3D prints · No. 03

A life-size bat on a 30 cm printer

An 84 cm baseball bat in red PETG, printed upright on a steel rod. Eight prints started, five of them stopped, and the barrel ended up in three pieces.

15 October 2026 · updated 16 October 2026 · Designed in CadQuery, printed on a Creality K1 Max · Short version ↓

PETG
22 h 20 min printing
about 680 g
8 prints started
84 cm

The bat from the side, drawn from its design profile, 840 mm. Left to right: knob and handle, taper, barrel with its cupped end. The thin lines are the two joints as designed, the dashed line the steel rod. The barrel later became three pieces.

A full-size adult bat is about 84 cm long. My printer's box is 30 cm on every side.

I wanted it full size and shaped like a real one, so it had to come out of that box in pieces and still hold together when swung.

Desk toys

The first job was to find a bat someone else had already got right. I didn't find one. The popular printable bats were desk toys: a 32 cm mini bat that, scaled up to 84 cm, gets a barrel 87 mm thick, because its proportions are a cartoon's. Another was 7.5 cm and low-poly, another a 70 cm stack of cones. The closest was MIHAI_T's full-size bat on MakerWorld: 85 cm, with an M8 threaded rod through the middle. I liked that rod, and it ended up in mine. But the bat never said what profile it used.

So I drew my own: a standard 33 inch bat, the classic shape, revolved in CadQuery, a Python library for CAD, from a table of diameters along its length. A name went across the barrel in raised graffiti letters with a drop shadow, every edge sloped at 45° so nothing needs support.

Render of the finished design: a red full-size baseball bat seen from the side, knob on the left, barrel on the right, with a pixelated patch on the barrel where the name is. — **Fig. 01** · The design, rendered: 84 cm, three pieces and a rod as designed, before any of it printed. Name pixelated.

Three pieces and a rod

The first plan was three upright pieces in PLA on a steel rod. But I wanted the bat tough on its own, without the rod, and PETG takes a hit better than PLA. A bat printed lying down is strongest, with every layer running along its length. Lying diagonally on a 30 cm bed, though, two pieces top out at 73.5 cm.

Standing up, the printer's height is the limit, so I let the size give instead. The tallest bat that fits in two upright pieces is 57.8 cm. That's a tee-ball bat for a kid. So: three pieces, full size.

Two red bats drawn from the side, one above the other and to the same scale: the upper one is short, the lower one about half as long again. A patch on each barrel is pixelated. — **Fig. 02** · To scale: the 57.8 cm bat that fits the printer's height in two pieces, above the full-size 84 cm one that needs three. Names pixelated.

So three upright pieces at full size. From the bottom up: the knob and handle, then the taper, where the handle swells into the barrel, with a short spigot at its lower end that plugs into the handle, then the barrel with its cupped end. They join where the bat bends least, 286 mm up from the knob, as far as the printer's height allows, and again at 558 mm, just before the barrel, where the bending is almost nothing.

Fig. 03 · What each version would have cost on the printer, as the slicer estimated it, with the plastic in grams. The printed spine was the expensive idea. The last row is the set of files that went to the printer.

But an upright print is weakest across its layers, and a bat bends across its layers every time it hits something. The plan for the first joint was a printed spine, an eight-sided bar printed lying down and glued into both pieces. On paper that joint had half to nine tenths of a solid handle's strength. Nobody measured it. That was enough for a bat on a shelf, and short of the one I wanted.

So the rod came back: an M8 threaded rod from a hardware shop, 13 dirhams, running through all three pieces. One steel nut sits in a printed nut holder inside the knob and another under the cup at the far end, so tightening them clamps the pieces end to end and every joint is squeezed instead of pulled. A short printed plug, the coupler, joins the taper to the barrel. Epoxy holds the nut holder and a cap over the knob, and a press-fit cup plug closes the far end.

Later the same day I had another idea: make the barrel itself the wrench, so the bat tightens when you spin it. A few minutes later I took it back. A bat that tightens by spinning loosens by spinning, every time it hits something. Both joints are keyed now, shaped so one piece can't turn against the next.

