Since COVID-19 hit, I moved out to my garage so I could work from home, and in doing so I realized how loud my network rack really is. My rack is not the prettiest rack, and to be frank I’ve never really been a fan of the sleek LED lit home lab. My aesthetic is more like something from a Tatooine droid shop: not disorganized, per se, but certainly not pretty.
I’ve really been wanting a Juniper switch for home, so I went shopping on ebay to see if I could find anything, and surprisingly, right now you can find EX2200 switches for really cheap — like, $75 shipped. I wondered then if I could modify the fans on the switch, and lo and behold, you can! So I bought a Juniper PoE switch, but the next part was purchasing the fans.
I came across the video below from Christian Scholz showing mostly how to replace the fans. It’s an excellent video, and shows you how to replace the two back chassis fans.
However, it doesn’t show you how to replace the fans for the power supply, largely because the power supply on non-PoE EX2200s doesn’t have fans (see YouTube preview above).
For the fans on a PoE EX2200, there’s actually just one fan, and if you want to replace it with a Noctua fan, it’s going to work, but it’s going to get a bit warm. One solution is to add a fan on the exhaust port that draws the air out, but it’s going to require a little modification for the chassis cover.
I’m going to briefly show what I did, but I didn’t think to take pictures during the process, but here’s a shot of what it looks like overall:
Backstory: I actually screwed up and got the wrong Noctua fan for the power supply fan. I thought it was a PWM for some reason, but it actually needed a FLX fan. However, I was able to still use the PWM fan. Here’s a breakdown of the above:
- Blue – This is the power supply fan, a Noctua NF-A4x20 FLX fan.
- Teal – This is a splice of the old cable adapter with the new using the Noctua omnijoin adapter set that comes with the fans (which is stupidly easy to use). Red and black wires matched up, and then I matched the yellow on the Noctua cable to the blue on the old fan plug.
- Green – This is the Noctua NF-A4x20 PWM fan that mistakenly bought. The fan runs at 100% all the time, but not an issue. While a mistake, the fan came with…
- Pink – Y-Adapter set that came with the PWM fan. I was able to plug this into the power plug port for the right FLX rear chassis fan (red), then plug that fan into the main y-adapter and then the PWM into the other port.
- Red – These are FLX fans.
- Purple – This was a port labeled “J9” that tried to use for power, but didn’t work (hence why I used the y-adapter).
Some notes I learned. For one, you’ll need to lift the power supply up in order to unscrew the screws:
I also had to drill a fairly large hole in the chassis cover so that I could fit the the cable in (see below). Had I not screwed up and spliced the PWM cable, the hole could have been smaller and I probably could have fit the y-adapter within the chassis.
I had to also modify the chassis for the PoE exhaust port by flattening the metal screen (so the attached fan didn’t rub against it) and I had to drill two additional holes so that the silicon screws could hold the fan down.
Finally, here’s a look from the CLI side:
Final Thoughts
- Overall, it’s working really well. It’s almost completely quiet — my EVE-NG server is actually louder than this thing.
- It’s not pretty, but the ugliness is hidden in the back.
- I DO NOT RECOMMEND YOU DO THIS TO PRODUCTION EQUIPMENT. I’m doing this to my home stuff, so I can live with it and the consequences, but I would never do this at work (too much work, TBH; better to buy an EX2300-C).
- Screwing the chassis cover back on will indeed be a little tighter, but with a little force you can get all the screws on.
- I don’t recommend using a drill to unscrew these (you can strip the screws pretty easily), but if you do, have a firm downward motion and screw/unscrew in bursts.
- I used to loathe the EX2200s for how slow they are, but on the 12.3R12.4 software, they seem to work well.
I HIGHLY recommend replacing the flash in that switch sooner than later. I have a few dozen EX2200 PoE switches in production and have had SEVERAL fail completely after power loss due to both partitions being corrupt on the flash. Even blowing them away and reflashing doesn’t always work.
Good call! Here’s the documentation if anyone needs it:
https://www.juniper.net/documentation/en_US/junos/topics/task/installation/poe-controller-software-upgrading.html
Thanks for writing this up, it helped me out.
Hi Jimmy,
Thank you for your write-up!
I just did a fan swap on an EX2200-48P-4G, which has 3 fans total, just like your 24-port model. I used three Sunon MagLev FLX, model KDE1204PKVX (Amazon Link: https://www.amazon.com/Sunon-40x40x20mm-pin-fan-KDE1204PKVX/dp/B006ODM76C). They have what I consider to be the best balance of CFM to noise (10.8 CFM, 27.5dB), so a bit higher flow than the Noctuas and still reasonably quiet.
My question/issue: for the CPU fan, which color wires (from fan to black molex connector) did you connect with scotch-locks? I did black to black, red to red and yellow (Sunon fan) to blue (molex connector). I’m now getting a “FPC 0 Fan 3 not spinning” error, even though the fan very much is spinning. Could my wire mapping be incorrect? Or perhaps it’s a mismatch of RPM capability? The Sunon maxes out at 8,200 RPM; the previous Delta maxed at 9,500.
I appreciate any insight you can think of. Thank you! – Trip
EDIT: Sorry, I meant *PSU* Fan, not CPU fan. Thanks again for any feedback you can lend!
> My question/issue: for the CPU fan, which color wires (from fan to black molex connector) did you connect with scotch-locks? I did black to black, red to red and yellow (Sunon fan) to blue (molex connector).
I used the Scotchlok’s that came with the Noctua set.
> I did black to black, red to red and yellow (Sunon fan) to blue (molex connector). I’m now getting a “FPC 0 Fan 3 not spinning” error, even though the fan very much is spinning. Could my wire mapping be incorrect? Or perhaps it’s a mismatch of RPM capability? The Sunon maxes out at 8,200 RPM; the previous Delta maxed at 9,500.
