No more bowden tube (or so I hope).

[bot]

Wish me luck. I'll wire it up tomorrow, and drill the hobbed bolt by hand. Dunno if the stepper will work, but I've got a bigger one if need be (same form factor, longer body).

[DeltaCon]

NIce construction! Don't you think the max printspeed will suffer much from the extra weight?

[critical_limit]

Which Motor/Gear you used?
5:1 or more?
Do you have a link to this combo?
What is the weight of the Motor/Gear?

[bot]

The print speed I use I this printer is already low. This is due to 0.9 degree steppers, only a 12v power supply, and attempting to reduce ringing to a minimum.

The motor and gearbox only weigh about 200g. The adapter is like 20g. The stepper is a nema11 from one stepper and the gearbox is 5:1.

[plasma]

Love Barrybot mod, Been using it for few months now and the results are great.

[teoman]

Looking good.

[bot]

Love Barrybot mod, Been using it for few months now and the results are great.

I didn't want to spend the 200 USD he was asking for his piece, when I wasn't sure it would be compatible with my fan holders. This is my own ABS adapter for an ezstruder. I'll share if anyone with a berrybot effector needs it.

[plasma]

If you could, that would be great. Thanks

[bot]

So the adapter worked out great. I'll attach the STL here for now, before releasing it officially after thorough testing. I've had to reduce the polygons for this STL over the one you see printed in the images above. The other file is about 80 MB and won't upload here. The extra facets are unneeded.

DirectDriveV1LowRes.stl

(2.48 MiB) Downloaded 262 times

The smaller motor had a rough spot in the gearbox that prevented it from working. It extruded filament fine, but would stall on the rough spot at max current. I bet one without the rough spot would work marginally well. The larger motor I had to use is only about 60 grams heavier and plenty of torque.

Adding the direct drive onto the effector means lowering jerk and accel values to prevent ringing. Alongside this, I've begun experimenting with RRF-dc42 elasticity compensation (M572) which is preventing corner blobs and infill mess when slowing down for the jerk values.

I'll post videos and photos once I've finished dialing it all in. I can print a 0.5 mm thick pillar with a 0.4 mm nozzle. The pillar breaks off if you breathe on it wrong, tho.

[Jimustanguitar]

The print speed I use I this printer is already low. This is due to 0.9 degree steppers...

Because of the Rambo's processing power, or something else?

[bot]

I'm actually using a duet with rrf. having 12v doesn't allow the current to build up as fast as higher voltages do. The higher the step frequency, the more this becomes a speed limitation. The most I can safely perform an edge-edge move with agressive acceleration is about 100-120mm/s. Now that I have the direct drive, I've got accelerations and jerk settings much much lower. This is turning out to be a slow printer, but a very very nice one.

[Generic Default]

Do you think it would be worth it to switch to 400/rev steppers since you're limited in speed and acceleration now? It looks like that entire effector plate (looks great by the way) will twist when accelerating because the motor is hanging off of one side. I built a big delta with leadscrews recently, I made sure to put extra thick CF tubes on it to keep rigidity. If you know the mass of the effector and the thickness of the arms and the acceleration, you can calculate how much deflection you will get.

[bot]

I have not had to re-calibrate the z height/endstop adjustments at all after adding the motor and filament driver. If the effector does deflect, it's not nough to notice any impact on first layer quality or height. These berrybot magnets are STRONG and there was never any play in the effector whatsoever. The long carbon arms themselves deflect quite a bit, causing ringing and vibrations, but the mechanics of the effector and carriages are rock solid. I do have 400/rev steppers on the axes. The extruder is a 200 step with a 5.18:1 gearbox on the front.

[bot]

I've discovered that there can be print artifacts caused by going too slowly with acceleration and jerk. There is a fine balance to be played. Now that I found out how slow I can go, I'm going to experiment with maximizing the time efficiency by slicing different speeds (normally I run everything at 50 mm/s. That won't cut it for this setup). So here are some pics of prints as I dial in the speeds and feeds (been watching too many CNC machining videos).

[JFettig]

I have gotten the best prints with really high acceleration settings, obviously that won't work for this situation, have you found the point you start getting ringing?

[bot]

You can see some ringing in the photo. That represents the least amount I can practically achieve without degrading performance of curved surfaces. I'm now trying. To increase acceleration as much as possible and use slicing speeds to keep the perimeters ring-free. I don't much like spending twice as long printing, even if the results are decent.

I'm around 2400-3000 accel right now.

[JFettig]

was hard to see much from the phone, even still I don't see an issue with that print other than the corners are a little large due to pressure buildup(even though its minimal) and deceleration/acceleration but its not bad looking at all. You can only get so much out of it before you start sacrificing one thing for another. That print looks better than most people ever get out of 3d printing. I have had a few people(engineers, machinists, designers, etc.) by the shop who are in awe of some of the worst prints I've ever made because they have never gotten anything anywhere close to that. The flying extruder might be the best option for a delta printer to get the best of both worlds - but then you can't have a heated chamber. Can't win them all.

I run 4000mm/s/s on my delta. Stock RMv2 firmware is something like 1600mm/s/s and I greatly benefited by increasing that to 3000. A poorly set up v2 will probably loose steps out at the extremes with that setting though.

[bot]

Oh for sure, I'm totally happy with the results in those photos. Problem is, the prints take SOoooo long at those settings. So I'm trying to find a way to make it fast when I can, but keep the ultra-low ringing as seen in these photos. I might have to accept a tiny bit more ringing to get acceptable print times.