims replaced as they were before dismantling

Here is the shaft

 

[Fergie 307]

To check headstock alignment you ideally want a ground test bar mounted on the end of the spindle but not supported at the other end. Run a dti along it mounted on the toolpost. You can also do a cut across collars. I use an inch diameter CDS tube about 18 inches long with two Delrin collars, each about 10mm wide and 40mm in diameter. One collar wants to be at the end of the tube, the other about a foot further down. Mount the tube in a chuck so one collar is close to the front of the chuck, do not support the other end of the tube. Take a light cut with a very sharp HSS tool by just running the saddle down the bed across the faces of both collars. Now measure them with a micrometer, calipers aren't really accurate enough. Ideally both collars will be the same size, if not the difference will tell you how the head is misaligned. Beauty of this method, or Rollies dad's which is similar, is that it will not be influenced by any inaccuracies in the chuck. Using the delrin collars avoids push away influencing the result as may be the case of you do a cut in steel. Very important to ensure that your saddle or slides are not moving about, so should be done with cross and top slides locked. and check carefully for any movement in the saddle. Also important to check there is no excessive play in the spindle bearings. If you do find the head is off then the usual way of correcting it is to loosen the mounting bolts and then insert shims of appropriate thickness between the flat sides of the headstock casting and the mating face of the bed, you will see where I mean. If you are lucky and the adjustment required is very small you may find you can get away without shims. It's a frustrating job as you can get it perfectly aligned by as soon as you torque up the bolts it will move again, so you tend to have to have several goes at it. usually it's a case of noting how much in moves when tightened up, then overcompensating by that amount so that tightening it up pulls it into line, if that makes sense. You are ideally aiming for a figure of 0.01mm or less difference between the collars. This is perfectly achievable with patience.

 

[Sideways]

I like the delrin collar around tube idea. As you say, cuts in steel with a long stick out can be affected by flex in the bar. Having sacrificial plastic to take text cuts off sounds like a neat alternative.
Cheers

 

[Fergie 307]

Easiest way to check for play in the saddle is to mount a dti on the flat of the rear bedway, a mag mount is ideal. Rest the pointer on the upper rear machined face of the saddle. Now use a piece of 4x2 or similar, suitably padded and insert it between the chip tray and the underside of the apron casting. Lever gently and see if the dti moves. if it does then this indicates too much clearance between the rest of the saddle and the bed, so The saddle is effectively see sawing on the front way. Very probably caused by a worn rear mounting strip, but can befriend to wear on the underside of the saddle itself.Repeat the rest with the dti on the front way. If this gives movement Then you have worn front mounting strips, or wear in the front way or it's matching surface on the saddle. there is really no definitive figure for how much play is acceptable. A Good guide is to run the saddle up and down the bed. If it runs very freely, with no resistance, then it is too loose. There should be a discernable, if slight, resistance. Good idea to do these measurements at the tailstock end first, where the bed is likely to have little wear, and repeat with the saddle near the chuck, the part of the bed most likely to be worn. Any significant difference in the readings at either end is an indication of bed wear.
Hope this is helpful.

 

[Fergie 307]

I like the delrin collar around tube idea. As you say, cuts in steel with a long stick out can be affected by flex in the bar. Having sacrificial plastic to take text cuts off sounds like a neat alternative.
Cheers

You can either bore them very accurately and make them a push fit, or stick them on with superglue doesn't really matter. CDS is Good because it tends to be pretty accurately sized and is very stiff but light so you don't get droop coming into play owing to it falling under its own weight, as you can with a solid bar. If you don't have a ground test bar, have a look to see if you have a gearbox reconditioner near you. They will probably sell you an old layshaft for next to nothing. May have been replaced owing to bearing wear, but still a nice precision ground bar which can be very useful mounted between centres for checking tailstock alignment for example.

 

[Sideways]

Light relief - roller skates

At 560 kilos, the L5 isn't that easy to move. In my garage, it will sit back to a wall in a corner. Anticipating the challenge of rolling into place and then having a means of pulling it forward from the wall if I need to do maintenance on the motor, I made up a pair of skates. They should be good for close on 2 tons the pair so ample. They are fabricated from hot rolled channel and the rollers are sold as spares for renovating pallet trucks. They have ball bearings and solid plastic tyres that avoid damage to the floor.

