View Full Version : Lapping

Paul Alciatore
11-14-2005, 11:46 AM
In the topic on Helical Gears someone mentioned lapping. I have long wanted to ask some questions about lapping.

I know that two surfaces can be lapped together with a fine abrasive. Are there other processes that are called lapping that do not use abrasives?

I also heard some express concerns about being able to completely clean up all the abrasive after lapping. Is this really a problem if we are just talking about two simple parts that can be easily cleaned as opposed to a mechanism that has a lot of places for the grit to hide? Does the grit embed itself in certain metals? In all metals?

Why is scraping recommended over lapping for fitting machinery? For making flats and straightedges?

Paul A.

Forrest Addy
11-14-2005, 12:43 PM
Lapping is the worse thing you can do to a pair of gears. The lapping concentrates where rubbing velocity is highest - at the tips and roots of the teeth. The pitch cylinder has nearly zero rubbing velocity and thus laps with difficulty. Lapping of gears in place is the surest way of making them noisier, less accurate, and shorter lived: its effect is to wear them out.

Lapping of gears has to be done under controlled conditions preferably in a ger lapping machine with a rapid cyclic axial displacement as a part of the process.

As you mentioned, after-lapping clean up is a vital step. No matter what elaborate precaustions are taken, lapping gears in place almost always requires a "last part" disassembly of the transmission then a thorough cleaning, flushing with clean solvent, followed by reassembly.

In my experience lapping is what is done in despiration before giving up on a noisy transmission and replacing gear sets.

11-14-2005, 12:58 PM
Hi Forrest,

I believe this is in response to a comment I made in another thread regarding the Bridgeport Model 3 right-angle head. Per the catalog, the spiral bevel gears in that assembly were hand-lapped for proper fit.

I assume this was done with each gear individually in a lapping jig, not by lapping one gear against the other.


11-14-2005, 01:03 PM
No expert here, but...

My understanding of lapping is that it's properly used to make a hole round and to-size (internal lap), a shaft round and to-size (external lap), or a surface flat (lapping plate). It's a controlled process, where the process is designed to generate the final result desired.

The idea of introducing lapping compound between two random moving parts (like gears) and hoping that the Right Thing would happen, never struck me as a very likely proposition.

11-14-2005, 01:25 PM
I knew guys who manufactured leadscrews and geartrains for measuring devices. Their machines were used to determine dimensions of those mysterious objects in Cuba which showed up on photos taken from U-2s. Turn the crank X times, that's Y microns, etc. They used to lap gears by running them together for a month or so with lapping compound on the teeth. I never saw them do it, though - they might have been feeding the junior engineers a ration of hooey.

I've lapped brass shafts to fit holes in zinc parts with no added abrasive, just oil.

I wouldn't put too weight on a Bridgeport advertising blurb. But if the gears were indeed lapped in some sort of jig, then mixing gears & pinions shouldn't be a problem, as they'd all be lapped to the same dimensions.

There's considerable sliding morion in a spiroid gear - not at all like a spur gear. Lapping a spiroid gear and pinion together might possibly work.

11-14-2005, 02:06 PM
Scraping and lapping are very different processes, imo not usually interchangeable. Like any good process to remove material, you the operator has to be in control. Using two mating parts with abrasives in between does not make things more accurate, is simply introduces/increases slop! I don't even know that that is called lapping - there is no control there. In my shop, where you can get at things to compare (split bearings, dovetails, etc) I use scraping, lapping is for cylinder bores where there is no way to access with a reference surface.

lapping is done against a lap or lapping plate - this is a (somewhat) sacrificial item. In the case of flat work, the control comes from keeping the lapping plate flat, in cylindrical work its the wear, feel and adjustment capabilities of the lap. All my homemade cylindrical laps are clad in copper. The abrasives become embedded and the copper quickly seats in. you get huge sensitivity on where things are tight and an very accurate bores are produced. I’ll try to post some pics of some homemade laps when at home. a great advantage of cylindrical lapping is that, if doen properly, it makes the bore very accurate - round and of the same dia - vs honing which is just affects finish.

you don't want to lap softer materials such as AL or bronzes as the abrasive becomes embedded, but CI or steel is fine. I've lapped just about every bore I’ve done it produces a superior finish. Having said that, I believe there is a lapping compound designed from bronze that breaks down and doesn’t embed – anyone know more about it?

scraping on the other hand is localized removal of material by carefully comparing - either against a mating part of reference such as a surface plate. For example, scraping one half of a dovetail to reference flats and then using it as a reference to scraping in the other half – this only works if you first scraped in one side to a reference.

You sometimes hear of the uninitiated making up hybrid of the two – throw in a little lapping compound and ‘seat’ the parts together. The only place I can see this working is a cone – like a valve. For dovetails or bearings all it does is create more clearance – it does nothing to bring things into flatness or alignment

[This message has been edited by Mcgyver (edited 11-14-2005).]

Paul F
11-14-2005, 03:26 PM
I would imagine that the application of grinding compounds to to mating, moving, parts is only a positive when the "problem" you are trying to solve is caused by those two parts having "high spots" that need to be worn down to allow full contact between the parts...

WIth two gears, I don't see a way to keep "good" teeth from being worn just like "bad" ones are though.

I've used fine compounds in gun parts... really only to "lay" the finish in the direction of of the movement of the part to make for smoother movement though.. not to try to "knock off" any high spots or improve a "fit".

Paul F.

11-14-2005, 04:43 PM
<font face="Verdana, Arial" size="2">Originally posted by Paul F:
I would imagine that the application of grinding compounds to to mating, moving, parts is only a positive when the "problem" you are trying to solve is caused by those two parts having "high spots" that need to be worn down to allow full contact between the parts...

Paul, I think you've fallen into the mental trap on this stuff - with scraping you get to identify the high spot and remove it. with lapping compound between two mating parts (as opposed to between a part and a lap) you are going to partially be removing the high spot and partially creating a low spot - the result may seem smooth action but that’s because you've built up clearance (or slop) not improved straightness/flatness/roundness etc

Mind you, I'm thinking bearing surfaces etc like a dovetail....it may be that in certain mechanism's that are small or inaccessible to a lap or scraper, creating a bit more clearance by abrading the parts together is exactly what the doctor ordered but still i think you are creating play rather than accuracy.

11-14-2005, 04:58 PM
Lapping done correctly is capable of producing far more accurate results than any other process. It is the method for hand finishing telescope mirrors to accuracies of a fraction of wavelength of light. With the right tools and abrasives as well as technique any level of accuracy can be obtained for certain tasks.

There is no generic method of lapping or generic abrasive or lap to use. It depends on the task at hand. To grind a mirror a pitch lap is used with cerium oxide in the final stages. To lap a tool a brass lap may be used with diamond dust.

<font face="Verdana, Arial" size="2">
you are going to partially be removing the high spot and partially creating a low spot</font>

Quite the opposite, lapping three surfaces together in rotation is used to produce truly flat surfaces.

Paul Alciatore
11-14-2005, 06:07 PM

Yes, my only real experience with the process is with telescope mirrors. But there you have a very hard substance, glass or pyrex or some other special glass and I think there is zero chance of any particles of the abrasive being embedded in that surface. Also, the mirror or lens in use does not slide against anything so any embedded abrasive would not be detected. Well, not that way anyway. It would degrade the image by scattering light. But is seems to be a non problem in that area.

I would not expect two surfaces, as in a dovetail slide, to become flatter if abrasive were introduced between them. They would become the same shape, but not necessairly flat or even straight. But I could see three surfaces that were lapped together in successive pairs 1-2, 2-3, 3-1, etc. and with proper random rotation, getting flatter as well as smoother. Just like you would compare three surfaces, not two, if you were scraping a flat from scratch. Optical workers do this with three blanks to make optical flats from scratch. The final polishing would be done with a pitch lap, but the initial work would be grinding (lapping) between the three blanks.

I'm not particularly interested in gears. I am not sure that this would even work on spur gears as the contact does not have that much sliding motion to it. It would not work well with a cylinder or rod in a bore because, as others have stated, it would only create more slop in the fit. Perhaps just a little if the fit has rough spots.

I am more interested in the process in general with any limitations or cautions that may go with it. Advantages vs drawbacks. Why would you choose to lap vs scrape? Or scrape vs lap?

Paul A.

11-14-2005, 06:14 PM
Evan, you are right, but your point is irrelevant to mine – perhaps I wasn’t clear enough but I don’t think I ever discarded proper lapping as an inaccurate technique. I've done a reasonable about of it as well as scraping. What I tried to be clear about was subject of the use of lapping compound between to mating parts. this is entirely different from lapping where a lap is used. Accurate results from lapping are not a fluke - its controlled by the lap, and that is not the case in abrading mating parts.

i don't really consider applying abrasives between two bearing surfaces to be lapping, ie there is no lap or control.

and to your quite the opposite, I raise you one quite the opposite. While the lap does its work and removes a high spot, part of the lap is also abraded (the low spot) While i understand this is not 1:1 with a copper/brass lap against ferrous, the context of the statement was in using lapping compound between mating parts, where if it was CI it would be pretty much 1:1

Also, can you really generate a flat from three surfaces via lapping? it makes perfect sense when scraping because you can control what you remove from what surface, but i see it being perpetually chasing your tail with lapping, while i don't speak from experience on this point, logically i don't think it can be done

[This message has been edited by Mcgyver (edited 11-15-2005).]