Annealing HSS

[FulmenParticipant @fulmen]

So I’m working on a project that involves machining down a 48mm drill in both ends. I know, nuts. Well, actually it’s balls. For ball bearings I mean. Long story.

 

The rear solid shaft and taper cut without too much problem, it was tough but cut well (sorry no pics). So I figured I could just anneal the tip with the oxy torch, but it just keeps on fighting me no matter what I do. I started with dull red (daylight), then tried again and again until I almost melted the darn thing. And it still eats inserts for breakfast, I just can’t get it anywhere as soft as the shank was.

I’m sure the interrupted cut is part of the problem, but what can I do about that? Silver solder some inserts into the flutes perhaps? Would be hard to get the fit right though.

[Andrew CrowParticipant @andrewcrow91475]

Hi Fulmen, why not set up an angle grinder to grind the end of the drill flat.

You could rough it out on the off hand grinder first

Obviously you would need to protect the lathe bed with some covers.

Andy.

[duncan webster 1Participant @duncanwebster1]

Without very specialist equipment forget annealing HSS.

[Andrew CrowParticipant @andrewcrow91475]

I’m not sure if HSS can be annealed unless returned to a molten condition, but I’ll await correction. 😅

[Andy StopfordParticipant @andystopford50521]

I’ve brazed bits of HSS onto mild steel shanks to make lathe tools – no hardening treatment after being heated red hot, and they still performed just fine. I don’t know if it’s possible to anneal the stuff, but if it is, I suspect you would need special equipment and techniques.

[JasonBModerator @jasonb]

And even if you could anneal it you are then left with how to reharden.

If it is for a semicircular bottomed hole to drop your balls into then consider a TCT woodworking router bit they will cut metal too. You can get quite large cove cutters.

[Fulmen]

On 3 October 2025 at 21:32 Fulmen Said:

 

 

The rear solid shaft and taper cut without too much problem, it was tough but cut well (sorry no pics).

Isn’t this because they are made from a different grade of steel welded to the HSS section?

John

[Nigel Graham 2Participant @nigelgraham2]

Of (following Jason’s suggestion), a round-ended slot-drill; or make a form tool from silver steel or gauge-plate.

[FulmenParticipant @fulmen]

@Jason: Thankfully I don’t need to reharden it, I just need to have the end machinable without destroying the rest of the drill. I need to fit a ball bearing to the end, so while I could grind off the bulk I do need to hit some fairly tight dimensions.

I have managed to soften the steel a fair bit, but it’s still too hard. I even tried cutting it with HSS in case it was the interrupted cut that was the issue, but it didn’t last long.

My best idea so far is to put it into a bucket of hot sand to get an even slower cooling rate.

[noel shelleyParticipant @noelshelley55608]

Sounds like you need a Clarkson radius grinding attachment may be ? Or is a ball race ? Grinding would seem the best bet. HSS, it’s in the name, It’s not made to soften with heat so annealing is unlikely to work. Noel.

[Julie AnnParticipant @julieann]

Noel beat me to it; HSS will remain hard enough to cut when red hot. So a quick heat and cool cycle isn’t going to achieve anything. Here is a quote showing a typical annealing profile for M42 HSS:

It is recommended that M42 is annealed after hot working and before re hardening. Heat to 850°C – 900°C at a rate of no more than 220°C per hour. Hold at temperature for a minimum of two hours or one hour per 25mm of thickness. Cool slowly to 700°C.

Julie

[Fulmen]

On Fulmen Said:

I need to fit a ball bearing to the end,

I think we were thinking you wanted a round end to the drill. But as Noel mentioned is it just a flat end with a spigot to accept a bearing to make a bearing guided cutter?

[FulmenParticipant @fulmen]

It’s actually a grinder for plastic waste., but it all boils down to turning the end down to 35mm to fit a ball bearing. We don’t have any tool grinders or budget for outsourcing the work.

I know it can take more heat than normal steel, but it’s not impervious to heat. Silly me assumed that as long as I keep below the austenitization temperature (around 1200°C IIRC) it should only get softer, but I guess it’s a bit more complicated than that. Since the drill is already wrecked I don’t have anything to loose (other than time), so I guess I’ll just have to try a slower cooling.

[peak4Participant @peak4]

I lashed up a toolpost grinder a few years ago, using a Bosch POF350 router motor body.
The later POF500 will also work well enough and is arguably better made; both use a standard drill collar size for mounting.
Mount here is just a bit of alloy plate with a hole, a  sawcut, clamping screw, and bar attached to one side to fit in a toolholder.

Rough the job out with an angle grinder and finish to size with something like this.
They come fairly cheap on eBay or local second hand stores etc.

Here seen grinding out the splines to a larger internal diameter in a motorcycle chain sprocket

Bill

[jaCK HobsonParticipant @jackhobson50760]

I guess you need to cool very slow – so a heat treat oven would be best.

Try heating up a massive bit of scrap metal red hot and bury it, with the red hot drill, in vermiculite or other high temp insulation.

I think the drill on its own, even in vermiculite, might cool too quick.

[John HaineParticipant @johnhaine32865]

IIRC Ivan Law (?) had instructions for tempering HSS in his book on screwcutting – softening HSS blanks to machine precise angles.

[Craig BrownParticipant @craigbrown60096]

Not of much help im afraid, but for your perusal. I recently did some rust removal on this very rusty drill bit. After removal from the solution you can quite clearly see the transition to HSS

[Julie AnnParticipant @julieann]

The professional advice seems clear; it is important to anneal using a slow rise to maximum temperature as well as as slow cooling. So why waste time implementing only slow cooling?

