After 2,100 hours of printing, I decided my Bambu Lab P2S deserved a little preventative maintenance.
And because I’m apparently incapable of making sensible decisions about when to perform maintenance, I decided to tackle this at 1am.
The logic was flawless: get the maintenance out of the way overnight, and I wouldn’t have to waste any of my precious non-working day messing around with the printer.
A quick job, a bit of preventative maintenance, and then off to bed.
Except, of course, I broke it.
So instead of spending today doing anything remotely productive, I spent it troubleshooting the maintenance I’d deliberately done at 1am to avoid spending today doing maintenance.
An absolutely flawless plan. No notes.
And the culprit behind this spectacular exercise in time management?
A few badly cut pieces of PTFE tubing.
And yes, it really was that simple.
The Setup#
My current 3D printing setup consists of:
- Bambu Lab P2S
- Two AMS 2 Pro units
- One AMS HT
- A 4-in-1 PTFE adapter
- An AMS filament buffer
- Genuine Bambu Lab PTFE tubing
That’s 9 filament slots across three AMS units, all feeding into the same printer.
It’s a fantastic setup, and after 2,100 hours of printing, I figured it was time to replace the PTFE tubing.
The original tubing had been through thousands of filament movements, loads, unloads and colour changes.
Replacing it seemed like sensible maintenance.
What could possibly go wrong?
Replacing All the PTFE Tubing#
I replaced all the external PTFE tubing between the AMS units, buffer, 4-in-1 adapter and printer.
I also replaced the PTFE tubing inside the P2S, leading towards the toolhead.
While I was there, I repositioned the 4-in-1 adapter to make it more stable and tidied up the tubing routes.
I made everything as short as reasonably possible, keeping the tubing in U-shaped or C-shaped runs rather than introducing sharp corners.
I tested each individual piece of tubing by manually feeding filament through it.
Everything felt perfectly smooth.
Excellent.
Job done.
Except it wasn’t.
The AMS Wouldn’t Feed Filament#
The first attempt to load filament failed.
The AMS started feeding normally, pushing filament through the external tubing and into the printer.
It travelled almost the entire distance to the toolhead.
Then, approximately four to eight inches short of the extruder, it stopped.
The AMS motor started struggling, making an unpleasant whirring noise.
Eventually, the printer gave up and tried to retract the filament.
Except it struggled to do that too.
It wasn’t just failing to feed. It was encountering resistance in both directions.
I tried again.
Same result.
And again.
Same result.
At this point, I’d managed to turn a perfectly functional printer into a very expensive filament transportation system that couldn’t quite reach its destination.
Maybe the AMS Just Needs One More Push?#
Initially, I wondered whether there was a way to adjust the AMS loading behaviour.
Could I increase the feed timeout?
Could I tell it to push another 100 mm?
Could I increase the number of retries?
Unfortunately, Bambu doesn’t expose those settings through the normal printer interface or Bambu Studio.
But it turned out I was looking in the wrong direction anyway.
The printer wasn’t simply giving up too early.
Something was physically preventing the filament from moving freely.
The Filament Buffer Was the Clue#
Watching the next loading attempt more closely, I noticed something interesting.
The spring-loaded mechanism in the AMS filament buffer was moving to the end of its travel.
And staying there.
Meanwhile, the AMS motor was audibly struggling to advance the filament.
The buffer is designed to accommodate differences in filament movement between the AMS and printer.
Seeing it reach the end of its travel, combined with the motor struggling, suggested that the filament path was experiencing excessive resistance.
The fact that the AMS also struggled to retract the filament reinforced that suspicion.
This probably wasn’t a software problem.
It was mechanical.
The Embarrassing Realisation#
Here’s where I have to admit something.
I’d replaced all the PTFE tubing using whatever cutting tools I could find.
Snips.
Wire cutters.
Anything sharp enough to cut plastic tubing.
And while they technically worked, the cuts weren’t particularly good.
Some ends were angled.
Others had been slightly squashed by the cutting action.
The tubing openings weren’t always perfectly circular.
I’d noticed this while cutting the tubing, but figured I could just use some needle-nose pliers to straighten the ends back out.
Because that’s good enough, right?
Spoiler: It was not good enough.
After all, I’d pushed filament through every piece manually, and everything seemed fine.
That was my mistake.