Cut-away render of the bat. A grey steel rod runs almost the full length inside the red shell. Yellow parts at each end are the nut holder in the knob and the cup plug, with nuts and washers clamping the rod. The lower half zooms in on the knob and the cup end. — **Fig. 04** · Cut through the middle. The rod (grey) is clamped by a nut and washer at each end. Yellow: the coupler at the barrel joint, the nut holder in the knob and the plug that closes the cup.

The washers I had at home were 30 mm across and 2 mm thick, so their seats were redrawn to fit. Then a check of every steel part against the plastic found a thin wall of plastic, 0.8 mm, across the rod hole in the knob, right where the rod has to pass. It came out of the model before anything printed.

Fig. 05 · How it goes together, drawn from the files: knob cap, nut holder, nut and washer, handle, taper, coupler, barrel, washer, nut, cup plug, and the rod through all of it. The name on the barrel faces away.

The rod runs down 80 cm of hole. A tight fit all the way would jam, and a loose one rattles. So the hole is 8.8 mm, loose, with short tight rings every 50 mm or so to hold the rod in the middle. How tight to make those rings depends on how this printer prints a small hole in this PETG, and I didn't know. So the first plate got a test bar with four holes in it: 8.1, 8.2, 8.3 and 8.4 mm.

Four minutes

Before anything printed I went through every file once more and found two problems. The small-parts plate had a 0.12 mm first layer, part of what tore up the first try of my Panamera sculpture, so it went to 0.24 mm. And the roof over the washer seat in the knob had to print across 30 mm of air with the rod hole through its middle, so the strands crossing the hole would have had nothing to land on. Now the roof prints whole, with a thin skin over the hole that I punch out with a screwdriver.

Four renders of red parts standing on a printer bed: the barrel with a pixelated patch on it, the long thin taper, the handle with its knob at the bottom, and the small parts plate with the coupler, the hexagonal nut holder, the knob cap, the cup plug and the test bar. — **Fig. 06** · The four files as they stand on the bed, drawn in CadQuery: barrel, taper, handle, and the small-parts plate. The barrel and the taper each use nearly the full 300 mm of height. Name pixelated.

The first plate started at 04:10 on Monday morning, 12 October. The bed and nozzle took their time heating. Four minutes of printing later, at 04:21, Obico, the add-on that watches the printer's camera for failed prints and had once stopped the Panamera, stopped this one too. I looked, and cancelled it. The outer loop of the brim, the flat ring the printer lays around a part to hold its first layer down, had run straight over the purge line, the stripe it draws at the front of the bed before every print to get the nozzle flowing. Where the two crossed it was a smeared, crushed mess.

Phone photo of the printer bed: a red brim loop around the outline of the small parts, crushed and smeared where it crosses the purge line at the front. — **Fig. 07** · The first plate, stopped at four minutes. The red loop is the brim, grown to about 18 mm by the slicer, and the smear at the front is the purge line it ran over.

The file said the brim was 5 mm. The slicer's automatic brim had quietly grown it to about 18 mm. And the printer draws its purge line 10 mm in front of the part without looking at the brim at all. The handle and taper files had the same clash waiting.

The fix was to stop letting the slicer choose: a fixed 7 mm brim, outer edge only, on all three files. The check I run before every print now measures the gap between brim and purge line. At 04:36 the first plate started again.

97 %

The plate got to 97 %. Then the coupler came off its brim, and the last layers went into the air as strings. At 08:29 Obico saw the strings and paused the print.

Printer camera frame: five red parts on the black bed, a tall cylinder at the left, a hexagonal holder, a flat disc, a flat bar and a wider cylinder at the right. — **Fig. 08** · The small-parts plate fifty minutes apart. The coupler is the tall cylinder at the left. By 08:56 only its brim ring is there, and the coupler is lying on the bed, printed to 97 %.

The same view fifty minutes later: the tall cylinder at the left is gone, only its brim ring remains, with a loose strand beside it. — **Fig. 08** · The small-parts plate fifty minutes apart. The coupler is the tall cylinder at the left. By 08:56 only its brim ring is there, and the coupler is lying on the bed, printed to 97 %.

A hand holding a short red printed cylinder with a hole through it. Its top edge is rough, with blobs and broken strands. — **Fig. 09** · The coupler, found on the bed. Intact apart from its top skin. It stayed: the top sanded flat, and a reprint file kept as the backup.