I had the same problem when I connected the initially used Noctua PWM fan (link in the post) which didn’t recognize the fan speed at all and gave me the same warning. It wasn’t until I purchased the FLX fan and put that in that I was getting the appropriate readings. Perhaps the fan you have is a PWM fan, or maybe something is wrong with it?
Thanks for such a quick reply, Jimmy.
I think I discovered the issue. The original PSU fan is a Delta FFB0412VHN-BF00 (data sheet here: https://www.delta-fan.com/Download/Spec/FFB0412VHN-BF00.pdf), which uses the yellow wire for locked rotor detection, *not* for speed sensing — understandable, given the implications if the single intake fan were to seize. Since our fans were of the speed-sensing type, that probably explains the error in phase detection/response detected by the switch.
One could get away with using a speed-sensing FLX 3pin fan by grounding the yellow wire, per Travis Illig’s post (https://www.paraesthesia.com/archive/2019/11/21/rotor-lock-fan-ground-yellow-wire/), but ultimately I may leave this in an office, so I’m going to play it safe and grab a fan with LR detection. I’ll post an update once I get it installed and tested!
Those Maglev, Audible from few meter ?
I’m using EX in my home, want it as quiet as possible.
For now I’ve disconnected all fans.
Thanks for the reply Jimmy.
So I dug a little deeper and discovered what I believe to be the issue: the original PSU fan, a Delta FFB0412VHN-BF00, uses the third wire (blue) for locked rotor detection, *not* speed-sensing, as is the case with the FLX fans we chose (Noctua and Sunon). That was the source of the “not spinning” errors.
A work-around for speed-sending FLX fans is leaving the third wire disconnected and ground each end (yellow from the fans, blue from the molex plug), but that would of course cripple actual failure discovery. Perhaps OK for home/lab use, but I plan on putting this back into a low-stress production role, so I ordered a proper replacement fan with LR detection, an NMB Technologies 1608VL-04W-B59-B00, which has the best balance of airflow, static pressure and noise that I could find (10.2 CFM, -31.4dB and about 5 Pa static pressure @10CFM), plenty of air movement but still WAY quieter, even at -31dB (like rustling leaves). I will install later this week and report back with my findings!
Thanks again for your suggestions, Trip
My one feedback to this is that I’m not seeing your issues with the Noctua FLX fan with all three cables connected.
Also, looking at the PSU fan specs:
https://www.delta-fan.com/Download/Spec/FFB0412VHN-F00.pdf
It looks like its sending a speed signal.
Either way, good luck!
Indeed, you may be correct. Either way, I hope to confirm by end of the week, and will report back.
Hi Jimmy,
I’m going through this same process right now and was wondering if you tested the noise levels without replacing the power supply fan? I’m not quite comfortable replacing power supply equipment, but my main goal is to lower the ambient noise my 2200 is making right now. Does replacing just the two exhaust fans make a noticeable different in noise levels?
Yes, that’s what I did initially was just replace the back fans, and the sound level was totally fine. However, I wanted to see what the difference would be with replacing the PS fan, and glad I did.
I’ve recently purchased a EX2200-48P-4G switch and I’m looking to replace the fans as well, what fan model did you wind up purchasing for the PSU fan? I’m looking at Sunon Maglevs since they have similar CFM spec and maybe operate 2dB higher, but they’re slightly cheaper, but I wanted to do a Noctua fan for the PSU since I can’t find any reference of a Sunon fan having lock detection besides ones that are in China and would take months to get here.
How did you raise the PSU section to access the PSU fan?
Got it. Extra screws 😉
I did the fan swap on my EX2200-4p.
3 x Noctua NF-A4x20 FLX fans installed to replace the factory ones.
After about 3 hours of operation these are my environment outputs. Looking fine from what I can see.
Class Item Status Measurement
Power FPC 0 Power Supply 0 OK
Temp FPC 0 CPU OK 30 degrees C / 86 degrees F
FPC 0 Exhaust Area OK 37 degrees C / 98 degrees F
FPC 0 EX-PFE1 OK 30 degrees C / 86 degrees F
FPC 0 Local Intake OK 20 degrees C / 68 degrees F
FPC 0 Remote Intake OK 22 degrees C / 71 degrees F
FPC 0 GEPHY1 OK 30 degrees C / 86 degrees F
FPC 0 GEPHY2 OK 30 degrees C / 86 degrees F
FPC 0 GEPHY3 OK 35 degrees C / 95 degrees F
FPC 0 GEPHY4 OK 25 degrees C / 77 degrees F
FPC 0 GEPHY5 OK 20 degrees C / 68 degrees F
FPC 0 GEPHY6 OK 20 degrees C / 68 degrees F
Fans FPC 0 Fan 1 OK Spinning at normal speed
FPC 0 Fan 2 OK Spinning at normal speed
FPC 0 PSU Fan OK Spinning at normal speed
Thanks for this post. It has rendered my way to noisy switch useful again. Hopefully It’ll see many months (years?) of hassle-free operation in my garage.
i have a EX2200-24p that I have lost poe on. im guessing it is a power supply issue…any idea where I can get a good schematic for the switch ?
Nice work – I have a pair of 4300 switches that sound like Death’s Benchgrinder.
Downside there is that the fans are in a plug-in module, and they use a blowing fan and a driven fan to measure airspeed. Long term plan is to design and 3D print something that takes a 40mm noctua fan, and fakes up the airspeed reply to the host.
A lot of this stuff is coming to End Of Support, which is nothing to do with End of Usefulness. So there will be more of them available over time at affordable prices.
Those Noctuas, will make the switch silent or still audible ?
I disconnected all fans now 🙂