Everything is self evident here. There were just some largish holes to drill - dead straight - axle pins to turn from mild steel and spacers to keep the rollers centred in the channel which I'd bought as an offcut. Holes were drilled on the mill as it's stiffer than even a good pillar drill.

The length of the C section spans the lathe's base front to back so these are easy to use once the base is lifted high enough to slide under.

The mild steel got some metalguard to slow down the inevitable but unimportant rust. The channel was wire brushed, heated with a gas torch and then wiped over with linseed oil. It makes an easy finish which lasts 1-2 years against the winter damp in the shop.

In normal use, the lathe sits on 4 adjustable anti vibration feet. The stiffness of the base is ample to keep the lathe happy. It doesn't need bolting down, just levelled side to side and front back.

 

[Fergie 307]

If you look at the pic of mine I have done much the same. I needed to lift it as I am very tall. The legs you see it standing on are cut down scaffold feet. These can be unbolted and swapped for five inch castors when the need arises, makes moving it around a doddle.
Oh and while I think about it, bed wipers. The felt ones tend to be pretty useless, the swarf just gets stuck in them and dragged up and down. I couldn't find anyone selling suitable plastic ones when I rebuilt mine. I got some 5mm thick leather from my local saddler. Easy to shape, and once they are on they soak up the oil and swell to give a nice tight fit, and spread the oil. They also seem to be too dense for stuff to get stuck in them. I think you can get plastic ones now on e bay, those for a 140 should also fit. Looks like you are doing a grand job. Lovely old machines and built like a tank.

 

[Sideways]

I found a chap on ebay selling 3D printed ones for a variety of machines. I've fitted a set of his.

Clutch and brake:
The L5 has a spring loaded clutch and brake concentric with the upper pulley driven by the motor V belts.

In the first image, you see it top left. Down and right is a hollow shaft which is the main spindle. The spindle bore on an L5 is small. Under an inch. I won't be feeding any two inch bar stock through the head on this. In this picture, the change wheels are removed so that you can see behind them.

Following are simple pictures of the component parts once the round nut at the end of the black spring is unscrewed and everything unthreaded.
Tightening the nut increases preload on the spring makes it harder for the clutch to slip.
Moving the clutch lever top left of the headstock operates the clutch and mechanically disengages drive into the head.

I'd be interested in other owners views on the state of the friction surfaces as I've never seen these in "new" condition. I'm not down to the metal yet, but think this is a job that I'll need to tackle before long.

 

[Sideways]

Wrapping up, the L5 was the first significant piece of metalworking kit I'd ever overhauled. The simplicity of the old thing being a blessing and a fun lesson in how machines used to be built.
We had to empty half my garage to roll it in to the furthest corner where its small footprint allows it to tuck under the shelving without making everything impossible to reach.
Seriously wood and metalwork in a singe garage is pushing it a bit !

A few chips just to prove that it works,

and a work in progress of the control box that I'll wire in eventually. This will give me an actual rpm readout directly off the spindle using a cheap RPM module and hall effect sensor as well as controlling the VFD with some sealed stainless "halo" switches to keep out the oil and coolant.

Thanks for reading. Please share your own overhauls. It's so useful to have some comparative pictures when you're working a machine for the first time.

 

[Fergie 307]

I found a chap on ebay selling 3D printed ones for a variety of machines. I've fitted a set of his.

Clutch and brake:
The L5 has a spring loaded clutch and brake concentric with the upper pulley driven by the motor V belts.
View attachment 126420

In the first image, you see it top left. Down and right is a hollow shaft which is the main spindle. The spindle bore on an L5 is small. Under an inch. I won't be feeding any two inch bar stock through the head on this. In this picture, the change wheels are removed so that you can see behind them.

Following are simple pictures of the component parts once the round nut at the end of the black spring is unscrewed and everything unthreaded.
Tightening the nut increases preload on the spring makes it harder for the clutch to slip.
Moving the clutch lever top left of the headstock operates the clutch and mechanically disengages drive into the head.
View attachment 126412
View attachment 126413
View attachment 126414
View attachment 126416
View attachment 126415
View attachment 126417
View attachment 126418
I'd be interested in other owners views on the state of the friction surfaces as I've never seen these in "new" condition. I'm not down to the metal yet, but think this is a job that I'll need to tackle before long.