My new check measured the brim against the purge line, and nothing measured the brim against the part. The files left a gap of 0.25 mm between the two, so they never bonded. The handle and taper files were sliced again with the brim attached to the part.

The one good result of the morning had come at 07:45, with the plate still printing. On the test bar, the 8.1 mm hole was the snuggest, with the least movement. That's tighter than my 8.2 mm guess. The long pieces were redrawn with 8.1 mm rings before any of them printed.

Render of a flat red bar with four round holes in a row, labelled 8.1, 8.2, 8.3 and 8.4, with the grey steel rod standing in the 8.1 hole. — **Fig. 10** · The test bar, drawn from the design file with the rod in the hole that won. On the printed bar the labels are engraved beside each hole.

Starved

The handle printed clean, from 09:24 to 13:51 on Monday, with its brim joined to the part this time. The rod went all the way through its rings. The taper started at 21:58 the same evening.

A hand holding the printed red handle by its knob against a grey wall, the long steel threaded rod pushed all the way through it and reaching far out of the top end. — **Fig. 11** · The handle off the bed, with the rod pushed all the way through. The 8.1 mm rings let it pass.

Fig. 12 · Four and a half hours in seven seconds: 09:24 to 13:51 on Monday, from the printer's camera.

At about 02:55, at 96 %, the taper paused: the filament had stopped feeding. On the spool, the turns on one side had stuck to each other, and the extruder couldn't pull them free. The spool had spent the whole day in the filament dryer on its PETG setting, and that's what I blamed first. I cut off the stuck turns, reloaded, pressed resume, and the taper finished at 03:22.

It came off the bed with one flaw: a ring just below the spigot, where the last layers before the pause had been starved of plastic. A reprint would have cost 4 h 48 min and 152 g. Instead the ring gets covered. A bat has a grip, and the grip will run over that joint.

The red taper lying on a dark granite counter, spigot end to the right with the steel rod sticking out of it. A fine line runs round the piece a little below the spigot. — **Fig. 13** · The taper the morning it came off, rod in the spigot. The line below the spigot is the ring from the pause.

In the air

The barrel started at 03:37 on Tuesday morning, 8 h 33 min of printing as sliced. By 06:26 the first letter was rising out of the surface, clean. At 08:29, the same minute as the day before, Obico paused the print.

The file was at 164.6 mm. A ruler along the barrel said about 144 mm, so it looked as if 20 mm of layers, about 100 of them, had gone nowhere: the same spool had jammed again and nothing had come out of the nozzle. The printer had carried on above the barrel, laying strings in the air.

Fig. 14 · Nearly five hours in eight seconds, 03:37 to the pause at 08:29, from the printer's timelapse. Name pixelated throughout. Watch the top at the end.

The red barrel standing on its brim on the printer bed, raised letters up its face pixelated, with a ragged fringe of loose strands at its top edge. — **Fig. 15** · The barrel after the pause. Left: 10 mm of strands the nozzle laid in the air. Right: under them, the real top, flat and clean, so the print could have gone on from there. Name pixelated.

Looking down into the top of the barrel: a flat surface of wavy gyroid infill with the round rod hole in the middle, and a few hairs of plastic at the rim. — **Fig. 15** · The barrel after the pause. Left: 10 mm of strands the nozzle laid in the air. Right: under them, the real top, flat and clean, so the print could have gone on from there. Name pixelated.

The top was flat and clean, so the print could continue from the real height. The plan was to let the printer measure it: bring the nozzle down over the wall in small steps with a sheet of paper on top, and the first step where the paper drags is the height. I sent the nozzle to 146 mm, two above the ruler's 144, as the first step.

The barrel was 155 mm tall, not 144. The gap was 10 mm of strings, not 20, and the nozzle pressed into the top of it.

The printer's toolhead sitting right on top of the red barrel, its nozzle buried in the barrel's top. The raised letters down the face are pixelated. — **Fig. 16** · Left: where the paper test ended, the toolhead on the barrel. Right: the tape afterwards, 4.6 at the brim and 20.2 at the top, so about 155 mm. The ruler had said 144. Name pixelated in both.

A hand holding the red barrel piece with a soft tape measure down it, the scale reading about 20 where the piece ends. A pixelated patch on the piece. — **Fig. 16** · Left: where the paper test ended, the toolhead on the barrel. Right: the tape afterwards, 4.6 at the brim and 20.2 at the top, so about 155 mm. The ruler had said 144. Name pixelated in both.

