Daniel's been out in his workshop again, marking inventory across what sounds like half the periodic table. Aluminum, stainless, wood, MDF, HDPE. He wants to zoom out from the carbide deep-dive we did before and map the whole landscape of engraving bits for handheld rotary tools. The geometries, the compositions, and how to pick when you need a legible, durable mark, not a work of art. He's running a Dremel Lite 7760, and he's skeptical of those twenty and thirty bit starter kits. Wants to curate a small set of quality bits for three jobs: ultra-fine lettering on PCBs, wider lettering on HDPE inventory boxes, and something in between.
So let's start by mapping the landscape of engraving bit geometries and what each one actually does to a surface.
The thing that strikes me first is that engraving bits are sold as if they're universal. One bit, any material, go. And the moment you try that on a bench with six different surfaces, you learn the lie. Aluminum galls, stainless laughs at you, HDPE melts and re-welds behind the cut, and MDF just turns to dust in a way that hides your line.
The material changes everything about how the cut happens. Hardness is the obvious variable, but chip formation matters just as much. Wood and MDF produce chips that clear easily. Aluminum produces chips that want to stick to the cutting edge. HDPE produces a chip that's really a melted ribbon, and if it doesn't get out of the way fast, it fuses back into the groove you just cut. So the geometry that works beautifully on maple will ruin a polyethylene box.
And then there's the tool itself. The Dremel Lite 7760 is a four volt tool. Eight thousand to twenty-five thousand RPM, one-eighth inch collet. You've got speed but almost no torque. The quick start guide recommends the number one-oh-six engraving bit, and it says right there in the manual that the tool stalls if you press too hard. Stainless is basically off the table at that power level.
Which means the real question isn't which bit is best. It's which geometry and which carbide grade, at what speed and what pressure, for which material. Three variables, and most people only shop for the first one.
Right. And the bit itself has more going on than most folks realize. You've got the head shape, the flute count, the helix angle, and the carbide grade. Four decisions hiding inside a thing that looks like a tiny metal toothpick.
Let's get into the weeds on geometry and grade, because that's where the real differences live. Head shape first. The fundamental shapes are cone, ball-nose, and pointed or diamond. A cone bit, like the classic one-oh-six, has a flat bottom and cuts a V-groove with a flat floor. That flat floor gives you a clean, wide line, which is exactly what you want for legible lettering on plastic or wood. The V-shape means the line gets wider as you go deeper, so you control stroke width with depth.
So with a cone bit, depth isn't just about how deep the mark goes, it's also about how bold the line reads. Shallow pass, fine line. Deeper pass, heavier stroke. Same bit, two different visual weights.
And that's a feature, not a bug. It means one bit can do the work of several if you're disciplined about depth control. But it also means that if your hand wanders, your line weight wanders with it. The bit doesn't hide your inconsistency.
And the ball-nose cuts a rounded channel. Why would you want that on metal?
Two reasons. A rounded groove avoids sharp corners at the bottom of the cut, which are stress risers. On aluminum or steel, a sharp V can initiate cracks over time, especially if the part flexes. And a ball-nose produces a smoother, shallower mark that's easier to control on hard materials where you can't go deep anyway. It's the bit you reach for when you want a mark that won't weaken the part.
That stress riser point matters more than people think. I've seen aluminum brackets fail along an engraved line because somebody used a sharp V-bit and cut too deep. The mark became a crack initiation site. It's not just cosmetic.
Right. And on anything that flexes or carries load, that's a real engineering concern. A ball-nose distributes the stress across a curved profile instead of concentrating it at a point. It's the difference between a mark that weakens the part and one that just labels it.
Then the pointed or diamond bits. Those are the fine detail tools. Tiny included angles, tiny tips. They're what you use on a PCB where you're cutting through copper and you don't want to dig into the fiberglass underneath.
And they're fragile. The tip on a really fine pointed bit is measured in tenths of a millimeter, and it does not forgive wobble. On a handheld tool with any runout at all, you can snap the tip before you've finished your first trace. Light touch, high speed, and accept that you'll replace them.