Unless yours is completely different to any I have seen before then unfortunately I think you are missing some bits. Do you know what the front of the pulley looks like If you remove the scored disc? I suspect you will find it has six holes around the inside face. These would hold the six clutch pins, which mate up with a fiction disc with six holes in it. Have you got a handbook for it? You can download one free as a pdf at vintagemachinery.org. If you look at the exploded diagram in the book yours is the older type clutch, the one that appears as the lower of the two sets of drawings of the two clutch options. The upper drawing is of the later high speed version. What you are looking at In your picture, with the two brass pins is actually the braking mechanism. Have a look at the drawings and compare with yours. I may have a spare disc, will have to have a rummage. Let me know when you've had a look, but if your pulley has the six holes then you are definitely missing the clutch plate and pins.

 

[Fergie 307]

Looked in my manual and seems to confirm what I remembered. You should have two sets of discs either side of the pulley. The ones with the two brass pins should be behind the pulley, with another set in front, which you appear to be missing. Not really a massive problem, I never really use the clutch on Mine as I have the lever on the front controlling the vfd, in the same way as your button box will. Only ever use the brake if I need to hold the chuck still, say to do a keyway. The brake on my vfd brings everything to a stop plenty quick enough.

 

[clogs]

A great read.....and a jolly good job.....another one saved....
I have a heavy lathe and also a Bridgeport J headed mill.....
The lathe was to low for comfort so I built a heavy stand/chassis with adjustable feet....
I also did the same for the mill but needed a raised duck board for that....
reason for the frames was so that both machines could be moved with a pallet truck on my own.....
The room was tight in the shop...also shared with Wadkin wood machines....
so for the bigger jobs it was a game of Chess to get everything in the right place.....
Thanks for ur trouble to show us the work done....always an interesting read....
partiq the gear repair...that kinda job always worried me....

 

[Sideways]

Cheers Fergie !
I do have a pdf of the manual. I will study the diagrams based on what you advise.
Looking at the part that has the rollers of the taper roller bearing still in it, I had assumed the friction material on one side was for drive, the opposite side for brake. Feedback like this is incredibly useful. Appreciate it !

 

[Fergie 307]

If you can get the drawing, and assuming that it's numbered the same as mine then the bits you appear to be missing are the two plates numbers 32 and 33, and the six pins number 34.

 

[Fergie 307]

The later machines have a separate brake inside the headstock. Long time since I have seen an old one but as I recall the brake and clutch functions are both performed by the discs either side of the pulley, depending which way you work the lever, there is no brake inside the head. So the ones you have showing, with two pins, are supposed to be behind the pulley, or maybe the pulley itself is on the wrong way round. The recessed section in the centre of the pulley should be plain on one side, and have six holes in the other side. The side with the holes is the front. This is where the clutch pins go, the pins I suggested you should check for wear. You must have wondered what I was on about as you haven't got any ! You can only put it back together as you found it, looks like someone has messed with it before. Very possible the clutch pins broke or whatever so they just took that part out. Have a look at the cross section of the clutch under maintenance. Confusingly, in my book anyway, the drawing is actually of the later type with two friction plates in front of the pulley, and none behind. Yours is very similar but only has one plate in front. Also looking at this drawing you should still have a cross pin as shown to the left of the collar D, part 26 on the exploded view. I will have a rummage through my spares tomorrow and see if I have any of the stuff you are missing.

 

[Sideways]

Here are the diagrams that I think Fergie and I are both looking at.
The gear head of the L5 complete (shows both my 9" L5 and the later 11" L5A)

Zooming in on the clutch and brake components

It does look like I'm one disk short from the stack in my picture a couple of posts above.



With use of a lathe and a mill, it may well be possible to replicate the missing metalwork if I can get a better photo or some dimensions for the parts.

The correct friction material may well be the biggest challenge.

This will keep me busy for a few days Fergie, thank you for pointing out the problem to me !