Why PTFE Cuts Matter#
PTFE tubing might seem like a fairly insignificant part of a 3D printer, but the quality of the cut matters.
Especially in a system with multiple AMS units, adapters, connectors and bends.
Angled Cuts#
If the tubing isn’t cut square, the end may not sit flush inside its connector.
That can leave an uneven transition where filament catches.
Crushed Tubing#
Using wire cutters or snips can compress the tubing before cutting it.
This can deform the opening and reduce its effective internal diameter.
Even a small restriction can increase friction.
Multiple Connections#
In my setup, filament passes through several sections of PTFE tubing and multiple connectors before reaching the extruder.
A slightly imperfect transition might not be enough to cause a problem by itself.
But several imperfect transitions can create cumulative resistance.
Why My Manual Tests Didn’t Catch It#
This was the particularly misleading part.
Every individual tube passed filament perfectly well when I tested it by hand.
But testing an isolated tube isn’t the same as feeding filament through the complete assembled system.
The filament has to pass through bends and transitions, with the AMS pushing it from some distance away.
A tube can feel fine individually while still contributing to excessive resistance in the complete path.
The Fix: Recutting the PTFE Tubing#
At this point, I remembered that I actually owned a proper PTFE tubing cutter.
Somewhere.
Naturally, I couldn’t immediately remember where I’d put it.
But the next step was obvious: revisit the tubing ends rather than keep trying to force the AMS to feed.
I went back through the PTFE tubing and recut the ends properly, paying attention to getting clean, square cuts without crushing the openings.
Then I reassembled the filament path and tried loading again.
And it fucking worked.
Just like that.
The AMS fed the filament through the system and into the toolhead successfully.
No struggling.
No buffer getting stuck at the end of its travel.
No horrible whirring noise.
No failed loading sequence.
After all that troubleshooting, the problem really was the quality of the cuts.
What I’d Do Differently#
If you’re planning to replace the PTFE tubing on your Bambu printer, here’s what I’d recommend.
1. Use a Proper PTFE Cutter#
Don’t use side cutters, wire cutters or ordinary snips.
Use a dedicated PTFE tubing cutter that produces clean, perpendicular cuts without deforming the tubing.
They’re inexpensive and can save a surprising amount of frustration.
2. Inspect Every Cut#
Before installing a tube, check that:
- The end is square.
- The opening is circular.
- There are no visible burrs.
- The tubing hasn’t been crushed.
- Filament passes through freely.
3. Don’t Make Every Tube as Short as Possible#
Shorter isn’t automatically better.
A slightly longer tube with a smooth, generous curve can create less friction than a short tube with a tight bend.
Allow enough slack for the toolhead to move freely.
4. Check Every Connection#
Make sure each PTFE tube is fully seated in its connector.
Pay particular attention to adapters, buffers and the connections around the printer’s internal filament path.
5. Replace Tubing in Stages#
This is probably the biggest procedural lesson.
I replaced practically every piece of tubing in one go.
When something went wrong, I had multiple possible causes to investigate.
Next time, I’ll replace a section, test it, and then move on to the next.
That makes troubleshooting considerably easier.
6. Pay Attention to the Buffer#
If your AMS is struggling to feed filament and the buffer spring reaches the end of its travel, don’t immediately assume there’s a software problem.
Check the filament path for excessive resistance.
And if retraction is difficult too, that’s another reason to suspect something mechanical.
The buffer behaviour isn’t proof of a particular obstruction, but it’s a useful diagnostic clue.
Final Thoughts#
After 2,100 hours of reliable printing, I decided to do some preventative maintenance.
I replaced all the PTFE tubing, improved the routing, repositioned the 4-in-1 adapter, and generally tried to make everything a little better.
Instead, I broke filament loading across my otherwise perfectly functional setup.
The AMS would push filament almost all the way to the toolhead, struggle, give up, and then struggle to retract it.
I considered feed timeouts, motor behaviour, buffer problems and tubing geometry.
In the end, the solution was simply to recut the PTFE tubing properly.
No firmware changes. No replacement motors. No new AMS hardware. Just clean, square cuts.
The lesson?
When working with PTFE tubing, the ends matter just as much as the routing.
And perhaps more importantly:
If it ain’t broke, make sure you’ve got the right bloody tools before you try to improve it.