Hi all,
I'm new to RYPN. I'm a volunteer on the steam crew at the Orange Empire Railway Museum in Perris, CA. As is the nature of steam locomotives, it seems that I spend a good amount of time lapping and repacking globe valves and needle valves on the locomotive.

I've had some questions floating around in my mind for a while, and I did a search of past RYPN topics, which turned up some interesting discussions but didn't completely answer my questions. I'm sure that some of you can probably shed some light on the following points:

1.) When placing new packing around the valve stem of a globe or needle valve, I have heard different opinions about how to cut the packing. Some people say to cut several strips of packing that are just long enough to make rings around the valve stem, then stack the rings with the seams in different positions. Others say that it is fine, or even better, to wrap a single long piece of packing around the valve stem in a spiral. I typically do the latter option, as it seems to me that the packing will seal better with no seams to permit leakage. When I do it, I wrap the spiral of packing such that it will be wound tighter, not looser, by the action of tightening the packing nut. What are your thoughts on the matter?

2.) How many layers of packing are ideal? Is it just a matter of "Get as much in as will reasonably fit"?

3.) When lapping valves, what grit do you typically start at? One old poster recommended starting at 240 grit to see what the surface looked like, then proceeding accordingly from there.

4.) How do you know when it's time to switch from one grit of lapping compound to the next finer grit when lapping a valve?

5.) Can you recommend any books to read on these topics? I wish there were an ICS book on valve repair, but I don't think they ever treated that subject....

Thanks a lot for your help!
Dan

Welcome to the craft. This is a job that will always need doing, so it's a good thing to find your own best way to do it. Here's what worked for me:

Build a set of forming tools like steel shafts which replicate the shape you want the valve seat to be and bushings to hold them central and perpendicular to the valve itself. If you get a deeply cut seat, this combined with an electric drill will make it much faster and easier. I also made bushings to replace the bonnet with something not threaded so I could manually rotate the valve itself against the seat in the correct alignment. generally, I like to start by forming the valve before working on the seat. This can be done in a lathe. Once the valve is in good form, lightly lap it using fine compound to show you where and how deep the cuts in the seat are. If light damage, lap the valve to the seat with the reasonably least coarse compound likely to work out the garf - then go finer for a good smooth finish. If deep, go to the forming tool and coarse compound. The steel won't wear as fast as the bronze, so the tool will last a long time and cut the seat well. Once the seat starts to get close to smooth, go finer until you get a good finish. Clean between lapping sessions, and always when going to a finer grade. What to start with and when to change is something you will figure out on your own after doing a couple hundred of them. You want to remove the least metal consistent with making a good seating surface efficiently.

Of course, this is for a limited fleet of similar locomotives so there's a limited number of similar valves to make lapping tools to fit. Few of us run a large and diverse fleet of steam any longer.

I like to stack packing rings cut to fit and use a little valve oil when installing them. Stagger the joints and put in as many as there's room to apply. When they compress, put in more. If the packing can't be compressed it won't seal. Letting packings blow destroys them, so leave your engine crews some margin to tighten as needed.

What to use? We can't get graphited asbestos rope any more. I generally stick with graphited something rope. Not sure how the white teflon stuff lasts in a steam application. It would be interesting to compile a list of what is used under what conditions here.

I throw my 2cents on the material.

I would recommend PackMaster #6 from Garlock

catalog here:

https://www.google.com/url?sa=t&rct=j&q ... 1273,d.amc

It is available in a variety of square sizes that fit in most valves. It has a good shelf life and since it is self lubricating does not need to be oiled.

If you have a blast cabinet available that uses glass beads, 120 grit preferred, I would recommend that you disassemble each at time of annual, blast all the parts for inspection and when you reassemble use a nickle based anti seize and lubricate all the threads and metal-on-metal mating surfaces including the stem and the back side of the gland.

Cutting separate rings with beveled ends works the best. The packing will tend to shorten in service and if you wrap the stem with one piece it will tend to leak more as the packing takes longer to find its home.

If you have access to a table saw you can make a grooved board to hold the packing as you cut the rings. use a modelers razor saw to cut a cutting guide at a 45 deg angle for each length. Use a sharp single edged razor blade to cut the rings.

It usually takes about an hour to complete a valve. Not this does not include any lapping. Since we have converted all of our valves to Nibco T276 and T376 AP (Alloy Plug) valves we don't do any lapping.

Remember everything wears out. There is a lifespan on every valve and fitting. When that comes into question, replace it.

Hi, helpful advice already on here but some extra points to numbers 3 and 4;

3: Depending on the condition of the valve. if the valve is heavily worn or damaged you will need to use a coarser grit to remove the damage and then work your way to a finer grit till you get the finish you require. We would recommend this lapping compound

4: When the face your lapping has completely cleaned up by which I mean is the same finish equally all over the part.

I have no idea about that product - I use old dented tin cans with rusty tops full of gray gritty mud ordered from McMaster. One can lasts for a very long time.......

If you have a question about whether both faces - seat and valve - completely seal, you can test them with Prussian Blue. One tube should last a lifetime. You use a very small amount to thinly coat the valve seating surface and assemble and close the valve against the seat - then you take it apart again. If there's a good transfer of the color to the seat all around, you are good to go. You can also test the valve with compressed shop air.

be careful lapping valves of any type. Lapping does not take the place of proper machining. If the seat is bad it should be faced to the proper angle or flat in a mill or lathe. The globe /taper/ whatever is on the stem is also machined. Lapping is only done for the final seating.

The problem with lapping to replace machining is that lapping will just replicate the already errors in the stem and seat. Compound will follow what is already in the seat, not cut a new angle or make the angle better. Compound can not possibly be applied in an absolute even coating as well. The only way to renew a seat is with a seat cutter or even seat grinder.

In the Santa Fe steam folio in discussing throttle valves it is stated that lapping is not to be used for renewing valve seats and that the proper cutters are used for this task. Lapping is for the final seating only.

Be careful with Teflon packing, in some steam circumstances it will extrude out.

Which reminds me - does anybody still make steam valves with replaceable valve ends and seats? Those make this work very easy - you can take the replaceable parts out and put them in the lathe in minutes, and when they are worn out you just install new ones.

We have a whole set of cutters and alignment bushings for all sizes of valves. No different than a hardware store cutter for old rubber washer seat water valves.

I would agree with other comments that getting the seat geometry correct should be task one, and then lapping is to finish the sealing surface.