What's new
What's new

Trying to get an Ra 32 finish without grinding...

DanASM

Hot Rolled
Joined
Mar 11, 2019
I have seen some parts lately that call for a Ra32 finish (4140). I cant find any info on improving surface finish with scotch brite pads or emery cloth.

The parts are not aerospace so I believe the finish is more cosmetic. I do not have a profilometer and I do not know if the customer does either. I know I can make the finish look acceptable from polishing but am not sure if a profilometer would agree with visual inspection.

Most things I read online have cemet inserts with cnc machines. I have none. Just manual lathes/Brownies. I want to know if anyone has experience trying to polish by hand with good results.

I have seen some of these automotive parts over the last year have some Ra32 callouts. If its on the O.D. its no big deal as I can have it centerless ground.

I am not sure if I should be passing on these jobs or if I should have at it. Any advice would be greatly appreciated.

Most of these parts would be 2nd op on a DV-59. I do have a grinding wheel attachment for a lathe but I have never seen it hooked up. I dont know if I can get it to mount on my DV-59 effectively.

Thanks in advance.
 
Yeah as long as you have your cutting speed up high enough you can pretty easily turn a 32Ra finish. If it's not quite shiny enough give it a quick spin with lightly oiled brown scotch-brite and it will look pretty damn good.
 
Yeah as long as you have your cutting speed up high enough you can pretty easily turn a 32Ra finish. If it's not quite shiny enough give it a quick spin with lightly oiled brown scotch-brite and it will look pretty damn good.

I see the 4140/42 comes in Annealed (cuts like 1018), pre heat treated 28/36, and then HT. People have better finishes in the harder materials with cemet (?) inserts. I dont use carbide much or any inserts in my machines. Mostly all HSS, some brazed carbide sticks here and there.

I like to machine Annealed, Cold Drawn, and use sharpe HSS tools.

It seems like I should run harder material to get a better finish with a different rake.

There is a lot of 4130/40/42 work I can get for my machines but I havent run much to get a good idea about it. It seems like just run it a little slower than normal with same tooling should work fine.

I may have to do a finish pass on my 2nd op with a carbide stick tool. Cranking handles. At least I wont have to send it out for grinding.
 
If you are running HSS, use the opposite. Slow it way down and take a very light cut with water based coolant. The chips will come off like foil and you'll get a very fine finish. I've gotten an 8 Ra easy that way before. Use a broadnose tool and kick the feed up if time is important. Crucial to keep the depth of cut light for best finish.

We had a thread about this just recently but I forget where it went. After hearing about this one of the guys here wanted to try it and made a video showing a pretty respectable finish.

BTW, 4140 is usually sold either annealed or quenched and tempered (aka pre-hard or "HT"). No third option. Annealed is pretty darn soft, HT - or Q&T - is the one that's usually in the high 20's to low 30's Rc. You want it harder than that, you pay someone to HT it.
 
Surface finish is nothing to do with how shiny it is. It has to do with nose radius vs feed rate, and how high the scallops left behind are. Which is average roughness. One other poster up top is right with the finish calculator.
 
Well I do not have the luxury of a constant feed rate due to cranking handles. No water based coolant either.

I get the nose radius and feed rates situation. This part doesnt make sense for me to take on it looks like. Too many problems could arise.

Thanks for the heads up guys
 
I see the 4140/42 comes in Annealed (cuts like 1018), pre heat treated 28/36, and then HT. People have better finishes in the harder materials with cemet (?) inserts. I dont use carbide much or any inserts in my machines. Mostly all HSS, some brazed carbide sticks here and there.

I like to machine Annealed, Cold Drawn, and use sharpe HSS tools.

It seems like I should run harder material to get a better finish with a different rake.

There is a lot of 4130/40/42 work I can get for my machines but I havent run much to get a good idea about it. It seems like just run it a little slower than normal with same tooling should work fine.

I may have to do a finish pass on my 2nd op with a carbide stick tool. Cranking handles. At least I wont have to send it out for grinding.

If you can get a lot of work why not get a proper Gauge, then you will know for shure.

https://www.mscdirect.com/browse/tn...rators/Surface-Roughness-Gages?navid=12107998
 
If you can get a lot of work why not get a proper Gauge, then you will know for shure.

https://www.mscdirect.com/browse/tn...rators/Surface-Roughness-Gages?navid=12107998

"MSCdirect.com is currently not available for your use."

