Notifications
Clear all

Extruder fan specification ?  

Page 4 / 13
  RSS
Omikron
(@omikron)
Estimable Member
Re: Extruder fan specification ?

Hmmm...

Some of these results don't seem right. To my knowledge the CFM rating of the Noctua meets or exceeds that of the original fan as long as the "Low-Noise Adapter" is not installed. When the LNA is installed (which is literally just a resistor). Without the LNA CFM should be 4.8, and 3.8CFM with the LNA.

How do you people have it hooked up?

Posted : 19/10/2016 4:09 am
ayourk
(@ayourk)
Reputable Member
Re: Extruder fan specification ?

Hmmm...

Some of these results don't seem right. To my knowledge the CFM rating of the Noctua meets or exceeds that of the original fan as long as the "Low-Noise Adapter" is not installed. When the LNA is installed (which is literally just a resistor). Without the LNA CFM should be 4.8, and 3.8CFM with the LNA.

How do you people have it hooked up?

I have it hooked up direct to the Rambo via Male/Female jumper wires. I will admit the overall rotation of the fan seems low. Maybe I need to edit the firmware to fully open up the speed of the fan?

The adapters from the fan connector were too big to fit in my cable chains setup, so I couldn't use them.

Dimensions PNG

and an 8 inch (200mm) or greater caliper is recommended.

Posted : 19/10/2016 4:32 am
David T.
(@david-t)
Noble Member
Re: Extruder fan specification ?

I cut the original fan from its wire and soldered noctua, no connectors. Generally I don't trust connectors placed on fast moving and accelerating parts.
I tested noctua on the table before mounting and compared with original fan. I could clearly feel that air flow coming from noctua is noticeably lower.
However, it works, no problems so far.

Posted : 19/10/2016 8:29 am
Patrik Rosén
(@patrik-rosen)
Reputable Member
Re: Extruder fan specification ?

Test done. Test conditions:

Both printers preheated to PLA (210/55), extruders homed and moved to X=10, Y=200, Z=50. Both printers in the same room, room temperature 23°C. Heaters turned off on both printers at the same time. Measured time until hotend fan went off.

Original fan extruder cooled down in 6m46s. Noctua fan extruder was at 69°C at the same moment.

Noctua fan extruder cooled down in 9m12s.

OK, thats cool! (hehe)

19 degrees "after".. that last degrees will be the slowest ones, so I would say that is pretty ok and pretty close to the original fan in comparison.
It would be nice to see the result if we could double the speed of the Noctua fan. Totally unnecessary though, since no one of us has got any problems after upgrade. Lets see what a long term test says!

I'm working on a small revision on my adapter, with some added features to simplify the cable routing. And I can also see a small gap between the adapter and the Noctua fan because of the rubber "feets" of the Noctua. I can feel a gentle breeze there so I've added a small flange there to prevent that. Will upload that to thingiverse as well when its printed, tested and considered done. At the moment there are soo much in line though, and since it works as it should now it has low priority. Damn I really need another printer 😀

One other potential improvement I could see to your design is to have the wind tunnel being a sleeve to eliminate the wind resistance of the internal screw holes.

The screws inside my tunnel doesn't affect/disturb the airflow in a grade which will affect the efficiency of the cooling I think.
Those screws was the one thing I discussed the most with my brother, and I trust his knowledge in this. There's also over-designing, which I dont want to step into, I'm already to close with this design hehe, To much things that's in the "Line", as long as it works as it should I'm satisfied with that.

Hmmm...

Some of these results don't seem right. To my knowledge the CFM rating of the Noctua meets or exceeds that of the original fan as long as the "Low-Noise Adapter" is not installed. When the LNA is installed (which is literally just a resistor). Without the LNA CFM should be 4.8, and 3.8CFM with the LNA.

How do you people have it hooked up?

I just cut the original cables as well, and soldered the Noctua onto that, I do also think that its moving really slow!
Can the firmware set the fanspeed through that voltage port? I guess that a less effective cooling could mean some long term problems?

Me on 3dhubs!
Posted : 19/10/2016 9:23 am
Omikron
(@omikron)
Estimable Member
Re: Extruder fan specification ?

Assuming the port on the mini Rambo is the same type as the one used by the nozzle cooling fan, then I imagine that the firmware can indeed control the voltage and fan speed.

The real question is if that port is already running at its maximum capacity, or if it being limited in firmware.

I'm traveling at the moment so I can't take a look at the code. Someone else want to take a peek?