So head shape sets the cross-section of your groove. What about flutes?
Flute count matters more than most people think, and it's where a lot of starter kits go wrong. A single-flute bit has one cutting edge and one big open channel for chips to escape. That aggressive chip evacuation is critical on HDPE, because the plastic melts from the friction of cutting, and if the melted chip doesn't get thrown clear immediately, it welds itself back into the groove. You end up with a ragged, lumpy line that looks like it was carved with a hot nail.
Which is what happens when someone grabs a four-flute bit off the rack and runs it on a polyethylene box.
Multi-flute bits give a smoother finish on metals because you're taking many tiny cuts per revolution instead of one bigger one. But on soft plastics, those extra flutes just clog. The chip has nowhere to go, it packs into the flute, the bit stops cutting and starts rubbing, and the rubbing generates more heat, which melts more plastic. It's a death spiral.
So for a low-torque handheld tool on mixed materials, single-flute is the safer default.
Most of the time, yes. The tradeoff is that a single-flute cuts more aggressively per revolution, so on a low-torque tool it can actually stall more easily if you push. But that's a pressure problem, not a flute problem. Let the single flute do the work and it clears its own path.
Now the carbide grades. This is where Daniel's earlier question about composition comes back. C2, C3, C4. What's actually different?
It's about grain size and cobalt content. C2 is the standard grade, typically six to ten percent cobalt binder with a coarser grain. The cobalt is the tough part that holds the carbide grains together. More cobalt means tougher, more resistant to impact and chipping, but it wears faster. C2 is the forgiving grade. It'll take abuse, shrug off vibration, and keep cutting even when your hand isn't perfectly steady.
And C4 at the other end?
C4 is micrograin. Very fine carbide particles, less cobalt, much harder and more wear-resistant. It holds an edge far longer, and it can be ground to a sharper point. But it's brittle. On a handheld tool with vibration and variable pressure, that brittleness is a real liability. Any wobble, any interrupted cut, any moment where the bit catches and releases, and the edge chips.
So the premium grade is actually the wrong choice for the four-volt Dremel.
For most handheld engraving, yes. C2 or a tough C3 is the sweet spot. C3 is a finer grain than C2, so it holds an edge a bit longer, but it's still tough enough to survive the vibration. C4 is for rigid machine spindles where the tool path is perfectly controlled and there's no hand in the loop. On a Dremel Lite, C4 is a fast way to turn a twenty-dollar bit into a ten-dollar bit with a chipped edge.
That's a useful correction to the usual instinct that higher number means better. In carbide grades, higher number means harder and more brittle. On a handheld tool, you want the one that bends rather than breaks.
And then there's the geometry-material match. On HDPE, you want a sharp single-flute cone bit at high RPM with light passes. The high speed cuts cleanly before heat builds up, the single flute throws the melted chip out, and the light passes mean you're not generating a big melted mass in the first place. On aluminum, a ball-nose or a flat-bottom cone at lower RPM with a lubricant. Even a wax stick rubbed on the surface helps. The lubricant stops the aluminum from galling, which is when the metal sticks to the cutting edge and tears instead of cutting.
Galling is the thing that makes aluminum feel harder than it is. The bit grabs, the metal smears, and suddenly you're not cutting, you're friction-welding.
And on wood and MDF, almost any sharp carbide works, but a wider cone gives cleaner edges on end grain. MDF is particularly forgiving because it's uniform and has no grain to tear, but it's also abrasive, so a cheaper bit will dull faster than you expect.
Which brings us to why the starter kits fail. Twenty or thirty bits for fifteen dollars sounds like a bargain, but the problem isn't the quantity. It's that the geometry is inconsistent. Off-center tips, uneven flutes, poorly ground edges. A bit with a tip that's five thousandths off center doesn't cut a clean V, it wobbles in a tiny circle and produces a ragged, chattered line. And when you've got thirty bits to choose from, you're tempted to keep swapping instead of learning the technique with one good bit.
