WS-24-400A at a glance

[lligetfa]

Unboxing was uneventful. No sign of damage. No missing parts. No fancy QSG but the only thing I needed to know is what IP it will come up with. I don't need no stinking QSG to tell me how to mount the ears, where to plug the power cord, and how to enter the URL. Speaking of ears, I was pleased to see that they can be turned sideway for wall mounting, which is how I mounted it. Given how deep this switch is, I would have liked to see it ship with a right-angle power cord. I'd also like to have some way to put a tie-wrap on it to secure it. I was pleased to see a beefy ground post/screw.

I was expecting a large light show with all the LEDs sequencing to blink out some obscure diagnostic info. What I got was one LED on a SFP port blink for a very short while. Could that be it? All booted up and ready to go? I walked over to my laptop and entered the IP in my browser, it was all booted up. Much faster to boot than the ES-24 it replaced.

I like the UI. Large enough text for old eyes on a laptop under less than ideal lighting. No complicated menu trees to navigate/hunt through and have the mouse pointer constantly slip off of.

Upgrading the firmware was child's play, much simpler than the obscure way with the ES-24. I was so looking forward to the uber-quick reboot that was talked about but mine got hung up and I had to pull the plug. Oh well... there will be more updates.

My expectations on the quietness of the fans were high. I realize the WS is not intended to be in a room with humans so take what I say with a grain of salt. It is much quieter that the ES was on 1.0 firmware. It is a little quieter than the ES on 1.0.1 beta firmware. What is noticeable though is the fans seem to fight each other and syncopate a little with a slight wah-wah that draws my ear. If I take the cover off they even right out but of course that is not an option. A short distance away or in another room, it is hardly noticeable.

I used to work for a Fortune 500 company as a network administrator where I could have tested all the advanced features but here at home I have a very simple network so I have not tried many of these features. I expect that they will work as advertised. At some point I will grow my home test bench but I won't have all the Fluke analyzers that were available to me.

To sum this up, I see this as very well built, (over-built actually) and I would have no hesitation to recommend it.

[sirhc]

The 90 degree power cord has been requested by many of you, my answer is OK!

HOWEVER I have to use up the ones I have first but I can switch to a 90 degree AC cord after those are gone!

I really like your idea of a place to zip tie the power cord on the back of the switch to prevent the power cord from falling out, especially if wall mounting like you are and in some of my older retro fitted tower cabinets the 19" rack is also vertical hanging. I will get this worked into the next batch of chassis we build but sadly I have 1,500 to burn through first. You would think I would have caught that since I put a strap hole on the ToughSwitch wall mount I sell through RF Armor, I guess I just have to many things going on which is why you guys are here to keep me in line.

I will work on the fan whirring sound, it is caused by the frequency of the pulse modulation we use to control the fan speed. Also our fans are currently running at a conservatively high speed which is decreased a bit in the next firmware release do out later today.

I also have Eric working on the SMTP service we are using so you do not have to downgrade Gmail security as you discovered.

The intermittent NTP config error you found is also being looked at but is hard to reproduce on our end.

Thanks Less.

[lligetfa]

The 90 degree power cord has been requested by many of you, my answer is OK!

HOWEVER I have to use up the ones I have first but I can switch to a 90 degree AC cord after those are gone!

Use them up on the 12 port which is not as deep and supply right angle on the 24 port.

On my last job I used to save the right angle power cords when I would ewaste gear. Saved all the really short and really long power cords too. One never knows when a Goldielocks situation comes up where you need a particular length.

[lligetfa]

I will work on the fan whirring sound, it is caused by the frequency of the pulse modulation we use to control the fan speed. Also our fans are currently running at a conservatively high speed which is decreased a bit in the next firmware release do out later today.

Before I got my ES-24, I had planned to put it in my wiring closet off the mudroom where my TS8 was but when it landed and I heard how loud it was, there was no way that would fly. Plan B then was to put it down in the crawlspace below the wiring closet. It was not long after that, when the wife started to complain. She kept asking me if I heard a train. I knew it was the ES she was hearing but not volunteering any more info than I had to, I just said no, I don't hear a train. To make a long story short, the ES got turned off and the TS8 got put back in service.

I tested out the WS in the mudroom the day I got it but the wife noticed the sound and so I knew it would have to go down into the crawlspace. Now I wonder if I could have gotten away with it in the wiring closet but I've already given up the shelf I was going to put it on. Oh well... it's cooler down there so the fans tend to stay around 6500 RPM. They have jumped up to 8000 but I could not hear them from the living space. The top shelf in the wiring closet is warmer and sure to kick the fans up a notch. The only downside to the crawlspace is I need 10 - 12 foot jumpers to reach the patch panel.