Alternatively someone can connect a multimeter to the wires or ports for the fan and see what the output voltage is while the cooling fan is running. That may give us a starting point.

Posted : 19/10/2016 12:44 pm
PJR
 PJR
(@pjr)
Antient Member Moderator
Re: Extruder fan specification ?

The extruder cooling fan speed is controlled by firmware (it is PWM output):

Configuration.prusa.h:
#define EXTRUDER_AUTO_FAN_SPEED 255 // == full speed

temperature.cpp:
:void setExtruderAutoFanState(int pin, bool state)
{
unsigned char newFanSpeed = (state != 0) ? EXTRUDER_AUTO_FAN_SPEED : 0;
// this idiom allows both digital and PWM fan outputs (see M42 handling).
pinMode(pin, OUTPUT);
digitalWrite(pin, newFanSpeed);
analogWrite(pin, newFanSpeed);
}

If would appear that there is very little in the way of real-time control of this fan.

I would suggest that not having sufficient cooling of the extruder will cause issues; softened filament will be drawn up into the extruder during retractions and we all know that this can cause blockages.

I would also believe that E3D have carried out some testing on the amount of cooling required to the heat sink to prevent issues and that they have built in a margin to play with. However, pushing those margins will most likely lead to problems and therefore I would recommend that any alternative solution must provide at least equal cooling to that as supplied.

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…

Posted : 19/10/2016 1:26 pm
ayourk
(@ayourk)
Reputable Member
Re: Extruder fan specification ?

So from what Peter is describing, M42 P8 S255 should turn the extruder fan on at full speed, and M42 P8 S0 should turn the extruder fan off. Doing testing this way vs heating up the hotend will prevent jams.

Dimensions PNG

and an 8 inch (200mm) or greater caliper is recommended.

Posted : 19/10/2016 2:07 pm
Madhu
(@madhu)
Eminent Member
Re: Extruder fan specification ?

All,

From the Noctua website, the NF-A4x10 FLX (12V version) when running at the max RPM of 4500 should provide an airflow of ~ 8.2 m3/hr (4.82 CFM) this is more than the airflow of the E3D fan which provides 4.1CFM.

If the Noctua fan is used with the Low Noise Adapter, then it will provide 3.88CFM.

~Madhu.

Posted : 19/10/2016 2:29 pm
Patrik Rosén
(@patrik-rosen)
Reputable Member
Re: Extruder fan specification ?

All,

From the Noctua website, the NF-A4x10 FLX (12V version) when running at the max RPM of 4500 should provide an airflow of ~ 8.2 m3/hr (4.82 CFM) this is more than the airflow of the E3D fan which provides 4.1CFM.

If the Noctua fan is used with the Low Noise Adapter, then it will provide 3.88CFM.

~Madhu.

Yes the specs of the fan says that it should be more than enough, but as everyone has noticed, it seems to run on a very low speed. Although from the time when I was building rendering-clusters etc I've experienced the same when going from 90mm to 140mm chassi fans for example. And especially Noctua makes very good fans. The cooldown time test is pure facts though, not sure why that difference happens. but I will continue to use the adapter and the noctua fan, to see if it causes any long term issues. I've printed 30+ hours now with it, and been switching between PETG, PLA and Filaflex with all kinds of different retraction settings, and its been working flawless so far.

Me on 3dhubs!
Posted : 19/10/2016 2:37 pm
Madhu
(@madhu)
Eminent Member
Re: Extruder fan specification ?

I dont have a Noctua fan with me... But, what I can do is program my Arduino to measure the frequency and pulse width of the stock E3D fan.

Anything specific you want me to measure?

Thanks,
Madhu.

Posted : 19/10/2016 2:45 pm
Madhu
(@madhu)
Eminent Member
Re: Extruder fan specification ?

Some real world data maybe:

and the chart at

~Madhu.

Posted : 19/10/2016 2:52 pm
ayourk
(@ayourk)
Reputable Member
Re: Extruder fan specification ?

Some real world data maybe:

and the chart at

~Madhu.

The Picture doesn't show up for some reason.

One of the things I'm suspecting is going on is that only 5V is getting sent to the hotend van, not 12V. I'd be interested to know how many amps are being put out on the hotend fan port from the Rambo.

Dimensions PNG

and an 8 inch (200mm) or greater caliper is recommended.

Posted : 19/10/2016 3:01 pm
Madhu
(@madhu)
Eminent Member
Re: Extruder fan specification ?



The Picture doesn't show up for some reason.