The barrel was still standing on its brim, but the print was over. A ruler held against a round barrel inside a dark printer had been 11 mm out, and I had used it to aim a hot nozzle. I cancelled the job at 09:05 and took the piece off.

A barrel in three

The barrel wasn't lost. Its bottom 155 mm were good, letters included. So the barrel model was cut at 154 mm and the top 128 mm became its own print, standing on its cut face. The bottom piece had to be sanded down to 154 mm so the letters would line up across the joint. The top started at 09:18.

At 11:14, 61 mm up, the runout sensor, which stops the print when the filament stops moving, stopped it. The spool again. The reload wouldn't feed, and at 14:22 I cancelled that one too. The barrel was now going to be three pieces, and a joint would cross the name twice. Those two joints are plain glued cuts with the rod through them, not keyed like the other two, and they sit in the barrel, where the ball hits.

A hand holding a short red printed cylinder with a soft tape measure running up it, reading about 6 cm. A pixelated patch at the left edge. — **Fig. 17** · The middle piece: 61 mm, where the spool stopped it the second time. The letter that crosses the cut lines up with the piece below. Name pixelated.

The model was cut again, at 60.6 mm, and a 67 mm top piece went to the printer. It started at 14:34 and stopped after 16 minutes. The spool. This time I pulled about 30 m off it by hand, in big loose loops on the floor, so the printer only had to pull from the pile. The piece started again at 15:13 and finished at 17:38, and the spool went in the bin with about 300 g still on it.

Looking along the red barrel from the cup end, held in one hand: two fine lines run round it where the pieces meet, and a third further down where the barrel meets the taper. — **Fig. 18** · The dry fit on Tuesday evening, all five pieces on the rod, seen from the cup end: the two barrel joints, and further down the taper joint. The name lines up across both.

Glued

The plan had been to glue only the nut holder and the knob cap and let the two nuts clamp the rest. That evening I glued every joint with 5-minute clear epoxy, nuts and washers on, and called it a day. One thing looked wrong: the top seemed a bit high. The rod stood proud of the cup, and the plug, which should sit flush over the nut, had been glued in on top of it. Either the rod was cut a little long or it wasn't all the way down into the knob nut. I didn't find out which. By the time I noticed, the glue was hard.

The cup end of the red bat held up beside a dark counter edge, with a short red stub standing out of the end, circled in red on the photo. — **Fig. 19** · The stub: the cup plug, pushed up by the rod underneath it. Circled on the phone that evening.

The next day I cut the rod stub off, flush with the plug.

White

The letters were always going to be a second colour. Red on red reads only when the light catches the edges. So on Wednesday night the barrel got masking tape around the name, with a window cut out so the letters stood up out of it, and the letters got satin white from a can of paint-and-primer.

Tape masks a flat edge. It doesn't seal against the foot of a raised letter, so the white went under the tape and onto the red around every letter. Two specks landed on the back of the barrel, and a lighter rectangle marks where a strip of tape had sat.

Close up on the foot of one white letter, circled in red: the white runs past the edge of the raised stroke onto the red, with specks of white further out on the barrel. — **Fig. 20** · Wednesday, 40 minutes after spraying. Left, circled: the foot of one letter, white where it belongs and white where it doesn't, with the rest of the name out of frame. Right: the back, with the seam, the two joint lines, two specks of overspray and the tape's shadow.

The back of the red barrel held over a tiled floor: a straight seam line down its length, two fine joint rings, two small white specks and a paler rectangle where tape had been. — **Fig. 20** · Wednesday, 40 minutes after spraying. Left, circled: the foot of one letter, white where it belongs and white where it doesn't, with the rest of the name out of frame. Right: the back, with the seam, the two joint lines, two specks of overspray and the tape's shadow.

The paint got its day to cure. Then each letter got a dark outline around the white, and the grip got its wrap: Formflex, white, wound from the knob to the barrel.

What I check now

Before a long print, I spin the spool by hand. The runout sensor only notices when the strand stops moving. A spool that drags starves the print quietly, layer by layer, and that is how 10 mm of a barrel ended up printed in the air.

It's done: one piece, 84 cm, a name down the barrel in white. Nobody has swung it yet.