The variety is a trap. You spend an hour trying six different bits on the same piece of HDPE, and every result is bad, so you conclude the tool is bad or your hand is bad. The real problem is that none of the six bits was ground well enough to tell you anything.
And a good bit teaches you. If the geometry is true and the grade is right, then the variable left is your hand. You can feel the difference between too much pressure and just enough. With a bad bit, every cut feels like a fight, and you can't tell what you're doing wrong.
So with that technical foundation, let's talk about what this means for someone actually marking a bench full of different materials. The practical reality of the Dremel Lite 7760 is that you have speed but not torque. This flips the usual advice. Most people, when a cut isn't working, push harder. On this tool, pushing harder is the worst thing you can do. The bit stops, the motor stalls, and you've learned nothing except that the tool has a thermal cutoff.
The manual says light pressure, and it means it. You're not carving, you're letting the bit skim the surface. The weight of the tool itself is almost enough. Your job is to guide it, not to bear down.
And RPM matching is the other half. High RPM for plastics, because you want to cut cleanly before heat builds. Lower RPM for metals, because too much speed on aluminum generates heat that causes galling, and on steel it just burns the bit. The Lite gives you eight thousand to twenty-five thousand, and for aluminum you want to be down near the bottom of that range.
For the PCB work, Daniel's first use case, what's the pick?
A pointed or small-angle cone bit, tip around half a millimeter, at high RPM with minimal pressure. The goal is to cut through the copper layer without digging into the fiberglass substrate. Copper is thin, thirty-five microns on a standard board, so you barely need any depth. A C3 grade is the right call here. It holds an edge long enough to do multiple boards, but it's not so brittle that the glass fibers in the board will chip it the way a C4 would.
The fiberglass is the hidden problem on PCBs. It's abrasive, and it's full of hard little glass fibers that will eat a cheap bit and shatter a brittle one.
If you go too deep, you lift the copper trace and tear it. The mark you want is just a clean separation in the copper, revealing the board underneath. Half a millimeter of depth is plenty.
Then the HDPE boxes, the wider lettering.
A larger cone bit, sixty or ninety degree included angle, single flute, moderate RPM, light passes. The wide angle gives you a legible stroke without having to cut deep, which matters because a deep cut in HDPE weakens the wall of the box. The single flute clears the melted chips before they re-weld, and the moderate speed keeps the heat down while still cutting cleanly.
I've seen HDPE engraved too deep. The box gets a groove that's almost through the wall, and then it cracks along the line the first time someone drops it.
Right. You're marking inventory, not carving a turkey. A shallow mark is a durable mark on plastic. The pigment in HDPE goes all the way through, so you don't need depth for contrast. You just need to break the surface.
The intermediate option for mixed materials?
A medium cone, around thirty to forty-five degrees included angle. That's the workhorse. Enough bite to mark wood and MDF cleanly, shallow enough to handle aluminum without galling if you use a lubricant, and still fine enough to produce legible lettering. If Daniel's going to buy three bits, this is the one that does the most work.
The kit starts to look like this: a pointed fine bit for PCBs, a medium cone for general use, a wide cone for the HDPE boxes, and maybe a ball-nose for metals if the aluminum work is regular enough to justify it.
The criteria for each one are the same. Consistent geometry, which means checking that the tip is centered when you spin it. A reputable carbide grade, C2 or C3. And a one-eighth inch shank to fit the Dremel Lite's collet. That's the whole spec. You don't need thirty bits. You need three or four that are true.
The non-artistic mindset actually simplifies the choice. For inventory marks, you don't need a beautiful groove. You need a legible, durable mark. That means depth control and repeatability matter more than aesthetics. You're looking for a bit that cuts predictably at a given speed and pressure, not one that produces a pretty swirl.
Predictability is exactly what the cheap kits don't give you. Every bit in the box cuts differently because none of them are ground the same. You can't build any muscle memory.
Which leaves the torque limitation. Even the best bit won't engrave stainless with a four-volt tool. The research confirms the Lite stalls on stainless, and no carbide grade fixes that. So part of choosing a bit is knowing when to stop blaming the bit and switch tools. A corded Dremel or a proper rotary engraver with some torque behind it.
That's the part people resist. They've got the bit, they've got the technique, and the tool just won't do the job. And they keep buying sharper bits, harder bits, more expensive bits, when the answer is a bigger motor.
Stainless is the wall. Aluminum you can manage with patience and a wax stick. Stainless just sits there while your four-volt motor whines and gives up.
I mean, I've seen it. The bit touches the surface, the RPM drops, and the tool stalls before the cut even engages. It's not a bit failure. It's a power failure.
Hilbert: The one-oh-six. It's not just a cone bit. Dremel's sold it for decades. It's the standard engraving cutter. And the trick with it is you run it wide open and let the weight of the tool do the cutting. Not your hand. The weight of the tool.
That's a useful distinction. The weight of the tool versus the pressure of the hand.
Hilbert: I lost money learning that. Late nineties, I had a side thing engraving dog tags and trophy plates. Dremel and a box of cheap carbide bits. Kept buying starter kits and blaming the bits. Pressed too hard, every time. The real problem was I was fighting a low-torque tool. You push, it stalls, you push harder, it stalls faster. And then you've got a drawer full of half-used bits and a pile of ruined plates.
The technique you're describing is basically letting the tool float on the surface.
Hilbert: Highest RPM, hold it like a pencil, and let the bit's own weight pull it into the cut. You're just steering. If it's not cutting, you don't push harder. You either change the speed or change the bit. Pushing harder is never the answer on a four-volt tool.
That directly addresses the stalling problem. Most people's instinct is exactly backwards.
Hilbert: I tried to engrave a stainless steel flask once. Wedding gift for a friend. The Dremel stalled so hard it spun the flask across the garage. Hit the wall and put a dent in the drywall. The flask was fine. The drywall wasn't.
That's the torque limitation in action. The bit caught, the motor couldn't turn it, and the reaction force went somewhere.
Hilbert: Here's what I want to know. The wide cone you're talking about for HDPE. The sixty or ninety degree one. If I ran that at high speed with the tool's own weight, could I put legible numbers on the inside of a plastic storage bin lid without melting through it?
The inside of a lid is a tricky surface because it's usually textured, and the texture will fight the tip.
But the technique would be the same. High RPM, light passes, single flute to clear the chips. The texture means you might need two passes to get a clean line, but it should work. The key is not stopping in one place, because that's when the heat builds and the plastic melts.
Hilbert: Two passes. Alright.
If the lid flexes while you're cutting, support it from underneath. A flexing surface changes your depth mid-cut, and that's when you get the ragged edge.
Hilbert: I've got a bin that needs a label. I'll try it.
The light-touch technique is the thing I keep coming back to. It's not intuitive. Everything about holding a tool says push. And the whole craft of the low-torque engraver is unlearning that.
Which ties into the bigger question of what's next for this tool category. As carbide grades and tool designs evolve, are we going to see bits specifically engineered for low-torque handheld tools? Geometry that's optimized for feather-light pressure, grades that survive vibration without sacrificing sharpness?
Or is the answer always a bigger motor? There's a ceiling on what a four-volt tool can do, and maybe the honest answer is that the Lite is a plastic and wood tool that can occasionally handle aluminum, and everything else needs a corded tool.
For the hobbyist marking a shop bench, the takeaway is that a few good bits and a light touch beat a drawer full of cheap ones. But the tool's limits are real, and knowing when to step up is part of the craft.
The cutting-room floor detail I liked was that the one-oh-six has been in Dremel's catalog for so long that it predates the entire Lite line by decades. It's the default engraving bit because it's been the default engraving bit since before anyone thought to put a four-volt motor in a rotary tool.
It still works, if you let it.
That's the open question I'd leave with. Whether the next generation of bits gets designed around the constraints of low-torque tools, or whether the tools themselves are the thing that has to change.
Thanks to Hilbert Flumingtop for producing.
This has been My Weird Prompts.
If you're curating your own engraving kit, we want to hear what works on your materials. Find us at my weird prompts dot com.
We'll be back soon.