Julie

[FulmenParticipant @fulmen]

1: Because I don’t want to anneal the entire drill.

2: Because it’s the best I can do. I have an oxy/acetylene cutting torch, that’s it.

Now I can live with some loss of hardness, it’s not going to cut steel anymore. But I’d rather not have it as soft as the tang was. And if they managed to get the tang soft without destroying the entire drill why can’t I do the same to the tip? I know, they are experts while I’m an idiot with a blow torch.

As for the slow heating it is mentioned in several articles for full annealing and austenitization, but not for the temper. I don’t know if it’s assumed for all heating or if it’s not really needed. They all mention the need for slow cooling in the 800-1000C range.

One source listed the soft anneal temp to 750C (standard temper is around 550C), which I have reached for sure. But they also recommend 2 or even 3 cycles with significant soak times, so it might be as simple as that.

 

@Craig: Interesting. Can you notice any significant difference in hardness around this line?

[Fulmen]

On Fulmen Said:

…

I know it can take more heat than normal steel, but it’s not impervious to heat. Silly me assumed that as long as I keep below the austenitization temperature (around 1200°C IIRC) it should only get softer, but I guess it’s a bit more complicated than that. …

Annealling HSS is very difficult without knowing the alloy’s profile and having a precision oven.

Julie explained what has to be done to anneal M42 in her #818662

It is recommended that M42 is annealed after hot working and before re hardening. Heat to 850°C – 900°C at a rate of no more than 220°C per hour. Hold at temperature for a minimum of two hours or one hour per 25mm of thickness. Cool slowly to 700°C.

So, your drill needs to go into an oven at 20°C, where it’s slowly heated to 900°C over about 4 hours.  Then held at 900°C for at least 2 hours, before being cooled slowly to 700°C.   The description is incomplete, but Julie’s snip shows annealing M42 takes at least 6 hours with a programmable oven! (This of course is intended to leave the alloy in good condition, not burnt, or internally shattered.)

Worth trying, but don’t expect meaningful softening with a haphazardly applied torch.  HSS is engineered to withstand prolonged high-temperatures.  It can be cut with carbide but it puts up a fight.  For that reason HSS is usually ground.

Most drills and saws are made from two alloys welded together.  The cutting part is expensive and somewhat brittle HSS.  The support is made of a tougher alloy, which is cheaper and absorbs shocks.  Most drill shanks are much softer than the bit.

What might work is a cheap Carbon Steel DIY store / woodworking drill.  Carbon steel anneals, hardens and tempers in the ordinary way.

Dave

 

[FulmenParticipant @fulmen]

Well, you’re no fun. Where’s the “can-do” and “hold my beer” attitude?

My first flawed assumption was that the shank was the same material as the tip. The other one was that “it’s just steel”. Usually temperature is enough to “get the job done” without significant soak times, but not here apparently. Seems I need a couple of hours at 750-900C to really soften it, and I just don’t have the equipment for that. And even if I did it would be a full anneal to the whole part that would soften the shank excessively.

I have reduced the hardness to a kinda-machinable state, and that might be enough if I didn’t have interrupted cuts. Especially if I can rough out the bulk with a grinder. So what do I do? I could:

1:  Just throw a pack of carbide inserts after the problem.
2: 1+ silver solder in some steel to get a uninterrupted cut.
3: Tie a angle grinder to the tool post and finish with carbide.
4: Spend a bit of time making a toolpost grinder from a pneumatic grinder.
5: Spend even more time building a TPG from a router.

Giving up isn’t an option just yet.

[noel shelleyParticipant @noelshelley55608]

Hi Fulmen, As I made my toast and thought as to what your intended purpose was for the drill, If I understand correctly, had you thought of using a slab milling cutter ? This could be mounted on a shaft between 2 bearings, could be removed for sharpening or different sizes could be used. It could be used in a small old horizontal mill. These can be bought second hand or you may find a machine shop that might give you some old ones. Just a thought. Noel.

[FulmenParticipant @fulmen]

Part of the reason for choosing the drill was the length, it needs to be around 200mm. And we didn’t have to order anything costly. But drills like this aren’t cheap to replace, so it would be nice if we didn’t break it for nothing.

If the drill doesn’t pan out I have a couple of other ideas, one being making a stack of cutters made from water cut hardox 450. It’s not as hard as tool steel but it should work on plastics. Another would be to machine it from something like EN24 with spring steel blades bolted on. We have some old circular saw blades that should be hard enough to work as blades.

[Chris CourtneyParticipant @chriscourtney72250]

Annealing High Speed steel is possible. As has been said previously the annealing temperature for most of the common high speed steels including M2 and M42 (high Cobolt grade) is between 870 and 900 deg C. It will need to be held there for a short while and then cooled very slowly. The cooling rate down to around 500 deg needs to be slower than 20 degrees per hour. This should result in a ferritic structure with finely distributed carbides. This should give a Brinell hardness around 250 HB which is fairly easily machined.
The problems are maintaining the slow cooling rate and preventing surface decarburisation.

Hardening again isn’t too difficult, heating to around 1200 degrees C and quenching. This is quite close to the melting point of the steel, and again surface decarburisation is an issue.

Annealing without a temperature controlled oven is pretty challenging, using a torch will probably result in it cooling too slowly and re hardening.

The actual metallurgy of High speed steels is very complicated and usually includes tempering cycles after hardening to improve toughness.

[noel shelleyParticipant @noelshelley55608]

The march of time has seen the shaper/planer fall by the wayside and the horizontal mill is now seldom seen. I have slab mills that are about 3″ dia and may be 3″ wide there are bigger (wider ) ones available – 2 of these would give the required width of about 200mm. Noel.

[Author]

Posts