The finished red bat lying along the picture on a wooden table, knob at the left: the handle and taper wound in white, the red barrel at the right with the name pixelated. — **Fig. 21** · Finished, on the table, knob to the left to match the drawing at the top. White wrap on the grip; the name on the barrel is pixelated. On a narrow screen it is shown in two halves, knob half above.

The short version

What went wrong, and what fixed it.

No model online was a realistic life-size bat.Drew my own from a standard 33 inch profile.
84 cm doesn't fit a 30 cm printer, and two upright pieces made a kid's bat.Three upright pieces at full size. Five, once the barrel split.
Upright layers are the weak direction for a bat.A steel M8 rod through the middle, clamped with a nut at each end.
A bat that tightens by spinning would also loosen by spinning.Both joints keyed so nothing can turn.
The brim grew onto the printer's purge line.A fixed 7 mm outer brim, and the print check now measures brim against purge.
The brim sat 0.25 mm away from the parts, and at 97 % the coupler came loose.Handle and taper sliced again with the brim attached to the part. Both held to the bed.
Nobody knew how tight an 8 mm rod hole prints in this PETG.A test bar with four holes. 8.1 mm won.
The red spool's turns had stuck together. It starved the taper at 96 % and left a ring.Cut off the stuck turns, reloaded, resumed. The ring stays, under the grip.
The same spool starved the barrel at about 155 mm, and the printer laid 10 mm of strings in the air.The missing top became its own piece.
I sent the nozzle down to 146 mm over a barrel that was 155 mm tall.Measure from far above in small steps, or don't move the head at all.
The spool stopped two more prints, at 61 mm and at 16 minutes.Pulled 30 m off by hand into loose loops. The barrel is three pieces on the rod.
The rod stood proud of the cup, and the plug was glued over it.Cut the stub flush the next day.
Masking tape doesn't seal against raised letters, so white landed on the red around them.A dark outline around each letter.

Built on

Where the ideas came from. Their files stay on their pages.

Full size baseball batMIHAI_T · MakerWorld · Standard Digital File License85 cm with an M8 rod inside. The closest bat I found, and the rod idea I built on.
Sedgwick Ave DisplayPedro Vergani, Kevin Burke · Google Fonts · SIL Open Font License 1.1The graffiti letters on the barrel.

How it was printed

Printer: Creality K1 Max, rooted, Klipper
Slicer: OrcaSlicer 2.4.2, profiles K1M 0.20 Standard for the long pieces and 0.12 Fine for the small parts
Material: PETG, red, 240 °C, bed 75 °C for the first layer then 72 °C
Walls, infill: 5 walls, gyroid: handle 40 %, taper 25 %, barrel 20 %
Pieces: Handle 286 mm, taper 272 mm plus an 18 mm spigot into the handle, barrel 282 mm, which ended up printed as three pieces of about 154, 61 and 67 mm, plus a coupler, a nut holder, a knob cap and a cup plug
Rod rings: 8.2 mm in the first files, 8.1 mm after the test bar
Hardware: M8 threaded rod, two M8 nuts, two 30 × 2 mm washers, 5-minute clear epoxy
Grip: Formflex wrap, wound from the knob to the barrel
Paint: Satin white paint-and-primer from a can on the letters, over masking tape, then a dark outline around each letter
Plan versus printer: About 22 h and 651 g planned. The printer's history says 22 h 20 min and about 680 g over 8 jobs, 5 of them cancelled, 3 of those kept as pieces

Settings that mattered

The profile’s value, then what this print used.

Brim auto changed to outer only, 7 mm
First layer, small parts 0.12 changed to 0.24 mm
Sparse infill, taper and barrel every layer changed to every 2nd layer
Brim gap (brim_object_gap) 0.25 changed to 0 mm

All strength figures in this post are calculated, not tested. The bat's outline follows typical 33 inch bat charts, not a measured bat. Grams are what the printer's job history says it fed, cancelled jobs included.

The files

The handle, taper, barrel and small parts as STLs, with previews and a README with the settings and the hardware list. The barrel is blank: the name is left out of the file.

Parts, STLbat.zip5.3 MB

License: CC BY-NC-SA 4.0. Credit: Abdallah El Belkasy, abdaishere.com/prints.

Off the bed