Can you check here: http://www.coolingtechnique.com/recensioni/74-ventole-rheobus/1153-recensione-ventole-noctua-a-series.html

Posted : 19/10/2016 3:03 pm
David T.
(@david-t)
Noble Member
Re: Extruder fan specification ?

One of the things I'm suspecting is going on is that only 5V is getting sent to the hotend van, not 12V.
As I have said, I tested the fan on the table from 12V PSU directly and compared to original E3D fan. Air flow is significantly lower from noctua fan. But I didn't measure it by any sofisticated method, don't expect exact CFM value, I just tried how it feels (on hand and face).

Posted : 19/10/2016 3:22 pm
Patrik Rosén
(@patrik-rosen)
Reputable Member
Re: Extruder fan specification ?

Some real world data maybe:

and the chart at

~Madhu.

The Picture doesn't show up for some reason.

One of the things I'm suspecting is going on is that only 5V is getting sent to the hotend van, not 12V. I'd be interested to know how many amps are being put out on the hotend fan port from the Rambo.

Yeah looking at / listening to that video makes me think the same, it really feels like the fan is going on half speed atm..

Me on 3dhubs!
Posted : 19/10/2016 3:44 pm
David T.
(@david-t)
Noble Member
Re: Extruder fan specification ?

I guess the problem is somewhere else - not CFM, but static pressure.

Original E3D fan is XYJ12B3010H, it has 4.08 CFM and 4.82 mmH2O static pressure.
Noctua fan is NF-A4x10 FLX, it has 4.83 CFM (a bit more) but only 1.78 mmH2O static pressure, that is almost 3 times less. And we need high static pressure because of obstacles in air path - hotend heatsink.
We'll see...

Posted : 19/10/2016 7:25 pm
christophe.p
(@christophe-p)
Member Moderator
Topic starter answered:
Re: Extruder fan specification ?

Hi guys,

Actually the real test would be to measure the efficiency on the extruder.

That's why I suggested to measure time to go from 200 °C down to 50 °C with both fan, the stock and the Noctua. That would give a measurable way to estimate efficiency of the actual design and of eventual evolutions, versus the stock design.

I need to get a Noctua, didn't take time yet to take care of it 🙁

I'm like Jon Snow, I know nothing.

Posted : 19/10/2016 11:13 pm
richard.l
(@richard-l)
Member Moderator
Re: Extruder fan specification ?

Might order one of these to try...

https://www.coolerguys.com/collections/40mm-fans/products/evercool-40x15mm-high-speed-fan-with-3-pin-connector-ec4015sh12ba

Posted : 20/10/2016 11:02 pm
StephanK
(@stephank)
Reputable Member
Re: Extruder fan specification ?

That's why I suggested to measure time to go from 200 °C down to 50 °C with both fan, the stock and the Noctua. That would give a measurable way to estimate efficiency of the actual design and of eventual evolutions, versus the stock design.

I should be getting a 30x30x10 fan rated 5 cfm on monday, so while i was messing around with new bearings today, i also I timed cooldown time of my stock fan today. Unfortunately, i used preheat to PLA starting at 215°C irc and didnt start timing at 200°C, but maybe it's still of use to somebody:

Original fan:
After 1 minute 18 seconds reached 150°C
after 2:37 (total) it reached 100°C
and finally after 5:20 it reached 50°C

(I have the nagging suspicion this new fan will be even louder than the original one, but we shall see.)

Posted : 21/10/2016 8:57 pm
David T.
(@david-t)
Noble Member
Re: Extruder fan specification ?

I'd be curious if there was a cool down time difference between the 2 different models of adapters.
Allright, here are the results.
One printer, the same noctua fan, the same initial conditions, I only changed fan adapters. Both adapters were printed from orange PETG and the same print settings (0.2mm layer, 2 shells, 10% infill).

Patrik's adapter cooldown time: 9m12s
Aaron's adapter cooldown time: 8m55s

My evaluation/comparison:

Aaron's adapter cools a bit (3%) better. The difference is probably caused by a bit cleaner air path (no cutouts for screws inside). Aaron's adapter also provides much better fit at fan side and it seems to me that also angle of the fan is slightly better.
On the other hand, Patrik's adapter provides much much much better fit at extruder side. To be honest, Aaron's adapter doesn't satisfy me at all when talking about mounting to extruder. It keeps falling apart when extruder does rapid movement changes. Maybe if mounting pins were a bit larger...., but I wouldn't trust them anyway.

Posted : 22/10/2016 1:58 pm
Page 4 / 13
Share: