Face turning Bronze castings - strange surface finish

I am machining bronze castings for the Bolton Marine triple expansion. Normally I can achieve good quality surface finishes using insert tooling (DCGT 070202 FL K10). The parts I am machining now are the 3 lower covers.

For a reason I cannot determine the face surface finish has ripples about .125" apart with a depth of .00025" using a .003" cut. As the photo shows there is an interrupted cut. The insert is new and the "sound" of cutting is good and the top slide and carriage are both locked when cutting. The cross side is set correctly. I have power cross feed on my Super Seven and first thought that was the source given the regularity of the ripples, even if I had very successfully skimmed my face plate with automatic feed.

I repeated the process and used manual cross feed and that result is what you see in the photo. I have never Any suggestions for understanding the cause and solving this problem is very welcome.

(PS: I have not tried a standard HSS tool bit as yet.

Thanks

Have you tried a different insert? No mention of the origin of the inserts - that might be pertinent.

Suggest you try that HSS cutter and report back.

This is the insert mike is using.

Yamaloy insert

Steve.

Edited By Steviegtr on 18/03/2020 14:12:59

It may be just down to inexperience on my part, but I nearly always find it harder to get a nice finish with carbide tooling. I rarely have the same level of issues with HSS.

[This posting has been removed]

Doubt it as your link is for a different code.

Carbide inserts are designed to operate at speeds and feeds so that the metal/tip interface becomes so hot that the metal softens. Given the lower co efficient of friction than steel, this may be more difficult with copper based alloys.

My suggestion would be to grind up a HSS tool and use that for the finish cut.

Before going for the finishing cut, it might be worth checking that any drive belts are in good condition, and not sending any vibrations through the drive train. Single phase motors are also known to give similar problems if the conditions: speed / load/rigidity/tooling are not correct

As ever, the devil is in the detail!

Howard.

Try a new tip just incase the one was damaged.

Also swing the tool around say 10deg

Pip on the end also suggests you are slightly low if same tool was used to face the spigot.

Inserts

Well just tried again & got the exact same number as mike posted.

Try reading his post again, the chipbreaker ref is FL not AL, Probably Taegutec as thats a common lower price brand that uses that suffix.

Normally one would invoke the following equation:

Ebay insert = crap shoot

However the linked to Yamaloy insert (even if they're not the ones being used) are sold by Cromwell, albeit at a slightly lower price. But they're on 2 days which makes me slightly suspicious as Cromwell customers are most likely to be commercial machine shops, not used to waiting.

From the picture the ridges seem quite obvious, but they're stated to be a bit over 2 tenths of the thou deep. At that sort of depth I don't think they'd be so obvious? Not sure it has a bearing on the finish, but 3 thou is a pretty shallow finishing cut.

I've not really had a problem turning bronze or gunmetal with inserts so difficult to make suggestions. But the wide spacing makes one think of a machine related problem rather than the basic cutting process or, possibly, the casting itself?

Andrew

If you see that sort of rippled finish on longitudinal turning, especially by compound slide, it *can* indicate a little too much clearance in the slide gib, so my first thought would be to look for such in the crossslide gib. You may already have checked that, of course.

Edited By Mick B1 on 18/03/2020 17:00:22

Maybe a machine issue? Difficult to tell.

Tony

Yes, that’s what I thought. Less than £3 per tip in a packet of ten. Might even have been less, if bought from bang good!

Could be anything - like a loose insert for instance. Let’s await some decent feedback.

My prime suspect would be the insert. Bronze is fairly tough stuff. The rather shiny finish shown in the picture suggests a degree of rubbing is taking place, despite the "good" sound. I'd go for a well sharpened HSS tool with some nose radius.

I've seen that effect caused by drive belt thrash !

[This posting has been removed]

The pattern seems to repeat evenly, as the op said at ) 0.125 spacing. How does the insert know it has cut that much over and over again? If it were a random pattern maybe somehow but not a neat repeat.
I'd say 3 prime suspects.
1/ cross slide screw, bent, or bearing problem or even unevenly made but would expect that to be related to the pitch ie 1/10 in.
2/ drive to the cross slide with a changewheel or other gear that is off centre or even dirty or damaged on one tooth.
3/ Spindle drive due to belt or back gear or pulley or bearings though all of these would probably be making a more frequent pattern repeat. The bearing cage may precess with a much slower rate than the spindle itself. There are reasons why top end lathes avoid back gear and use incredibly expensive bearings.

edit - just re-read the fist post and see he did manual feed in the end so 2 & 3 discounted. I'm impressed he has managed such a good finish manually as slight changes to hand pressure and feed rate show up. When I hand feed I use two hands and can get marks where I change grip, not necessarily at exactly one turn.

Edited By Bazyle on 18/03/2020 17:34:38

I have experienced this before myself. Facing on the Smart & Brown model A, I was getting exactly that finish. Now I don't. The cross slide leadscrew was unevenly worn, especially in the middle and the nut was bad too. Backlash was possibly 0.025", maybe more. I bought a new leadscrew with two new bronze nuts on ebay which was sold as 8tpi, originally for some other make of lathe, but the threaded length meant it could be modified. Unfortunately it turned out to be 3mm pitch and the vendor kindly refunded me my money and said keep it. Eventually, I decided to use it on the imperial lathe, as there was little prospect of ever getting an imperial threaded leadscrew, especially a new one. The metric leadscrew now resides in the lathe with an additional modification using both nuts for superior antibacklash. It is set for 0.001" backlash over the 7" of movement. The new dial we made has 118 divisions instead of 125. The error per turn with 118 divisions is 0.0001102", which is not significant if you measure the size when getting close to the finishing cut.                                                                                                                                                                 Checking the gibs would be worthwhile as well as leadscrew nut backlash, plus leadscrew axial play.

Edited By old mart on 18/03/2020 19:19:43