I need to string a few more cable runs to put all my cameras in service. Right now the WS is powering a Pico, three 24V cams, and two 48V UVC Pro. It is nice to see the wattage stats that the ES did not show. I have one cam on an AGW that I need to hardwire and four more cams that are waiting on wiring.

[lligetfa]

I will work on the fan whirring sound, it is caused by the frequency of the pulse modulation we use to control the fan speed. Also our fans are currently running at a conservatively high speed which is decreased a bit in the next firmware release do out later today.

As mentioned, the WS replaced an ES that I had mounted on the same wall. When I mounted the WS, I had to put it at the top of the wall because I had a few jumpers coming down through the floor that were short. Also the WS is deeper than the ES so I ended up flipping it around so it faced the other direction. Flipping it put the air intake at the top so it was sucking in warmer air and fighting natural convection.

I had left the ES sitting on the floor after taking it down and when Ubiquiti released the latest firmware for it, I just plugged it in so I could flash it. Anyway... I noticed that the ES was quieter than how I remembered it and wondered if the coolness of the floor may be the reason. I took down the WS, flipped it around so the intake was on the bottom, and mounted it closer to the floor.

The difference was significant. I don't know if gravity pulling down on the fans were a factor, if working with rather than against natural convection was a factor, or the cooler air closer to the floor. Maybe all of the above. Anyway, the fans run much smoother and even when the RPMs go up I cannot hear it upstairs.

[sirhc]

Well we are currently bake testing the WS-24-400A model in an oven with I think a 330 watt load on it.

The first run through the oven was a breeze for the switch at 60C so Dave is upping the annie I think to either 65C or 75C right now.

Our Chipset is rated to run up to 125C but there are other components on the board that are not rated that high.

I think the next lowest component is the Ethernet transformers which are rated to like 75C but they really do not generate much internal heat and they are really far away from components that do generate heat and most components are rated at less than they tested them at (typical engineering safety range) to increase MTBF rating estimates. And obviously most MTBF on new components are just that ESTIMATES because the product has not been around that long. WHich is another reason I went with this product as it has been on the market for 5+ years so the MTBF is pretty accurate.

Where as Ubiquiti selected a power supply Gospower/Cospel that was designed for forced air cooling where as I selected a power supply that was rated at 400watts convection cooled at 50C but we have forced air cooling. We actually easily pulled well over 500 watts from our power supply but as I said it is rated at 400 "convection" cooled which means it is capable of more if it is in a cool environment or has forced air cooled but we have to advertise it to the CE declaration of the manufacturer or re-test and certify it for more....ehh 400 watts is plenty big enough!

Anyway I would not be surprised to see our switch be just fine in 70C environments or more but we hope to increase the suggested operating temperature soon when testing is completed.

That being said the fan speeds could be turned even further down but it is already very quiet compared to many other POE switches on the market and it is designed for towers not desks so cooler is better for a higher MTBF anyway.

We are selling switches but I had hoped to get more interest and support from the WISP community since I "did" listen and put the features in people wanted?

I need to move 3,000 switches by end of March or I may just decide not to continue developing more equipment lines and just sell what we already designed until no one wants them anymore and make a few dollars. Obviously we will continue to sell what is already designed so no worries of us disappearing but I will we continue to develop more lines??? Well that is up to how well this line does.

[lligetfa]

Don't get me wrong, I understand the intended use for the product and that it was not meant to be installed in the living space of an open concept home. If the fans had not had a little bit of wah-wah variation to them they would go unnoticed by most folk. Maybe I am the only one to notice it? I was just pleasantly surprised that flipping it around made such a difference. I only offered up that tidbit for others that may be wall mounting. I think having the intake on the bottom and working with convection is a best practice but in the case of the WS not a necessity for adequate cooling.

The specs on the WS are impressive and would instill confidence when installed in cabinets in any orientation and without artificial cooling.

I hope you sell lots of them and create more great products.

[sirhc]

LOL you got me wrong Les I was taking the opportunity to talk about the switches ability to deal with heat.

I was not taking offense to anything you said!

I figured you pull the soap box out so I would stand up on it and take full advantage to tell people what we are doing and how rugged they are!

[rockhead]

FWIW I was a little surprised at the fan noise on my 12 port after all I'd read about how quiet they were. I have not, however, experienced any of the edgeswitches so I can't compare there. It is certainly quiet compared to a couple of cisco widgets I've got, one router I cannot be in the same room with even after the fans rev down, lol.

I'm certainly glad to have an extended temp switch in the box that hit -24°C last night .

If anybody is dying for an RA power connector they should buy a few, I'm sure http://www.linhawstore.com/ would have 'em for cheap along with a zillion other tech stores.

[lligetfa]

With firmware version 1.0.7 the fans are so quiet that I had to go down in the crawlspace to hear them. Too bad I gave up the top shelf where I wanted to put it.