So I could be mistaken about what you linked to .... but there is a low-cost method. You can buy nickel-plated roughness sample plates that have a variety of surface finishes. They used to be pretty cheap. It's not a $4,000 gage but better than guessing.

You can use your fingernail to compare, but for max sensitivity use your front teeth, like a bunny. Your teeth are more sensitive than your fingernails.

Seriously :D

Agree that surface finish and shininess are not the same, what I found was that you tailor what you do to the customer. The ones that mean "shiny" when they put a small number on the print get shiny, and the ones that know what they are doing get what they put on the print. Sometimes you have to interpret these things ...
 
While it is true that the calculators for nose radius and feed rate are very useful, they don't always tell the whole story. You still need to be making it "shiny" or you're likely producing a torn finish. This does show up on the profilometer as a significant difference in Ra. For steel, "shiny" is also significantly more resistant to rusting in my experience.
 
"MSCdirect.com is currently not available for your use."

So I could be mistaken about what you linked to .... but there is a low-cost method. You can buy nickel-plated roughness sample plates that have a variety of surface finishes. They used to be pretty cheap. It's not a $4,000 gage but better than guessing.

You can use your fingernail to compare, but for max sensitivity use your front teeth, like a bunny. Your teeth are more sensitive than your fingernails.

Seriously :D

Agree that surface finish and shininess are not the same, what I found was that you tailor what you do to the customer. The ones that mean "shiny" when they put a small number on the print get shiny, and the ones that know what they are doing get what they put on the print. Sometimes you have to interpret these things ...

I have a roughness sample plate. It would still be up to me to determine by vision if it is good or not. I have seen profilometers contradict what the eye says is good. I do not want these issues to come up so I will not put myself in that situation.

I figured I could do a finish pass and skim .005 with a carbide brazed stick. Then polish with scotch brite or emery cloth.

If the finish still isnt good enough at that point it would not be profitable. It may take too long to do the skim pass and polish just in cycle time to make the part affordable anyway. Someone with a cnc lathe would bang this out easily with no headaches.
 
It would be nice if you could swing down a Supfina attachment, maybe from the back of the cross slide ?

Drop it on there and let it polish for 15 seconds or so as the spindle rotates slowly, flood coolant.
 
It would be nice if you could swing down a Supfina attachment, maybe from the back of the cross slide ?

Drop it on there and let it polish for 15 seconds or so as the spindle rotates slowly, flood coolant.

I do have a grinding wheel attachment for lathes. I pulled it out 6 months ago and didnt see how I could mount it. If I remember correctly I need to make a mounting plate for it. I will have to pull it out again and take a look at it. It seems like I can get some good use out of it if I can get it mounted right.

Where there is a will there is a way.

This part has 2 center drill marks on the ends for 2nd op I imagine. Its a 4.5" long part. Probably chucked up like that for the grinding/finish pass. It says the centering holes are optional but were put on the print years ago for a reason. Someone probably made this part the same way I am quoting it.
 
I do have a grinding wheel attachment for lathes. I pulled it out 6 months ago and didnt see how I could mount it. If I remember correctly I need to make a mounting plate for it. I will have to pull it out again and take a look at it. It seems like I can get some good use out of it if I can get it mounted right.

Where there is a will there is a way.

I did NOT say "Grinding Wheel".

I said "Supfina".

Google is your friend.

FWIW they appear to be local to you (somewhat)
 
I think you may be overthinking this

A 32 finish is a nice finish, but does not require grinding. I personally would think a 16 would.

If you are turning with a decent nose radius and a fine feed, providing your machine is not imparting any glitches[like if you were using the compound manually with no finesse, you should meet a 32 with at most a quick hit with fine emer

I mean a 63 is a standard machined finish
 
Just a few thoughts.
Is the drawing per ASME / ANSI Y14.5? If so the Surface Texture (not surface finish) is per ASME B46.1
This can be a complicated subject if you read the whole spec.
Are their and modifiers in the callout MAX? LAY? TURN,GRIND,HONE? Sample length? Sub surface imperfections not allowed?

32Ra is not very specific, which is good for manufacturing flexibility. I would compare with a tactile specimen using a #2 pencil.
 








 
Back
Top